1IIIIIIIIIIlr'lllillllll~fIllil\~Olili~llliillrllrlllllllllll3 1822 00070 8784

UNIVERSITYOF CALIFORNIA

SAN DIEGOi t m >1"\0~ »>:., V:;.· ,j

Lexical Cr~~~~c:tion, Semantic Me~ory, and Reading Comprehension

A dissertation submitted in partial satisfaction of the

requirements for the degree rector of Pnilosophy in

Psychology

by

Billy E. Vaughn-:

Committee in charge:

Professor Michael Cole, O'lairProfessor Jean Mandler, Co-ChairProfessor David RumelhartProfessor Hugh MehanProfessor Roy D'Andrade

1986

The dissertation of Billy E. Vaughn is approved,and it is acceptable in quality and form for

.publication on microfilm:

V-1 Co-chair:r:€rson

UuduL L7 Chairperson

University of California, San Diego

1986

iii

ToAunt Barbara,

who just might be the WIZARD.

iv

TABLEOF CO~~ENTS

list of FiguresPage

vii

list of Tables viii

Acknowl edgener. ts ix

Vita .............................................................. xi

FUblicatior xi

Fields of Stu:::ly xi

Abstr act xii

Chapter I

A Rich ar:d QJick Theory of Lexical Og ar i zat.Lon ar.d Access

Semartic Memory, Lexical Crg ar i zat ion , ard Readir:g

Compreher sror 6

Computer s ir: Educatior: 20

Chapter II 26A Study of Lexical Crg ar.i zat.Iot- ar d Readirg

Compr ehen si.or 26

Method 31

Subjects 31

Appar atus ard !v:ater ials 31

Desigr. 40

Procedur es .................................................... 41

Resul ts ......................................................... 45

Pr etest Anal yse s 46

v

Trials to Criter ior .

Posttest results usirg Pretest scores as Covariates .

Posttest ccmpar isors usirg the Trials to Criter Ion

covar iates .

Free Reeall .

Di scuasi on .

Chapter III ~ .

Conclusion .

Refer er ces .

Apper:dix

Rescue Micro'MJrld Courseware .

Per forrnar.ce Record s ard Eval uatior of MJdel .

5261

61

6369

73

73

76

79

79

84

Table

1

2

3

4

5.

6

7

8

9

10

11

12

13

14

LIST CF TABLES

The Instruction by Semantic List Type factorsin the design .

Taxonomic List Stimuli .........•............

Mixed List Stimuli .•.•'...•..•.••••.•.•......

Thematic List Stimuli .... ;•.•.......•.......

Page

2935

36

37

Pretest and Post test Raw Percent Correct onthe Spellirig~est ....•...•.•.•..•..•........ 47

Pretest and Posttest ,Raw Percent Correct onthe Vocabulary Discrimination Test ..••..•...

Pretest and Posttest Raw Percent Correct onthe Vocabulary Definition Test ............•.

Pretest and Post test Raw Percent Correct onthe Open~ended Test for Appropriate WordResponses .

Trials to criterion as a function ofinstruction and list type ...••...•...•......

Adjusted Percent Correct Scores on theSpelling Test .

Adjusted Percent Correct Scores on theVocabulary Discrimination Test .

Adjusted Percent Correct Scores on theDefini tion Test e' •••••••••••••••••••• e· •••••••

Recall and clustering performance as afunction of insruction and list type .

Average number of categories recalled andaverage number of items recalled forInstructional Group as a funrition of Speed ofInstruction and Semantic List Type .

viii

48

49

50

53

55

57

59

64

65

ACKNCWLEI:GEMENTS

The combination of the department of psychology and the

Laboratory of Comparative I-lumanCognition created a remarkable

environment for me to develop my expertise in Cognitive Psychology. I

ow= a tremendous amount to Mike Cole and Peg Griffin for the

construction of a zone of proximal developnent that al Ioved me to study

mainstream psychology without completely losing my soul. Jean Mandler's

patience and dedication will never be forgotten. A special thanks t 0

IUd tvi2han for his friendship and dependability. Edvun Hutchins and Roy

D'Andrade both helped me discover cognitive Anthropology. As an

undergraduate at this university, my first exposure to cognitive

psychology was in a course taught by by Iav id Rumelhart. I worked with

him closely in a year long course in artificial intelligence during

graduate school. He will never know how much he influenced the direction

of my thinking.

There have been many others, mentioned in order in which they

entered my life; each have contributed to the social accomplishment of

this task. Alfred Galloway, Norma Johnson, the Gardons, tvlrs. Barbara

Iugger Anderson (God rest her soul), the Peyers, David Floyd, fvlaggie

Elliot, Elliot Weinstein, Lonnie Anderson, Linda Amerman (God rest her

soul), Laura fv1.artin, Ken Traupn ann , ::helley Stokes, Judy 3nith, Katie

King, Mary tvtGinnis, S1eila Broyles, Lerman Penson, f-A,argaret Armstrong,

Charles Cooper, Jim 1tiertsch, Eileen Conway, and Nark Wallen.

- .

A special thanks to v,tyatt Casey, who taught me apple

microcomputer graphics and other pascal programming tricks. The

students w110vor-ked with me on this project included Deni se Stevens,

Juri Yoshida, Ellen Schi l l.er , Peter Kurata, Tom Cougher , Ha Tran, and

Cedric Lewis.

Thanks to Ir , John Griffith of the San Diego In i f't ed School.

District, who permitted me to enter the classrooms at Kennedy and Knox

el enerrtar y schools. The people at Knox were extremely helpful and

understanding. The children in both schools were very patient and

diligent subjects. tvlt. Erie Bapt i st church is thanked for its

receptiveness. Without the use of their facilities during school

holiday and vacation periods, I might still be collecting data.

lastly, I would like to extend my appreciation to my family. My

wife Liz is the most sensitive and understanding per son I have come to

know and love dearly. Per patience and admiration will be treasured

forever. My daughter Lindsay has endured a great deal during my

struggle to complete my dissertation. It is obvious that she is a strong

person. I love her dearly. Iav id Vaughn is much too young to understand

vklat his father has vor-ked so hard to accomplish for his sake.

VITA

March 2, 1951 -- Born -- Houston, Texas

1978 -

1980 -

1982 -1986 -

Bachelor of Arts, Social PsychologyLhiversity of California, San DiegoMasters of Arts, Clinical PsychologyCalifornia State Chiversity, Long BeachMasters of Arts, Cognitive Psychology

Doctor of Philosophy, Cognitive PsychologyLhiversity of California, San Diego

PUBLICATIONS

Vaughn, B. E. ( 1985). Computer NetlMJrks and Education.The Quarterly Newsletter of the Laboratory of Comparative~uman Cognition, 7(4), p .-13r -

Vaughn, B. R. (in collaboration with the Laboratory of ComparativeHuman Cognition) (in press). Cross-cul tural Psychology andEdu~ation. American Psychologist.

F IELrs CF STUDY

Major Fields: Cognitive Psychology, Cognitive Sociology and Computer~ience

ABSTRACTOF THEDISSERTATION

Lexical C1"ganization, Semantic Vemory, and Reading Comprehension

by

Billy E. Vaughn

Doctor of Fhilosophy in Psychology

lliiversityof California, San Diego, 1986

Professor t-lichael Cole, Chair

Professor Jean Mandler, Co-Chair

A microcomputer task was designed on the basis of a theory of

how lexical items can be related and how these lexical organizing

factors facilitate semantic memory, vocabulary instruction, and reading

comprehension. The task was employed in a study that ex anined three

issues: (1) the utility of the microcomputer lexical decision task for

pre-reading instruction, (2) the benefits of presenting the task under

speeded constraints, and (3) the differential effects of varying

semantic structure on lexical organization and reading comprehension.

Twovocabulary treatments, chosen to contrast varying anounts of speeded

instructional constraints, and a control condition were used with fourth

graders who were reading at grade level or below. Speed of instruction

was studied in interaction with three types of semantic 'MJrd list

structures (taxonomic, thematic, and a mix of the two). The training

produced significantly higher scores on two vocabulary tests and one of

t'MJ reading comprehension measures, indicating that the instruction had

a significant effect on both comprehension and vocabulary learning.

AIthough the literature suggests that recall of thematic lists is

greater than for taxonomic lists, no differences were found in the

present study. In addition, semantic list differences did not

differentially affect performance on the various tests of vocabulary

knowledge and reading comprehension.