Research ArticleA Tangible Programming Tool for Children toCultivate Computational Thinking

Danli Wang1 Tingting Wang1 and Zhen Liu2

1 Beijing Key Lab of Human-Computer Interaction Institute of Software Chinese Academy of Sciences Beijing 100190 China2 Chongqing Medicine Exchange Co Ltd Chongqing 404100 China

Correspondence should be addressed to Danli Wang danliiscasaccn

Received 21 August 2013 Accepted 16 December 2013 Published 25 February 2014

Academic Editors C Esposito and O Greevy

Copyright copy 2014 Danli Wang et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Game and creation are activities which have good potential for computational thinking skills In this paper we present T-Mazean economical tangible programming tool for children aged 5ndash9 to build computer programs in maze games by placing woodenblocksThrough the use of computer vision technology T-Maze provides a live programming interface with real-time graphical andvoice feedback We conducted a user study with 7 children using T-Maze to play two levels of maze-escape games and create theirownmazesThe results show that T-Maze is not only easy to use but also has the potential to help children cultivate computationalthinking like abstraction problem decomposition and creativity

1 Introduction

Computational thinking (CT) is a term coined byWing [1] todescribe a set of thinking skills habits and approaches thatare integral to solving complex problems using a computerand widely applicable in the information society There isgrowing consensus [2] that computational thinking is afundamental skill that everyone needs to succeed in ourcomplex and technological culture CT makes it possiblefor children to improve the analytical ability that may behelpful in both STEM (science technology engineering andmathematics) subjects and many other professional areaseven in daily life [3] The advantage of this early exposureto CT is that it will help children to build a solid foundationof algorithmic and data structuresmdashthe basic nuts and boltsof the mechanics of computer programming [4] Computerprogramming is an excellent way to develop computationalthinking skills [5] Grover takes programming as a usefulway for kids to learn problem solving and computationalthinking [3] Research has indicated that learning how toprogram computers can have a positive andmeasurable effecton childrenrsquos achievement not only in areas such as mathand science but also in language skills creativity and socialemotional interaction [6 7] However the fact remains that

programming for children is still just plain hard and requiresstrong motivation on the childrsquos part in order to succeed [5]Thus the programming environment must strongly motivatethe subject by placing it in a context that is so compelling andmeaningful to the child that heshe does not give up

Tangible programming is tomake programming an activ-ity that is accessible to the hands andminds of young childrenby making it more direct and less abstract Tangible pro-gramming may have an appeal even to experienced abstractthinkers [8] By combining computer programming andtangible interaction tangible programming allows childrento manipulate ldquocodesrdquo directly which makes programmingmore appealing Besides using physical objects to interactwith computer is easier to involve children in the process[9] Research involving tangible interaction and children hasoften focused on how tangibles might support or improvelearning compared to more traditional methods [10 11]This research is mainly about the capability of tangibleprogramming to help children with some CT skills Thephysical manipulation acts like a scaffold between real worldand virtual world which may contribute to the abstraction inCTWe assume that tangible programming is also an efficientapproach to teach children about CT

Hindawi Publishing Corporatione Scientific World JournalVolume 2014 Article ID 428080 10 pageshttpdxdoiorg1011552014428080

2 The Scientific World Journal

In this paper we propose T-Maze a tangible program-ming environment designed to allow children to build com-puter programs by manipulating a set of wooden blockswhich are interconnected by magnets We conduct a studyinvolving 7 participants (aged 5ndash9) using T-Maze to playmaze-escape games and create their own mazes to explorehow CT concepts take shape for children in a tangibleprogramming environment The results may provide a lensthrough which one can consider the implications for learningand teaching computational thinking

2 Background and Related Work

21 Computational Thinking (CT) Computational thinkingis closely related to if not the same as the original notions ofprocedural thinking developed by Sey inMindstorms [12] Ina 2006 article Wing discussed computational thinking as ldquoaway of solving problems designing systems and understand-ing human behavior that draws on concepts fundamental tocomputer sciencerdquo [1] To successfully broaden the awarenessof computer science efforts must be made to lay the founda-tions of CT in an early stage of childrenrsquos development [13]The impact of CT on disciplines covers philosophy physicseducation and so forth [2] and CT concepts have been usedas a basic element for several efforts aimed at more precisedeeper and wider interpretation of computing [14]

Research regarding the implementation of computationalthinking skills in informal education provides valuableinsights This includes attention to K-12 curriculum generaleducation at colleges and universities and interdisciplinaryresearch [15ndash17]TheNational Academiesrsquo Computer Scienceand Telecommunications Board held a series of workshopson ldquoComputational Thinking for Everyonerdquo with a focus onidentifying the fundamental concepts of computer sciencethat can be taught to K-12 students The first workshopreport [18] provides multiple perspectives on computationalthinking The International Working Group on Computa-tional Thinking [19] points to several successful projects thatuse simulation and modeling robotics and computer gamedesign to teach abstraction automation and analysis As theynote these kinds of activities also involve an iterative designrefinement and reflection process that is central to creative aswell as computational thinking [20]

Research involving graphical programming and CT nur-turing has also gotten some progress Researchers used Aliceas a tool to support the development of algorithmic thinkingproblem solving and event processing [4 21 22] It is shownthat Alice does help children in the development of abstractthinking result analyzing and automation understanding[22] And Scratch was chosen as an appropriate platformfor teaching student CT through musical coding [23] Thevisual feedback that students get from Alice and Scratchallows them to relate the program to the action they seeon the screen and helps them refine their programs [4]The promising performance of graphical programming inchildrenrsquos computational ability development serves as agood spur for us to look into similar potential of tangibleprogramming

CT involves defining understanding and solving prob-lems reasoning at multiple levels of abstraction understand-ing and applying automation and analyzing the appropriate-ness of the abstractions made [22] However there is yet noconsensus on how to assess these skills [18]Therefore in thispaper particular attention is paid on the potential capabilityof a tangible programming tool to help children cultivatesome CT skills like abstraction automation problem decom-position and analysis We describe the results in regard toour research questions but we do not draw conclusions aboutlearning outcomes with regard to computational thinkinginstead we try to get to know about the computationalconcepts in childrenrsquos programming activity

22 Tangible Programming Since the 1960s a large numberof programming languages targeted at novice users havebeen created [24] Novice computer programming systemsaim to ease or eliminate the process of learning languagesyntax Beyond syntax there are many specific conceptualhurdles faced by novice programmers [25] A relatively recentapproach to ease the learning process has been the creation oftangible programming languages [9]

Electronic Blocks [26] are physical Lego blocks designedto allow young children (aged 3ndash8) to create Lego forms withinteresting behaviors It consists of three types of buildingblocks sensor block as input logic block to compute behav-ior block as output Each block is embedded with a processoror other essential electronic devices which dramaticallyincrease the cost of each piece of the token and the entiresystem Besides the syntax is simple with relatively limitedblocks available

Tangible Programming Bricks [8] are Lego blocks thatcan be stacked together to form programs Each block couldaccept a single card allowing users to communicate withother blocks via IR transmission By stacking blocks togetherwith accompanying cards users can teach toy cars to dancekitchen users to program microwaves and toy trains to reactto signals But the problem is that the way to manipulate theblocks is complicated for children and the programming isin a procedural style with a lack of advanced programmingconstructs like conditional and so forth

Tern [11 27] is a tangible programming language designedto allow children to use a collection of interlocking woodenblocks to build physical computer programs Each blockrepresents an action a flow-of-control construct a parameteror a sensor value The blocks are low cost but children needto manually use a camera to capture the image of blocksequence which is then transferred to computer to compileThus without immediate feedbacks about the outcomes ofthe physical programs real-time test and debug are notsupported in Tern

Toque [28] uses concrete real world cooking scenarios asprogramming metaphors to support an accessible program-ming learning experience User can control the cooking pro-cess and simulate some cookingmotions by bodymovementslike stir-fry action and so forthThere is a distinct proceduralworkflow in this programming experience but not sufficientfor programming learning in practice

The Scientific World Journal 3

TurTan [29] is a desktop tangible programming systemdesigned for Turtle Geometry which is based on LOGO Ituses a camera to capture and recognize the real-time positionof fingers and objects on the desktop However comparedwith blocks the interactive desktop is expensive and not easyto control for children

Recently there is a more generally defined tangibleprogramming work fromMIT called Twinkle Programmingwith Color [30] It uses a color sensor to read colors fromphysical objects drawings or collages And those colors aremapped to certain outputs like sounds graphics or roboticmovements It is remarkable that the objects can be anythingat hands but the system seems not programming-rich

From the above we find that many programming toolsare high cost because of the embedded electronic equipment(eg Electronic Blocks Tangible Programming bricks) Wiiand Interactive Desktop (eg Toque TurTan) That makesthemnot easily available for children in developing countriesTern is low cost and has rich programming concepts butchildren need to manually take photos of the blocks toactually run the program which is especially not appropriatefor young children To this end we are devoted to providea low-cost tangible programming environment appropriatefor children Also as to computational thinking there is littleresearch about the computational skills nurturing capabilityof these programming tools

In this paper we present T-Maze for children (aged 5to 9) to build programs by constructing physical blocks andintroduce it more in terms of how computational thinkingwas considered in the tool T-Maze uses wooden blocksto represent basic programming commands and computervision technology to convert physical programs into digitalcode automatically Sensors are integrated into the system toinvite interaction and attempt to let children experience theevent trigger mechanism

3 Implementation of T-Maze

31 T-Maze Overview T-Maze [31] is a tangible program-ming tool for children (ages 5ndash9) to build programs to playmultilevel maze-escape games and create their own mazesby manipulating a collection of wooden blocks In a maze-escape game children connect wooden blocks to control avirtual avatar in a grid world on screen to escape from amaze To create a new maze children use the same set ofblocks representing awholemazemap and the passable pathsFigure 1 shows the main part of T-Maze

32 Tangible Programming Blocks Tangible programmingblocks are inexpensive and durable cubes with no embeddedelectronics or power supplies We hypothesized that the useof familiar objects (wooden cubes) would transform an unfa-miliar and potentially intimidating activity like computerprogramming into an inviting and playful experience Thereare five kinds of programming blocks (Figure 2) in T-Mazestart and end block sensor block normal block (passablepath block used in maze creation) direction block and loopand numeric block For each block two little button magnets

Figure 1 T-Maze overview sensors wooden blocks camera andsoftware

are attached on each of the two opposite sides The magnetattraction and repulsion can somehow tell children whetherthe block of now is placed in a correct angle or used with thecorrect sideThe way that magnets interconnect the blocks ina straight line helps children arrange their program snippetsand also is beneficial to computer vision processing

For each type of blocks the symbols are different Butgenerally each symbol has three parts computer visionidentifier (the circular black-and-white symbols printed oneach block) simple descriptive texts and an icon indicatingthe function

33 Sensors T-Maze has three sensors temperature sensorlight senor and button sensor (Figure 3) The three sensorsare represented by three sensor blocks (Figure 2(c)) whichgenerate three corresponding squares in virtual mazemap onscreen In a program execution when the avatar reaches oneof these squares in the maze the child must do somethingwith the sensors (eg cover a light sensor) to allow the avatarto proceed Sensors are introduced here to invite interactionbetween children and the system meanwhile attempt tohelp children get familiar with event trigger mechanism inprogramming

34 Computer Vision System T-Maze uses TopCode [32]computer vision identifier (the circular black-and-whitesymbols printed on each block in Figure 2) to convert phys-ical programs into digital code automatically The identifiersallow the computer vision system to identify each block anddetermine its position in relation to the other blocks A digitalcamera with an image resolution set to 1280 by 960 pixels isattached to a laptop PC A programming area approximately40 cm wide by 30 cm high can be reliably compiled as longas the area is white or light-colored Captured images aretransferred to the host computer through a USB connectionand saved as JPEG images on the file systemWith this image

4 The Scientific World Journal

(a) Start and end block (b) Sensor block (c) Normal block

(d) Loop and numeric block (e) Direction block

Figure 2 Tangible programming blocks

Button sensor

Temperature sensor

Light sensor

Figure 3 Physical sensors in T-Maze

the compiler converts a programdirectly into virtualmachinecode

35 Game Modules T-Maze has two game modules MazeEscape and Maze Creation Game and creation are activitieswhich have good potential for computational thinking skills[33] Mazes are compelling and meaningful to most childrenfor their experience in school or day-to-day life The mazemetaphor is used becausewe think that placing programmingin this context could motivate the child so that heshe doesnot give up

Problem analysis takes place during the process of testingand debugging their programs [22] The computer visionsystem captured images automatically enabling T-Maze togive real-time graphical and voice feedbacks which helpchildren with the analysis For example in Maze Escape(Figure 4) when a block is placed correctly the background

Green arrow prompt

Figure 4 Interface of one level maze-escape game

of the occupied square will change into a green arrowOtherwise the smiley on the upper left screen will turnupset along with a voice prompting the possible solutions likeldquomaybe I need a sensorrdquo or ldquodid I walk too many steps in thisdirectionrdquo In Maze Creation (Figure 5) the real-time visualfeedback tells children ldquowhat you see is what you getrdquo andprovides them a live creating experience

4 CT in T-Maze Programming

CT is a comprehensive term that covers abstraction automa-tion analysis creativity and so forth [14] In this paper weselect to focus on several concepts that can be observed inT-Maze

41 Abstraction Abstraction ability is to find appropriatelevel of detail to define and solve a problem [34] With T-Maze children could simulate the abstract model of certainmazes in realitymap themselves into the virtual characters onscreen and control the charactersrsquo behaviors For example toplay a maze-escape game in T-Maze children need to builda path for their avatars using the physical blocks In order toknowhow these blocks function children need to think abouthow to perceive the world coordinates of the virtualmaze andhow to map the behaviors in real world like move straight

The Scientific World Journal 5

(a) Maze creating area (b) User interface of Maze Creation

Figure 5 The creating area (a) and user interface (b) of Maze Creation

or make a left turn into the virtual charactersrsquo behaviors inthe programs As to the sensors there are three layers ofabstraction-reality mapping that is sensor cells in virtualmaze physical sensor blocks and physical sensors

42 Automation Automation is a labor saving process inwhich a computer is instructed to execute a set of repetitivetasks It is much more efficiently compared to the processingpower of a human and the automated execution of processby machine is going to change everything [35] In this lightcomputer programs are ldquoautomations of abstractionsrdquo Theprogram children built in T-Maze automates the simulationof escaping from maze using a run loop which updates thefeedbacks on screen and interacts with users on the basis ofdesigned rules for sensors

43 Problem Decomposition and Analysis Decompositionmay take the form of stripping down a problem to whatis believed to be its bare essentials [22] Breaking problemsdown into smaller parts that may be more easily solved andanalyzing to reuse them can enable and simplify the resolv-ing of more complicated and larger-scale problems Andanalysis is a reflective practice which validates whether theabstractions are correct In programming theywould analyzewhether the avatarsrsquo behavior is expected and whether thereare some conditions that are not taken into account duringthe abstraction phase Children also engage in analysis whenthey judge whether their abstractions are efficient Thisanalysis may help them optimize and find a better solutionto the problem

In T-Maze the entry of the maze is marked with greentexts and the exit is marked with red texts Children areguided to get a blueprint of the path to be built from thesemarkers After knowing the ports where the path starts andends children have to connect them by ldquopavingrdquo the cellsin mazes with programming blocks During this processchildren are encouraged to analyze and compare their plansand choose the one that leads to the shortest path To thisend the feasible path in one maze is not unique And severalsections of the paths are designed to be straight in one

direction where the loop block may be a better choiceHowever it is not worthy to use the loop block if the straightsection of the path has less than four cells Though this is alittle ticklish for children it is where the analysis is neededAlso T-Maze gives real-time feedbacks which help childrenwith the analysis when testing and debugging their programs

44 Creativity Creativity is both a comprehensive capabilityof humanbeings and an important part of theCT skill set [14]We try to develop childrenrsquos creative skill by stimulating theircuriosity especially the creative imagination Creation is atthe root of creative thinking [20] The tangible programmingtool should enable children to process any abstractingsimulating and creating activities freely The freedom mayopen childrenrsquos minds a key point in promoting childrenrsquoscreativity

And as such the Maze Creation module is designed forthis purpose which offers children freedom to create anyshape of mazes as they like It is interesting to see whatchildren would come up with from the creating

5 User Study

We have conducted several studies to evaluate T-Maze interms of its usabilityThese studies involved 20 children (aged5ndash9) using T-Maze to program in a laboratory environmentIn the latest study we focused on the potential capabilityof T-Maze to nurture computational thinking in childrenrsquosprogramming activities

51 Research Design

511 ResearchQuestions For this study wewere interested inseveral research questions with an emphasis on the last one

(1) Is T-Maze easy to use for children(2) Do children like programmingwith T-Maze andwhat

are they interested in(3) How do the four CT concepts take shape for children

in programming

6 The Scientific World Journal

(a) (b)

Figure 6 The simple (a) and complex (b) maze-escape games in the study

Though the first two questions have been examined inprevious studies we wanted to be sure that children arecomfortable to involve in our study every time Consideringthe diversity of users and usage scenarios we regard theusability of T-Maze as a long-term test subject especiallywhen some new modifications or improvements need to beverified Also we expect usable feedbacks from children ifany to help us improve the tool since it is always preferableto have more children like T-Maze

512 Tasks We conducted two sessions in the study mazeescape and maze creation In maze escape session childrenwere asked to complete two different levels of maze-escapegames which were designed in consideration of individualability difference as well as the occasions where problemdecomposition and analysis probably happen To play amaze-escape game children had to find a passable path in themaze for the avatar to traverse through by building programswith wooden programming blocks The simple maze-escapegame only had one passable path and would need 9 blocks atmost (without the use of Loop blocks) while the complex onehadmultiple passable paths and would need variable numberof blocks according to the length of chosen path (Figure 6)For a maze creation task children were given no restrictionsto create anymaze they liked by constructing newmazemapswith the same set of blocks There was no time limitation inboth sessions Throughout the study we had a comfortableseat in front of the table for children to sit when they wereworking on the tasks as well as a rest area for them to wait orrest

513 Interviews and Questions For the interviews after thetwo main sessions we wanted to know childrenrsquos opinionsabout the tool their programming experience or anythingthat remained in their mind So the questions asked in theinterviews were mostly without model answers as followsldquoWhat other things do you think the blocks can dordquo ldquoDo youhave any other games that can be played with those blocksrdquoldquoHave you ever noticed any sensors or something like that inyour liferdquo The questions are listed in a questionnaire whichalso contains another two questions about the usability of T-Maze and one about the invitation for our next study recruit

Reasonably we did not expect children to becomecomputational thinkers after the one-off practices but wehoped to find some clues that might answer our questionsand preferably got some implications for teaching childrencomputational thinking

52 Participants and Apparatus Seven children (aged 8 inaverage) 5 girls and 2 boys participated in this studyThey allhad some experience in computer use but none of them hadever known about programming or used any programmingtools

The apparatus was a set of the latest version of T-Mazeplaced on an only table in a laboratory room The table witha chair was in the center of the room and we also made aresting area at the right corner beside the door One of thetwo cameras was set up in front left of the table while theother one was behind the table to the left Each time onechild went to the table to complete the tasks with hisher backtowards others waiting in the resting area so that they wouldnot interrupt the experiment

53 Procedure The study was conducted in four stagesdemonstration practice main sessions and interviewsBefore we started all children watched a demonstrationabout how to play a simple maze-escape game and create anew maze We introduced how to use the loop blocks andencouraged children to find the shortest path as they canThen they were given some minutes to practice freely Afterthey were ready to start the main sessions of the study began

Without time limitation children could involve in theirtasks as much time as they like but they were asked to give asign when they were about to begin or to finish and we notedthe time for records In maze-escape session children wereexplained to complete a simple escape task first and a complexone afterward Next in maze creation session children weretold to freely create any maze as they liked When theyfinished the creation we encouraged them to introduce theirmazes to others hoping to invite some creative talking Afterthemain sessions we conducted one-on-one semi-structuredinterviewswith children andwe taped all the conversation forlater analysis

The Scientific World Journal 7

Table 1 The time (unit minute) and blocks children used in the study

NumberMaze escape session Maze creation session

Simple ComplexTime (min) Blocks Time (min) Blocks Time (min) Blocks

1 350 900 700 1800 900 18002 300 900 500 1900 1000 26003 450 900 700 2100 1300 18004 450 900 900 2200 800 16005 300 900 800 2000 1100 31006 400 800 900 2100 1000 26007 400 900 800 1800 900 2500Avg 378 886 757 1986 986 2286

We collected qualitative data in the form of observationnotes photographs and videotape from which some quanti-tative data was also derivedWe also collected childrenrsquos work(the mazes they created) Three researchers were involved inthis study two collected data while the third acted in the roleof lead interpreter

54 Results and Discussion We comprehensively analyzedthe observation notes tapes and interview records andfound answers to the research questions from the analysisof childrenrsquos words and behaviors as well as the statisticalanalysis

For the first question (easy to use) we were interested inwhether children could create their own computer programsBased on observation notes and an analysis of videotapewe found that children were able to manipulate the tangibleblocks to form their own programs We allowed childrento take as much time as they wanted to complete a taskHowever we were lucky to see that they neither seemed totake their time nor to rush through the experiment whichgave us the reasonable and valid data in Table 1 Childrenwere able to accomplish both the simple and complex maze-escape tasks within similar period of time Also they wereeven more engaged in maze creation session and spent moretime creating and exploring All children finished the twosessions with their freely-created mazes some of which hadinteresting shapes or complex structure with sensor blocksused

In interviews when asked whether the tool was easy touse 2 children chose ldquovery easyrdquo 4 chose ldquoeasyrdquo and one choseldquonormalrdquo

For the second research question (interesting to playwith) according to the records and videotape childrenalways showed their interests in sensors 3 children askedquestions about the sensors like ldquowhat are these used forrdquoldquowhat if I pressed it longerrdquo ldquoonly three of themrdquo Moreoverchildren were very excited in maze creation section andshowed the richest behaviors For example one boy said ldquoIrsquomgoing to make the most difficult maze that none of themcan get throughrdquo and another boy asked ldquowhat if the blocksare not enough to create my mazerdquo As part of the mazecreation session we asked children not only to demonstratetheir mazes but also to show others the blocks they used to

program it In many cases children gathered together to thetable just stood by watching and praised for others In othercases however children asked for a competition to provetheir own

In interviews when asked whether the system was funall children confirmed their enjoyment with T-Maze andexpressed thewish to participate again in the future 4 of themvoted for the maze creation session to be their favorite

For the last question (CT concepts in programming)we found some cue that T-Maze has the potential to raisechildrenrsquos awareness about CT and help them understand CTto some extent

(i) AbstractionThemazemetaphor used inT-Maze promptedchildren to relate their experience about real-world maze tothe virtual maze on the screen They needed the process ofabstraction to narrow the escape problem down to some-thing that could be implemented on the computer using T-Maze as well as mapping the physical blocks to the virtualsquares in maze map on screen Restrictions imposed by theprogramming environment include an upper bound on thenumber of blocks and a limit on the size of the programmingarea (40 cm width by 30 cm height) Abstraction also tookplace as children designed avatars to react to a limited set ofconditions that may be encountered in the real-world mazesuch as some roadblocks represented by sensors

The fact that children all accomplished the tasks success-fully was inspiring and we indeed found something relatedfrom the observation In maze-escape session when theavatar came across a temperature sensor square in the mazewe found 3 children attempting to heat up the physicalsensors with their breath Later in the interviews we askedwhy did they do that and one said ldquoEr my mum alwayswarm my hand with her breath in winterrdquo Other two saidthat they always knew their breath was warmer than theirhands The same situation happened with light sensor where4 children glanced at the gap between their hands and thelight sensor they were covering and muttered somethingabout the light or the avatar on screen One girl said ldquoit isdark now go quicklyrdquo and another boy asked to himselfthat ldquois it dark enoughrdquo The interesting thing was that oneboy bended his body upon the sensor to block the lightbecause he thought it would be quicker to see the outcome

8 The Scientific World Journal

(a) A girl with her maze (b) The most complicated maze

Figure 7 Two mazes collected from maze creation session a girl with her maze (a) and the complicated maze created by a boy (b)

Another remarkable domain where abstraction takes place ismaze design and creation Because the ldquowhat you see is whatyou getrdquo creation process children could then test complexabstractions quickly and precisely For example to create anewmaze the child designed amap where each square couldbe represented by certain programming blocks Then hesheselected only those appropriate for the virtual world in hishermind

(ii) Automation Automation occurs as the system executeschildrenrsquos programs The ldquoprogramrdquo itself automated ldquostep-ping throughrdquo and updated virtual avatarrsquos location anddirection (representing the veer) at each step As to this termwe did not expect too much actually it was an attempt tofind whether children could realize the automatic processAnd we indeed found some cue through the study Forexample before the execution one boy said ldquolisten to myinstructions and gordquo Another girl jumped for joy when shesaw the animation of her avatar traversing through the mazeBesides we saw 5 children stood up from their seats whenthe execution began and kept standing by waiting for thesuccessful moment

In interviews 4 children expressed their enjoyment whilewatching the virtual avatar traversing through the mazeautomatically under their instructions This was also backedup in the videotape from which we observed that 3 of themclapped out of joy and cheered on their avatarswhilewatchingthe animation

(iii) Problem Decomposition and Analysis T-Maze providesa platform that enables live programming by allowing thecode to be adjusted in real time As the session proceededwe moved beyond programs consisting of simple sequencesof actions and introduced more advanced concepts such asloops and numeric parameter values Through these activi-ties we found evidence that children could engage these con-cepts reasoning about possible outcomes of different blocksto build and test every piece of solution until they finallysolved the maze-escape problem Children also performedanalysis when they decidedwhether or not the avatar behavedas expected If the avatar was stuck or ldquomisbehavedrdquo it mayeither mean that their implementation of their control idea isfaulty or that some blocks were misplaced

Mostly the process of analysis happened implicitly butwecould still find some track from the observation For exampleone child in his complex maze-escape mission determinedthat he needed five forward blocks to reach the first turningAfter learning the loop syntax he realized that the largestvalue of a numeric block is five and the loop is limited to 5times the most so he needed two more forward blocks Thechild then built a program snippet as

(Start Loop + 5 + Forward + End Loop) 997888rarr Forward997888rarr Forward

In maze creation session we found 4 children chose tolocate the entrance and exit point of a maze at first and thentried to link them through while others just placed the blocksas they wish and were happy-go-lucky with any shape ofmazes Besides we observed a sense of ldquohabitrdquo in childrenrsquosprogramming that 5 children would add several blocks intothe sequence once when it was a simple straight line whileattempted one step at a time in a complex situation (a cornera sensor square or a cross) when they would adjust currentplacement and decide what to do next while observing thefeedbacks

(iv) Creativity To create a maze programming became amedium for childrenrsquos personal and creative expression inthe design of their mazes children engaged their fantasiesand built relationships with other pockets of reality that wentbeyond common thoughts As we analyzed childrenrsquos workcollected in the study several interesting mazes surprised usOne girl created her maze shaping like a snake (Figure 7(a))but she described it as the Great Wall coming out from hertextbook And the boywho claimed tomake themost difficultmaze used 31 blocks constructing the most complex fish-shape maze (Figure 7(b)) During the creating he put asidethe end block until he finished all the paths then counted thenumber of cells in each path trying to find the longest onewhere the end block was appended

Generally the mazes children created are different fromwhat we designed for T-Maze intrinsically and the descrip-tions children gave in interviews were also unexpected Forexample one boy created a maze with only one straight linewhile he described it as an airport runway And he talked

The Scientific World Journal 9

about his experience with his father at the airport last weekright off the reel

55 Summary Though the study is relatively small scale wecan tell from the results that T-Maze has some computationalconcepts in its programming activities and children indeedhandle those experienceswellThis study suggests farther thatT-Maze is easy to use for children and they are able to buildtheir own programs in playing and creating activities with T-Maze The emphasis is that T-Maze has the potential to helpraise childrenrsquos awareness about CT concepts

In interviews we also wanted to get some inspirationsfrom children with questions like ldquowhat other things do youthink the blocks can dordquo or ldquodo you think of any other gamesthat can be played with those blocksrdquo Two children said theywould like to control a flying plane with the blocks and 3others wished to help their mothers with some householdduties like cooking cleaning or furnishing their roomsTheir unlimited imagination inspired us in the revision andupgrade of T-Maze in the future

6 Conclusion and Future Work

T-Maze is designed for children aged 5ndash9 to help cultivatetheir computational thinking skills It allows children to playmaze-escape games and create their own mazes by buildingcomputer programs out of wooden blocks We conducted auser study to verify its ease of use and the potential capabilityto help children with computational thinking The resultsshowed that T-Maze has the promising potential to helpchildren cultivate some computational thinking awarenesslike abstraction automation problem decomposition andanalysis

In the future we will develop more games and scenariosdesign more kinds of programming blocks and enrich thetypes of sensors More and further user studies are neededto improve T-Maze for better user experience and skillsnurturing capability Besides as an upgrade we may providea collaborative tangible programming tool for children toexperience the fun of cooperation

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank the children from nearbykindergartens that took part in the study The authors grate-fully acknowledge financial support from the Major StateBasic Research Development Program of China under Grantno 2013CB328805 the National Natural Science Foundationof China under Grant no 60970090 and no 61272325 theCooperation Project of Chinese Academy of Sciences andFoshan City under Grant no 2012YS04 and the NationalScience and Technology Supported Program under Grant no2012BAH19F01

References

[1] J M Wing ldquoComputational thinkingrdquo Communications of theACM vol 49 no 3 pp 33ndash35 2006

[2] A Bundy ldquoComputational thinking is pervasiverdquo Journal ofScientific and Practical Computing vol 1 no 2 pp 67ndash69 2007

[3] S Grover ldquoComputational Thinking Programming andthe Google App Inventorrdquo 2010 httpwwwthesmartbeancomlibrarycomputational-thinking-programming-and-the-google-app-inventor

[4] S Grover ldquoExpanding the Technology Curriculum to IncludeFoundational Elements of Computer Science for K-8rdquo 2009httpwwwthesmartbeancommagazinehome-schoolingexpanding-the-technology-curriculum-foundational-ele-ments-of-computer-science-for-k-8

[5] G Orr ldquoComputational thinking through programming andalgorithmic artrdquo in Proceedings of the SIGGRAPH 2009 Talks(SIGGRAPH rsquo09) vol 1 p 1 ACM 2009

[6] M U Bers Blocks to Robots Learning with Technology in theEarly Childhood Classroom Teachers College New York NYUSA 2008

[7] D H Clements ldquoThe future of educational computing researchthe case of computer programmingrdquo Information Technology inChildhood Education Annual vol 1999 no 1 pp 147ndash179 1999

[8] T S McNerney Tangible programming bricks an approach tomaking programming accessible to everyone [MS thesis] MITCambridge Cambridge Mass USA 2000

[9] T S McNerney ldquoFrom turtles to tangible programming bricksexplorations in physical language designrdquo Personal and Ubiqui-tous Computing vol 8 no 5 pp 326ndash337 2004

[10] J Fails A Druin M Guha G Chipman S Simms and WChuraman ldquoChildrsquos play a comparison of desktop and physicalinteractive environmentsrdquo in Proceedings of the InteractionDesign and Children (IDC rsquo05) pp 48ndash55 ACM 2005

[11] M S Horn E T Solovey R J Crouser and R J K JacobldquoComparing the use of tangible and graphical programminglanguages for informal science educationrdquo in Proceedings of theSIGCHI Conference on Human Factors in Computing Systems(CHI rsquo09) pp 975ndash984 ACM 2009

[12] S PapertMindstorms Children Computers and Powerful IdeasBasic Books New York NY USA 1981

[13] G H Fletcher and J J Lu ldquoEducation human computing skillsrethinking the K-12 experiencerdquo Communications of the ACMvol 52 no 2 pp 23ndash25 2009

[14] P Curzon J Peckham A Settle E Roberts and H TaylorldquoComputational thinking(CT) on weaving it inrdquo in Proceedingsof the 14th Annual ACM SIGCSE Conference on Innovation andTechnology in Computer Science Education (ITiCSE rsquo09) pp201ndash202 ACM 2009

[15] P B Henderson ldquoUbiquitous computational thinkingrdquo Com-puter vol 42 no 10 pp 100ndash102 2009

[16] J J Lu and G H L Fletcher ldquoThinking about computationalthinkingrdquo in Proceedings of the 40th ACM Technical Symposiumon Computer Science Education (SIGCSE rsquo09) pp 260ndash264ACM 2009

[17] D Moursund ldquoComputational Thinkingrdquo IAE-Pedia 2009httpiae-pediaorgComputational Thinking

[18] National Research Council ldquoCommittee for the workshops oncomputational thinking 2010rdquo Report of a Workshop on theScope and Nature of Computational Thinking The NationalAcademies Press Washington DC USA 2010

10 The Scientific World Journal

[19] IWG Computational Thinking for Youth Education Develop-ment Center Newton Mass USA 2010

[20] M Resnick ldquoAll I really need to know (about creative thinking)I learned (by studying how children learn) in kindergartenrdquo inProceedings of the 6th Conference on Creativity amp Cognition (CCrsquo07) pp 1ndash6 ACM 2007

[21] J Dougherty ldquoConcept visualization in CS0 using ALICErdquoJournal of Computing Science in Colleges vol 22 no 3 pp 145ndash152 2007

[22] I Lee F Martin J Denner et al ldquoComputational thinking foryouth in practicerdquo ACM Inroads vol 2 no 1 pp 32ndash37 2011

[23] A Ruthmann J M Heines G R Greher P Laidler and CSaulters II ldquoTeaching computational thinking through musicallive coding in Scratchrdquo in Proceedings of the 41st ACM TechnicalSymposium on Computer Science Education (SIGCSE rsquo10) pp351ndash355 ACM 2010

[24] C Kelleher and R Pausch ldquoLowering the barriers to pro-gramming a taxonomy of programming environments andlanguages for novice programmersrdquo ACM Computing Surveysvol 37 no 2 pp 83ndash137 2005

[25] C Rader C Brand and C Lewis ldquoDegrees of comprehensionchildrenrsquos understanding of a visual programming environ-mentrdquo in Proceedings of the Conference on Human Factors inComputing Systems (CHI rsquo97) pp 351ndash358 ACM 1997

[26] PWyeth and H C Purchase ldquoTangible programming elementsfor young childrenrdquo in Proceedings of the Extended Abstracts onHuman Factors in Computing Systems (CHI rsquo02) pp 774ndash775ACM 2002

[27] M S Horn and R J K Jacob ldquoTangible programming in theclassroomwith ternrdquo inProceedings of the ExtendedAbstracts onHuman Factors in Computing Systems (CHI Trends Interactivity)(CHI rsquo07) pp 1965ndash1970 ACM 2007

[28] S Tarkan V Sazawal A Druin et al ldquoToque designing acooking-based programming language for and with childrenrdquoin Proceedings of the 28th Annual Conference on Human Factorsin Computing Systems (CHI rsquo10) pp 2417ndash2426 ACM 2010

[29] D Gallardo C F Julia and S Jorda ldquoTurTan a tangibleprogramming language for creative explorationrdquo in Proceedingsof the 3rd IEEE InternationalWorkshop onHorizontal InteractiveHuman Computer System (TABLETOP rsquo08) pp 89ndash92 IEEEComputer Society Press 2008

[30] J Silver andE Rosenbaum ldquoTwinkle programmingwith colorrdquoin Proceedings of the 4th International Conference on TangibleEmbedded and Embodied Interaction (TEI rsquo10) pp 383ndash384ACM 2010

[31] D L Wang C Zhang and H A Wang ldquoT-Maze a tangibleprogramming tool for childrenrdquo in Proceedings of the 10thInternational Conference on Interaction Design and Children(IDC rsquo11) pp 127ndash135 ACM 2011

[32] M S Horn ldquoTopCode Tangible Object Placement Codesrdquo2012 httpuserseecsnorthwesternedusimmhorntopcodes

[33] K Howland ldquoSupporting the development of multimodalwriting and computational thinking skills through computergame creationrdquo in Proceedings of the IEEE Symposium on VisualLanguages and Human-Centric Computing (VLHCC rsquo09) pp250ndash251 IEEE Computer Society Press 2009

[34] J LrsquoHeureux D Boisvert K Sanghera and R Cohen ldquoITproblem solving an implementation of computational thinkingin information technologyrdquo in Proceedings of the 13th AnnualConference on Information Technology Education (SIGITE rsquo12)pp 183ndash188 ACM 2012

[35] M Guzdial ldquoEducation paving the way for computationalthinkingrdquo Communications of the ACM vol 51 no 8 pp 25ndash27 2008

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

International Journal of

ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia

International Journal of

Biomedical Imaging

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArtificialNeural Systems

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational Intelligence and Neuroscience

Industrial EngineeringJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

2 The Scientific World Journal

In this paper we propose T-Maze a tangible program-ming environment designed to allow children to build com-puter programs by manipulating a set of wooden blockswhich are interconnected by magnets We conduct a studyinvolving 7 participants (aged 5ndash9) using T-Maze to playmaze-escape games and create their own mazes to explorehow CT concepts take shape for children in a tangibleprogramming environment The results may provide a lensthrough which one can consider the implications for learningand teaching computational thinking

2 Background and Related Work

21 Computational Thinking (CT) Computational thinkingis closely related to if not the same as the original notions ofprocedural thinking developed by Sey inMindstorms [12] Ina 2006 article Wing discussed computational thinking as ldquoaway of solving problems designing systems and understand-ing human behavior that draws on concepts fundamental tocomputer sciencerdquo [1] To successfully broaden the awarenessof computer science efforts must be made to lay the founda-tions of CT in an early stage of childrenrsquos development [13]The impact of CT on disciplines covers philosophy physicseducation and so forth [2] and CT concepts have been usedas a basic element for several efforts aimed at more precisedeeper and wider interpretation of computing [14]

Research regarding the implementation of computationalthinking skills in informal education provides valuableinsights This includes attention to K-12 curriculum generaleducation at colleges and universities and interdisciplinaryresearch [15ndash17]TheNational Academiesrsquo Computer Scienceand Telecommunications Board held a series of workshopson ldquoComputational Thinking for Everyonerdquo with a focus onidentifying the fundamental concepts of computer sciencethat can be taught to K-12 students The first workshopreport [18] provides multiple perspectives on computationalthinking The International Working Group on Computa-tional Thinking [19] points to several successful projects thatuse simulation and modeling robotics and computer gamedesign to teach abstraction automation and analysis As theynote these kinds of activities also involve an iterative designrefinement and reflection process that is central to creative aswell as computational thinking [20]

Research involving graphical programming and CT nur-turing has also gotten some progress Researchers used Aliceas a tool to support the development of algorithmic thinkingproblem solving and event processing [4 21 22] It is shownthat Alice does help children in the development of abstractthinking result analyzing and automation understanding[22] And Scratch was chosen as an appropriate platformfor teaching student CT through musical coding [23] Thevisual feedback that students get from Alice and Scratchallows them to relate the program to the action they seeon the screen and helps them refine their programs [4]The promising performance of graphical programming inchildrenrsquos computational ability development serves as agood spur for us to look into similar potential of tangibleprogramming

CT involves defining understanding and solving prob-lems reasoning at multiple levels of abstraction understand-ing and applying automation and analyzing the appropriate-ness of the abstractions made [22] However there is yet noconsensus on how to assess these skills [18]Therefore in thispaper particular attention is paid on the potential capabilityof a tangible programming tool to help children cultivatesome CT skills like abstraction automation problem decom-position and analysis We describe the results in regard toour research questions but we do not draw conclusions aboutlearning outcomes with regard to computational thinkinginstead we try to get to know about the computationalconcepts in childrenrsquos programming activity

22 Tangible Programming Since the 1960s a large numberof programming languages targeted at novice users havebeen created [24] Novice computer programming systemsaim to ease or eliminate the process of learning languagesyntax Beyond syntax there are many specific conceptualhurdles faced by novice programmers [25] A relatively recentapproach to ease the learning process has been the creation oftangible programming languages [9]

Electronic Blocks [26] are physical Lego blocks designedto allow young children (aged 3ndash8) to create Lego forms withinteresting behaviors It consists of three types of buildingblocks sensor block as input logic block to compute behav-ior block as output Each block is embedded with a processoror other essential electronic devices which dramaticallyincrease the cost of each piece of the token and the entiresystem Besides the syntax is simple with relatively limitedblocks available

Tangible Programming Bricks [8] are Lego blocks thatcan be stacked together to form programs Each block couldaccept a single card allowing users to communicate withother blocks via IR transmission By stacking blocks togetherwith accompanying cards users can teach toy cars to dancekitchen users to program microwaves and toy trains to reactto signals But the problem is that the way to manipulate theblocks is complicated for children and the programming isin a procedural style with a lack of advanced programmingconstructs like conditional and so forth

Tern [11 27] is a tangible programming language designedto allow children to use a collection of interlocking woodenblocks to build physical computer programs Each blockrepresents an action a flow-of-control construct a parameteror a sensor value The blocks are low cost but children needto manually use a camera to capture the image of blocksequence which is then transferred to computer to compileThus without immediate feedbacks about the outcomes ofthe physical programs real-time test and debug are notsupported in Tern

Toque [28] uses concrete real world cooking scenarios asprogramming metaphors to support an accessible program-ming learning experience User can control the cooking pro-cess and simulate some cookingmotions by bodymovementslike stir-fry action and so forthThere is a distinct proceduralworkflow in this programming experience but not sufficientfor programming learning in practice

The Scientific World Journal 3

TurTan [29] is a desktop tangible programming systemdesigned for Turtle Geometry which is based on LOGO Ituses a camera to capture and recognize the real-time positionof fingers and objects on the desktop However comparedwith blocks the interactive desktop is expensive and not easyto control for children

Recently there is a more generally defined tangibleprogramming work fromMIT called Twinkle Programmingwith Color [30] It uses a color sensor to read colors fromphysical objects drawings or collages And those colors aremapped to certain outputs like sounds graphics or roboticmovements It is remarkable that the objects can be anythingat hands but the system seems not programming-rich

From the above we find that many programming toolsare high cost because of the embedded electronic equipment(eg Electronic Blocks Tangible Programming bricks) Wiiand Interactive Desktop (eg Toque TurTan) That makesthemnot easily available for children in developing countriesTern is low cost and has rich programming concepts butchildren need to manually take photos of the blocks toactually run the program which is especially not appropriatefor young children To this end we are devoted to providea low-cost tangible programming environment appropriatefor children Also as to computational thinking there is littleresearch about the computational skills nurturing capabilityof these programming tools

In this paper we present T-Maze for children (aged 5to 9) to build programs by constructing physical blocks andintroduce it more in terms of how computational thinkingwas considered in the tool T-Maze uses wooden blocksto represent basic programming commands and computervision technology to convert physical programs into digitalcode automatically Sensors are integrated into the system toinvite interaction and attempt to let children experience theevent trigger mechanism

3 Implementation of T-Maze

31 T-Maze Overview T-Maze [31] is a tangible program-ming tool for children (ages 5ndash9) to build programs to playmultilevel maze-escape games and create their own mazesby manipulating a collection of wooden blocks In a maze-escape game children connect wooden blocks to control avirtual avatar in a grid world on screen to escape from amaze To create a new maze children use the same set ofblocks representing awholemazemap and the passable pathsFigure 1 shows the main part of T-Maze

32 Tangible Programming Blocks Tangible programmingblocks are inexpensive and durable cubes with no embeddedelectronics or power supplies We hypothesized that the useof familiar objects (wooden cubes) would transform an unfa-miliar and potentially intimidating activity like computerprogramming into an inviting and playful experience Thereare five kinds of programming blocks (Figure 2) in T-Mazestart and end block sensor block normal block (passablepath block used in maze creation) direction block and loopand numeric block For each block two little button magnets

Figure 1 T-Maze overview sensors wooden blocks camera andsoftware

are attached on each of the two opposite sides The magnetattraction and repulsion can somehow tell children whetherthe block of now is placed in a correct angle or used with thecorrect sideThe way that magnets interconnect the blocks ina straight line helps children arrange their program snippetsand also is beneficial to computer vision processing

For each type of blocks the symbols are different Butgenerally each symbol has three parts computer visionidentifier (the circular black-and-white symbols printed oneach block) simple descriptive texts and an icon indicatingthe function

33 Sensors T-Maze has three sensors temperature sensorlight senor and button sensor (Figure 3) The three sensorsare represented by three sensor blocks (Figure 2(c)) whichgenerate three corresponding squares in virtual mazemap onscreen In a program execution when the avatar reaches oneof these squares in the maze the child must do somethingwith the sensors (eg cover a light sensor) to allow the avatarto proceed Sensors are introduced here to invite interactionbetween children and the system meanwhile attempt tohelp children get familiar with event trigger mechanism inprogramming

34 Computer Vision System T-Maze uses TopCode [32]computer vision identifier (the circular black-and-whitesymbols printed on each block in Figure 2) to convert phys-ical programs into digital code automatically The identifiersallow the computer vision system to identify each block anddetermine its position in relation to the other blocks A digitalcamera with an image resolution set to 1280 by 960 pixels isattached to a laptop PC A programming area approximately40 cm wide by 30 cm high can be reliably compiled as longas the area is white or light-colored Captured images aretransferred to the host computer through a USB connectionand saved as JPEG images on the file systemWith this image

4 The Scientific World Journal

(a) Start and end block (b) Sensor block (c) Normal block

(d) Loop and numeric block (e) Direction block

Figure 2 Tangible programming blocks

Button sensor

Temperature sensor

Light sensor

Figure 3 Physical sensors in T-Maze

the compiler converts a programdirectly into virtualmachinecode

35 Game Modules T-Maze has two game modules MazeEscape and Maze Creation Game and creation are activitieswhich have good potential for computational thinking skills[33] Mazes are compelling and meaningful to most childrenfor their experience in school or day-to-day life The mazemetaphor is used becausewe think that placing programmingin this context could motivate the child so that heshe doesnot give up

Problem analysis takes place during the process of testingand debugging their programs [22] The computer visionsystem captured images automatically enabling T-Maze togive real-time graphical and voice feedbacks which helpchildren with the analysis For example in Maze Escape(Figure 4) when a block is placed correctly the background

Green arrow prompt

Figure 4 Interface of one level maze-escape game

of the occupied square will change into a green arrowOtherwise the smiley on the upper left screen will turnupset along with a voice prompting the possible solutions likeldquomaybe I need a sensorrdquo or ldquodid I walk too many steps in thisdirectionrdquo In Maze Creation (Figure 5) the real-time visualfeedback tells children ldquowhat you see is what you getrdquo andprovides them a live creating experience

4 CT in T-Maze Programming

CT is a comprehensive term that covers abstraction automa-tion analysis creativity and so forth [14] In this paper weselect to focus on several concepts that can be observed inT-Maze

41 Abstraction Abstraction ability is to find appropriatelevel of detail to define and solve a problem [34] With T-Maze children could simulate the abstract model of certainmazes in realitymap themselves into the virtual characters onscreen and control the charactersrsquo behaviors For example toplay a maze-escape game in T-Maze children need to builda path for their avatars using the physical blocks In order toknowhow these blocks function children need to think abouthow to perceive the world coordinates of the virtualmaze andhow to map the behaviors in real world like move straight

The Scientific World Journal 5

(a) Maze creating area (b) User interface of Maze Creation

Figure 5 The creating area (a) and user interface (b) of Maze Creation

or make a left turn into the virtual charactersrsquo behaviors inthe programs As to the sensors there are three layers ofabstraction-reality mapping that is sensor cells in virtualmaze physical sensor blocks and physical sensors

42 Automation Automation is a labor saving process inwhich a computer is instructed to execute a set of repetitivetasks It is much more efficiently compared to the processingpower of a human and the automated execution of processby machine is going to change everything [35] In this lightcomputer programs are ldquoautomations of abstractionsrdquo Theprogram children built in T-Maze automates the simulationof escaping from maze using a run loop which updates thefeedbacks on screen and interacts with users on the basis ofdesigned rules for sensors

43 Problem Decomposition and Analysis Decompositionmay take the form of stripping down a problem to whatis believed to be its bare essentials [22] Breaking problemsdown into smaller parts that may be more easily solved andanalyzing to reuse them can enable and simplify the resolv-ing of more complicated and larger-scale problems Andanalysis is a reflective practice which validates whether theabstractions are correct In programming theywould analyzewhether the avatarsrsquo behavior is expected and whether thereare some conditions that are not taken into account duringthe abstraction phase Children also engage in analysis whenthey judge whether their abstractions are efficient Thisanalysis may help them optimize and find a better solutionto the problem

In T-Maze the entry of the maze is marked with greentexts and the exit is marked with red texts Children areguided to get a blueprint of the path to be built from thesemarkers After knowing the ports where the path starts andends children have to connect them by ldquopavingrdquo the cellsin mazes with programming blocks During this processchildren are encouraged to analyze and compare their plansand choose the one that leads to the shortest path To thisend the feasible path in one maze is not unique And severalsections of the paths are designed to be straight in one

direction where the loop block may be a better choiceHowever it is not worthy to use the loop block if the straightsection of the path has less than four cells Though this is alittle ticklish for children it is where the analysis is neededAlso T-Maze gives real-time feedbacks which help childrenwith the analysis when testing and debugging their programs

44 Creativity Creativity is both a comprehensive capabilityof humanbeings and an important part of theCT skill set [14]We try to develop childrenrsquos creative skill by stimulating theircuriosity especially the creative imagination Creation is atthe root of creative thinking [20] The tangible programmingtool should enable children to process any abstractingsimulating and creating activities freely The freedom mayopen childrenrsquos minds a key point in promoting childrenrsquoscreativity

And as such the Maze Creation module is designed forthis purpose which offers children freedom to create anyshape of mazes as they like It is interesting to see whatchildren would come up with from the creating

5 User Study

We have conducted several studies to evaluate T-Maze interms of its usabilityThese studies involved 20 children (aged5ndash9) using T-Maze to program in a laboratory environmentIn the latest study we focused on the potential capabilityof T-Maze to nurture computational thinking in childrenrsquosprogramming activities

51 Research Design

511 ResearchQuestions For this study wewere interested inseveral research questions with an emphasis on the last one

(1) Is T-Maze easy to use for children(2) Do children like programmingwith T-Maze andwhat

are they interested in(3) How do the four CT concepts take shape for children

in programming

6 The Scientific World Journal

(a) (b)

Figure 6 The simple (a) and complex (b) maze-escape games in the study

Though the first two questions have been examined inprevious studies we wanted to be sure that children arecomfortable to involve in our study every time Consideringthe diversity of users and usage scenarios we regard theusability of T-Maze as a long-term test subject especiallywhen some new modifications or improvements need to beverified Also we expect usable feedbacks from children ifany to help us improve the tool since it is always preferableto have more children like T-Maze

512 Tasks We conducted two sessions in the study mazeescape and maze creation In maze escape session childrenwere asked to complete two different levels of maze-escapegames which were designed in consideration of individualability difference as well as the occasions where problemdecomposition and analysis probably happen To play amaze-escape game children had to find a passable path in themaze for the avatar to traverse through by building programswith wooden programming blocks The simple maze-escapegame only had one passable path and would need 9 blocks atmost (without the use of Loop blocks) while the complex onehadmultiple passable paths and would need variable numberof blocks according to the length of chosen path (Figure 6)For a maze creation task children were given no restrictionsto create anymaze they liked by constructing newmazemapswith the same set of blocks There was no time limitation inboth sessions Throughout the study we had a comfortableseat in front of the table for children to sit when they wereworking on the tasks as well as a rest area for them to wait orrest

513 Interviews and Questions For the interviews after thetwo main sessions we wanted to know childrenrsquos opinionsabout the tool their programming experience or anythingthat remained in their mind So the questions asked in theinterviews were mostly without model answers as followsldquoWhat other things do you think the blocks can dordquo ldquoDo youhave any other games that can be played with those blocksrdquoldquoHave you ever noticed any sensors or something like that inyour liferdquo The questions are listed in a questionnaire whichalso contains another two questions about the usability of T-Maze and one about the invitation for our next study recruit

Reasonably we did not expect children to becomecomputational thinkers after the one-off practices but wehoped to find some clues that might answer our questionsand preferably got some implications for teaching childrencomputational thinking

52 Participants and Apparatus Seven children (aged 8 inaverage) 5 girls and 2 boys participated in this studyThey allhad some experience in computer use but none of them hadever known about programming or used any programmingtools

The apparatus was a set of the latest version of T-Mazeplaced on an only table in a laboratory room The table witha chair was in the center of the room and we also made aresting area at the right corner beside the door One of thetwo cameras was set up in front left of the table while theother one was behind the table to the left Each time onechild went to the table to complete the tasks with hisher backtowards others waiting in the resting area so that they wouldnot interrupt the experiment

53 Procedure The study was conducted in four stagesdemonstration practice main sessions and interviewsBefore we started all children watched a demonstrationabout how to play a simple maze-escape game and create anew maze We introduced how to use the loop blocks andencouraged children to find the shortest path as they canThen they were given some minutes to practice freely Afterthey were ready to start the main sessions of the study began

Without time limitation children could involve in theirtasks as much time as they like but they were asked to give asign when they were about to begin or to finish and we notedthe time for records In maze-escape session children wereexplained to complete a simple escape task first and a complexone afterward Next in maze creation session children weretold to freely create any maze as they liked When theyfinished the creation we encouraged them to introduce theirmazes to others hoping to invite some creative talking Afterthemain sessions we conducted one-on-one semi-structuredinterviewswith children andwe taped all the conversation forlater analysis

The Scientific World Journal 7

Table 1 The time (unit minute) and blocks children used in the study

NumberMaze escape session Maze creation session

Simple ComplexTime (min) Blocks Time (min) Blocks Time (min) Blocks

1 350 900 700 1800 900 18002 300 900 500 1900 1000 26003 450 900 700 2100 1300 18004 450 900 900 2200 800 16005 300 900 800 2000 1100 31006 400 800 900 2100 1000 26007 400 900 800 1800 900 2500Avg 378 886 757 1986 986 2286

We collected qualitative data in the form of observationnotes photographs and videotape from which some quanti-tative data was also derivedWe also collected childrenrsquos work(the mazes they created) Three researchers were involved inthis study two collected data while the third acted in the roleof lead interpreter

54 Results and Discussion We comprehensively analyzedthe observation notes tapes and interview records andfound answers to the research questions from the analysisof childrenrsquos words and behaviors as well as the statisticalanalysis

For the first question (easy to use) we were interested inwhether children could create their own computer programsBased on observation notes and an analysis of videotapewe found that children were able to manipulate the tangibleblocks to form their own programs We allowed childrento take as much time as they wanted to complete a taskHowever we were lucky to see that they neither seemed totake their time nor to rush through the experiment whichgave us the reasonable and valid data in Table 1 Childrenwere able to accomplish both the simple and complex maze-escape tasks within similar period of time Also they wereeven more engaged in maze creation session and spent moretime creating and exploring All children finished the twosessions with their freely-created mazes some of which hadinteresting shapes or complex structure with sensor blocksused

In interviews when asked whether the tool was easy touse 2 children chose ldquovery easyrdquo 4 chose ldquoeasyrdquo and one choseldquonormalrdquo

For the second research question (interesting to playwith) according to the records and videotape childrenalways showed their interests in sensors 3 children askedquestions about the sensors like ldquowhat are these used forrdquoldquowhat if I pressed it longerrdquo ldquoonly three of themrdquo Moreoverchildren were very excited in maze creation section andshowed the richest behaviors For example one boy said ldquoIrsquomgoing to make the most difficult maze that none of themcan get throughrdquo and another boy asked ldquowhat if the blocksare not enough to create my mazerdquo As part of the mazecreation session we asked children not only to demonstratetheir mazes but also to show others the blocks they used to

program it In many cases children gathered together to thetable just stood by watching and praised for others In othercases however children asked for a competition to provetheir own

In interviews when asked whether the system was funall children confirmed their enjoyment with T-Maze andexpressed thewish to participate again in the future 4 of themvoted for the maze creation session to be their favorite

For the last question (CT concepts in programming)we found some cue that T-Maze has the potential to raisechildrenrsquos awareness about CT and help them understand CTto some extent

(i) AbstractionThemazemetaphor used inT-Maze promptedchildren to relate their experience about real-world maze tothe virtual maze on the screen They needed the process ofabstraction to narrow the escape problem down to some-thing that could be implemented on the computer using T-Maze as well as mapping the physical blocks to the virtualsquares in maze map on screen Restrictions imposed by theprogramming environment include an upper bound on thenumber of blocks and a limit on the size of the programmingarea (40 cm width by 30 cm height) Abstraction also tookplace as children designed avatars to react to a limited set ofconditions that may be encountered in the real-world mazesuch as some roadblocks represented by sensors

The fact that children all accomplished the tasks success-fully was inspiring and we indeed found something relatedfrom the observation In maze-escape session when theavatar came across a temperature sensor square in the mazewe found 3 children attempting to heat up the physicalsensors with their breath Later in the interviews we askedwhy did they do that and one said ldquoEr my mum alwayswarm my hand with her breath in winterrdquo Other two saidthat they always knew their breath was warmer than theirhands The same situation happened with light sensor where4 children glanced at the gap between their hands and thelight sensor they were covering and muttered somethingabout the light or the avatar on screen One girl said ldquoit isdark now go quicklyrdquo and another boy asked to himselfthat ldquois it dark enoughrdquo The interesting thing was that oneboy bended his body upon the sensor to block the lightbecause he thought it would be quicker to see the outcome

8 The Scientific World Journal

(a) A girl with her maze (b) The most complicated maze

Figure 7 Two mazes collected from maze creation session a girl with her maze (a) and the complicated maze created by a boy (b)

Another remarkable domain where abstraction takes place ismaze design and creation Because the ldquowhat you see is whatyou getrdquo creation process children could then test complexabstractions quickly and precisely For example to create anewmaze the child designed amap where each square couldbe represented by certain programming blocks Then hesheselected only those appropriate for the virtual world in hishermind

(ii) Automation Automation occurs as the system executeschildrenrsquos programs The ldquoprogramrdquo itself automated ldquostep-ping throughrdquo and updated virtual avatarrsquos location anddirection (representing the veer) at each step As to this termwe did not expect too much actually it was an attempt tofind whether children could realize the automatic processAnd we indeed found some cue through the study Forexample before the execution one boy said ldquolisten to myinstructions and gordquo Another girl jumped for joy when shesaw the animation of her avatar traversing through the mazeBesides we saw 5 children stood up from their seats whenthe execution began and kept standing by waiting for thesuccessful moment

In interviews 4 children expressed their enjoyment whilewatching the virtual avatar traversing through the mazeautomatically under their instructions This was also backedup in the videotape from which we observed that 3 of themclapped out of joy and cheered on their avatarswhilewatchingthe animation

(iii) Problem Decomposition and Analysis T-Maze providesa platform that enables live programming by allowing thecode to be adjusted in real time As the session proceededwe moved beyond programs consisting of simple sequencesof actions and introduced more advanced concepts such asloops and numeric parameter values Through these activi-ties we found evidence that children could engage these con-cepts reasoning about possible outcomes of different blocksto build and test every piece of solution until they finallysolved the maze-escape problem Children also performedanalysis when they decidedwhether or not the avatar behavedas expected If the avatar was stuck or ldquomisbehavedrdquo it mayeither mean that their implementation of their control idea isfaulty or that some blocks were misplaced

Mostly the process of analysis happened implicitly butwecould still find some track from the observation For exampleone child in his complex maze-escape mission determinedthat he needed five forward blocks to reach the first turningAfter learning the loop syntax he realized that the largestvalue of a numeric block is five and the loop is limited to 5times the most so he needed two more forward blocks Thechild then built a program snippet as

(Start Loop + 5 + Forward + End Loop) 997888rarr Forward997888rarr Forward

In maze creation session we found 4 children chose tolocate the entrance and exit point of a maze at first and thentried to link them through while others just placed the blocksas they wish and were happy-go-lucky with any shape ofmazes Besides we observed a sense of ldquohabitrdquo in childrenrsquosprogramming that 5 children would add several blocks intothe sequence once when it was a simple straight line whileattempted one step at a time in a complex situation (a cornera sensor square or a cross) when they would adjust currentplacement and decide what to do next while observing thefeedbacks

(iv) Creativity To create a maze programming became amedium for childrenrsquos personal and creative expression inthe design of their mazes children engaged their fantasiesand built relationships with other pockets of reality that wentbeyond common thoughts As we analyzed childrenrsquos workcollected in the study several interesting mazes surprised usOne girl created her maze shaping like a snake (Figure 7(a))but she described it as the Great Wall coming out from hertextbook And the boywho claimed tomake themost difficultmaze used 31 blocks constructing the most complex fish-shape maze (Figure 7(b)) During the creating he put asidethe end block until he finished all the paths then counted thenumber of cells in each path trying to find the longest onewhere the end block was appended

Generally the mazes children created are different fromwhat we designed for T-Maze intrinsically and the descrip-tions children gave in interviews were also unexpected Forexample one boy created a maze with only one straight linewhile he described it as an airport runway And he talked

The Scientific World Journal 9

about his experience with his father at the airport last weekright off the reel

55 Summary Though the study is relatively small scale wecan tell from the results that T-Maze has some computationalconcepts in its programming activities and children indeedhandle those experienceswellThis study suggests farther thatT-Maze is easy to use for children and they are able to buildtheir own programs in playing and creating activities with T-Maze The emphasis is that T-Maze has the potential to helpraise childrenrsquos awareness about CT concepts

In interviews we also wanted to get some inspirationsfrom children with questions like ldquowhat other things do youthink the blocks can dordquo or ldquodo you think of any other gamesthat can be played with those blocksrdquo Two children said theywould like to control a flying plane with the blocks and 3others wished to help their mothers with some householdduties like cooking cleaning or furnishing their roomsTheir unlimited imagination inspired us in the revision andupgrade of T-Maze in the future

6 Conclusion and Future Work

T-Maze is designed for children aged 5ndash9 to help cultivatetheir computational thinking skills It allows children to playmaze-escape games and create their own mazes by buildingcomputer programs out of wooden blocks We conducted auser study to verify its ease of use and the potential capabilityto help children with computational thinking The resultsshowed that T-Maze has the promising potential to helpchildren cultivate some computational thinking awarenesslike abstraction automation problem decomposition andanalysis

In the future we will develop more games and scenariosdesign more kinds of programming blocks and enrich thetypes of sensors More and further user studies are neededto improve T-Maze for better user experience and skillsnurturing capability Besides as an upgrade we may providea collaborative tangible programming tool for children toexperience the fun of cooperation

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank the children from nearbykindergartens that took part in the study The authors grate-fully acknowledge financial support from the Major StateBasic Research Development Program of China under Grantno 2013CB328805 the National Natural Science Foundationof China under Grant no 60970090 and no 61272325 theCooperation Project of Chinese Academy of Sciences andFoshan City under Grant no 2012YS04 and the NationalScience and Technology Supported Program under Grant no2012BAH19F01

References

[1] J M Wing ldquoComputational thinkingrdquo Communications of theACM vol 49 no 3 pp 33ndash35 2006

[2] A Bundy ldquoComputational thinking is pervasiverdquo Journal ofScientific and Practical Computing vol 1 no 2 pp 67ndash69 2007

[3] S Grover ldquoComputational Thinking Programming andthe Google App Inventorrdquo 2010 httpwwwthesmartbeancomlibrarycomputational-thinking-programming-and-the-google-app-inventor

[4] S Grover ldquoExpanding the Technology Curriculum to IncludeFoundational Elements of Computer Science for K-8rdquo 2009httpwwwthesmartbeancommagazinehome-schoolingexpanding-the-technology-curriculum-foundational-ele-ments-of-computer-science-for-k-8

[5] G Orr ldquoComputational thinking through programming andalgorithmic artrdquo in Proceedings of the SIGGRAPH 2009 Talks(SIGGRAPH rsquo09) vol 1 p 1 ACM 2009

[6] M U Bers Blocks to Robots Learning with Technology in theEarly Childhood Classroom Teachers College New York NYUSA 2008

[7] D H Clements ldquoThe future of educational computing researchthe case of computer programmingrdquo Information Technology inChildhood Education Annual vol 1999 no 1 pp 147ndash179 1999

[8] T S McNerney Tangible programming bricks an approach tomaking programming accessible to everyone [MS thesis] MITCambridge Cambridge Mass USA 2000

[9] T S McNerney ldquoFrom turtles to tangible programming bricksexplorations in physical language designrdquo Personal and Ubiqui-tous Computing vol 8 no 5 pp 326ndash337 2004

[10] J Fails A Druin M Guha G Chipman S Simms and WChuraman ldquoChildrsquos play a comparison of desktop and physicalinteractive environmentsrdquo in Proceedings of the InteractionDesign and Children (IDC rsquo05) pp 48ndash55 ACM 2005

[11] M S Horn E T Solovey R J Crouser and R J K JacobldquoComparing the use of tangible and graphical programminglanguages for informal science educationrdquo in Proceedings of theSIGCHI Conference on Human Factors in Computing Systems(CHI rsquo09) pp 975ndash984 ACM 2009

[12] S PapertMindstorms Children Computers and Powerful IdeasBasic Books New York NY USA 1981

[13] G H Fletcher and J J Lu ldquoEducation human computing skillsrethinking the K-12 experiencerdquo Communications of the ACMvol 52 no 2 pp 23ndash25 2009

[14] P Curzon J Peckham A Settle E Roberts and H TaylorldquoComputational thinking(CT) on weaving it inrdquo in Proceedingsof the 14th Annual ACM SIGCSE Conference on Innovation andTechnology in Computer Science Education (ITiCSE rsquo09) pp201ndash202 ACM 2009

[15] P B Henderson ldquoUbiquitous computational thinkingrdquo Com-puter vol 42 no 10 pp 100ndash102 2009

[16] J J Lu and G H L Fletcher ldquoThinking about computationalthinkingrdquo in Proceedings of the 40th ACM Technical Symposiumon Computer Science Education (SIGCSE rsquo09) pp 260ndash264ACM 2009

[17] D Moursund ldquoComputational Thinkingrdquo IAE-Pedia 2009httpiae-pediaorgComputational Thinking

[18] National Research Council ldquoCommittee for the workshops oncomputational thinking 2010rdquo Report of a Workshop on theScope and Nature of Computational Thinking The NationalAcademies Press Washington DC USA 2010

10 The Scientific World Journal

[19] IWG Computational Thinking for Youth Education Develop-ment Center Newton Mass USA 2010

[20] M Resnick ldquoAll I really need to know (about creative thinking)I learned (by studying how children learn) in kindergartenrdquo inProceedings of the 6th Conference on Creativity amp Cognition (CCrsquo07) pp 1ndash6 ACM 2007

[21] J Dougherty ldquoConcept visualization in CS0 using ALICErdquoJournal of Computing Science in Colleges vol 22 no 3 pp 145ndash152 2007

[22] I Lee F Martin J Denner et al ldquoComputational thinking foryouth in practicerdquo ACM Inroads vol 2 no 1 pp 32ndash37 2011

[23] A Ruthmann J M Heines G R Greher P Laidler and CSaulters II ldquoTeaching computational thinking through musicallive coding in Scratchrdquo in Proceedings of the 41st ACM TechnicalSymposium on Computer Science Education (SIGCSE rsquo10) pp351ndash355 ACM 2010

[24] C Kelleher and R Pausch ldquoLowering the barriers to pro-gramming a taxonomy of programming environments andlanguages for novice programmersrdquo ACM Computing Surveysvol 37 no 2 pp 83ndash137 2005

[25] C Rader C Brand and C Lewis ldquoDegrees of comprehensionchildrenrsquos understanding of a visual programming environ-mentrdquo in Proceedings of the Conference on Human Factors inComputing Systems (CHI rsquo97) pp 351ndash358 ACM 1997

[26] PWyeth and H C Purchase ldquoTangible programming elementsfor young childrenrdquo in Proceedings of the Extended Abstracts onHuman Factors in Computing Systems (CHI rsquo02) pp 774ndash775ACM 2002

[27] M S Horn and R J K Jacob ldquoTangible programming in theclassroomwith ternrdquo inProceedings of the ExtendedAbstracts onHuman Factors in Computing Systems (CHI Trends Interactivity)(CHI rsquo07) pp 1965ndash1970 ACM 2007

[28] S Tarkan V Sazawal A Druin et al ldquoToque designing acooking-based programming language for and with childrenrdquoin Proceedings of the 28th Annual Conference on Human Factorsin Computing Systems (CHI rsquo10) pp 2417ndash2426 ACM 2010

[29] D Gallardo C F Julia and S Jorda ldquoTurTan a tangibleprogramming language for creative explorationrdquo in Proceedingsof the 3rd IEEE InternationalWorkshop onHorizontal InteractiveHuman Computer System (TABLETOP rsquo08) pp 89ndash92 IEEEComputer Society Press 2008

[30] J Silver andE Rosenbaum ldquoTwinkle programmingwith colorrdquoin Proceedings of the 4th International Conference on TangibleEmbedded and Embodied Interaction (TEI rsquo10) pp 383ndash384ACM 2010

[31] D L Wang C Zhang and H A Wang ldquoT-Maze a tangibleprogramming tool for childrenrdquo in Proceedings of the 10thInternational Conference on Interaction Design and Children(IDC rsquo11) pp 127ndash135 ACM 2011

[32] M S Horn ldquoTopCode Tangible Object Placement Codesrdquo2012 httpuserseecsnorthwesternedusimmhorntopcodes

[33] K Howland ldquoSupporting the development of multimodalwriting and computational thinking skills through computergame creationrdquo in Proceedings of the IEEE Symposium on VisualLanguages and Human-Centric Computing (VLHCC rsquo09) pp250ndash251 IEEE Computer Society Press 2009

[34] J LrsquoHeureux D Boisvert K Sanghera and R Cohen ldquoITproblem solving an implementation of computational thinkingin information technologyrdquo in Proceedings of the 13th AnnualConference on Information Technology Education (SIGITE rsquo12)pp 183ndash188 ACM 2012

[35] M Guzdial ldquoEducation paving the way for computationalthinkingrdquo Communications of the ACM vol 51 no 8 pp 25ndash27 2008

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

International Journal of

ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia

International Journal of

Biomedical Imaging

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArtificialNeural Systems

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational Intelligence and Neuroscience

Industrial EngineeringJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World Journal 3

TurTan [29] is a desktop tangible programming systemdesigned for Turtle Geometry which is based on LOGO Ituses a camera to capture and recognize the real-time positionof fingers and objects on the desktop However comparedwith blocks the interactive desktop is expensive and not easyto control for children

Recently there is a more generally defined tangibleprogramming work fromMIT called Twinkle Programmingwith Color [30] It uses a color sensor to read colors fromphysical objects drawings or collages And those colors aremapped to certain outputs like sounds graphics or roboticmovements It is remarkable that the objects can be anythingat hands but the system seems not programming-rich

From the above we find that many programming toolsare high cost because of the embedded electronic equipment(eg Electronic Blocks Tangible Programming bricks) Wiiand Interactive Desktop (eg Toque TurTan) That makesthemnot easily available for children in developing countriesTern is low cost and has rich programming concepts butchildren need to manually take photos of the blocks toactually run the program which is especially not appropriatefor young children To this end we are devoted to providea low-cost tangible programming environment appropriatefor children Also as to computational thinking there is littleresearch about the computational skills nurturing capabilityof these programming tools

In this paper we present T-Maze for children (aged 5to 9) to build programs by constructing physical blocks andintroduce it more in terms of how computational thinkingwas considered in the tool T-Maze uses wooden blocksto represent basic programming commands and computervision technology to convert physical programs into digitalcode automatically Sensors are integrated into the system toinvite interaction and attempt to let children experience theevent trigger mechanism

3 Implementation of T-Maze

31 T-Maze Overview T-Maze [31] is a tangible program-ming tool for children (ages 5ndash9) to build programs to playmultilevel maze-escape games and create their own mazesby manipulating a collection of wooden blocks In a maze-escape game children connect wooden blocks to control avirtual avatar in a grid world on screen to escape from amaze To create a new maze children use the same set ofblocks representing awholemazemap and the passable pathsFigure 1 shows the main part of T-Maze

32 Tangible Programming Blocks Tangible programmingblocks are inexpensive and durable cubes with no embeddedelectronics or power supplies We hypothesized that the useof familiar objects (wooden cubes) would transform an unfa-miliar and potentially intimidating activity like computerprogramming into an inviting and playful experience Thereare five kinds of programming blocks (Figure 2) in T-Mazestart and end block sensor block normal block (passablepath block used in maze creation) direction block and loopand numeric block For each block two little button magnets

Figure 1 T-Maze overview sensors wooden blocks camera andsoftware

are attached on each of the two opposite sides The magnetattraction and repulsion can somehow tell children whetherthe block of now is placed in a correct angle or used with thecorrect sideThe way that magnets interconnect the blocks ina straight line helps children arrange their program snippetsand also is beneficial to computer vision processing

For each type of blocks the symbols are different Butgenerally each symbol has three parts computer visionidentifier (the circular black-and-white symbols printed oneach block) simple descriptive texts and an icon indicatingthe function

33 Sensors T-Maze has three sensors temperature sensorlight senor and button sensor (Figure 3) The three sensorsare represented by three sensor blocks (Figure 2(c)) whichgenerate three corresponding squares in virtual mazemap onscreen In a program execution when the avatar reaches oneof these squares in the maze the child must do somethingwith the sensors (eg cover a light sensor) to allow the avatarto proceed Sensors are introduced here to invite interactionbetween children and the system meanwhile attempt tohelp children get familiar with event trigger mechanism inprogramming

34 Computer Vision System T-Maze uses TopCode [32]computer vision identifier (the circular black-and-whitesymbols printed on each block in Figure 2) to convert phys-ical programs into digital code automatically The identifiersallow the computer vision system to identify each block anddetermine its position in relation to the other blocks A digitalcamera with an image resolution set to 1280 by 960 pixels isattached to a laptop PC A programming area approximately40 cm wide by 30 cm high can be reliably compiled as longas the area is white or light-colored Captured images aretransferred to the host computer through a USB connectionand saved as JPEG images on the file systemWith this image

4 The Scientific World Journal

(a) Start and end block (b) Sensor block (c) Normal block

(d) Loop and numeric block (e) Direction block

Figure 2 Tangible programming blocks

Button sensor

Temperature sensor

Light sensor

Figure 3 Physical sensors in T-Maze

the compiler converts a programdirectly into virtualmachinecode

35 Game Modules T-Maze has two game modules MazeEscape and Maze Creation Game and creation are activitieswhich have good potential for computational thinking skills[33] Mazes are compelling and meaningful to most childrenfor their experience in school or day-to-day life The mazemetaphor is used becausewe think that placing programmingin this context could motivate the child so that heshe doesnot give up

Problem analysis takes place during the process of testingand debugging their programs [22] The computer visionsystem captured images automatically enabling T-Maze togive real-time graphical and voice feedbacks which helpchildren with the analysis For example in Maze Escape(Figure 4) when a block is placed correctly the background

Green arrow prompt

Figure 4 Interface of one level maze-escape game

of the occupied square will change into a green arrowOtherwise the smiley on the upper left screen will turnupset along with a voice prompting the possible solutions likeldquomaybe I need a sensorrdquo or ldquodid I walk too many steps in thisdirectionrdquo In Maze Creation (Figure 5) the real-time visualfeedback tells children ldquowhat you see is what you getrdquo andprovides them a live creating experience

4 CT in T-Maze Programming

CT is a comprehensive term that covers abstraction automa-tion analysis creativity and so forth [14] In this paper weselect to focus on several concepts that can be observed inT-Maze

41 Abstraction Abstraction ability is to find appropriatelevel of detail to define and solve a problem [34] With T-Maze children could simulate the abstract model of certainmazes in realitymap themselves into the virtual characters onscreen and control the charactersrsquo behaviors For example toplay a maze-escape game in T-Maze children need to builda path for their avatars using the physical blocks In order toknowhow these blocks function children need to think abouthow to perceive the world coordinates of the virtualmaze andhow to map the behaviors in real world like move straight

The Scientific World Journal 5

(a) Maze creating area (b) User interface of Maze Creation

Figure 5 The creating area (a) and user interface (b) of Maze Creation

or make a left turn into the virtual charactersrsquo behaviors inthe programs As to the sensors there are three layers ofabstraction-reality mapping that is sensor cells in virtualmaze physical sensor blocks and physical sensors

42 Automation Automation is a labor saving process inwhich a computer is instructed to execute a set of repetitivetasks It is much more efficiently compared to the processingpower of a human and the automated execution of processby machine is going to change everything [35] In this lightcomputer programs are ldquoautomations of abstractionsrdquo Theprogram children built in T-Maze automates the simulationof escaping from maze using a run loop which updates thefeedbacks on screen and interacts with users on the basis ofdesigned rules for sensors

43 Problem Decomposition and Analysis Decompositionmay take the form of stripping down a problem to whatis believed to be its bare essentials [22] Breaking problemsdown into smaller parts that may be more easily solved andanalyzing to reuse them can enable and simplify the resolv-ing of more complicated and larger-scale problems Andanalysis is a reflective practice which validates whether theabstractions are correct In programming theywould analyzewhether the avatarsrsquo behavior is expected and whether thereare some conditions that are not taken into account duringthe abstraction phase Children also engage in analysis whenthey judge whether their abstractions are efficient Thisanalysis may help them optimize and find a better solutionto the problem

In T-Maze the entry of the maze is marked with greentexts and the exit is marked with red texts Children areguided to get a blueprint of the path to be built from thesemarkers After knowing the ports where the path starts andends children have to connect them by ldquopavingrdquo the cellsin mazes with programming blocks During this processchildren are encouraged to analyze and compare their plansand choose the one that leads to the shortest path To thisend the feasible path in one maze is not unique And severalsections of the paths are designed to be straight in one

direction where the loop block may be a better choiceHowever it is not worthy to use the loop block if the straightsection of the path has less than four cells Though this is alittle ticklish for children it is where the analysis is neededAlso T-Maze gives real-time feedbacks which help childrenwith the analysis when testing and debugging their programs

44 Creativity Creativity is both a comprehensive capabilityof humanbeings and an important part of theCT skill set [14]We try to develop childrenrsquos creative skill by stimulating theircuriosity especially the creative imagination Creation is atthe root of creative thinking [20] The tangible programmingtool should enable children to process any abstractingsimulating and creating activities freely The freedom mayopen childrenrsquos minds a key point in promoting childrenrsquoscreativity

And as such the Maze Creation module is designed forthis purpose which offers children freedom to create anyshape of mazes as they like It is interesting to see whatchildren would come up with from the creating

5 User Study

We have conducted several studies to evaluate T-Maze interms of its usabilityThese studies involved 20 children (aged5ndash9) using T-Maze to program in a laboratory environmentIn the latest study we focused on the potential capabilityof T-Maze to nurture computational thinking in childrenrsquosprogramming activities

51 Research Design

511 ResearchQuestions For this study wewere interested inseveral research questions with an emphasis on the last one

(1) Is T-Maze easy to use for children(2) Do children like programmingwith T-Maze andwhat

are they interested in(3) How do the four CT concepts take shape for children

in programming

6 The Scientific World Journal

(a) (b)

Figure 6 The simple (a) and complex (b) maze-escape games in the study

Though the first two questions have been examined inprevious studies we wanted to be sure that children arecomfortable to involve in our study every time Consideringthe diversity of users and usage scenarios we regard theusability of T-Maze as a long-term test subject especiallywhen some new modifications or improvements need to beverified Also we expect usable feedbacks from children ifany to help us improve the tool since it is always preferableto have more children like T-Maze

512 Tasks We conducted two sessions in the study mazeescape and maze creation In maze escape session childrenwere asked to complete two different levels of maze-escapegames which were designed in consideration of individualability difference as well as the occasions where problemdecomposition and analysis probably happen To play amaze-escape game children had to find a passable path in themaze for the avatar to traverse through by building programswith wooden programming blocks The simple maze-escapegame only had one passable path and would need 9 blocks atmost (without the use of Loop blocks) while the complex onehadmultiple passable paths and would need variable numberof blocks according to the length of chosen path (Figure 6)For a maze creation task children were given no restrictionsto create anymaze they liked by constructing newmazemapswith the same set of blocks There was no time limitation inboth sessions Throughout the study we had a comfortableseat in front of the table for children to sit when they wereworking on the tasks as well as a rest area for them to wait orrest

513 Interviews and Questions For the interviews after thetwo main sessions we wanted to know childrenrsquos opinionsabout the tool their programming experience or anythingthat remained in their mind So the questions asked in theinterviews were mostly without model answers as followsldquoWhat other things do you think the blocks can dordquo ldquoDo youhave any other games that can be played with those blocksrdquoldquoHave you ever noticed any sensors or something like that inyour liferdquo The questions are listed in a questionnaire whichalso contains another two questions about the usability of T-Maze and one about the invitation for our next study recruit

Reasonably we did not expect children to becomecomputational thinkers after the one-off practices but wehoped to find some clues that might answer our questionsand preferably got some implications for teaching childrencomputational thinking

52 Participants and Apparatus Seven children (aged 8 inaverage) 5 girls and 2 boys participated in this studyThey allhad some experience in computer use but none of them hadever known about programming or used any programmingtools

The apparatus was a set of the latest version of T-Mazeplaced on an only table in a laboratory room The table witha chair was in the center of the room and we also made aresting area at the right corner beside the door One of thetwo cameras was set up in front left of the table while theother one was behind the table to the left Each time onechild went to the table to complete the tasks with hisher backtowards others waiting in the resting area so that they wouldnot interrupt the experiment

53 Procedure The study was conducted in four stagesdemonstration practice main sessions and interviewsBefore we started all children watched a demonstrationabout how to play a simple maze-escape game and create anew maze We introduced how to use the loop blocks andencouraged children to find the shortest path as they canThen they were given some minutes to practice freely Afterthey were ready to start the main sessions of the study began

Without time limitation children could involve in theirtasks as much time as they like but they were asked to give asign when they were about to begin or to finish and we notedthe time for records In maze-escape session children wereexplained to complete a simple escape task first and a complexone afterward Next in maze creation session children weretold to freely create any maze as they liked When theyfinished the creation we encouraged them to introduce theirmazes to others hoping to invite some creative talking Afterthemain sessions we conducted one-on-one semi-structuredinterviewswith children andwe taped all the conversation forlater analysis

The Scientific World Journal 7

Table 1 The time (unit minute) and blocks children used in the study

NumberMaze escape session Maze creation session

Simple ComplexTime (min) Blocks Time (min) Blocks Time (min) Blocks

1 350 900 700 1800 900 18002 300 900 500 1900 1000 26003 450 900 700 2100 1300 18004 450 900 900 2200 800 16005 300 900 800 2000 1100 31006 400 800 900 2100 1000 26007 400 900 800 1800 900 2500Avg 378 886 757 1986 986 2286

We collected qualitative data in the form of observationnotes photographs and videotape from which some quanti-tative data was also derivedWe also collected childrenrsquos work(the mazes they created) Three researchers were involved inthis study two collected data while the third acted in the roleof lead interpreter

54 Results and Discussion We comprehensively analyzedthe observation notes tapes and interview records andfound answers to the research questions from the analysisof childrenrsquos words and behaviors as well as the statisticalanalysis

For the first question (easy to use) we were interested inwhether children could create their own computer programsBased on observation notes and an analysis of videotapewe found that children were able to manipulate the tangibleblocks to form their own programs We allowed childrento take as much time as they wanted to complete a taskHowever we were lucky to see that they neither seemed totake their time nor to rush through the experiment whichgave us the reasonable and valid data in Table 1 Childrenwere able to accomplish both the simple and complex maze-escape tasks within similar period of time Also they wereeven more engaged in maze creation session and spent moretime creating and exploring All children finished the twosessions with their freely-created mazes some of which hadinteresting shapes or complex structure with sensor blocksused

In interviews when asked whether the tool was easy touse 2 children chose ldquovery easyrdquo 4 chose ldquoeasyrdquo and one choseldquonormalrdquo

For the second research question (interesting to playwith) according to the records and videotape childrenalways showed their interests in sensors 3 children askedquestions about the sensors like ldquowhat are these used forrdquoldquowhat if I pressed it longerrdquo ldquoonly three of themrdquo Moreoverchildren were very excited in maze creation section andshowed the richest behaviors For example one boy said ldquoIrsquomgoing to make the most difficult maze that none of themcan get throughrdquo and another boy asked ldquowhat if the blocksare not enough to create my mazerdquo As part of the mazecreation session we asked children not only to demonstratetheir mazes but also to show others the blocks they used to

program it In many cases children gathered together to thetable just stood by watching and praised for others In othercases however children asked for a competition to provetheir own

In interviews when asked whether the system was funall children confirmed their enjoyment with T-Maze andexpressed thewish to participate again in the future 4 of themvoted for the maze creation session to be their favorite

For the last question (CT concepts in programming)we found some cue that T-Maze has the potential to raisechildrenrsquos awareness about CT and help them understand CTto some extent

(i) AbstractionThemazemetaphor used inT-Maze promptedchildren to relate their experience about real-world maze tothe virtual maze on the screen They needed the process ofabstraction to narrow the escape problem down to some-thing that could be implemented on the computer using T-Maze as well as mapping the physical blocks to the virtualsquares in maze map on screen Restrictions imposed by theprogramming environment include an upper bound on thenumber of blocks and a limit on the size of the programmingarea (40 cm width by 30 cm height) Abstraction also tookplace as children designed avatars to react to a limited set ofconditions that may be encountered in the real-world mazesuch as some roadblocks represented by sensors

The fact that children all accomplished the tasks success-fully was inspiring and we indeed found something relatedfrom the observation In maze-escape session when theavatar came across a temperature sensor square in the mazewe found 3 children attempting to heat up the physicalsensors with their breath Later in the interviews we askedwhy did they do that and one said ldquoEr my mum alwayswarm my hand with her breath in winterrdquo Other two saidthat they always knew their breath was warmer than theirhands The same situation happened with light sensor where4 children glanced at the gap between their hands and thelight sensor they were covering and muttered somethingabout the light or the avatar on screen One girl said ldquoit isdark now go quicklyrdquo and another boy asked to himselfthat ldquois it dark enoughrdquo The interesting thing was that oneboy bended his body upon the sensor to block the lightbecause he thought it would be quicker to see the outcome

8 The Scientific World Journal

(a) A girl with her maze (b) The most complicated maze

Figure 7 Two mazes collected from maze creation session a girl with her maze (a) and the complicated maze created by a boy (b)

Another remarkable domain where abstraction takes place ismaze design and creation Because the ldquowhat you see is whatyou getrdquo creation process children could then test complexabstractions quickly and precisely For example to create anewmaze the child designed amap where each square couldbe represented by certain programming blocks Then hesheselected only those appropriate for the virtual world in hishermind

(ii) Automation Automation occurs as the system executeschildrenrsquos programs The ldquoprogramrdquo itself automated ldquostep-ping throughrdquo and updated virtual avatarrsquos location anddirection (representing the veer) at each step As to this termwe did not expect too much actually it was an attempt tofind whether children could realize the automatic processAnd we indeed found some cue through the study Forexample before the execution one boy said ldquolisten to myinstructions and gordquo Another girl jumped for joy when shesaw the animation of her avatar traversing through the mazeBesides we saw 5 children stood up from their seats whenthe execution began and kept standing by waiting for thesuccessful moment

In interviews 4 children expressed their enjoyment whilewatching the virtual avatar traversing through the mazeautomatically under their instructions This was also backedup in the videotape from which we observed that 3 of themclapped out of joy and cheered on their avatarswhilewatchingthe animation

(iii) Problem Decomposition and Analysis T-Maze providesa platform that enables live programming by allowing thecode to be adjusted in real time As the session proceededwe moved beyond programs consisting of simple sequencesof actions and introduced more advanced concepts such asloops and numeric parameter values Through these activi-ties we found evidence that children could engage these con-cepts reasoning about possible outcomes of different blocksto build and test every piece of solution until they finallysolved the maze-escape problem Children also performedanalysis when they decidedwhether or not the avatar behavedas expected If the avatar was stuck or ldquomisbehavedrdquo it mayeither mean that their implementation of their control idea isfaulty or that some blocks were misplaced

Mostly the process of analysis happened implicitly butwecould still find some track from the observation For exampleone child in his complex maze-escape mission determinedthat he needed five forward blocks to reach the first turningAfter learning the loop syntax he realized that the largestvalue of a numeric block is five and the loop is limited to 5times the most so he needed two more forward blocks Thechild then built a program snippet as

(Start Loop + 5 + Forward + End Loop) 997888rarr Forward997888rarr Forward

In maze creation session we found 4 children chose tolocate the entrance and exit point of a maze at first and thentried to link them through while others just placed the blocksas they wish and were happy-go-lucky with any shape ofmazes Besides we observed a sense of ldquohabitrdquo in childrenrsquosprogramming that 5 children would add several blocks intothe sequence once when it was a simple straight line whileattempted one step at a time in a complex situation (a cornera sensor square or a cross) when they would adjust currentplacement and decide what to do next while observing thefeedbacks

(iv) Creativity To create a maze programming became amedium for childrenrsquos personal and creative expression inthe design of their mazes children engaged their fantasiesand built relationships with other pockets of reality that wentbeyond common thoughts As we analyzed childrenrsquos workcollected in the study several interesting mazes surprised usOne girl created her maze shaping like a snake (Figure 7(a))but she described it as the Great Wall coming out from hertextbook And the boywho claimed tomake themost difficultmaze used 31 blocks constructing the most complex fish-shape maze (Figure 7(b)) During the creating he put asidethe end block until he finished all the paths then counted thenumber of cells in each path trying to find the longest onewhere the end block was appended

Generally the mazes children created are different fromwhat we designed for T-Maze intrinsically and the descrip-tions children gave in interviews were also unexpected Forexample one boy created a maze with only one straight linewhile he described it as an airport runway And he talked

The Scientific World Journal 9

about his experience with his father at the airport last weekright off the reel

55 Summary Though the study is relatively small scale wecan tell from the results that T-Maze has some computationalconcepts in its programming activities and children indeedhandle those experienceswellThis study suggests farther thatT-Maze is easy to use for children and they are able to buildtheir own programs in playing and creating activities with T-Maze The emphasis is that T-Maze has the potential to helpraise childrenrsquos awareness about CT concepts

In interviews we also wanted to get some inspirationsfrom children with questions like ldquowhat other things do youthink the blocks can dordquo or ldquodo you think of any other gamesthat can be played with those blocksrdquo Two children said theywould like to control a flying plane with the blocks and 3others wished to help their mothers with some householdduties like cooking cleaning or furnishing their roomsTheir unlimited imagination inspired us in the revision andupgrade of T-Maze in the future

6 Conclusion and Future Work

T-Maze is designed for children aged 5ndash9 to help cultivatetheir computational thinking skills It allows children to playmaze-escape games and create their own mazes by buildingcomputer programs out of wooden blocks We conducted auser study to verify its ease of use and the potential capabilityto help children with computational thinking The resultsshowed that T-Maze has the promising potential to helpchildren cultivate some computational thinking awarenesslike abstraction automation problem decomposition andanalysis

In the future we will develop more games and scenariosdesign more kinds of programming blocks and enrich thetypes of sensors More and further user studies are neededto improve T-Maze for better user experience and skillsnurturing capability Besides as an upgrade we may providea collaborative tangible programming tool for children toexperience the fun of cooperation

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank the children from nearbykindergartens that took part in the study The authors grate-fully acknowledge financial support from the Major StateBasic Research Development Program of China under Grantno 2013CB328805 the National Natural Science Foundationof China under Grant no 60970090 and no 61272325 theCooperation Project of Chinese Academy of Sciences andFoshan City under Grant no 2012YS04 and the NationalScience and Technology Supported Program under Grant no2012BAH19F01

References

[1] J M Wing ldquoComputational thinkingrdquo Communications of theACM vol 49 no 3 pp 33ndash35 2006

[2] A Bundy ldquoComputational thinking is pervasiverdquo Journal ofScientific and Practical Computing vol 1 no 2 pp 67ndash69 2007

[3] S Grover ldquoComputational Thinking Programming andthe Google App Inventorrdquo 2010 httpwwwthesmartbeancomlibrarycomputational-thinking-programming-and-the-google-app-inventor

[4] S Grover ldquoExpanding the Technology Curriculum to IncludeFoundational Elements of Computer Science for K-8rdquo 2009httpwwwthesmartbeancommagazinehome-schoolingexpanding-the-technology-curriculum-foundational-ele-ments-of-computer-science-for-k-8

[5] G Orr ldquoComputational thinking through programming andalgorithmic artrdquo in Proceedings of the SIGGRAPH 2009 Talks(SIGGRAPH rsquo09) vol 1 p 1 ACM 2009

[6] M U Bers Blocks to Robots Learning with Technology in theEarly Childhood Classroom Teachers College New York NYUSA 2008

[7] D H Clements ldquoThe future of educational computing researchthe case of computer programmingrdquo Information Technology inChildhood Education Annual vol 1999 no 1 pp 147ndash179 1999

[8] T S McNerney Tangible programming bricks an approach tomaking programming accessible to everyone [MS thesis] MITCambridge Cambridge Mass USA 2000

[9] T S McNerney ldquoFrom turtles to tangible programming bricksexplorations in physical language designrdquo Personal and Ubiqui-tous Computing vol 8 no 5 pp 326ndash337 2004

[10] J Fails A Druin M Guha G Chipman S Simms and WChuraman ldquoChildrsquos play a comparison of desktop and physicalinteractive environmentsrdquo in Proceedings of the InteractionDesign and Children (IDC rsquo05) pp 48ndash55 ACM 2005

[11] M S Horn E T Solovey R J Crouser and R J K JacobldquoComparing the use of tangible and graphical programminglanguages for informal science educationrdquo in Proceedings of theSIGCHI Conference on Human Factors in Computing Systems(CHI rsquo09) pp 975ndash984 ACM 2009

[12] S PapertMindstorms Children Computers and Powerful IdeasBasic Books New York NY USA 1981

[13] G H Fletcher and J J Lu ldquoEducation human computing skillsrethinking the K-12 experiencerdquo Communications of the ACMvol 52 no 2 pp 23ndash25 2009

[14] P Curzon J Peckham A Settle E Roberts and H TaylorldquoComputational thinking(CT) on weaving it inrdquo in Proceedingsof the 14th Annual ACM SIGCSE Conference on Innovation andTechnology in Computer Science Education (ITiCSE rsquo09) pp201ndash202 ACM 2009

[15] P B Henderson ldquoUbiquitous computational thinkingrdquo Com-puter vol 42 no 10 pp 100ndash102 2009

[16] J J Lu and G H L Fletcher ldquoThinking about computationalthinkingrdquo in Proceedings of the 40th ACM Technical Symposiumon Computer Science Education (SIGCSE rsquo09) pp 260ndash264ACM 2009

[17] D Moursund ldquoComputational Thinkingrdquo IAE-Pedia 2009httpiae-pediaorgComputational Thinking

[18] National Research Council ldquoCommittee for the workshops oncomputational thinking 2010rdquo Report of a Workshop on theScope and Nature of Computational Thinking The NationalAcademies Press Washington DC USA 2010

10 The Scientific World Journal

[19] IWG Computational Thinking for Youth Education Develop-ment Center Newton Mass USA 2010

[20] M Resnick ldquoAll I really need to know (about creative thinking)I learned (by studying how children learn) in kindergartenrdquo inProceedings of the 6th Conference on Creativity amp Cognition (CCrsquo07) pp 1ndash6 ACM 2007

[21] J Dougherty ldquoConcept visualization in CS0 using ALICErdquoJournal of Computing Science in Colleges vol 22 no 3 pp 145ndash152 2007

[22] I Lee F Martin J Denner et al ldquoComputational thinking foryouth in practicerdquo ACM Inroads vol 2 no 1 pp 32ndash37 2011

[23] A Ruthmann J M Heines G R Greher P Laidler and CSaulters II ldquoTeaching computational thinking through musicallive coding in Scratchrdquo in Proceedings of the 41st ACM TechnicalSymposium on Computer Science Education (SIGCSE rsquo10) pp351ndash355 ACM 2010

[24] C Kelleher and R Pausch ldquoLowering the barriers to pro-gramming a taxonomy of programming environments andlanguages for novice programmersrdquo ACM Computing Surveysvol 37 no 2 pp 83ndash137 2005

[25] C Rader C Brand and C Lewis ldquoDegrees of comprehensionchildrenrsquos understanding of a visual programming environ-mentrdquo in Proceedings of the Conference on Human Factors inComputing Systems (CHI rsquo97) pp 351ndash358 ACM 1997

[26] PWyeth and H C Purchase ldquoTangible programming elementsfor young childrenrdquo in Proceedings of the Extended Abstracts onHuman Factors in Computing Systems (CHI rsquo02) pp 774ndash775ACM 2002

[27] M S Horn and R J K Jacob ldquoTangible programming in theclassroomwith ternrdquo inProceedings of the ExtendedAbstracts onHuman Factors in Computing Systems (CHI Trends Interactivity)(CHI rsquo07) pp 1965ndash1970 ACM 2007

[28] S Tarkan V Sazawal A Druin et al ldquoToque designing acooking-based programming language for and with childrenrdquoin Proceedings of the 28th Annual Conference on Human Factorsin Computing Systems (CHI rsquo10) pp 2417ndash2426 ACM 2010

[29] D Gallardo C F Julia and S Jorda ldquoTurTan a tangibleprogramming language for creative explorationrdquo in Proceedingsof the 3rd IEEE InternationalWorkshop onHorizontal InteractiveHuman Computer System (TABLETOP rsquo08) pp 89ndash92 IEEEComputer Society Press 2008

[30] J Silver andE Rosenbaum ldquoTwinkle programmingwith colorrdquoin Proceedings of the 4th International Conference on TangibleEmbedded and Embodied Interaction (TEI rsquo10) pp 383ndash384ACM 2010

[31] D L Wang C Zhang and H A Wang ldquoT-Maze a tangibleprogramming tool for childrenrdquo in Proceedings of the 10thInternational Conference on Interaction Design and Children(IDC rsquo11) pp 127ndash135 ACM 2011

[32] M S Horn ldquoTopCode Tangible Object Placement Codesrdquo2012 httpuserseecsnorthwesternedusimmhorntopcodes

[33] K Howland ldquoSupporting the development of multimodalwriting and computational thinking skills through computergame creationrdquo in Proceedings of the IEEE Symposium on VisualLanguages and Human-Centric Computing (VLHCC rsquo09) pp250ndash251 IEEE Computer Society Press 2009

[34] J LrsquoHeureux D Boisvert K Sanghera and R Cohen ldquoITproblem solving an implementation of computational thinkingin information technologyrdquo in Proceedings of the 13th AnnualConference on Information Technology Education (SIGITE rsquo12)pp 183ndash188 ACM 2012

[35] M Guzdial ldquoEducation paving the way for computationalthinkingrdquo Communications of the ACM vol 51 no 8 pp 25ndash27 2008

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

International Journal of

ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia

International Journal of

Biomedical Imaging

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArtificialNeural Systems

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational Intelligence and Neuroscience

Industrial EngineeringJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

4 The Scientific World Journal

(a) Start and end block (b) Sensor block (c) Normal block

(d) Loop and numeric block (e) Direction block

Figure 2 Tangible programming blocks

Button sensor

Temperature sensor

Light sensor

Figure 3 Physical sensors in T-Maze

the compiler converts a programdirectly into virtualmachinecode

35 Game Modules T-Maze has two game modules MazeEscape and Maze Creation Game and creation are activitieswhich have good potential for computational thinking skills[33] Mazes are compelling and meaningful to most childrenfor their experience in school or day-to-day life The mazemetaphor is used becausewe think that placing programmingin this context could motivate the child so that heshe doesnot give up

Problem analysis takes place during the process of testingand debugging their programs [22] The computer visionsystem captured images automatically enabling T-Maze togive real-time graphical and voice feedbacks which helpchildren with the analysis For example in Maze Escape(Figure 4) when a block is placed correctly the background

Green arrow prompt

Figure 4 Interface of one level maze-escape game

of the occupied square will change into a green arrowOtherwise the smiley on the upper left screen will turnupset along with a voice prompting the possible solutions likeldquomaybe I need a sensorrdquo or ldquodid I walk too many steps in thisdirectionrdquo In Maze Creation (Figure 5) the real-time visualfeedback tells children ldquowhat you see is what you getrdquo andprovides them a live creating experience

4 CT in T-Maze Programming

CT is a comprehensive term that covers abstraction automa-tion analysis creativity and so forth [14] In this paper weselect to focus on several concepts that can be observed inT-Maze

41 Abstraction Abstraction ability is to find appropriatelevel of detail to define and solve a problem [34] With T-Maze children could simulate the abstract model of certainmazes in realitymap themselves into the virtual characters onscreen and control the charactersrsquo behaviors For example toplay a maze-escape game in T-Maze children need to builda path for their avatars using the physical blocks In order toknowhow these blocks function children need to think abouthow to perceive the world coordinates of the virtualmaze andhow to map the behaviors in real world like move straight

The Scientific World Journal 5

(a) Maze creating area (b) User interface of Maze Creation

Figure 5 The creating area (a) and user interface (b) of Maze Creation

or make a left turn into the virtual charactersrsquo behaviors inthe programs As to the sensors there are three layers ofabstraction-reality mapping that is sensor cells in virtualmaze physical sensor blocks and physical sensors

42 Automation Automation is a labor saving process inwhich a computer is instructed to execute a set of repetitivetasks It is much more efficiently compared to the processingpower of a human and the automated execution of processby machine is going to change everything [35] In this lightcomputer programs are ldquoautomations of abstractionsrdquo Theprogram children built in T-Maze automates the simulationof escaping from maze using a run loop which updates thefeedbacks on screen and interacts with users on the basis ofdesigned rules for sensors

43 Problem Decomposition and Analysis Decompositionmay take the form of stripping down a problem to whatis believed to be its bare essentials [22] Breaking problemsdown into smaller parts that may be more easily solved andanalyzing to reuse them can enable and simplify the resolv-ing of more complicated and larger-scale problems Andanalysis is a reflective practice which validates whether theabstractions are correct In programming theywould analyzewhether the avatarsrsquo behavior is expected and whether thereare some conditions that are not taken into account duringthe abstraction phase Children also engage in analysis whenthey judge whether their abstractions are efficient Thisanalysis may help them optimize and find a better solutionto the problem

In T-Maze the entry of the maze is marked with greentexts and the exit is marked with red texts Children areguided to get a blueprint of the path to be built from thesemarkers After knowing the ports where the path starts andends children have to connect them by ldquopavingrdquo the cellsin mazes with programming blocks During this processchildren are encouraged to analyze and compare their plansand choose the one that leads to the shortest path To thisend the feasible path in one maze is not unique And severalsections of the paths are designed to be straight in one

direction where the loop block may be a better choiceHowever it is not worthy to use the loop block if the straightsection of the path has less than four cells Though this is alittle ticklish for children it is where the analysis is neededAlso T-Maze gives real-time feedbacks which help childrenwith the analysis when testing and debugging their programs

44 Creativity Creativity is both a comprehensive capabilityof humanbeings and an important part of theCT skill set [14]We try to develop childrenrsquos creative skill by stimulating theircuriosity especially the creative imagination Creation is atthe root of creative thinking [20] The tangible programmingtool should enable children to process any abstractingsimulating and creating activities freely The freedom mayopen childrenrsquos minds a key point in promoting childrenrsquoscreativity

And as such the Maze Creation module is designed forthis purpose which offers children freedom to create anyshape of mazes as they like It is interesting to see whatchildren would come up with from the creating

5 User Study

We have conducted several studies to evaluate T-Maze interms of its usabilityThese studies involved 20 children (aged5ndash9) using T-Maze to program in a laboratory environmentIn the latest study we focused on the potential capabilityof T-Maze to nurture computational thinking in childrenrsquosprogramming activities

51 Research Design

511 ResearchQuestions For this study wewere interested inseveral research questions with an emphasis on the last one

(1) Is T-Maze easy to use for children(2) Do children like programmingwith T-Maze andwhat

are they interested in(3) How do the four CT concepts take shape for children

in programming

6 The Scientific World Journal

(a) (b)

Figure 6 The simple (a) and complex (b) maze-escape games in the study

Though the first two questions have been examined inprevious studies we wanted to be sure that children arecomfortable to involve in our study every time Consideringthe diversity of users and usage scenarios we regard theusability of T-Maze as a long-term test subject especiallywhen some new modifications or improvements need to beverified Also we expect usable feedbacks from children ifany to help us improve the tool since it is always preferableto have more children like T-Maze

512 Tasks We conducted two sessions in the study mazeescape and maze creation In maze escape session childrenwere asked to complete two different levels of maze-escapegames which were designed in consideration of individualability difference as well as the occasions where problemdecomposition and analysis probably happen To play amaze-escape game children had to find a passable path in themaze for the avatar to traverse through by building programswith wooden programming blocks The simple maze-escapegame only had one passable path and would need 9 blocks atmost (without the use of Loop blocks) while the complex onehadmultiple passable paths and would need variable numberof blocks according to the length of chosen path (Figure 6)For a maze creation task children were given no restrictionsto create anymaze they liked by constructing newmazemapswith the same set of blocks There was no time limitation inboth sessions Throughout the study we had a comfortableseat in front of the table for children to sit when they wereworking on the tasks as well as a rest area for them to wait orrest

513 Interviews and Questions For the interviews after thetwo main sessions we wanted to know childrenrsquos opinionsabout the tool their programming experience or anythingthat remained in their mind So the questions asked in theinterviews were mostly without model answers as followsldquoWhat other things do you think the blocks can dordquo ldquoDo youhave any other games that can be played with those blocksrdquoldquoHave you ever noticed any sensors or something like that inyour liferdquo The questions are listed in a questionnaire whichalso contains another two questions about the usability of T-Maze and one about the invitation for our next study recruit

Reasonably we did not expect children to becomecomputational thinkers after the one-off practices but wehoped to find some clues that might answer our questionsand preferably got some implications for teaching childrencomputational thinking

52 Participants and Apparatus Seven children (aged 8 inaverage) 5 girls and 2 boys participated in this studyThey allhad some experience in computer use but none of them hadever known about programming or used any programmingtools

The apparatus was a set of the latest version of T-Mazeplaced on an only table in a laboratory room The table witha chair was in the center of the room and we also made aresting area at the right corner beside the door One of thetwo cameras was set up in front left of the table while theother one was behind the table to the left Each time onechild went to the table to complete the tasks with hisher backtowards others waiting in the resting area so that they wouldnot interrupt the experiment

53 Procedure The study was conducted in four stagesdemonstration practice main sessions and interviewsBefore we started all children watched a demonstrationabout how to play a simple maze-escape game and create anew maze We introduced how to use the loop blocks andencouraged children to find the shortest path as they canThen they were given some minutes to practice freely Afterthey were ready to start the main sessions of the study began

Without time limitation children could involve in theirtasks as much time as they like but they were asked to give asign when they were about to begin or to finish and we notedthe time for records In maze-escape session children wereexplained to complete a simple escape task first and a complexone afterward Next in maze creation session children weretold to freely create any maze as they liked When theyfinished the creation we encouraged them to introduce theirmazes to others hoping to invite some creative talking Afterthemain sessions we conducted one-on-one semi-structuredinterviewswith children andwe taped all the conversation forlater analysis

The Scientific World Journal 7

Table 1 The time (unit minute) and blocks children used in the study

NumberMaze escape session Maze creation session

Simple ComplexTime (min) Blocks Time (min) Blocks Time (min) Blocks

1 350 900 700 1800 900 18002 300 900 500 1900 1000 26003 450 900 700 2100 1300 18004 450 900 900 2200 800 16005 300 900 800 2000 1100 31006 400 800 900 2100 1000 26007 400 900 800 1800 900 2500Avg 378 886 757 1986 986 2286

We collected qualitative data in the form of observationnotes photographs and videotape from which some quanti-tative data was also derivedWe also collected childrenrsquos work(the mazes they created) Three researchers were involved inthis study two collected data while the third acted in the roleof lead interpreter

54 Results and Discussion We comprehensively analyzedthe observation notes tapes and interview records andfound answers to the research questions from the analysisof childrenrsquos words and behaviors as well as the statisticalanalysis

For the first question (easy to use) we were interested inwhether children could create their own computer programsBased on observation notes and an analysis of videotapewe found that children were able to manipulate the tangibleblocks to form their own programs We allowed childrento take as much time as they wanted to complete a taskHowever we were lucky to see that they neither seemed totake their time nor to rush through the experiment whichgave us the reasonable and valid data in Table 1 Childrenwere able to accomplish both the simple and complex maze-escape tasks within similar period of time Also they wereeven more engaged in maze creation session and spent moretime creating and exploring All children finished the twosessions with their freely-created mazes some of which hadinteresting shapes or complex structure with sensor blocksused

In interviews when asked whether the tool was easy touse 2 children chose ldquovery easyrdquo 4 chose ldquoeasyrdquo and one choseldquonormalrdquo

For the second research question (interesting to playwith) according to the records and videotape childrenalways showed their interests in sensors 3 children askedquestions about the sensors like ldquowhat are these used forrdquoldquowhat if I pressed it longerrdquo ldquoonly three of themrdquo Moreoverchildren were very excited in maze creation section andshowed the richest behaviors For example one boy said ldquoIrsquomgoing to make the most difficult maze that none of themcan get throughrdquo and another boy asked ldquowhat if the blocksare not enough to create my mazerdquo As part of the mazecreation session we asked children not only to demonstratetheir mazes but also to show others the blocks they used to

program it In many cases children gathered together to thetable just stood by watching and praised for others In othercases however children asked for a competition to provetheir own

In interviews when asked whether the system was funall children confirmed their enjoyment with T-Maze andexpressed thewish to participate again in the future 4 of themvoted for the maze creation session to be their favorite

For the last question (CT concepts in programming)we found some cue that T-Maze has the potential to raisechildrenrsquos awareness about CT and help them understand CTto some extent

(i) AbstractionThemazemetaphor used inT-Maze promptedchildren to relate their experience about real-world maze tothe virtual maze on the screen They needed the process ofabstraction to narrow the escape problem down to some-thing that could be implemented on the computer using T-Maze as well as mapping the physical blocks to the virtualsquares in maze map on screen Restrictions imposed by theprogramming environment include an upper bound on thenumber of blocks and a limit on the size of the programmingarea (40 cm width by 30 cm height) Abstraction also tookplace as children designed avatars to react to a limited set ofconditions that may be encountered in the real-world mazesuch as some roadblocks represented by sensors

The fact that children all accomplished the tasks success-fully was inspiring and we indeed found something relatedfrom the observation In maze-escape session when theavatar came across a temperature sensor square in the mazewe found 3 children attempting to heat up the physicalsensors with their breath Later in the interviews we askedwhy did they do that and one said ldquoEr my mum alwayswarm my hand with her breath in winterrdquo Other two saidthat they always knew their breath was warmer than theirhands The same situation happened with light sensor where4 children glanced at the gap between their hands and thelight sensor they were covering and muttered somethingabout the light or the avatar on screen One girl said ldquoit isdark now go quicklyrdquo and another boy asked to himselfthat ldquois it dark enoughrdquo The interesting thing was that oneboy bended his body upon the sensor to block the lightbecause he thought it would be quicker to see the outcome

8 The Scientific World Journal

(a) A girl with her maze (b) The most complicated maze

Figure 7 Two mazes collected from maze creation session a girl with her maze (a) and the complicated maze created by a boy (b)

Another remarkable domain where abstraction takes place ismaze design and creation Because the ldquowhat you see is whatyou getrdquo creation process children could then test complexabstractions quickly and precisely For example to create anewmaze the child designed amap where each square couldbe represented by certain programming blocks Then hesheselected only those appropriate for the virtual world in hishermind

(ii) Automation Automation occurs as the system executeschildrenrsquos programs The ldquoprogramrdquo itself automated ldquostep-ping throughrdquo and updated virtual avatarrsquos location anddirection (representing the veer) at each step As to this termwe did not expect too much actually it was an attempt tofind whether children could realize the automatic processAnd we indeed found some cue through the study Forexample before the execution one boy said ldquolisten to myinstructions and gordquo Another girl jumped for joy when shesaw the animation of her avatar traversing through the mazeBesides we saw 5 children stood up from their seats whenthe execution began and kept standing by waiting for thesuccessful moment

In interviews 4 children expressed their enjoyment whilewatching the virtual avatar traversing through the mazeautomatically under their instructions This was also backedup in the videotape from which we observed that 3 of themclapped out of joy and cheered on their avatarswhilewatchingthe animation

(iii) Problem Decomposition and Analysis T-Maze providesa platform that enables live programming by allowing thecode to be adjusted in real time As the session proceededwe moved beyond programs consisting of simple sequencesof actions and introduced more advanced concepts such asloops and numeric parameter values Through these activi-ties we found evidence that children could engage these con-cepts reasoning about possible outcomes of different blocksto build and test every piece of solution until they finallysolved the maze-escape problem Children also performedanalysis when they decidedwhether or not the avatar behavedas expected If the avatar was stuck or ldquomisbehavedrdquo it mayeither mean that their implementation of their control idea isfaulty or that some blocks were misplaced

Mostly the process of analysis happened implicitly butwecould still find some track from the observation For exampleone child in his complex maze-escape mission determinedthat he needed five forward blocks to reach the first turningAfter learning the loop syntax he realized that the largestvalue of a numeric block is five and the loop is limited to 5times the most so he needed two more forward blocks Thechild then built a program snippet as

(Start Loop + 5 + Forward + End Loop) 997888rarr Forward997888rarr Forward

In maze creation session we found 4 children chose tolocate the entrance and exit point of a maze at first and thentried to link them through while others just placed the blocksas they wish and were happy-go-lucky with any shape ofmazes Besides we observed a sense of ldquohabitrdquo in childrenrsquosprogramming that 5 children would add several blocks intothe sequence once when it was a simple straight line whileattempted one step at a time in a complex situation (a cornera sensor square or a cross) when they would adjust currentplacement and decide what to do next while observing thefeedbacks

(iv) Creativity To create a maze programming became amedium for childrenrsquos personal and creative expression inthe design of their mazes children engaged their fantasiesand built relationships with other pockets of reality that wentbeyond common thoughts As we analyzed childrenrsquos workcollected in the study several interesting mazes surprised usOne girl created her maze shaping like a snake (Figure 7(a))but she described it as the Great Wall coming out from hertextbook And the boywho claimed tomake themost difficultmaze used 31 blocks constructing the most complex fish-shape maze (Figure 7(b)) During the creating he put asidethe end block until he finished all the paths then counted thenumber of cells in each path trying to find the longest onewhere the end block was appended

Generally the mazes children created are different fromwhat we designed for T-Maze intrinsically and the descrip-tions children gave in interviews were also unexpected Forexample one boy created a maze with only one straight linewhile he described it as an airport runway And he talked

The Scientific World Journal 9

about his experience with his father at the airport last weekright off the reel

55 Summary Though the study is relatively small scale wecan tell from the results that T-Maze has some computationalconcepts in its programming activities and children indeedhandle those experienceswellThis study suggests farther thatT-Maze is easy to use for children and they are able to buildtheir own programs in playing and creating activities with T-Maze The emphasis is that T-Maze has the potential to helpraise childrenrsquos awareness about CT concepts

In interviews we also wanted to get some inspirationsfrom children with questions like ldquowhat other things do youthink the blocks can dordquo or ldquodo you think of any other gamesthat can be played with those blocksrdquo Two children said theywould like to control a flying plane with the blocks and 3others wished to help their mothers with some householdduties like cooking cleaning or furnishing their roomsTheir unlimited imagination inspired us in the revision andupgrade of T-Maze in the future

6 Conclusion and Future Work

T-Maze is designed for children aged 5ndash9 to help cultivatetheir computational thinking skills It allows children to playmaze-escape games and create their own mazes by buildingcomputer programs out of wooden blocks We conducted auser study to verify its ease of use and the potential capabilityto help children with computational thinking The resultsshowed that T-Maze has the promising potential to helpchildren cultivate some computational thinking awarenesslike abstraction automation problem decomposition andanalysis

In the future we will develop more games and scenariosdesign more kinds of programming blocks and enrich thetypes of sensors More and further user studies are neededto improve T-Maze for better user experience and skillsnurturing capability Besides as an upgrade we may providea collaborative tangible programming tool for children toexperience the fun of cooperation

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank the children from nearbykindergartens that took part in the study The authors grate-fully acknowledge financial support from the Major StateBasic Research Development Program of China under Grantno 2013CB328805 the National Natural Science Foundationof China under Grant no 60970090 and no 61272325 theCooperation Project of Chinese Academy of Sciences andFoshan City under Grant no 2012YS04 and the NationalScience and Technology Supported Program under Grant no2012BAH19F01

References

[1] J M Wing ldquoComputational thinkingrdquo Communications of theACM vol 49 no 3 pp 33ndash35 2006

[2] A Bundy ldquoComputational thinking is pervasiverdquo Journal ofScientific and Practical Computing vol 1 no 2 pp 67ndash69 2007

[3] S Grover ldquoComputational Thinking Programming andthe Google App Inventorrdquo 2010 httpwwwthesmartbeancomlibrarycomputational-thinking-programming-and-the-google-app-inventor

[4] S Grover ldquoExpanding the Technology Curriculum to IncludeFoundational Elements of Computer Science for K-8rdquo 2009httpwwwthesmartbeancommagazinehome-schoolingexpanding-the-technology-curriculum-foundational-ele-ments-of-computer-science-for-k-8

[5] G Orr ldquoComputational thinking through programming andalgorithmic artrdquo in Proceedings of the SIGGRAPH 2009 Talks(SIGGRAPH rsquo09) vol 1 p 1 ACM 2009

[6] M U Bers Blocks to Robots Learning with Technology in theEarly Childhood Classroom Teachers College New York NYUSA 2008

[7] D H Clements ldquoThe future of educational computing researchthe case of computer programmingrdquo Information Technology inChildhood Education Annual vol 1999 no 1 pp 147ndash179 1999

[8] T S McNerney Tangible programming bricks an approach tomaking programming accessible to everyone [MS thesis] MITCambridge Cambridge Mass USA 2000

[9] T S McNerney ldquoFrom turtles to tangible programming bricksexplorations in physical language designrdquo Personal and Ubiqui-tous Computing vol 8 no 5 pp 326ndash337 2004

[10] J Fails A Druin M Guha G Chipman S Simms and WChuraman ldquoChildrsquos play a comparison of desktop and physicalinteractive environmentsrdquo in Proceedings of the InteractionDesign and Children (IDC rsquo05) pp 48ndash55 ACM 2005

[11] M S Horn E T Solovey R J Crouser and R J K JacobldquoComparing the use of tangible and graphical programminglanguages for informal science educationrdquo in Proceedings of theSIGCHI Conference on Human Factors in Computing Systems(CHI rsquo09) pp 975ndash984 ACM 2009

[12] S PapertMindstorms Children Computers and Powerful IdeasBasic Books New York NY USA 1981

[13] G H Fletcher and J J Lu ldquoEducation human computing skillsrethinking the K-12 experiencerdquo Communications of the ACMvol 52 no 2 pp 23ndash25 2009

[14] P Curzon J Peckham A Settle E Roberts and H TaylorldquoComputational thinking(CT) on weaving it inrdquo in Proceedingsof the 14th Annual ACM SIGCSE Conference on Innovation andTechnology in Computer Science Education (ITiCSE rsquo09) pp201ndash202 ACM 2009

[15] P B Henderson ldquoUbiquitous computational thinkingrdquo Com-puter vol 42 no 10 pp 100ndash102 2009

[16] J J Lu and G H L Fletcher ldquoThinking about computationalthinkingrdquo in Proceedings of the 40th ACM Technical Symposiumon Computer Science Education (SIGCSE rsquo09) pp 260ndash264ACM 2009

[17] D Moursund ldquoComputational Thinkingrdquo IAE-Pedia 2009httpiae-pediaorgComputational Thinking

[18] National Research Council ldquoCommittee for the workshops oncomputational thinking 2010rdquo Report of a Workshop on theScope and Nature of Computational Thinking The NationalAcademies Press Washington DC USA 2010

10 The Scientific World Journal

[19] IWG Computational Thinking for Youth Education Develop-ment Center Newton Mass USA 2010

[20] M Resnick ldquoAll I really need to know (about creative thinking)I learned (by studying how children learn) in kindergartenrdquo inProceedings of the 6th Conference on Creativity amp Cognition (CCrsquo07) pp 1ndash6 ACM 2007

[21] J Dougherty ldquoConcept visualization in CS0 using ALICErdquoJournal of Computing Science in Colleges vol 22 no 3 pp 145ndash152 2007

[22] I Lee F Martin J Denner et al ldquoComputational thinking foryouth in practicerdquo ACM Inroads vol 2 no 1 pp 32ndash37 2011

[23] A Ruthmann J M Heines G R Greher P Laidler and CSaulters II ldquoTeaching computational thinking through musicallive coding in Scratchrdquo in Proceedings of the 41st ACM TechnicalSymposium on Computer Science Education (SIGCSE rsquo10) pp351ndash355 ACM 2010

[24] C Kelleher and R Pausch ldquoLowering the barriers to pro-gramming a taxonomy of programming environments andlanguages for novice programmersrdquo ACM Computing Surveysvol 37 no 2 pp 83ndash137 2005

[25] C Rader C Brand and C Lewis ldquoDegrees of comprehensionchildrenrsquos understanding of a visual programming environ-mentrdquo in Proceedings of the Conference on Human Factors inComputing Systems (CHI rsquo97) pp 351ndash358 ACM 1997

[26] PWyeth and H C Purchase ldquoTangible programming elementsfor young childrenrdquo in Proceedings of the Extended Abstracts onHuman Factors in Computing Systems (CHI rsquo02) pp 774ndash775ACM 2002

[27] M S Horn and R J K Jacob ldquoTangible programming in theclassroomwith ternrdquo inProceedings of the ExtendedAbstracts onHuman Factors in Computing Systems (CHI Trends Interactivity)(CHI rsquo07) pp 1965ndash1970 ACM 2007

[28] S Tarkan V Sazawal A Druin et al ldquoToque designing acooking-based programming language for and with childrenrdquoin Proceedings of the 28th Annual Conference on Human Factorsin Computing Systems (CHI rsquo10) pp 2417ndash2426 ACM 2010

[29] D Gallardo C F Julia and S Jorda ldquoTurTan a tangibleprogramming language for creative explorationrdquo in Proceedingsof the 3rd IEEE InternationalWorkshop onHorizontal InteractiveHuman Computer System (TABLETOP rsquo08) pp 89ndash92 IEEEComputer Society Press 2008

[30] J Silver andE Rosenbaum ldquoTwinkle programmingwith colorrdquoin Proceedings of the 4th International Conference on TangibleEmbedded and Embodied Interaction (TEI rsquo10) pp 383ndash384ACM 2010

[31] D L Wang C Zhang and H A Wang ldquoT-Maze a tangibleprogramming tool for childrenrdquo in Proceedings of the 10thInternational Conference on Interaction Design and Children(IDC rsquo11) pp 127ndash135 ACM 2011

[32] M S Horn ldquoTopCode Tangible Object Placement Codesrdquo2012 httpuserseecsnorthwesternedusimmhorntopcodes

[33] K Howland ldquoSupporting the development of multimodalwriting and computational thinking skills through computergame creationrdquo in Proceedings of the IEEE Symposium on VisualLanguages and Human-Centric Computing (VLHCC rsquo09) pp250ndash251 IEEE Computer Society Press 2009

[34] J LrsquoHeureux D Boisvert K Sanghera and R Cohen ldquoITproblem solving an implementation of computational thinkingin information technologyrdquo in Proceedings of the 13th AnnualConference on Information Technology Education (SIGITE rsquo12)pp 183ndash188 ACM 2012

[35] M Guzdial ldquoEducation paving the way for computationalthinkingrdquo Communications of the ACM vol 51 no 8 pp 25ndash27 2008

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

International Journal of

ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia

International Journal of

Biomedical Imaging

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArtificialNeural Systems

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational Intelligence and Neuroscience

Industrial EngineeringJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World Journal 5

(a) Maze creating area (b) User interface of Maze Creation

Figure 5 The creating area (a) and user interface (b) of Maze Creation

or make a left turn into the virtual charactersrsquo behaviors inthe programs As to the sensors there are three layers ofabstraction-reality mapping that is sensor cells in virtualmaze physical sensor blocks and physical sensors

42 Automation Automation is a labor saving process inwhich a computer is instructed to execute a set of repetitivetasks It is much more efficiently compared to the processingpower of a human and the automated execution of processby machine is going to change everything [35] In this lightcomputer programs are ldquoautomations of abstractionsrdquo Theprogram children built in T-Maze automates the simulationof escaping from maze using a run loop which updates thefeedbacks on screen and interacts with users on the basis ofdesigned rules for sensors

43 Problem Decomposition and Analysis Decompositionmay take the form of stripping down a problem to whatis believed to be its bare essentials [22] Breaking problemsdown into smaller parts that may be more easily solved andanalyzing to reuse them can enable and simplify the resolv-ing of more complicated and larger-scale problems Andanalysis is a reflective practice which validates whether theabstractions are correct In programming theywould analyzewhether the avatarsrsquo behavior is expected and whether thereare some conditions that are not taken into account duringthe abstraction phase Children also engage in analysis whenthey judge whether their abstractions are efficient Thisanalysis may help them optimize and find a better solutionto the problem

In T-Maze the entry of the maze is marked with greentexts and the exit is marked with red texts Children areguided to get a blueprint of the path to be built from thesemarkers After knowing the ports where the path starts andends children have to connect them by ldquopavingrdquo the cellsin mazes with programming blocks During this processchildren are encouraged to analyze and compare their plansand choose the one that leads to the shortest path To thisend the feasible path in one maze is not unique And severalsections of the paths are designed to be straight in one

direction where the loop block may be a better choiceHowever it is not worthy to use the loop block if the straightsection of the path has less than four cells Though this is alittle ticklish for children it is where the analysis is neededAlso T-Maze gives real-time feedbacks which help childrenwith the analysis when testing and debugging their programs

44 Creativity Creativity is both a comprehensive capabilityof humanbeings and an important part of theCT skill set [14]We try to develop childrenrsquos creative skill by stimulating theircuriosity especially the creative imagination Creation is atthe root of creative thinking [20] The tangible programmingtool should enable children to process any abstractingsimulating and creating activities freely The freedom mayopen childrenrsquos minds a key point in promoting childrenrsquoscreativity

And as such the Maze Creation module is designed forthis purpose which offers children freedom to create anyshape of mazes as they like It is interesting to see whatchildren would come up with from the creating

5 User Study

We have conducted several studies to evaluate T-Maze interms of its usabilityThese studies involved 20 children (aged5ndash9) using T-Maze to program in a laboratory environmentIn the latest study we focused on the potential capabilityof T-Maze to nurture computational thinking in childrenrsquosprogramming activities

51 Research Design

511 ResearchQuestions For this study wewere interested inseveral research questions with an emphasis on the last one

(1) Is T-Maze easy to use for children(2) Do children like programmingwith T-Maze andwhat

are they interested in(3) How do the four CT concepts take shape for children

in programming

6 The Scientific World Journal

(a) (b)

Figure 6 The simple (a) and complex (b) maze-escape games in the study

Though the first two questions have been examined inprevious studies we wanted to be sure that children arecomfortable to involve in our study every time Consideringthe diversity of users and usage scenarios we regard theusability of T-Maze as a long-term test subject especiallywhen some new modifications or improvements need to beverified Also we expect usable feedbacks from children ifany to help us improve the tool since it is always preferableto have more children like T-Maze

512 Tasks We conducted two sessions in the study mazeescape and maze creation In maze escape session childrenwere asked to complete two different levels of maze-escapegames which were designed in consideration of individualability difference as well as the occasions where problemdecomposition and analysis probably happen To play amaze-escape game children had to find a passable path in themaze for the avatar to traverse through by building programswith wooden programming blocks The simple maze-escapegame only had one passable path and would need 9 blocks atmost (without the use of Loop blocks) while the complex onehadmultiple passable paths and would need variable numberof blocks according to the length of chosen path (Figure 6)For a maze creation task children were given no restrictionsto create anymaze they liked by constructing newmazemapswith the same set of blocks There was no time limitation inboth sessions Throughout the study we had a comfortableseat in front of the table for children to sit when they wereworking on the tasks as well as a rest area for them to wait orrest

513 Interviews and Questions For the interviews after thetwo main sessions we wanted to know childrenrsquos opinionsabout the tool their programming experience or anythingthat remained in their mind So the questions asked in theinterviews were mostly without model answers as followsldquoWhat other things do you think the blocks can dordquo ldquoDo youhave any other games that can be played with those blocksrdquoldquoHave you ever noticed any sensors or something like that inyour liferdquo The questions are listed in a questionnaire whichalso contains another two questions about the usability of T-Maze and one about the invitation for our next study recruit

Reasonably we did not expect children to becomecomputational thinkers after the one-off practices but wehoped to find some clues that might answer our questionsand preferably got some implications for teaching childrencomputational thinking

52 Participants and Apparatus Seven children (aged 8 inaverage) 5 girls and 2 boys participated in this studyThey allhad some experience in computer use but none of them hadever known about programming or used any programmingtools

The apparatus was a set of the latest version of T-Mazeplaced on an only table in a laboratory room The table witha chair was in the center of the room and we also made aresting area at the right corner beside the door One of thetwo cameras was set up in front left of the table while theother one was behind the table to the left Each time onechild went to the table to complete the tasks with hisher backtowards others waiting in the resting area so that they wouldnot interrupt the experiment

53 Procedure The study was conducted in four stagesdemonstration practice main sessions and interviewsBefore we started all children watched a demonstrationabout how to play a simple maze-escape game and create anew maze We introduced how to use the loop blocks andencouraged children to find the shortest path as they canThen they were given some minutes to practice freely Afterthey were ready to start the main sessions of the study began

Without time limitation children could involve in theirtasks as much time as they like but they were asked to give asign when they were about to begin or to finish and we notedthe time for records In maze-escape session children wereexplained to complete a simple escape task first and a complexone afterward Next in maze creation session children weretold to freely create any maze as they liked When theyfinished the creation we encouraged them to introduce theirmazes to others hoping to invite some creative talking Afterthemain sessions we conducted one-on-one semi-structuredinterviewswith children andwe taped all the conversation forlater analysis

The Scientific World Journal 7

Table 1 The time (unit minute) and blocks children used in the study

NumberMaze escape session Maze creation session

Simple ComplexTime (min) Blocks Time (min) Blocks Time (min) Blocks

1 350 900 700 1800 900 18002 300 900 500 1900 1000 26003 450 900 700 2100 1300 18004 450 900 900 2200 800 16005 300 900 800 2000 1100 31006 400 800 900 2100 1000 26007 400 900 800 1800 900 2500Avg 378 886 757 1986 986 2286

We collected qualitative data in the form of observationnotes photographs and videotape from which some quanti-tative data was also derivedWe also collected childrenrsquos work(the mazes they created) Three researchers were involved inthis study two collected data while the third acted in the roleof lead interpreter

54 Results and Discussion We comprehensively analyzedthe observation notes tapes and interview records andfound answers to the research questions from the analysisof childrenrsquos words and behaviors as well as the statisticalanalysis

For the first question (easy to use) we were interested inwhether children could create their own computer programsBased on observation notes and an analysis of videotapewe found that children were able to manipulate the tangibleblocks to form their own programs We allowed childrento take as much time as they wanted to complete a taskHowever we were lucky to see that they neither seemed totake their time nor to rush through the experiment whichgave us the reasonable and valid data in Table 1 Childrenwere able to accomplish both the simple and complex maze-escape tasks within similar period of time Also they wereeven more engaged in maze creation session and spent moretime creating and exploring All children finished the twosessions with their freely-created mazes some of which hadinteresting shapes or complex structure with sensor blocksused

In interviews when asked whether the tool was easy touse 2 children chose ldquovery easyrdquo 4 chose ldquoeasyrdquo and one choseldquonormalrdquo

For the second research question (interesting to playwith) according to the records and videotape childrenalways showed their interests in sensors 3 children askedquestions about the sensors like ldquowhat are these used forrdquoldquowhat if I pressed it longerrdquo ldquoonly three of themrdquo Moreoverchildren were very excited in maze creation section andshowed the richest behaviors For example one boy said ldquoIrsquomgoing to make the most difficult maze that none of themcan get throughrdquo and another boy asked ldquowhat if the blocksare not enough to create my mazerdquo As part of the mazecreation session we asked children not only to demonstratetheir mazes but also to show others the blocks they used to

program it In many cases children gathered together to thetable just stood by watching and praised for others In othercases however children asked for a competition to provetheir own

In interviews when asked whether the system was funall children confirmed their enjoyment with T-Maze andexpressed thewish to participate again in the future 4 of themvoted for the maze creation session to be their favorite

For the last question (CT concepts in programming)we found some cue that T-Maze has the potential to raisechildrenrsquos awareness about CT and help them understand CTto some extent

(i) AbstractionThemazemetaphor used inT-Maze promptedchildren to relate their experience about real-world maze tothe virtual maze on the screen They needed the process ofabstraction to narrow the escape problem down to some-thing that could be implemented on the computer using T-Maze as well as mapping the physical blocks to the virtualsquares in maze map on screen Restrictions imposed by theprogramming environment include an upper bound on thenumber of blocks and a limit on the size of the programmingarea (40 cm width by 30 cm height) Abstraction also tookplace as children designed avatars to react to a limited set ofconditions that may be encountered in the real-world mazesuch as some roadblocks represented by sensors

The fact that children all accomplished the tasks success-fully was inspiring and we indeed found something relatedfrom the observation In maze-escape session when theavatar came across a temperature sensor square in the mazewe found 3 children attempting to heat up the physicalsensors with their breath Later in the interviews we askedwhy did they do that and one said ldquoEr my mum alwayswarm my hand with her breath in winterrdquo Other two saidthat they always knew their breath was warmer than theirhands The same situation happened with light sensor where4 children glanced at the gap between their hands and thelight sensor they were covering and muttered somethingabout the light or the avatar on screen One girl said ldquoit isdark now go quicklyrdquo and another boy asked to himselfthat ldquois it dark enoughrdquo The interesting thing was that oneboy bended his body upon the sensor to block the lightbecause he thought it would be quicker to see the outcome

8 The Scientific World Journal

(a) A girl with her maze (b) The most complicated maze

Figure 7 Two mazes collected from maze creation session a girl with her maze (a) and the complicated maze created by a boy (b)

Another remarkable domain where abstraction takes place ismaze design and creation Because the ldquowhat you see is whatyou getrdquo creation process children could then test complexabstractions quickly and precisely For example to create anewmaze the child designed amap where each square couldbe represented by certain programming blocks Then hesheselected only those appropriate for the virtual world in hishermind

(ii) Automation Automation occurs as the system executeschildrenrsquos programs The ldquoprogramrdquo itself automated ldquostep-ping throughrdquo and updated virtual avatarrsquos location anddirection (representing the veer) at each step As to this termwe did not expect too much actually it was an attempt tofind whether children could realize the automatic processAnd we indeed found some cue through the study Forexample before the execution one boy said ldquolisten to myinstructions and gordquo Another girl jumped for joy when shesaw the animation of her avatar traversing through the mazeBesides we saw 5 children stood up from their seats whenthe execution began and kept standing by waiting for thesuccessful moment

In interviews 4 children expressed their enjoyment whilewatching the virtual avatar traversing through the mazeautomatically under their instructions This was also backedup in the videotape from which we observed that 3 of themclapped out of joy and cheered on their avatarswhilewatchingthe animation

(iii) Problem Decomposition and Analysis T-Maze providesa platform that enables live programming by allowing thecode to be adjusted in real time As the session proceededwe moved beyond programs consisting of simple sequencesof actions and introduced more advanced concepts such asloops and numeric parameter values Through these activi-ties we found evidence that children could engage these con-cepts reasoning about possible outcomes of different blocksto build and test every piece of solution until they finallysolved the maze-escape problem Children also performedanalysis when they decidedwhether or not the avatar behavedas expected If the avatar was stuck or ldquomisbehavedrdquo it mayeither mean that their implementation of their control idea isfaulty or that some blocks were misplaced

Mostly the process of analysis happened implicitly butwecould still find some track from the observation For exampleone child in his complex maze-escape mission determinedthat he needed five forward blocks to reach the first turningAfter learning the loop syntax he realized that the largestvalue of a numeric block is five and the loop is limited to 5times the most so he needed two more forward blocks Thechild then built a program snippet as

(Start Loop + 5 + Forward + End Loop) 997888rarr Forward997888rarr Forward

In maze creation session we found 4 children chose tolocate the entrance and exit point of a maze at first and thentried to link them through while others just placed the blocksas they wish and were happy-go-lucky with any shape ofmazes Besides we observed a sense of ldquohabitrdquo in childrenrsquosprogramming that 5 children would add several blocks intothe sequence once when it was a simple straight line whileattempted one step at a time in a complex situation (a cornera sensor square or a cross) when they would adjust currentplacement and decide what to do next while observing thefeedbacks

(iv) Creativity To create a maze programming became amedium for childrenrsquos personal and creative expression inthe design of their mazes children engaged their fantasiesand built relationships with other pockets of reality that wentbeyond common thoughts As we analyzed childrenrsquos workcollected in the study several interesting mazes surprised usOne girl created her maze shaping like a snake (Figure 7(a))but she described it as the Great Wall coming out from hertextbook And the boywho claimed tomake themost difficultmaze used 31 blocks constructing the most complex fish-shape maze (Figure 7(b)) During the creating he put asidethe end block until he finished all the paths then counted thenumber of cells in each path trying to find the longest onewhere the end block was appended

Generally the mazes children created are different fromwhat we designed for T-Maze intrinsically and the descrip-tions children gave in interviews were also unexpected Forexample one boy created a maze with only one straight linewhile he described it as an airport runway And he talked

The Scientific World Journal 9

about his experience with his father at the airport last weekright off the reel

55 Summary Though the study is relatively small scale wecan tell from the results that T-Maze has some computationalconcepts in its programming activities and children indeedhandle those experienceswellThis study suggests farther thatT-Maze is easy to use for children and they are able to buildtheir own programs in playing and creating activities with T-Maze The emphasis is that T-Maze has the potential to helpraise childrenrsquos awareness about CT concepts

In interviews we also wanted to get some inspirationsfrom children with questions like ldquowhat other things do youthink the blocks can dordquo or ldquodo you think of any other gamesthat can be played with those blocksrdquo Two children said theywould like to control a flying plane with the blocks and 3others wished to help their mothers with some householdduties like cooking cleaning or furnishing their roomsTheir unlimited imagination inspired us in the revision andupgrade of T-Maze in the future

6 Conclusion and Future Work

T-Maze is designed for children aged 5ndash9 to help cultivatetheir computational thinking skills It allows children to playmaze-escape games and create their own mazes by buildingcomputer programs out of wooden blocks We conducted auser study to verify its ease of use and the potential capabilityto help children with computational thinking The resultsshowed that T-Maze has the promising potential to helpchildren cultivate some computational thinking awarenesslike abstraction automation problem decomposition andanalysis

In the future we will develop more games and scenariosdesign more kinds of programming blocks and enrich thetypes of sensors More and further user studies are neededto improve T-Maze for better user experience and skillsnurturing capability Besides as an upgrade we may providea collaborative tangible programming tool for children toexperience the fun of cooperation

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank the children from nearbykindergartens that took part in the study The authors grate-fully acknowledge financial support from the Major StateBasic Research Development Program of China under Grantno 2013CB328805 the National Natural Science Foundationof China under Grant no 60970090 and no 61272325 theCooperation Project of Chinese Academy of Sciences andFoshan City under Grant no 2012YS04 and the NationalScience and Technology Supported Program under Grant no2012BAH19F01

References

[1] J M Wing ldquoComputational thinkingrdquo Communications of theACM vol 49 no 3 pp 33ndash35 2006

[2] A Bundy ldquoComputational thinking is pervasiverdquo Journal ofScientific and Practical Computing vol 1 no 2 pp 67ndash69 2007

[3] S Grover ldquoComputational Thinking Programming andthe Google App Inventorrdquo 2010 httpwwwthesmartbeancomlibrarycomputational-thinking-programming-and-the-google-app-inventor

[4] S Grover ldquoExpanding the Technology Curriculum to IncludeFoundational Elements of Computer Science for K-8rdquo 2009httpwwwthesmartbeancommagazinehome-schoolingexpanding-the-technology-curriculum-foundational-ele-ments-of-computer-science-for-k-8

[5] G Orr ldquoComputational thinking through programming andalgorithmic artrdquo in Proceedings of the SIGGRAPH 2009 Talks(SIGGRAPH rsquo09) vol 1 p 1 ACM 2009

[6] M U Bers Blocks to Robots Learning with Technology in theEarly Childhood Classroom Teachers College New York NYUSA 2008

[7] D H Clements ldquoThe future of educational computing researchthe case of computer programmingrdquo Information Technology inChildhood Education Annual vol 1999 no 1 pp 147ndash179 1999

[8] T S McNerney Tangible programming bricks an approach tomaking programming accessible to everyone [MS thesis] MITCambridge Cambridge Mass USA 2000

[9] T S McNerney ldquoFrom turtles to tangible programming bricksexplorations in physical language designrdquo Personal and Ubiqui-tous Computing vol 8 no 5 pp 326ndash337 2004

[10] J Fails A Druin M Guha G Chipman S Simms and WChuraman ldquoChildrsquos play a comparison of desktop and physicalinteractive environmentsrdquo in Proceedings of the InteractionDesign and Children (IDC rsquo05) pp 48ndash55 ACM 2005

[11] M S Horn E T Solovey R J Crouser and R J K JacobldquoComparing the use of tangible and graphical programminglanguages for informal science educationrdquo in Proceedings of theSIGCHI Conference on Human Factors in Computing Systems(CHI rsquo09) pp 975ndash984 ACM 2009

[12] S PapertMindstorms Children Computers and Powerful IdeasBasic Books New York NY USA 1981

[13] G H Fletcher and J J Lu ldquoEducation human computing skillsrethinking the K-12 experiencerdquo Communications of the ACMvol 52 no 2 pp 23ndash25 2009

[14] P Curzon J Peckham A Settle E Roberts and H TaylorldquoComputational thinking(CT) on weaving it inrdquo in Proceedingsof the 14th Annual ACM SIGCSE Conference on Innovation andTechnology in Computer Science Education (ITiCSE rsquo09) pp201ndash202 ACM 2009

[15] P B Henderson ldquoUbiquitous computational thinkingrdquo Com-puter vol 42 no 10 pp 100ndash102 2009

[16] J J Lu and G H L Fletcher ldquoThinking about computationalthinkingrdquo in Proceedings of the 40th ACM Technical Symposiumon Computer Science Education (SIGCSE rsquo09) pp 260ndash264ACM 2009

[17] D Moursund ldquoComputational Thinkingrdquo IAE-Pedia 2009httpiae-pediaorgComputational Thinking

[18] National Research Council ldquoCommittee for the workshops oncomputational thinking 2010rdquo Report of a Workshop on theScope and Nature of Computational Thinking The NationalAcademies Press Washington DC USA 2010

10 The Scientific World Journal

[19] IWG Computational Thinking for Youth Education Develop-ment Center Newton Mass USA 2010

[20] M Resnick ldquoAll I really need to know (about creative thinking)I learned (by studying how children learn) in kindergartenrdquo inProceedings of the 6th Conference on Creativity amp Cognition (CCrsquo07) pp 1ndash6 ACM 2007

[21] J Dougherty ldquoConcept visualization in CS0 using ALICErdquoJournal of Computing Science in Colleges vol 22 no 3 pp 145ndash152 2007

[22] I Lee F Martin J Denner et al ldquoComputational thinking foryouth in practicerdquo ACM Inroads vol 2 no 1 pp 32ndash37 2011

[23] A Ruthmann J M Heines G R Greher P Laidler and CSaulters II ldquoTeaching computational thinking through musicallive coding in Scratchrdquo in Proceedings of the 41st ACM TechnicalSymposium on Computer Science Education (SIGCSE rsquo10) pp351ndash355 ACM 2010

[24] C Kelleher and R Pausch ldquoLowering the barriers to pro-gramming a taxonomy of programming environments andlanguages for novice programmersrdquo ACM Computing Surveysvol 37 no 2 pp 83ndash137 2005

[25] C Rader C Brand and C Lewis ldquoDegrees of comprehensionchildrenrsquos understanding of a visual programming environ-mentrdquo in Proceedings of the Conference on Human Factors inComputing Systems (CHI rsquo97) pp 351ndash358 ACM 1997

[26] PWyeth and H C Purchase ldquoTangible programming elementsfor young childrenrdquo in Proceedings of the Extended Abstracts onHuman Factors in Computing Systems (CHI rsquo02) pp 774ndash775ACM 2002

[27] M S Horn and R J K Jacob ldquoTangible programming in theclassroomwith ternrdquo inProceedings of the ExtendedAbstracts onHuman Factors in Computing Systems (CHI Trends Interactivity)(CHI rsquo07) pp 1965ndash1970 ACM 2007

[28] S Tarkan V Sazawal A Druin et al ldquoToque designing acooking-based programming language for and with childrenrdquoin Proceedings of the 28th Annual Conference on Human Factorsin Computing Systems (CHI rsquo10) pp 2417ndash2426 ACM 2010

[29] D Gallardo C F Julia and S Jorda ldquoTurTan a tangibleprogramming language for creative explorationrdquo in Proceedingsof the 3rd IEEE InternationalWorkshop onHorizontal InteractiveHuman Computer System (TABLETOP rsquo08) pp 89ndash92 IEEEComputer Society Press 2008

[30] J Silver andE Rosenbaum ldquoTwinkle programmingwith colorrdquoin Proceedings of the 4th International Conference on TangibleEmbedded and Embodied Interaction (TEI rsquo10) pp 383ndash384ACM 2010

[31] D L Wang C Zhang and H A Wang ldquoT-Maze a tangibleprogramming tool for childrenrdquo in Proceedings of the 10thInternational Conference on Interaction Design and Children(IDC rsquo11) pp 127ndash135 ACM 2011

[32] M S Horn ldquoTopCode Tangible Object Placement Codesrdquo2012 httpuserseecsnorthwesternedusimmhorntopcodes

[33] K Howland ldquoSupporting the development of multimodalwriting and computational thinking skills through computergame creationrdquo in Proceedings of the IEEE Symposium on VisualLanguages and Human-Centric Computing (VLHCC rsquo09) pp250ndash251 IEEE Computer Society Press 2009

[34] J LrsquoHeureux D Boisvert K Sanghera and R Cohen ldquoITproblem solving an implementation of computational thinkingin information technologyrdquo in Proceedings of the 13th AnnualConference on Information Technology Education (SIGITE rsquo12)pp 183ndash188 ACM 2012

[35] M Guzdial ldquoEducation paving the way for computationalthinkingrdquo Communications of the ACM vol 51 no 8 pp 25ndash27 2008

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

International Journal of

ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia

International Journal of

Biomedical Imaging

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArtificialNeural Systems

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational Intelligence and Neuroscience

Industrial EngineeringJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

6 The Scientific World Journal

(a) (b)

Figure 6 The simple (a) and complex (b) maze-escape games in the study

Though the first two questions have been examined inprevious studies we wanted to be sure that children arecomfortable to involve in our study every time Consideringthe diversity of users and usage scenarios we regard theusability of T-Maze as a long-term test subject especiallywhen some new modifications or improvements need to beverified Also we expect usable feedbacks from children ifany to help us improve the tool since it is always preferableto have more children like T-Maze

512 Tasks We conducted two sessions in the study mazeescape and maze creation In maze escape session childrenwere asked to complete two different levels of maze-escapegames which were designed in consideration of individualability difference as well as the occasions where problemdecomposition and analysis probably happen To play amaze-escape game children had to find a passable path in themaze for the avatar to traverse through by building programswith wooden programming blocks The simple maze-escapegame only had one passable path and would need 9 blocks atmost (without the use of Loop blocks) while the complex onehadmultiple passable paths and would need variable numberof blocks according to the length of chosen path (Figure 6)For a maze creation task children were given no restrictionsto create anymaze they liked by constructing newmazemapswith the same set of blocks There was no time limitation inboth sessions Throughout the study we had a comfortableseat in front of the table for children to sit when they wereworking on the tasks as well as a rest area for them to wait orrest

513 Interviews and Questions For the interviews after thetwo main sessions we wanted to know childrenrsquos opinionsabout the tool their programming experience or anythingthat remained in their mind So the questions asked in theinterviews were mostly without model answers as followsldquoWhat other things do you think the blocks can dordquo ldquoDo youhave any other games that can be played with those blocksrdquoldquoHave you ever noticed any sensors or something like that inyour liferdquo The questions are listed in a questionnaire whichalso contains another two questions about the usability of T-Maze and one about the invitation for our next study recruit

Reasonably we did not expect children to becomecomputational thinkers after the one-off practices but wehoped to find some clues that might answer our questionsand preferably got some implications for teaching childrencomputational thinking

52 Participants and Apparatus Seven children (aged 8 inaverage) 5 girls and 2 boys participated in this studyThey allhad some experience in computer use but none of them hadever known about programming or used any programmingtools

The apparatus was a set of the latest version of T-Mazeplaced on an only table in a laboratory room The table witha chair was in the center of the room and we also made aresting area at the right corner beside the door One of thetwo cameras was set up in front left of the table while theother one was behind the table to the left Each time onechild went to the table to complete the tasks with hisher backtowards others waiting in the resting area so that they wouldnot interrupt the experiment

53 Procedure The study was conducted in four stagesdemonstration practice main sessions and interviewsBefore we started all children watched a demonstrationabout how to play a simple maze-escape game and create anew maze We introduced how to use the loop blocks andencouraged children to find the shortest path as they canThen they were given some minutes to practice freely Afterthey were ready to start the main sessions of the study began

Without time limitation children could involve in theirtasks as much time as they like but they were asked to give asign when they were about to begin or to finish and we notedthe time for records In maze-escape session children wereexplained to complete a simple escape task first and a complexone afterward Next in maze creation session children weretold to freely create any maze as they liked When theyfinished the creation we encouraged them to introduce theirmazes to others hoping to invite some creative talking Afterthemain sessions we conducted one-on-one semi-structuredinterviewswith children andwe taped all the conversation forlater analysis

The Scientific World Journal 7

Table 1 The time (unit minute) and blocks children used in the study

NumberMaze escape session Maze creation session

Simple ComplexTime (min) Blocks Time (min) Blocks Time (min) Blocks

1 350 900 700 1800 900 18002 300 900 500 1900 1000 26003 450 900 700 2100 1300 18004 450 900 900 2200 800 16005 300 900 800 2000 1100 31006 400 800 900 2100 1000 26007 400 900 800 1800 900 2500Avg 378 886 757 1986 986 2286

We collected qualitative data in the form of observationnotes photographs and videotape from which some quanti-tative data was also derivedWe also collected childrenrsquos work(the mazes they created) Three researchers were involved inthis study two collected data while the third acted in the roleof lead interpreter

54 Results and Discussion We comprehensively analyzedthe observation notes tapes and interview records andfound answers to the research questions from the analysisof childrenrsquos words and behaviors as well as the statisticalanalysis

For the first question (easy to use) we were interested inwhether children could create their own computer programsBased on observation notes and an analysis of videotapewe found that children were able to manipulate the tangibleblocks to form their own programs We allowed childrento take as much time as they wanted to complete a taskHowever we were lucky to see that they neither seemed totake their time nor to rush through the experiment whichgave us the reasonable and valid data in Table 1 Childrenwere able to accomplish both the simple and complex maze-escape tasks within similar period of time Also they wereeven more engaged in maze creation session and spent moretime creating and exploring All children finished the twosessions with their freely-created mazes some of which hadinteresting shapes or complex structure with sensor blocksused

In interviews when asked whether the tool was easy touse 2 children chose ldquovery easyrdquo 4 chose ldquoeasyrdquo and one choseldquonormalrdquo

For the second research question (interesting to playwith) according to the records and videotape childrenalways showed their interests in sensors 3 children askedquestions about the sensors like ldquowhat are these used forrdquoldquowhat if I pressed it longerrdquo ldquoonly three of themrdquo Moreoverchildren were very excited in maze creation section andshowed the richest behaviors For example one boy said ldquoIrsquomgoing to make the most difficult maze that none of themcan get throughrdquo and another boy asked ldquowhat if the blocksare not enough to create my mazerdquo As part of the mazecreation session we asked children not only to demonstratetheir mazes but also to show others the blocks they used to

program it In many cases children gathered together to thetable just stood by watching and praised for others In othercases however children asked for a competition to provetheir own

In interviews when asked whether the system was funall children confirmed their enjoyment with T-Maze andexpressed thewish to participate again in the future 4 of themvoted for the maze creation session to be their favorite

For the last question (CT concepts in programming)we found some cue that T-Maze has the potential to raisechildrenrsquos awareness about CT and help them understand CTto some extent

(i) AbstractionThemazemetaphor used inT-Maze promptedchildren to relate their experience about real-world maze tothe virtual maze on the screen They needed the process ofabstraction to narrow the escape problem down to some-thing that could be implemented on the computer using T-Maze as well as mapping the physical blocks to the virtualsquares in maze map on screen Restrictions imposed by theprogramming environment include an upper bound on thenumber of blocks and a limit on the size of the programmingarea (40 cm width by 30 cm height) Abstraction also tookplace as children designed avatars to react to a limited set ofconditions that may be encountered in the real-world mazesuch as some roadblocks represented by sensors

The fact that children all accomplished the tasks success-fully was inspiring and we indeed found something relatedfrom the observation In maze-escape session when theavatar came across a temperature sensor square in the mazewe found 3 children attempting to heat up the physicalsensors with their breath Later in the interviews we askedwhy did they do that and one said ldquoEr my mum alwayswarm my hand with her breath in winterrdquo Other two saidthat they always knew their breath was warmer than theirhands The same situation happened with light sensor where4 children glanced at the gap between their hands and thelight sensor they were covering and muttered somethingabout the light or the avatar on screen One girl said ldquoit isdark now go quicklyrdquo and another boy asked to himselfthat ldquois it dark enoughrdquo The interesting thing was that oneboy bended his body upon the sensor to block the lightbecause he thought it would be quicker to see the outcome

8 The Scientific World Journal

(a) A girl with her maze (b) The most complicated maze

Figure 7 Two mazes collected from maze creation session a girl with her maze (a) and the complicated maze created by a boy (b)

Another remarkable domain where abstraction takes place ismaze design and creation Because the ldquowhat you see is whatyou getrdquo creation process children could then test complexabstractions quickly and precisely For example to create anewmaze the child designed amap where each square couldbe represented by certain programming blocks Then hesheselected only those appropriate for the virtual world in hishermind

(ii) Automation Automation occurs as the system executeschildrenrsquos programs The ldquoprogramrdquo itself automated ldquostep-ping throughrdquo and updated virtual avatarrsquos location anddirection (representing the veer) at each step As to this termwe did not expect too much actually it was an attempt tofind whether children could realize the automatic processAnd we indeed found some cue through the study Forexample before the execution one boy said ldquolisten to myinstructions and gordquo Another girl jumped for joy when shesaw the animation of her avatar traversing through the mazeBesides we saw 5 children stood up from their seats whenthe execution began and kept standing by waiting for thesuccessful moment

In interviews 4 children expressed their enjoyment whilewatching the virtual avatar traversing through the mazeautomatically under their instructions This was also backedup in the videotape from which we observed that 3 of themclapped out of joy and cheered on their avatarswhilewatchingthe animation

(iii) Problem Decomposition and Analysis T-Maze providesa platform that enables live programming by allowing thecode to be adjusted in real time As the session proceededwe moved beyond programs consisting of simple sequencesof actions and introduced more advanced concepts such asloops and numeric parameter values Through these activi-ties we found evidence that children could engage these con-cepts reasoning about possible outcomes of different blocksto build and test every piece of solution until they finallysolved the maze-escape problem Children also performedanalysis when they decidedwhether or not the avatar behavedas expected If the avatar was stuck or ldquomisbehavedrdquo it mayeither mean that their implementation of their control idea isfaulty or that some blocks were misplaced

Mostly the process of analysis happened implicitly butwecould still find some track from the observation For exampleone child in his complex maze-escape mission determinedthat he needed five forward blocks to reach the first turningAfter learning the loop syntax he realized that the largestvalue of a numeric block is five and the loop is limited to 5times the most so he needed two more forward blocks Thechild then built a program snippet as

(Start Loop + 5 + Forward + End Loop) 997888rarr Forward997888rarr Forward

In maze creation session we found 4 children chose tolocate the entrance and exit point of a maze at first and thentried to link them through while others just placed the blocksas they wish and were happy-go-lucky with any shape ofmazes Besides we observed a sense of ldquohabitrdquo in childrenrsquosprogramming that 5 children would add several blocks intothe sequence once when it was a simple straight line whileattempted one step at a time in a complex situation (a cornera sensor square or a cross) when they would adjust currentplacement and decide what to do next while observing thefeedbacks

(iv) Creativity To create a maze programming became amedium for childrenrsquos personal and creative expression inthe design of their mazes children engaged their fantasiesand built relationships with other pockets of reality that wentbeyond common thoughts As we analyzed childrenrsquos workcollected in the study several interesting mazes surprised usOne girl created her maze shaping like a snake (Figure 7(a))but she described it as the Great Wall coming out from hertextbook And the boywho claimed tomake themost difficultmaze used 31 blocks constructing the most complex fish-shape maze (Figure 7(b)) During the creating he put asidethe end block until he finished all the paths then counted thenumber of cells in each path trying to find the longest onewhere the end block was appended

Generally the mazes children created are different fromwhat we designed for T-Maze intrinsically and the descrip-tions children gave in interviews were also unexpected Forexample one boy created a maze with only one straight linewhile he described it as an airport runway And he talked

The Scientific World Journal 9

about his experience with his father at the airport last weekright off the reel

55 Summary Though the study is relatively small scale wecan tell from the results that T-Maze has some computationalconcepts in its programming activities and children indeedhandle those experienceswellThis study suggests farther thatT-Maze is easy to use for children and they are able to buildtheir own programs in playing and creating activities with T-Maze The emphasis is that T-Maze has the potential to helpraise childrenrsquos awareness about CT concepts

In interviews we also wanted to get some inspirationsfrom children with questions like ldquowhat other things do youthink the blocks can dordquo or ldquodo you think of any other gamesthat can be played with those blocksrdquo Two children said theywould like to control a flying plane with the blocks and 3others wished to help their mothers with some householdduties like cooking cleaning or furnishing their roomsTheir unlimited imagination inspired us in the revision andupgrade of T-Maze in the future

6 Conclusion and Future Work

T-Maze is designed for children aged 5ndash9 to help cultivatetheir computational thinking skills It allows children to playmaze-escape games and create their own mazes by buildingcomputer programs out of wooden blocks We conducted auser study to verify its ease of use and the potential capabilityto help children with computational thinking The resultsshowed that T-Maze has the promising potential to helpchildren cultivate some computational thinking awarenesslike abstraction automation problem decomposition andanalysis

In the future we will develop more games and scenariosdesign more kinds of programming blocks and enrich thetypes of sensors More and further user studies are neededto improve T-Maze for better user experience and skillsnurturing capability Besides as an upgrade we may providea collaborative tangible programming tool for children toexperience the fun of cooperation

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank the children from nearbykindergartens that took part in the study The authors grate-fully acknowledge financial support from the Major StateBasic Research Development Program of China under Grantno 2013CB328805 the National Natural Science Foundationof China under Grant no 60970090 and no 61272325 theCooperation Project of Chinese Academy of Sciences andFoshan City under Grant no 2012YS04 and the NationalScience and Technology Supported Program under Grant no2012BAH19F01

References

[1] J M Wing ldquoComputational thinkingrdquo Communications of theACM vol 49 no 3 pp 33ndash35 2006

[2] A Bundy ldquoComputational thinking is pervasiverdquo Journal ofScientific and Practical Computing vol 1 no 2 pp 67ndash69 2007

[3] S Grover ldquoComputational Thinking Programming andthe Google App Inventorrdquo 2010 httpwwwthesmartbeancomlibrarycomputational-thinking-programming-and-the-google-app-inventor

[4] S Grover ldquoExpanding the Technology Curriculum to IncludeFoundational Elements of Computer Science for K-8rdquo 2009httpwwwthesmartbeancommagazinehome-schoolingexpanding-the-technology-curriculum-foundational-ele-ments-of-computer-science-for-k-8

[5] G Orr ldquoComputational thinking through programming andalgorithmic artrdquo in Proceedings of the SIGGRAPH 2009 Talks(SIGGRAPH rsquo09) vol 1 p 1 ACM 2009

[6] M U Bers Blocks to Robots Learning with Technology in theEarly Childhood Classroom Teachers College New York NYUSA 2008

[7] D H Clements ldquoThe future of educational computing researchthe case of computer programmingrdquo Information Technology inChildhood Education Annual vol 1999 no 1 pp 147ndash179 1999

[8] T S McNerney Tangible programming bricks an approach tomaking programming accessible to everyone [MS thesis] MITCambridge Cambridge Mass USA 2000

[9] T S McNerney ldquoFrom turtles to tangible programming bricksexplorations in physical language designrdquo Personal and Ubiqui-tous Computing vol 8 no 5 pp 326ndash337 2004

[10] J Fails A Druin M Guha G Chipman S Simms and WChuraman ldquoChildrsquos play a comparison of desktop and physicalinteractive environmentsrdquo in Proceedings of the InteractionDesign and Children (IDC rsquo05) pp 48ndash55 ACM 2005

[11] M S Horn E T Solovey R J Crouser and R J K JacobldquoComparing the use of tangible and graphical programminglanguages for informal science educationrdquo in Proceedings of theSIGCHI Conference on Human Factors in Computing Systems(CHI rsquo09) pp 975ndash984 ACM 2009

[12] S PapertMindstorms Children Computers and Powerful IdeasBasic Books New York NY USA 1981

[13] G H Fletcher and J J Lu ldquoEducation human computing skillsrethinking the K-12 experiencerdquo Communications of the ACMvol 52 no 2 pp 23ndash25 2009

[14] P Curzon J Peckham A Settle E Roberts and H TaylorldquoComputational thinking(CT) on weaving it inrdquo in Proceedingsof the 14th Annual ACM SIGCSE Conference on Innovation andTechnology in Computer Science Education (ITiCSE rsquo09) pp201ndash202 ACM 2009

[15] P B Henderson ldquoUbiquitous computational thinkingrdquo Com-puter vol 42 no 10 pp 100ndash102 2009

[16] J J Lu and G H L Fletcher ldquoThinking about computationalthinkingrdquo in Proceedings of the 40th ACM Technical Symposiumon Computer Science Education (SIGCSE rsquo09) pp 260ndash264ACM 2009

[17] D Moursund ldquoComputational Thinkingrdquo IAE-Pedia 2009httpiae-pediaorgComputational Thinking

[18] National Research Council ldquoCommittee for the workshops oncomputational thinking 2010rdquo Report of a Workshop on theScope and Nature of Computational Thinking The NationalAcademies Press Washington DC USA 2010

10 The Scientific World Journal

[19] IWG Computational Thinking for Youth Education Develop-ment Center Newton Mass USA 2010

[20] M Resnick ldquoAll I really need to know (about creative thinking)I learned (by studying how children learn) in kindergartenrdquo inProceedings of the 6th Conference on Creativity amp Cognition (CCrsquo07) pp 1ndash6 ACM 2007

[21] J Dougherty ldquoConcept visualization in CS0 using ALICErdquoJournal of Computing Science in Colleges vol 22 no 3 pp 145ndash152 2007

[22] I Lee F Martin J Denner et al ldquoComputational thinking foryouth in practicerdquo ACM Inroads vol 2 no 1 pp 32ndash37 2011

[23] A Ruthmann J M Heines G R Greher P Laidler and CSaulters II ldquoTeaching computational thinking through musicallive coding in Scratchrdquo in Proceedings of the 41st ACM TechnicalSymposium on Computer Science Education (SIGCSE rsquo10) pp351ndash355 ACM 2010

[24] C Kelleher and R Pausch ldquoLowering the barriers to pro-gramming a taxonomy of programming environments andlanguages for novice programmersrdquo ACM Computing Surveysvol 37 no 2 pp 83ndash137 2005

[25] C Rader C Brand and C Lewis ldquoDegrees of comprehensionchildrenrsquos understanding of a visual programming environ-mentrdquo in Proceedings of the Conference on Human Factors inComputing Systems (CHI rsquo97) pp 351ndash358 ACM 1997

[26] PWyeth and H C Purchase ldquoTangible programming elementsfor young childrenrdquo in Proceedings of the Extended Abstracts onHuman Factors in Computing Systems (CHI rsquo02) pp 774ndash775ACM 2002

[27] M S Horn and R J K Jacob ldquoTangible programming in theclassroomwith ternrdquo inProceedings of the ExtendedAbstracts onHuman Factors in Computing Systems (CHI Trends Interactivity)(CHI rsquo07) pp 1965ndash1970 ACM 2007

[28] S Tarkan V Sazawal A Druin et al ldquoToque designing acooking-based programming language for and with childrenrdquoin Proceedings of the 28th Annual Conference on Human Factorsin Computing Systems (CHI rsquo10) pp 2417ndash2426 ACM 2010

[29] D Gallardo C F Julia and S Jorda ldquoTurTan a tangibleprogramming language for creative explorationrdquo in Proceedingsof the 3rd IEEE InternationalWorkshop onHorizontal InteractiveHuman Computer System (TABLETOP rsquo08) pp 89ndash92 IEEEComputer Society Press 2008

[30] J Silver andE Rosenbaum ldquoTwinkle programmingwith colorrdquoin Proceedings of the 4th International Conference on TangibleEmbedded and Embodied Interaction (TEI rsquo10) pp 383ndash384ACM 2010

[31] D L Wang C Zhang and H A Wang ldquoT-Maze a tangibleprogramming tool for childrenrdquo in Proceedings of the 10thInternational Conference on Interaction Design and Children(IDC rsquo11) pp 127ndash135 ACM 2011

[32] M S Horn ldquoTopCode Tangible Object Placement Codesrdquo2012 httpuserseecsnorthwesternedusimmhorntopcodes

[33] K Howland ldquoSupporting the development of multimodalwriting and computational thinking skills through computergame creationrdquo in Proceedings of the IEEE Symposium on VisualLanguages and Human-Centric Computing (VLHCC rsquo09) pp250ndash251 IEEE Computer Society Press 2009

[34] J LrsquoHeureux D Boisvert K Sanghera and R Cohen ldquoITproblem solving an implementation of computational thinkingin information technologyrdquo in Proceedings of the 13th AnnualConference on Information Technology Education (SIGITE rsquo12)pp 183ndash188 ACM 2012

[35] M Guzdial ldquoEducation paving the way for computationalthinkingrdquo Communications of the ACM vol 51 no 8 pp 25ndash27 2008

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

International Journal of

ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia

International Journal of

Biomedical Imaging

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArtificialNeural Systems

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational Intelligence and Neuroscience

Industrial EngineeringJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World Journal 7

Table 1 The time (unit minute) and blocks children used in the study

NumberMaze escape session Maze creation session

Simple ComplexTime (min) Blocks Time (min) Blocks Time (min) Blocks

1 350 900 700 1800 900 18002 300 900 500 1900 1000 26003 450 900 700 2100 1300 18004 450 900 900 2200 800 16005 300 900 800 2000 1100 31006 400 800 900 2100 1000 26007 400 900 800 1800 900 2500Avg 378 886 757 1986 986 2286

We collected qualitative data in the form of observationnotes photographs and videotape from which some quanti-tative data was also derivedWe also collected childrenrsquos work(the mazes they created) Three researchers were involved inthis study two collected data while the third acted in the roleof lead interpreter

54 Results and Discussion We comprehensively analyzedthe observation notes tapes and interview records andfound answers to the research questions from the analysisof childrenrsquos words and behaviors as well as the statisticalanalysis

For the first question (easy to use) we were interested inwhether children could create their own computer programsBased on observation notes and an analysis of videotapewe found that children were able to manipulate the tangibleblocks to form their own programs We allowed childrento take as much time as they wanted to complete a taskHowever we were lucky to see that they neither seemed totake their time nor to rush through the experiment whichgave us the reasonable and valid data in Table 1 Childrenwere able to accomplish both the simple and complex maze-escape tasks within similar period of time Also they wereeven more engaged in maze creation session and spent moretime creating and exploring All children finished the twosessions with their freely-created mazes some of which hadinteresting shapes or complex structure with sensor blocksused

In interviews when asked whether the tool was easy touse 2 children chose ldquovery easyrdquo 4 chose ldquoeasyrdquo and one choseldquonormalrdquo

For the second research question (interesting to playwith) according to the records and videotape childrenalways showed their interests in sensors 3 children askedquestions about the sensors like ldquowhat are these used forrdquoldquowhat if I pressed it longerrdquo ldquoonly three of themrdquo Moreoverchildren were very excited in maze creation section andshowed the richest behaviors For example one boy said ldquoIrsquomgoing to make the most difficult maze that none of themcan get throughrdquo and another boy asked ldquowhat if the blocksare not enough to create my mazerdquo As part of the mazecreation session we asked children not only to demonstratetheir mazes but also to show others the blocks they used to

program it In many cases children gathered together to thetable just stood by watching and praised for others In othercases however children asked for a competition to provetheir own

In interviews when asked whether the system was funall children confirmed their enjoyment with T-Maze andexpressed thewish to participate again in the future 4 of themvoted for the maze creation session to be their favorite

For the last question (CT concepts in programming)we found some cue that T-Maze has the potential to raisechildrenrsquos awareness about CT and help them understand CTto some extent

(i) AbstractionThemazemetaphor used inT-Maze promptedchildren to relate their experience about real-world maze tothe virtual maze on the screen They needed the process ofabstraction to narrow the escape problem down to some-thing that could be implemented on the computer using T-Maze as well as mapping the physical blocks to the virtualsquares in maze map on screen Restrictions imposed by theprogramming environment include an upper bound on thenumber of blocks and a limit on the size of the programmingarea (40 cm width by 30 cm height) Abstraction also tookplace as children designed avatars to react to a limited set ofconditions that may be encountered in the real-world mazesuch as some roadblocks represented by sensors

The fact that children all accomplished the tasks success-fully was inspiring and we indeed found something relatedfrom the observation In maze-escape session when theavatar came across a temperature sensor square in the mazewe found 3 children attempting to heat up the physicalsensors with their breath Later in the interviews we askedwhy did they do that and one said ldquoEr my mum alwayswarm my hand with her breath in winterrdquo Other two saidthat they always knew their breath was warmer than theirhands The same situation happened with light sensor where4 children glanced at the gap between their hands and thelight sensor they were covering and muttered somethingabout the light or the avatar on screen One girl said ldquoit isdark now go quicklyrdquo and another boy asked to himselfthat ldquois it dark enoughrdquo The interesting thing was that oneboy bended his body upon the sensor to block the lightbecause he thought it would be quicker to see the outcome

8 The Scientific World Journal

(a) A girl with her maze (b) The most complicated maze

Figure 7 Two mazes collected from maze creation session a girl with her maze (a) and the complicated maze created by a boy (b)

Another remarkable domain where abstraction takes place ismaze design and creation Because the ldquowhat you see is whatyou getrdquo creation process children could then test complexabstractions quickly and precisely For example to create anewmaze the child designed amap where each square couldbe represented by certain programming blocks Then hesheselected only those appropriate for the virtual world in hishermind

(ii) Automation Automation occurs as the system executeschildrenrsquos programs The ldquoprogramrdquo itself automated ldquostep-ping throughrdquo and updated virtual avatarrsquos location anddirection (representing the veer) at each step As to this termwe did not expect too much actually it was an attempt tofind whether children could realize the automatic processAnd we indeed found some cue through the study Forexample before the execution one boy said ldquolisten to myinstructions and gordquo Another girl jumped for joy when shesaw the animation of her avatar traversing through the mazeBesides we saw 5 children stood up from their seats whenthe execution began and kept standing by waiting for thesuccessful moment

In interviews 4 children expressed their enjoyment whilewatching the virtual avatar traversing through the mazeautomatically under their instructions This was also backedup in the videotape from which we observed that 3 of themclapped out of joy and cheered on their avatarswhilewatchingthe animation

(iii) Problem Decomposition and Analysis T-Maze providesa platform that enables live programming by allowing thecode to be adjusted in real time As the session proceededwe moved beyond programs consisting of simple sequencesof actions and introduced more advanced concepts such asloops and numeric parameter values Through these activi-ties we found evidence that children could engage these con-cepts reasoning about possible outcomes of different blocksto build and test every piece of solution until they finallysolved the maze-escape problem Children also performedanalysis when they decidedwhether or not the avatar behavedas expected If the avatar was stuck or ldquomisbehavedrdquo it mayeither mean that their implementation of their control idea isfaulty or that some blocks were misplaced

Mostly the process of analysis happened implicitly butwecould still find some track from the observation For exampleone child in his complex maze-escape mission determinedthat he needed five forward blocks to reach the first turningAfter learning the loop syntax he realized that the largestvalue of a numeric block is five and the loop is limited to 5times the most so he needed two more forward blocks Thechild then built a program snippet as

(Start Loop + 5 + Forward + End Loop) 997888rarr Forward997888rarr Forward

In maze creation session we found 4 children chose tolocate the entrance and exit point of a maze at first and thentried to link them through while others just placed the blocksas they wish and were happy-go-lucky with any shape ofmazes Besides we observed a sense of ldquohabitrdquo in childrenrsquosprogramming that 5 children would add several blocks intothe sequence once when it was a simple straight line whileattempted one step at a time in a complex situation (a cornera sensor square or a cross) when they would adjust currentplacement and decide what to do next while observing thefeedbacks

(iv) Creativity To create a maze programming became amedium for childrenrsquos personal and creative expression inthe design of their mazes children engaged their fantasiesand built relationships with other pockets of reality that wentbeyond common thoughts As we analyzed childrenrsquos workcollected in the study several interesting mazes surprised usOne girl created her maze shaping like a snake (Figure 7(a))but she described it as the Great Wall coming out from hertextbook And the boywho claimed tomake themost difficultmaze used 31 blocks constructing the most complex fish-shape maze (Figure 7(b)) During the creating he put asidethe end block until he finished all the paths then counted thenumber of cells in each path trying to find the longest onewhere the end block was appended

Generally the mazes children created are different fromwhat we designed for T-Maze intrinsically and the descrip-tions children gave in interviews were also unexpected Forexample one boy created a maze with only one straight linewhile he described it as an airport runway And he talked

The Scientific World Journal 9

about his experience with his father at the airport last weekright off the reel

55 Summary Though the study is relatively small scale wecan tell from the results that T-Maze has some computationalconcepts in its programming activities and children indeedhandle those experienceswellThis study suggests farther thatT-Maze is easy to use for children and they are able to buildtheir own programs in playing and creating activities with T-Maze The emphasis is that T-Maze has the potential to helpraise childrenrsquos awareness about CT concepts

In interviews we also wanted to get some inspirationsfrom children with questions like ldquowhat other things do youthink the blocks can dordquo or ldquodo you think of any other gamesthat can be played with those blocksrdquo Two children said theywould like to control a flying plane with the blocks and 3others wished to help their mothers with some householdduties like cooking cleaning or furnishing their roomsTheir unlimited imagination inspired us in the revision andupgrade of T-Maze in the future

6 Conclusion and Future Work

T-Maze is designed for children aged 5ndash9 to help cultivatetheir computational thinking skills It allows children to playmaze-escape games and create their own mazes by buildingcomputer programs out of wooden blocks We conducted auser study to verify its ease of use and the potential capabilityto help children with computational thinking The resultsshowed that T-Maze has the promising potential to helpchildren cultivate some computational thinking awarenesslike abstraction automation problem decomposition andanalysis

In the future we will develop more games and scenariosdesign more kinds of programming blocks and enrich thetypes of sensors More and further user studies are neededto improve T-Maze for better user experience and skillsnurturing capability Besides as an upgrade we may providea collaborative tangible programming tool for children toexperience the fun of cooperation

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank the children from nearbykindergartens that took part in the study The authors grate-fully acknowledge financial support from the Major StateBasic Research Development Program of China under Grantno 2013CB328805 the National Natural Science Foundationof China under Grant no 60970090 and no 61272325 theCooperation Project of Chinese Academy of Sciences andFoshan City under Grant no 2012YS04 and the NationalScience and Technology Supported Program under Grant no2012BAH19F01

References

[1] J M Wing ldquoComputational thinkingrdquo Communications of theACM vol 49 no 3 pp 33ndash35 2006

[2] A Bundy ldquoComputational thinking is pervasiverdquo Journal ofScientific and Practical Computing vol 1 no 2 pp 67ndash69 2007

[3] S Grover ldquoComputational Thinking Programming andthe Google App Inventorrdquo 2010 httpwwwthesmartbeancomlibrarycomputational-thinking-programming-and-the-google-app-inventor

[4] S Grover ldquoExpanding the Technology Curriculum to IncludeFoundational Elements of Computer Science for K-8rdquo 2009httpwwwthesmartbeancommagazinehome-schoolingexpanding-the-technology-curriculum-foundational-ele-ments-of-computer-science-for-k-8

[5] G Orr ldquoComputational thinking through programming andalgorithmic artrdquo in Proceedings of the SIGGRAPH 2009 Talks(SIGGRAPH rsquo09) vol 1 p 1 ACM 2009

[6] M U Bers Blocks to Robots Learning with Technology in theEarly Childhood Classroom Teachers College New York NYUSA 2008

[7] D H Clements ldquoThe future of educational computing researchthe case of computer programmingrdquo Information Technology inChildhood Education Annual vol 1999 no 1 pp 147ndash179 1999

[8] T S McNerney Tangible programming bricks an approach tomaking programming accessible to everyone [MS thesis] MITCambridge Cambridge Mass USA 2000

[9] T S McNerney ldquoFrom turtles to tangible programming bricksexplorations in physical language designrdquo Personal and Ubiqui-tous Computing vol 8 no 5 pp 326ndash337 2004

[10] J Fails A Druin M Guha G Chipman S Simms and WChuraman ldquoChildrsquos play a comparison of desktop and physicalinteractive environmentsrdquo in Proceedings of the InteractionDesign and Children (IDC rsquo05) pp 48ndash55 ACM 2005

[11] M S Horn E T Solovey R J Crouser and R J K JacobldquoComparing the use of tangible and graphical programminglanguages for informal science educationrdquo in Proceedings of theSIGCHI Conference on Human Factors in Computing Systems(CHI rsquo09) pp 975ndash984 ACM 2009

[12] S PapertMindstorms Children Computers and Powerful IdeasBasic Books New York NY USA 1981

[13] G H Fletcher and J J Lu ldquoEducation human computing skillsrethinking the K-12 experiencerdquo Communications of the ACMvol 52 no 2 pp 23ndash25 2009

[14] P Curzon J Peckham A Settle E Roberts and H TaylorldquoComputational thinking(CT) on weaving it inrdquo in Proceedingsof the 14th Annual ACM SIGCSE Conference on Innovation andTechnology in Computer Science Education (ITiCSE rsquo09) pp201ndash202 ACM 2009

[15] P B Henderson ldquoUbiquitous computational thinkingrdquo Com-puter vol 42 no 10 pp 100ndash102 2009

[16] J J Lu and G H L Fletcher ldquoThinking about computationalthinkingrdquo in Proceedings of the 40th ACM Technical Symposiumon Computer Science Education (SIGCSE rsquo09) pp 260ndash264ACM 2009

[17] D Moursund ldquoComputational Thinkingrdquo IAE-Pedia 2009httpiae-pediaorgComputational Thinking

[18] National Research Council ldquoCommittee for the workshops oncomputational thinking 2010rdquo Report of a Workshop on theScope and Nature of Computational Thinking The NationalAcademies Press Washington DC USA 2010

10 The Scientific World Journal

[19] IWG Computational Thinking for Youth Education Develop-ment Center Newton Mass USA 2010

[20] M Resnick ldquoAll I really need to know (about creative thinking)I learned (by studying how children learn) in kindergartenrdquo inProceedings of the 6th Conference on Creativity amp Cognition (CCrsquo07) pp 1ndash6 ACM 2007

[21] J Dougherty ldquoConcept visualization in CS0 using ALICErdquoJournal of Computing Science in Colleges vol 22 no 3 pp 145ndash152 2007

[22] I Lee F Martin J Denner et al ldquoComputational thinking foryouth in practicerdquo ACM Inroads vol 2 no 1 pp 32ndash37 2011

[23] A Ruthmann J M Heines G R Greher P Laidler and CSaulters II ldquoTeaching computational thinking through musicallive coding in Scratchrdquo in Proceedings of the 41st ACM TechnicalSymposium on Computer Science Education (SIGCSE rsquo10) pp351ndash355 ACM 2010

[24] C Kelleher and R Pausch ldquoLowering the barriers to pro-gramming a taxonomy of programming environments andlanguages for novice programmersrdquo ACM Computing Surveysvol 37 no 2 pp 83ndash137 2005

[25] C Rader C Brand and C Lewis ldquoDegrees of comprehensionchildrenrsquos understanding of a visual programming environ-mentrdquo in Proceedings of the Conference on Human Factors inComputing Systems (CHI rsquo97) pp 351ndash358 ACM 1997

[26] PWyeth and H C Purchase ldquoTangible programming elementsfor young childrenrdquo in Proceedings of the Extended Abstracts onHuman Factors in Computing Systems (CHI rsquo02) pp 774ndash775ACM 2002

[27] M S Horn and R J K Jacob ldquoTangible programming in theclassroomwith ternrdquo inProceedings of the ExtendedAbstracts onHuman Factors in Computing Systems (CHI Trends Interactivity)(CHI rsquo07) pp 1965ndash1970 ACM 2007

[28] S Tarkan V Sazawal A Druin et al ldquoToque designing acooking-based programming language for and with childrenrdquoin Proceedings of the 28th Annual Conference on Human Factorsin Computing Systems (CHI rsquo10) pp 2417ndash2426 ACM 2010

[29] D Gallardo C F Julia and S Jorda ldquoTurTan a tangibleprogramming language for creative explorationrdquo in Proceedingsof the 3rd IEEE InternationalWorkshop onHorizontal InteractiveHuman Computer System (TABLETOP rsquo08) pp 89ndash92 IEEEComputer Society Press 2008

[30] J Silver andE Rosenbaum ldquoTwinkle programmingwith colorrdquoin Proceedings of the 4th International Conference on TangibleEmbedded and Embodied Interaction (TEI rsquo10) pp 383ndash384ACM 2010

[31] D L Wang C Zhang and H A Wang ldquoT-Maze a tangibleprogramming tool for childrenrdquo in Proceedings of the 10thInternational Conference on Interaction Design and Children(IDC rsquo11) pp 127ndash135 ACM 2011

[32] M S Horn ldquoTopCode Tangible Object Placement Codesrdquo2012 httpuserseecsnorthwesternedusimmhorntopcodes

[33] K Howland ldquoSupporting the development of multimodalwriting and computational thinking skills through computergame creationrdquo in Proceedings of the IEEE Symposium on VisualLanguages and Human-Centric Computing (VLHCC rsquo09) pp250ndash251 IEEE Computer Society Press 2009

[34] J LrsquoHeureux D Boisvert K Sanghera and R Cohen ldquoITproblem solving an implementation of computational thinkingin information technologyrdquo in Proceedings of the 13th AnnualConference on Information Technology Education (SIGITE rsquo12)pp 183ndash188 ACM 2012

[35] M Guzdial ldquoEducation paving the way for computationalthinkingrdquo Communications of the ACM vol 51 no 8 pp 25ndash27 2008

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

International Journal of

ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia

International Journal of

Biomedical Imaging

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArtificialNeural Systems

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational Intelligence and Neuroscience

Industrial EngineeringJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

8 The Scientific World Journal

(a) A girl with her maze (b) The most complicated maze

Figure 7 Two mazes collected from maze creation session a girl with her maze (a) and the complicated maze created by a boy (b)

Another remarkable domain where abstraction takes place ismaze design and creation Because the ldquowhat you see is whatyou getrdquo creation process children could then test complexabstractions quickly and precisely For example to create anewmaze the child designed amap where each square couldbe represented by certain programming blocks Then hesheselected only those appropriate for the virtual world in hishermind

(ii) Automation Automation occurs as the system executeschildrenrsquos programs The ldquoprogramrdquo itself automated ldquostep-ping throughrdquo and updated virtual avatarrsquos location anddirection (representing the veer) at each step As to this termwe did not expect too much actually it was an attempt tofind whether children could realize the automatic processAnd we indeed found some cue through the study Forexample before the execution one boy said ldquolisten to myinstructions and gordquo Another girl jumped for joy when shesaw the animation of her avatar traversing through the mazeBesides we saw 5 children stood up from their seats whenthe execution began and kept standing by waiting for thesuccessful moment

In interviews 4 children expressed their enjoyment whilewatching the virtual avatar traversing through the mazeautomatically under their instructions This was also backedup in the videotape from which we observed that 3 of themclapped out of joy and cheered on their avatarswhilewatchingthe animation

(iii) Problem Decomposition and Analysis T-Maze providesa platform that enables live programming by allowing thecode to be adjusted in real time As the session proceededwe moved beyond programs consisting of simple sequencesof actions and introduced more advanced concepts such asloops and numeric parameter values Through these activi-ties we found evidence that children could engage these con-cepts reasoning about possible outcomes of different blocksto build and test every piece of solution until they finallysolved the maze-escape problem Children also performedanalysis when they decidedwhether or not the avatar behavedas expected If the avatar was stuck or ldquomisbehavedrdquo it mayeither mean that their implementation of their control idea isfaulty or that some blocks were misplaced

Mostly the process of analysis happened implicitly butwecould still find some track from the observation For exampleone child in his complex maze-escape mission determinedthat he needed five forward blocks to reach the first turningAfter learning the loop syntax he realized that the largestvalue of a numeric block is five and the loop is limited to 5times the most so he needed two more forward blocks Thechild then built a program snippet as

(Start Loop + 5 + Forward + End Loop) 997888rarr Forward997888rarr Forward

In maze creation session we found 4 children chose tolocate the entrance and exit point of a maze at first and thentried to link them through while others just placed the blocksas they wish and were happy-go-lucky with any shape ofmazes Besides we observed a sense of ldquohabitrdquo in childrenrsquosprogramming that 5 children would add several blocks intothe sequence once when it was a simple straight line whileattempted one step at a time in a complex situation (a cornera sensor square or a cross) when they would adjust currentplacement and decide what to do next while observing thefeedbacks

(iv) Creativity To create a maze programming became amedium for childrenrsquos personal and creative expression inthe design of their mazes children engaged their fantasiesand built relationships with other pockets of reality that wentbeyond common thoughts As we analyzed childrenrsquos workcollected in the study several interesting mazes surprised usOne girl created her maze shaping like a snake (Figure 7(a))but she described it as the Great Wall coming out from hertextbook And the boywho claimed tomake themost difficultmaze used 31 blocks constructing the most complex fish-shape maze (Figure 7(b)) During the creating he put asidethe end block until he finished all the paths then counted thenumber of cells in each path trying to find the longest onewhere the end block was appended

Generally the mazes children created are different fromwhat we designed for T-Maze intrinsically and the descrip-tions children gave in interviews were also unexpected Forexample one boy created a maze with only one straight linewhile he described it as an airport runway And he talked

The Scientific World Journal 9

about his experience with his father at the airport last weekright off the reel

55 Summary Though the study is relatively small scale wecan tell from the results that T-Maze has some computationalconcepts in its programming activities and children indeedhandle those experienceswellThis study suggests farther thatT-Maze is easy to use for children and they are able to buildtheir own programs in playing and creating activities with T-Maze The emphasis is that T-Maze has the potential to helpraise childrenrsquos awareness about CT concepts

In interviews we also wanted to get some inspirationsfrom children with questions like ldquowhat other things do youthink the blocks can dordquo or ldquodo you think of any other gamesthat can be played with those blocksrdquo Two children said theywould like to control a flying plane with the blocks and 3others wished to help their mothers with some householdduties like cooking cleaning or furnishing their roomsTheir unlimited imagination inspired us in the revision andupgrade of T-Maze in the future

6 Conclusion and Future Work

T-Maze is designed for children aged 5ndash9 to help cultivatetheir computational thinking skills It allows children to playmaze-escape games and create their own mazes by buildingcomputer programs out of wooden blocks We conducted auser study to verify its ease of use and the potential capabilityto help children with computational thinking The resultsshowed that T-Maze has the promising potential to helpchildren cultivate some computational thinking awarenesslike abstraction automation problem decomposition andanalysis

In the future we will develop more games and scenariosdesign more kinds of programming blocks and enrich thetypes of sensors More and further user studies are neededto improve T-Maze for better user experience and skillsnurturing capability Besides as an upgrade we may providea collaborative tangible programming tool for children toexperience the fun of cooperation

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank the children from nearbykindergartens that took part in the study The authors grate-fully acknowledge financial support from the Major StateBasic Research Development Program of China under Grantno 2013CB328805 the National Natural Science Foundationof China under Grant no 60970090 and no 61272325 theCooperation Project of Chinese Academy of Sciences andFoshan City under Grant no 2012YS04 and the NationalScience and Technology Supported Program under Grant no2012BAH19F01

References

[1] J M Wing ldquoComputational thinkingrdquo Communications of theACM vol 49 no 3 pp 33ndash35 2006

[2] A Bundy ldquoComputational thinking is pervasiverdquo Journal ofScientific and Practical Computing vol 1 no 2 pp 67ndash69 2007

[3] S Grover ldquoComputational Thinking Programming andthe Google App Inventorrdquo 2010 httpwwwthesmartbeancomlibrarycomputational-thinking-programming-and-the-google-app-inventor

[4] S Grover ldquoExpanding the Technology Curriculum to IncludeFoundational Elements of Computer Science for K-8rdquo 2009httpwwwthesmartbeancommagazinehome-schoolingexpanding-the-technology-curriculum-foundational-ele-ments-of-computer-science-for-k-8

[5] G Orr ldquoComputational thinking through programming andalgorithmic artrdquo in Proceedings of the SIGGRAPH 2009 Talks(SIGGRAPH rsquo09) vol 1 p 1 ACM 2009

[6] M U Bers Blocks to Robots Learning with Technology in theEarly Childhood Classroom Teachers College New York NYUSA 2008

[7] D H Clements ldquoThe future of educational computing researchthe case of computer programmingrdquo Information Technology inChildhood Education Annual vol 1999 no 1 pp 147ndash179 1999

[8] T S McNerney Tangible programming bricks an approach tomaking programming accessible to everyone [MS thesis] MITCambridge Cambridge Mass USA 2000

[9] T S McNerney ldquoFrom turtles to tangible programming bricksexplorations in physical language designrdquo Personal and Ubiqui-tous Computing vol 8 no 5 pp 326ndash337 2004

[10] J Fails A Druin M Guha G Chipman S Simms and WChuraman ldquoChildrsquos play a comparison of desktop and physicalinteractive environmentsrdquo in Proceedings of the InteractionDesign and Children (IDC rsquo05) pp 48ndash55 ACM 2005

[11] M S Horn E T Solovey R J Crouser and R J K JacobldquoComparing the use of tangible and graphical programminglanguages for informal science educationrdquo in Proceedings of theSIGCHI Conference on Human Factors in Computing Systems(CHI rsquo09) pp 975ndash984 ACM 2009

[12] S PapertMindstorms Children Computers and Powerful IdeasBasic Books New York NY USA 1981

[13] G H Fletcher and J J Lu ldquoEducation human computing skillsrethinking the K-12 experiencerdquo Communications of the ACMvol 52 no 2 pp 23ndash25 2009

[14] P Curzon J Peckham A Settle E Roberts and H TaylorldquoComputational thinking(CT) on weaving it inrdquo in Proceedingsof the 14th Annual ACM SIGCSE Conference on Innovation andTechnology in Computer Science Education (ITiCSE rsquo09) pp201ndash202 ACM 2009

[15] P B Henderson ldquoUbiquitous computational thinkingrdquo Com-puter vol 42 no 10 pp 100ndash102 2009

[16] J J Lu and G H L Fletcher ldquoThinking about computationalthinkingrdquo in Proceedings of the 40th ACM Technical Symposiumon Computer Science Education (SIGCSE rsquo09) pp 260ndash264ACM 2009

[17] D Moursund ldquoComputational Thinkingrdquo IAE-Pedia 2009httpiae-pediaorgComputational Thinking

[18] National Research Council ldquoCommittee for the workshops oncomputational thinking 2010rdquo Report of a Workshop on theScope and Nature of Computational Thinking The NationalAcademies Press Washington DC USA 2010

10 The Scientific World Journal

[19] IWG Computational Thinking for Youth Education Develop-ment Center Newton Mass USA 2010

[20] M Resnick ldquoAll I really need to know (about creative thinking)I learned (by studying how children learn) in kindergartenrdquo inProceedings of the 6th Conference on Creativity amp Cognition (CCrsquo07) pp 1ndash6 ACM 2007

[21] J Dougherty ldquoConcept visualization in CS0 using ALICErdquoJournal of Computing Science in Colleges vol 22 no 3 pp 145ndash152 2007

[22] I Lee F Martin J Denner et al ldquoComputational thinking foryouth in practicerdquo ACM Inroads vol 2 no 1 pp 32ndash37 2011

[23] A Ruthmann J M Heines G R Greher P Laidler and CSaulters II ldquoTeaching computational thinking through musicallive coding in Scratchrdquo in Proceedings of the 41st ACM TechnicalSymposium on Computer Science Education (SIGCSE rsquo10) pp351ndash355 ACM 2010

[24] C Kelleher and R Pausch ldquoLowering the barriers to pro-gramming a taxonomy of programming environments andlanguages for novice programmersrdquo ACM Computing Surveysvol 37 no 2 pp 83ndash137 2005

[25] C Rader C Brand and C Lewis ldquoDegrees of comprehensionchildrenrsquos understanding of a visual programming environ-mentrdquo in Proceedings of the Conference on Human Factors inComputing Systems (CHI rsquo97) pp 351ndash358 ACM 1997

[26] PWyeth and H C Purchase ldquoTangible programming elementsfor young childrenrdquo in Proceedings of the Extended Abstracts onHuman Factors in Computing Systems (CHI rsquo02) pp 774ndash775ACM 2002

[27] M S Horn and R J K Jacob ldquoTangible programming in theclassroomwith ternrdquo inProceedings of the ExtendedAbstracts onHuman Factors in Computing Systems (CHI Trends Interactivity)(CHI rsquo07) pp 1965ndash1970 ACM 2007

[28] S Tarkan V Sazawal A Druin et al ldquoToque designing acooking-based programming language for and with childrenrdquoin Proceedings of the 28th Annual Conference on Human Factorsin Computing Systems (CHI rsquo10) pp 2417ndash2426 ACM 2010

[29] D Gallardo C F Julia and S Jorda ldquoTurTan a tangibleprogramming language for creative explorationrdquo in Proceedingsof the 3rd IEEE InternationalWorkshop onHorizontal InteractiveHuman Computer System (TABLETOP rsquo08) pp 89ndash92 IEEEComputer Society Press 2008

[30] J Silver andE Rosenbaum ldquoTwinkle programmingwith colorrdquoin Proceedings of the 4th International Conference on TangibleEmbedded and Embodied Interaction (TEI rsquo10) pp 383ndash384ACM 2010

[31] D L Wang C Zhang and H A Wang ldquoT-Maze a tangibleprogramming tool for childrenrdquo in Proceedings of the 10thInternational Conference on Interaction Design and Children(IDC rsquo11) pp 127ndash135 ACM 2011

[32] M S Horn ldquoTopCode Tangible Object Placement Codesrdquo2012 httpuserseecsnorthwesternedusimmhorntopcodes

[33] K Howland ldquoSupporting the development of multimodalwriting and computational thinking skills through computergame creationrdquo in Proceedings of the IEEE Symposium on VisualLanguages and Human-Centric Computing (VLHCC rsquo09) pp250ndash251 IEEE Computer Society Press 2009

[34] J LrsquoHeureux D Boisvert K Sanghera and R Cohen ldquoITproblem solving an implementation of computational thinkingin information technologyrdquo in Proceedings of the 13th AnnualConference on Information Technology Education (SIGITE rsquo12)pp 183ndash188 ACM 2012

[35] M Guzdial ldquoEducation paving the way for computationalthinkingrdquo Communications of the ACM vol 51 no 8 pp 25ndash27 2008

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

International Journal of

ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia

International Journal of

Biomedical Imaging

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArtificialNeural Systems

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational Intelligence and Neuroscience

Industrial EngineeringJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World Journal 9

about his experience with his father at the airport last weekright off the reel

55 Summary Though the study is relatively small scale wecan tell from the results that T-Maze has some computationalconcepts in its programming activities and children indeedhandle those experienceswellThis study suggests farther thatT-Maze is easy to use for children and they are able to buildtheir own programs in playing and creating activities with T-Maze The emphasis is that T-Maze has the potential to helpraise childrenrsquos awareness about CT concepts

In interviews we also wanted to get some inspirationsfrom children with questions like ldquowhat other things do youthink the blocks can dordquo or ldquodo you think of any other gamesthat can be played with those blocksrdquo Two children said theywould like to control a flying plane with the blocks and 3others wished to help their mothers with some householdduties like cooking cleaning or furnishing their roomsTheir unlimited imagination inspired us in the revision andupgrade of T-Maze in the future

6 Conclusion and Future Work

T-Maze is designed for children aged 5ndash9 to help cultivatetheir computational thinking skills It allows children to playmaze-escape games and create their own mazes by buildingcomputer programs out of wooden blocks We conducted auser study to verify its ease of use and the potential capabilityto help children with computational thinking The resultsshowed that T-Maze has the promising potential to helpchildren cultivate some computational thinking awarenesslike abstraction automation problem decomposition andanalysis

In the future we will develop more games and scenariosdesign more kinds of programming blocks and enrich thetypes of sensors More and further user studies are neededto improve T-Maze for better user experience and skillsnurturing capability Besides as an upgrade we may providea collaborative tangible programming tool for children toexperience the fun of cooperation

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank the children from nearbykindergartens that took part in the study The authors grate-fully acknowledge financial support from the Major StateBasic Research Development Program of China under Grantno 2013CB328805 the National Natural Science Foundationof China under Grant no 60970090 and no 61272325 theCooperation Project of Chinese Academy of Sciences andFoshan City under Grant no 2012YS04 and the NationalScience and Technology Supported Program under Grant no2012BAH19F01

References

[1] J M Wing ldquoComputational thinkingrdquo Communications of theACM vol 49 no 3 pp 33ndash35 2006

[2] A Bundy ldquoComputational thinking is pervasiverdquo Journal ofScientific and Practical Computing vol 1 no 2 pp 67ndash69 2007

[3] S Grover ldquoComputational Thinking Programming andthe Google App Inventorrdquo 2010 httpwwwthesmartbeancomlibrarycomputational-thinking-programming-and-the-google-app-inventor

[4] S Grover ldquoExpanding the Technology Curriculum to IncludeFoundational Elements of Computer Science for K-8rdquo 2009httpwwwthesmartbeancommagazinehome-schoolingexpanding-the-technology-curriculum-foundational-ele-ments-of-computer-science-for-k-8

[5] G Orr ldquoComputational thinking through programming andalgorithmic artrdquo in Proceedings of the SIGGRAPH 2009 Talks(SIGGRAPH rsquo09) vol 1 p 1 ACM 2009

[6] M U Bers Blocks to Robots Learning with Technology in theEarly Childhood Classroom Teachers College New York NYUSA 2008

[7] D H Clements ldquoThe future of educational computing researchthe case of computer programmingrdquo Information Technology inChildhood Education Annual vol 1999 no 1 pp 147ndash179 1999

[8] T S McNerney Tangible programming bricks an approach tomaking programming accessible to everyone [MS thesis] MITCambridge Cambridge Mass USA 2000

[9] T S McNerney ldquoFrom turtles to tangible programming bricksexplorations in physical language designrdquo Personal and Ubiqui-tous Computing vol 8 no 5 pp 326ndash337 2004

[10] J Fails A Druin M Guha G Chipman S Simms and WChuraman ldquoChildrsquos play a comparison of desktop and physicalinteractive environmentsrdquo in Proceedings of the InteractionDesign and Children (IDC rsquo05) pp 48ndash55 ACM 2005

[11] M S Horn E T Solovey R J Crouser and R J K JacobldquoComparing the use of tangible and graphical programminglanguages for informal science educationrdquo in Proceedings of theSIGCHI Conference on Human Factors in Computing Systems(CHI rsquo09) pp 975ndash984 ACM 2009

[12] S PapertMindstorms Children Computers and Powerful IdeasBasic Books New York NY USA 1981

[13] G H Fletcher and J J Lu ldquoEducation human computing skillsrethinking the K-12 experiencerdquo Communications of the ACMvol 52 no 2 pp 23ndash25 2009

[14] P Curzon J Peckham A Settle E Roberts and H TaylorldquoComputational thinking(CT) on weaving it inrdquo in Proceedingsof the 14th Annual ACM SIGCSE Conference on Innovation andTechnology in Computer Science Education (ITiCSE rsquo09) pp201ndash202 ACM 2009

[15] P B Henderson ldquoUbiquitous computational thinkingrdquo Com-puter vol 42 no 10 pp 100ndash102 2009

[16] J J Lu and G H L Fletcher ldquoThinking about computationalthinkingrdquo in Proceedings of the 40th ACM Technical Symposiumon Computer Science Education (SIGCSE rsquo09) pp 260ndash264ACM 2009

[17] D Moursund ldquoComputational Thinkingrdquo IAE-Pedia 2009httpiae-pediaorgComputational Thinking

[18] National Research Council ldquoCommittee for the workshops oncomputational thinking 2010rdquo Report of a Workshop on theScope and Nature of Computational Thinking The NationalAcademies Press Washington DC USA 2010

10 The Scientific World Journal

[19] IWG Computational Thinking for Youth Education Develop-ment Center Newton Mass USA 2010

[20] M Resnick ldquoAll I really need to know (about creative thinking)I learned (by studying how children learn) in kindergartenrdquo inProceedings of the 6th Conference on Creativity amp Cognition (CCrsquo07) pp 1ndash6 ACM 2007

[21] J Dougherty ldquoConcept visualization in CS0 using ALICErdquoJournal of Computing Science in Colleges vol 22 no 3 pp 145ndash152 2007

[22] I Lee F Martin J Denner et al ldquoComputational thinking foryouth in practicerdquo ACM Inroads vol 2 no 1 pp 32ndash37 2011

[23] A Ruthmann J M Heines G R Greher P Laidler and CSaulters II ldquoTeaching computational thinking through musicallive coding in Scratchrdquo in Proceedings of the 41st ACM TechnicalSymposium on Computer Science Education (SIGCSE rsquo10) pp351ndash355 ACM 2010

[24] C Kelleher and R Pausch ldquoLowering the barriers to pro-gramming a taxonomy of programming environments andlanguages for novice programmersrdquo ACM Computing Surveysvol 37 no 2 pp 83ndash137 2005

[25] C Rader C Brand and C Lewis ldquoDegrees of comprehensionchildrenrsquos understanding of a visual programming environ-mentrdquo in Proceedings of the Conference on Human Factors inComputing Systems (CHI rsquo97) pp 351ndash358 ACM 1997

[26] PWyeth and H C Purchase ldquoTangible programming elementsfor young childrenrdquo in Proceedings of the Extended Abstracts onHuman Factors in Computing Systems (CHI rsquo02) pp 774ndash775ACM 2002

[27] M S Horn and R J K Jacob ldquoTangible programming in theclassroomwith ternrdquo inProceedings of the ExtendedAbstracts onHuman Factors in Computing Systems (CHI Trends Interactivity)(CHI rsquo07) pp 1965ndash1970 ACM 2007

[28] S Tarkan V Sazawal A Druin et al ldquoToque designing acooking-based programming language for and with childrenrdquoin Proceedings of the 28th Annual Conference on Human Factorsin Computing Systems (CHI rsquo10) pp 2417ndash2426 ACM 2010

[29] D Gallardo C F Julia and S Jorda ldquoTurTan a tangibleprogramming language for creative explorationrdquo in Proceedingsof the 3rd IEEE InternationalWorkshop onHorizontal InteractiveHuman Computer System (TABLETOP rsquo08) pp 89ndash92 IEEEComputer Society Press 2008

[30] J Silver andE Rosenbaum ldquoTwinkle programmingwith colorrdquoin Proceedings of the 4th International Conference on TangibleEmbedded and Embodied Interaction (TEI rsquo10) pp 383ndash384ACM 2010

[31] D L Wang C Zhang and H A Wang ldquoT-Maze a tangibleprogramming tool for childrenrdquo in Proceedings of the 10thInternational Conference on Interaction Design and Children(IDC rsquo11) pp 127ndash135 ACM 2011

[32] M S Horn ldquoTopCode Tangible Object Placement Codesrdquo2012 httpuserseecsnorthwesternedusimmhorntopcodes

[33] K Howland ldquoSupporting the development of multimodalwriting and computational thinking skills through computergame creationrdquo in Proceedings of the IEEE Symposium on VisualLanguages and Human-Centric Computing (VLHCC rsquo09) pp250ndash251 IEEE Computer Society Press 2009

[34] J LrsquoHeureux D Boisvert K Sanghera and R Cohen ldquoITproblem solving an implementation of computational thinkingin information technologyrdquo in Proceedings of the 13th AnnualConference on Information Technology Education (SIGITE rsquo12)pp 183ndash188 ACM 2012

[35] M Guzdial ldquoEducation paving the way for computationalthinkingrdquo Communications of the ACM vol 51 no 8 pp 25ndash27 2008

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

International Journal of

ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia

International Journal of

Biomedical Imaging

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArtificialNeural Systems

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational Intelligence and Neuroscience

Industrial EngineeringJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

10 The Scientific World Journal

[19] IWG Computational Thinking for Youth Education Develop-ment Center Newton Mass USA 2010

[20] M Resnick ldquoAll I really need to know (about creative thinking)I learned (by studying how children learn) in kindergartenrdquo inProceedings of the 6th Conference on Creativity amp Cognition (CCrsquo07) pp 1ndash6 ACM 2007

[21] J Dougherty ldquoConcept visualization in CS0 using ALICErdquoJournal of Computing Science in Colleges vol 22 no 3 pp 145ndash152 2007

[22] I Lee F Martin J Denner et al ldquoComputational thinking foryouth in practicerdquo ACM Inroads vol 2 no 1 pp 32ndash37 2011

[23] A Ruthmann J M Heines G R Greher P Laidler and CSaulters II ldquoTeaching computational thinking through musicallive coding in Scratchrdquo in Proceedings of the 41st ACM TechnicalSymposium on Computer Science Education (SIGCSE rsquo10) pp351ndash355 ACM 2010

[24] C Kelleher and R Pausch ldquoLowering the barriers to pro-gramming a taxonomy of programming environments andlanguages for novice programmersrdquo ACM Computing Surveysvol 37 no 2 pp 83ndash137 2005

[25] C Rader C Brand and C Lewis ldquoDegrees of comprehensionchildrenrsquos understanding of a visual programming environ-mentrdquo in Proceedings of the Conference on Human Factors inComputing Systems (CHI rsquo97) pp 351ndash358 ACM 1997

[26] PWyeth and H C Purchase ldquoTangible programming elementsfor young childrenrdquo in Proceedings of the Extended Abstracts onHuman Factors in Computing Systems (CHI rsquo02) pp 774ndash775ACM 2002

[27] M S Horn and R J K Jacob ldquoTangible programming in theclassroomwith ternrdquo inProceedings of the ExtendedAbstracts onHuman Factors in Computing Systems (CHI Trends Interactivity)(CHI rsquo07) pp 1965ndash1970 ACM 2007

[28] S Tarkan V Sazawal A Druin et al ldquoToque designing acooking-based programming language for and with childrenrdquoin Proceedings of the 28th Annual Conference on Human Factorsin Computing Systems (CHI rsquo10) pp 2417ndash2426 ACM 2010

[29] D Gallardo C F Julia and S Jorda ldquoTurTan a tangibleprogramming language for creative explorationrdquo in Proceedingsof the 3rd IEEE InternationalWorkshop onHorizontal InteractiveHuman Computer System (TABLETOP rsquo08) pp 89ndash92 IEEEComputer Society Press 2008

[30] J Silver andE Rosenbaum ldquoTwinkle programmingwith colorrdquoin Proceedings of the 4th International Conference on TangibleEmbedded and Embodied Interaction (TEI rsquo10) pp 383ndash384ACM 2010

[31] D L Wang C Zhang and H A Wang ldquoT-Maze a tangibleprogramming tool for childrenrdquo in Proceedings of the 10thInternational Conference on Interaction Design and Children(IDC rsquo11) pp 127ndash135 ACM 2011

[32] M S Horn ldquoTopCode Tangible Object Placement Codesrdquo2012 httpuserseecsnorthwesternedusimmhorntopcodes

[33] K Howland ldquoSupporting the development of multimodalwriting and computational thinking skills through computergame creationrdquo in Proceedings of the IEEE Symposium on VisualLanguages and Human-Centric Computing (VLHCC rsquo09) pp250ndash251 IEEE Computer Society Press 2009

[34] J LrsquoHeureux D Boisvert K Sanghera and R Cohen ldquoITproblem solving an implementation of computational thinkingin information technologyrdquo in Proceedings of the 13th AnnualConference on Information Technology Education (SIGITE rsquo12)pp 183ndash188 ACM 2012

[35] M Guzdial ldquoEducation paving the way for computationalthinkingrdquo Communications of the ACM vol 51 no 8 pp 25ndash27 2008

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

International Journal of

ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia

International Journal of

Biomedical Imaging

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArtificialNeural Systems

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational Intelligence and Neuroscience

Industrial EngineeringJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

International Journal of

ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia

International Journal of

Biomedical Imaging

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArtificialNeural Systems

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational Intelligence and Neuroscience

Industrial EngineeringJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014