Gardner, Babu, Ghosh, Sherwood, Chen 2007 - Neurophysiology of Pre Hens Ion. III. Representation of Object Features in Posterior Parietal Cortex of the Macaque Monkey

8/8/2019 Gardner, Babu, Ghosh, Sherwood, Chen 2007 - Neurophysiology of Pre Hens Ion. III. Representation of Object Feat

1/24

doi:10.1152/jn.00609.200798:3708-3730, 2007. First published 17 October 2007;J NeurophysiolJessie ChenEsther P. Gardner, K. Srinivasa Babu, Soumya Ghosh, Adam Sherwood andMacaque MonkeyObject Features in Posterior Parietal Cortex of theNeurophysiology of Prehension. III. Representation of

You might find this additional info useful...

90 articles, 38 of which can be accessed free at:This article cites

http://jn.physiology.org/content/98/6/3708.full.html#ref-list-1

3 other HighWire hosted articlesThis article has been cited by

[PDF][Full Text][Abstract]

, December, 2009; 102 (6): 3433-3446.J NeurophysiolLeighton B. N. Hinkley, Leah A. Krubitzer, Jeff Padberg and Elizabeth A. DisbrowVisual-Manual Exploration and Posterior Parietal Cortex in Humans

[PDF][Full Text][Abstract], December, 2009; 102 (6): 3310-3328.J Neurophysiol

Jessie Chen, Shari D. Reitzen, Jane B. Kohlenstein and Esther P. GardnerCortex of the Macaque MonkeyNeural Representation of Hand Kinematics During Prehension in Posterior Parietal

[PDF][Full Text][Abstract]

, September, 29 2010; 30 (39): 12918-12935.J. Neurosci.KrubitzerJeffrey Padberg, Gregg Recanzone, James Engle, Dylan Cooke, Adam Goldring and LeahPlasticityLesions in Posterior Parietal Area 5 in Monkeys Result in Rapid Behavioral and Cortical

including high resolution figures, can be found at:Updated information and services

http://jn.physiology.org/content/98/6/3708.full.html

can be found at:Journal of NeurophysiologyaboutAdditional material and information

http://www.the-aps.org/publications/jn

This infomation is current as of January 14, 2011.

American Physiological Society. ISSN: 0022-3077, ESSN: 1522-1598. Visit our website at http://www.the-aps.org/.(monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright 2007 by the

publishes original articles on the function of the nervous system. It is published 12 times a yearJournal of Neurophysiology
http://jn.physiology.org/content/98/6/3708.full.html#ref-list-1http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://www.jneurosci.org/cgi/reprint/30/39/12918http://www.jneurosci.org/cgi/reprint/30/39/12918http://www.jneurosci.org/cgi/content/full/30/39/12918http://www.jneurosci.org/cgi/content/abstract/30/39/12918http://www.jneurosci.org/cgi/content/full/30/39/12918http://www.jneurosci.org/cgi/reprint/30/39/12918http://www.jneurosci.org/cgi/content/abstract/30/39/12918http://www.jneurosci.org/cgi/content/full/30/39/12918http://jn.physiology.org/content/98/6/3708.full.htmlhttp://jn.physiology.org/content/98/6/3708.full.htmlhttp://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://www.jneurosci.org/cgi/reprint/30/39/12918http://www.jneurosci.org/cgi/content/full/30/39/12918http://www.jneurosci.org/cgi/content/abstract/30/39/12918http://jn.physiology.org/content/98/6/3708.full.html#ref-list-1


2/24

Neurophysiology of Prehension. III. Representation of Object Features in

Posterior Parietal Cortex of the Macaque Monkey

Esther P. Gardner, K. Srinivasa Babu, Soumya Ghosh, Adam Sherwood, and Jessie Chen

Department of Physiology and Neuroscience, New York University School of Medicine, New York, New York

Submitted 29 May 2007; accepted in final form 12 October 2007

Gardner EP, Babu KS, Ghosh S, Sherwood A, Chen J. Neurophys-iology of prehension. III. Representation of object features in posteriorparietal cortex of the macaque monkey. J Neurophysiol 98: 37083730,2007. First published October 17, 2007; doi:10.1152/jn.00609.2007.Neurons in posterior parietal cortex (PPC) may serve both proprio-ceptive and exteroceptive functions during prehension, signaling handactions and object properties. To assess these roles, we used digital videorecordings to analyze responses of 83 hand-manipulation neurons in area5 as monkeys grasped and lifted objects that differed in shape (round andrectangular), size (large and small spheres), and location (identical rect-

angular blocks placed lateral and medial to the shoulder). The taskcontained seven stagesapproach, contact, grasp, lift, hold, lower, re-laxplus a pretrial interval. The four test objects evoked similar spiketrains and mean rate profiles that rose significantly above baseline fromapproach through lift, with peak activity at contact. Although represen-tation by the spike train of specific hand actions was stronger thandistinctions between grasped objects, 34% of these neurons showedstatistically significant effects of object properties or hand postures onfiring rates. Somatosensory input from the hand played an important roleas firing rates diverged most prominently on contact as grasp wassecured. The small spheregrasped with the most flexed hand postureevoked the highest firing rates in 43% of the population. Twenty-onepercent distinguished spheres that differed in size and weight, and 14%discriminated spheres from rectangular blocks. Location in the workspacemodulated response amplitude as objects placed across the midline

evoked higher firing rates than positions lateral to the shoulder. Weconclude that area 5 neurons, like those in area AIP, integrate objectfeatures, hand actions, and grasp postures during prehension.

I N T R O D U C T I O N

The ability to recognize objects placed in the hand using thesense of touch is a fundamental property of the somatic sensorysystem. This cognitive property, called stereognosis, is se-verely impaired in humans with lesions of the parietal lobe,particularly the posterior parietal regions (reviewed in Binkof-ski et al. 2001; Milner and Goodale 1995; Pause and Freund1989; Pause et al. 1989). These patients fail to shape and orient

the hand properly to grasp objects and misdirect the arm duringreaching. They use abnormally high levels of grip force whenan object is placed in their hand and are unable to direct thefingers when asked to evaluate its size and shape.

In earlier studies of neurons in the hand representation ofprimary somatosensory (S-I) and posterior parietal cortex(PPC), we used digital video recordings of hand kinematics tocorrelate neuronal spike trains to specific actions performed bythe hand as monkeys grasped and lifted objects in a trainedprehension task (Debowy et al. 2001; Gardner et al. 1999,

2002, 2007a,b; Ro et al. 2000). Those studies demonstratedthat activity of hand manipulation neurons in areas 5 and7b/AIP of PPC preceded the onset of firing in S-I, peaking asthe hand contacted and grasped the object. In this report, weconsider how the firing rates of these neurons are influenced byproperties of the grasped object. Our prehension task wasdesigned to measure the sensitivity of cortical neurons to theintrinsic size and shape parameters of the grasped objects, andto extrinsic features such as location. Four objects were tested

in each session in interleaved trials; they varied in shape (roundand rectangular), size (large and small), and position in theworkspace (lateral or medial to the shoulder). Two rectanglesof identical size and shape placed at different positions in theworkspace tested whether neural activity reflected object loca-tion or geometry. If neural activity reflected shape primitives,then firing rates should be independent of an objects location.However, if firing rates signaled location, then neighboringobjects should evoke similar responses, whereas those at op-posite ends of the workspace should evoke different firing rateseven if they have the same shape. Furthermore, if neurons inthe hand area are tuned to a specific location, as are cells inmore medial regions of area 5 (Kalaska et al. 1983), theirresponses should become progressively weaker as objects aredisplaced from the preferred position.

The data presented in this report indicate that neural re-sponses in area 5 are related to both the hand posture andintrinsic properties of the grasped object. Object selectivity isexpressed as an alteration in the gain of neural responses, andtheir duration, rather than the presence or absence of activityduring prehension.

M E T H O D S

Neurophysiological and behavioral data were obtained from twoadult rhesus monkeys ( Macaca mulatta) trained to perform a prehen-sion task; these animals were also used in earlier studies of PPCneurons (Gardner et al. 2007a,b). Experimental protocols were re-

viewed and approved by the New York University Medical CenterInstitutional Animal Care and Use Committee (IACUC) and are inaccordance with the guiding principles for the care and use ofexperimental animals approved by the Councils of the AmericanPhysiological Society, the National Research Council, and the Societyfor Neuroscience.

Prehension task

The monkeys were trained in a grasp-and-lift task to manipulateobjects placed at defined locations in the workspace. The objects were

Address for reprint requests and other correspondence: E. P. Gardner,Dept. of Physiology and Neuroscience, New York University School ofMedicine, 550 First Ave., MSB 442, New York, NY 10016 (E-mail: [email protected]).

The costs of publication of this article were defrayed in part by the paymentof page charges. The article must therefore be hereby marked advertisementin accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

J Neurophysiol 98: 37083730, 2007.First published October 17, 2007; doi:10.1152/jn.00609.2007.

3708 0022-3077/07 $8.00 Copyright 2007 The American Physiological Society www.jn.org


3/24

a set of four aluminum or brass knobs mounted on a box placed 2224cm in front of the animal as shown in Fig. 1. We tested two round andtwo rectangular knobs in each session. The rectangular knobs mea-sured 20 20 40 mm; the round knobs were 15 and 30-mm-diamspheres. Additional weights placed inside the shape box were used tocompensate for the difference in object mass. The total load lifted was108 g (small round), 137 g (rectangular block), and 242 g (large

round). The knobs were typically held between the fingers and palmusing a whole-hand power grasp or in a semi-precision grip betweenthe thumb and digits 2 and 3. The animals could view the workspaceand used visual guidance to position their hand on the objects.

The knob arrangement on the shape box in sessions testing the righthand is illustrated in Figs. 1 and 2. The knob shafts were positioned(1) slightly medial to the left shoulder, (2) at the midline, (3) alignedto the right shoulder, and (4) lateral to the right shoulder. The roundknobs were placed in the center positions in the earliest sessions (Fig.2A), with the two rectangular blocks at the left and right ends of theshape box (knobs 1 and 4). In the middle study period, the rectangleswere placed centrally, and the round knobs at the end locations (Fig.2B). Round and rectangular knobs were alternated in the late sessions(Fig. 2, Cand D). During recordings from the right hemisphere, whenthe left hand was tested, the shape box was shifted to the left side of

the chair. Knob 2 was aligned with the left shoulder, the round knobswere placed centrally, and the rectangles sited at the outer positions(Fig. 9).

Individual knobs were tested in blocks of 210 trials; the order oftesting was determined by the system computers random numbergenerator. The location of the four knobs in the workspace wasrepresented on the computer monitor by four identical black icons;one icon was flashed in red to indicate which knob should be lifted onthat trial (Fig. 1A). The animal had to reach to the specified knob,grasp and lift it until an upper stop was contacted. If the correct object

was lifted and held in place, the animal received a juice reward; if hechose the wrong object, he was not rewarded on that trial. Althoughthe visual cues directed the animals attention to a specific object oneach trial, he selected the particular grasp postures used to accomplishthe task goals, and the timing of the task stages. Each monkeydeveloped its own grasp strategy that was natural, comfortable, andfluid and was used repeatedly during the period of study.

The task comprised a succession of stages characterized by aspecific goal for each action, a unique pattern of underlying muscleactivity expressed as kinematic behavior, and a transient mechanicalevent signaling goal completion and transition to the next stage.We divided the task into seven stagesapproach, contact, grasp,lift, hold, lower, relaxplus an intertrial interval that precededapproach (stage 0). Stages 13 were required for object acquisi-tion, stages 4 and 5 for manipulation, and stages 6 and 7 for release

of the object.We monitored hand kinematics during the task using digital video

(DV) recordings of the animals behavior synchronized to neuronal

2. Lift

1. Knob Cue

4. Incorrect Response

3. Correct Response

00:16:10:20Approach00:16:10:23Preshape00:16:10:25Contact00:16:10:27Grasp

Knob 1Knob 2

Knob 3Knob 4

Unit H17094-131-3

15:31:17

16:20:23

15:18:27

15:14:18

Knob 1

Knob 2

Knob 3

Knob 4

A

B

C

FIG. 1. Monkeys performed a trained prehension task. A: task cues presented to the animal on a computer monitor. The flashing red icon indicates that theanimal should lift knob 3. Feedback is provided during lift by upward movement of the icon representing the knob selected on that trial; correct performanceis signal by a white icon at the top position. B: kinematics of the approach, contact, and grasp stages traced from video images captured during task performanceby monkey H17094. C: digital video images captured during task performance illustrating the hand posture used to grasp each knob. Time code specifiesminute:second:frame number (30 fps) in the original DV records. Recordings from clip 1, unit H17094-131-3.

3709OBJECT REPRESENTATION IN PPC AREA 5

J Neurophysiol VOL 98 DECEMBER 2007 www.jn.org


4/24

spike trains, as previously described (Debowy et al. 2001, 2002;Gardner et al. 1999, 2002, 2007a,b; Ro et al. 1998). Three DVcamcorders provided lateral, frontal, and overhead images of themonkey and the workspace at 29.97 frame/s. The onset of each taskstage was measured from the time code of the matching videoframe by visual observation, and/or by tracings of the hand posturein successive video images (Fig. 1B). Event time codes werestored in spreadsheets and were used subsequently as markers fordisplay in burst analyses, alignment of neural responses in rastersand PSTHs and for bracketing task stages in statistical analyses offiring rates.

Recording and data analysis techniques

Extracellular single-unit recordings from neurons in area 5 weremade in the left hemisphere of monkeys H17094 and N18588 and inthe right hemisphere ofN18588 as described in Gardner et al. (2007a);the specific recording locations are illustrated in Fig. 3. Spike trainswere digitized at 16-bit resolution, 48 kHz, or 12-bit resolution, 32kHz by the DV camcorders and stored as an audio trace together withvideo records of the hand actions. Video clips of the animalsbehavior and the digitized spike trains were downloaded to the labcomputers and stored as both QuickTime files, and in Audio

Interchange file format (AIFF) for quantitative analyses of firingpatterns. As video and spike trains were recorded and digitizedsimultaneously, both datasets spanned the same time interval.Hence knowledge of the time code of each video frame in the clipprovided a precise way to locate the matching firing patterns.Similarly, measurements of the timing of spikes with respect to theonset of the audio data sample placed each spike in a preciselydesignated video frame.

The spike trains of each neuron were analyzed using customdesigned software (Sherwood et al. 2006) written in Igor Pro 5(WaveMetrics). Neural responses during the task were screened ini-tially with burst analysis graphs (Fig. 4) that provide a continuousrecord of neural and behavioral events within each video clip. Thistechnique automatically detects periods of high neural activity inwhich the firing rate exceeds one SD above the mean rate per 100-msbin (white burst threshold trace); pulses on the green burst tracemark the mean firing rate and duration of these epochs. Reversecorrelation of the burst onset, peak, and end times with the matchingvideo images of the monkeys behavior highlighted the behaviors towhich the neuron was most responsive.

The single trial responses depicted in burst analysis traces werecomplemented by more standard multi-trial analyses. Spike rastersand peristimulus time histograms (PSTHs), aligned to the video frame

33:59:20 33:43:25 33:27:0935:13:23

3 4 12

03:41:29 05:35:01 00:28:07 00:10:14

4 3 21

29:34:06 30:33:1429:20:11 30:23:15

3 4 1 2

49:24:06 47:54:1749:17:19 50:08:16A

B

C

D

Right rectangle Large round Small round Left rectangle

4 3 2 1

AA A

FIG. 2. Video images from sample trials illustrating the knob arrangement and grasp postures used by monkey H17094 during the period of study. Large

numbers in the bottom left cornerof each image denote the knob position on the shape box. Note the similarity in hand postures across days and sessions. A: unitH17094-10-2.4 : The rectangles were placed at the outer margins (knobs 1 and 4), and the round knobs in the center. The corresponding neural responses fromare shown in Figs. 4 and 5; letters designate the matching trials in the burst analysis traces in Fig. 4. B: unit H17094-70-4.2: The round knobs were placed atthe outer margins, and the rectangles in the center. C: unit H17094-129-2. D: unit H17094-131-3. Round and rectangular knobs were alternated on the shapebox. Spike density plots and average firing rate graphs for the neurons in BD are shown in Fig. 6, BD. Images in A and B were captured in Hi8, those in Cand D in DV.

3710 E. P. GARDNER, K. S. BABU, S. GHOSH, A. SHERWOOD, AND J. CHEN



5/24

onset times of hand contact with the knob, were used to measure theconsistency and reliability of neural responses (Fig. 5). PSTHs weresmoothed to create spike density functions of firing rate as a functionof time using a 50-ms window displaced every 10 ms (de Lafuenteand Romo 2006).

The task-evoked spike trains were subdivided into stages linked tothe actions viewed in the video records, allowing us to calculate meanfiring rates per stage for each trial. Trials were grouped by knob andreach style to compute average firing rate profiles for each knob(Fig. 6), and for statistical analyses. A repeated-measures ANOVAmodel (StatView, SAS Institute) analyzed whether there wassignificant modulation of firing rates across the seven task stagesand the pretrial interval for each trial (F-test, P 0.05), and

whether the knob identity (size, shape and/or location) was asignificant between subjects factor.

We also evaluated neural tuning to specific objects using a varietyof selectivity indices. Object selectivity among the four test knobs wasevaluated during each task stage using the preference index (PI)

PI 100 * [(n (K1 K2 . . . Kn)/Kmax)/(n 1)] (1)

where n the number of objects tested, Ki the response to object

i, and Kmax the strongest response during a particular action. The

PI has been used to measure object selectivity of neurons in areaAIP and in frontal motor areas (Murata et al. 2000; Raos et al.2004, 2006; Umilta et al. 2007) and to evaluate hand musclesynergies in monkeys during similar tasks (Brochier et al. 2004).

5 mm

5 mm

5 mm

Tracks

Area 4

Area 2

Area 5

Area 7b/AIP

Monkey H17094

Monkey N18588Right Hemisphere

Monkey N18588Left Hemisphere

FIG. 3. Cortical recording sites in monkeys N18588 and H17094. Colored markers denote sites where task-related neurons were recorded; dots, electrode entry

points of other tracks. Recordings in the right hemisphere of N18588 were performed using 3- or 4-electrode arrays centered at the marked sites.




6/24

As in these other neural studies, we also computed the differentialPI (dPI) for each task stage

dPI 100 * (max min)/(maxmin) (2)

where max and min are the maximum and minimum responses perstage regardless of which object evoked that activity.

The effects of object size, shape, and location on neural responseswere assessed using similar indices. The size index compared re-sponses to the two round knobs

Size Index 100 * (small large)/(small large) (3)

The place index compared responses to the two rectangularknobs

Place Index 100 * (right left)/(right left) (4)

Positive values indicate preferences for the small round and rightrectangular knobs; negative values denote preferences for the largeround and left rectangular knobs.

To test the role of object shape primitives, we pooled trials of thelarge and small spheres (round) and the left and right rectangularblocks (rectangle)

Shape Index 100 * (round rectangle)/(round rectangle) (5)

R E S U L T S

This report describes the prehension responses of 83 neuronsrecorded in the hand representation of area 5 in the posteriorparietal cortex of two monkeys (left hemisphere, n 52; righthemisphere, n 31). These cells were tested systematicallywith four objects and comprise a subset of the populationdescribed in our earlier studies of PPC (Gardner et al. 2007a).All of the neurons were studied over a minimum of 50 trials

and showed highly significant modulation of firing rates acrossthe task stages (P 0.0001). The mean number of trials perneuron in monkey H17094 was 123.6 7.8, evenly distributedamong the four knobs (mean 30.9 1.1 trial/knob). Themean number of trials per neuron in monkey N18588 was89.6 4.4; mean total trial/knob ranged from 16.6 1.5 forthe small sphere to 25.9 0.8 for the large sphere with anaverage per knob of 22.4 0.7 trials. We also analyzedresponses of 32 neurons in adjacent regions of S-I cortex andarea AIP/7b in these animals (Table 1).

Both animals performed the task using a continuous se-quence of hand movements from approach through lift. Objectacquisition and manipulatory actions were similar in timecourse to the coordination of grip and load forces reported in

0

50

100

150

200

250

Spikespers

ec

0

50

100

150

200

250

Spikespersec

Clip 1 00:49:16:04 Unit H17094-10-2.4

Large round Right rectangle

120 125 130 135 140 145 150

Time in clip (sec)

Knob

Task stage

Burst

Burst threshold

Small round

Right

rectangle Large round

90 95 100 105 110 115 120

Task with right hand

A

J

A

AD E

H

L

B C B C D E

K B C

F G

FIG. 4. Burst analysis graphs of continuous neural and behavioral activity recorded from unit H17094-10-2.4 during a 60-s period. The spike train was binnedin 100-ms intervals (blue graph) to compute continuous firing rates. Yellow task stage trace: each stepped yellow pyramid marks a single trial. Upward deflectionsdenote the start of stages 14 (approach through lift); downward deflections mark the onset of stages 58 (hold through release). Neural responses began atapproach and peaked at contact. The animal did not relax the grasp on some trials of the right rectangle (green H, L) and the large sphere (red E) but simplylifted the knob again; neural responses were much weaker on these incomplete trials. Orange knob trace: downward pulses that span the contact through lower

stages indicate the knob location on the shape box and the duration of hand contact. The pulse amplitude is proportional to the knob distance from the left edgeof the box. White burst threshold trace: firing rate set 1 SD above the mean rate during the entire 2.5-min video clip. Green burst trace: upward pulses markperiods when continuous firing rates exceeded the burst threshold; the burst pulse amplitude indicates the mean firing rate during this interval. The burst tracehas been displaced by 90 spike/s to improve readability. Images keyed to the grasp stage of specific trials are shown in Fig. 2.




7/24

studies of precision grip in humans (Johansson 1996; Johans-son and Cole 1992; Westling and Johansson 1984, 1987). Theanimals differed in the hand tested and the grasp postures usedto acquire objects.

Area 5 neurons respond most vigorously to acquisition ofdiverse objects

Monkey H17094 was trained to perform the task with theright hand as recordings were made in the left hemisphere. Heapproached the knobs from above, targeting the right side ofeach knob (Fig. 1B). The objects were grasped with the palmplaced to the right, and the fingers aligned parallel to the frontalplane (Fig. 1C). Although the left and right rectangles wereoften positioned at opposite ends of the shape box, they weregrasped with the same hand posture with the digits extended.The round knobs were grasped with greater flexion of the digitswith the ulnar fingers placed below the knob. The small roundknob was usually clasped at its base in a semi-precision gripbetween the thumb and digits 2 and 3 and rarely contacted thepalm; the large sphere contacted all 5 digits and the palm pads

(Fig. 2). Both spheres were scooped up during lift with the loadforce concentrated on the digit shafts; the tight grip and flexedhand posture was maintained through the hold stage. The handpostures used in the initial tracks (Fig. 2A) were preserved almostunchanged throughout the period of study regardless of the actuallocation of the knobs on the shape box.

Figure 4 illustrates continuous spike trains and markers of

the task stages recorded over a 60-s period during the sessionshown in Fig. 2A. Each of the stepped pyramids in the yellowtask stage trace denotes a trial of the prehension task. Theneural responses began at or before the onset of approach(initial step in the task stage trace). The firing rate increasedabruptly as the animal projected the arm toward the targetobject with peak activity at contact, and then subsided duringlift (highest pulse in the task trace). The spike trains decayed tolow levels during holding, and returned to or below baseline inthe late task stages as grasp was relaxed. Spontaneous activitybetween trials was weak or absent.

Although the temporal pattern of the neural response wassimilar on each trial, the peak firing rate and response durationdiffered depending on the object tested (orange knob trace), the

0

20

40

60

80

100

0

20

40

60

80

100

0

20

40

60

80

100

0

20

40

60

80

100

0

20

40

60

80

100

0

20

40

60

80

100

0

20

40

60

80

100

0

20

40

60

80

100

-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000

-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000Time re contact (ms)

-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000Time re contact (ms)

A Left rectangle (Knob 1) C Small round (Knob 2)

B Right rectangle (Knob 4) D Large round (Knob 3)

II

II

II

II

II

I

II

III

III

IIII

IIII

III

III

IIIIIIIIIIIIIIIII

II

IIIIII

II

II

III

I I

II

IIIIII

II

II

III

I I

II

II

IIII

III

II

II

I I

II

II

II

II

II

II

II

III

II

II

II

II

II

II

II

III

I

II

II

II

II

II

II

II

I

II

II

II

II

II

II

II

III IIIIIIIIIIIIIIIIII I I II I I I I I I I I I I I I I II II I I I I I I I I I I II II IIIIIIIIIIIIIIII IIII

IIIII II I I I I IIIIIIIIIIIIIIIIIIII I I I I I II II IIIIII I I II

I IIIIIIIIIIII I I I I IIIIIIIIIIIIIIII I I I

IIIII III I IIIIIIIIIIIIIIII IIII I I I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII III I II

IIIIIII III II IIIIIII I IIIIIIIIIIIIIIIIIIIII I I I I I II II II

I I I I I I I I II II IIIIIIIIIIIIII I I II I I I I I I I I II IIII IIIIII I I I I I I I I I I I I I I

I I I II II IIIIII I I II III III II II IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII I I I II IIIIIIII IIIIIIIIIIIIIIIIIIII I I IIIIIIIIIIIIIIIIIIIIII I I IIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIII II I I I I II I I I

I I IIIIIIIIIIIIIIIIII I I I I III III III I I I II II I IIIIIIIIIIIIIIIIIIIII II I III III II I I I I I II I I IIIIIIIIIIIIIIIIIIII II II I I I I I I I

I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIII I II I I I I I I II IIIIIIIIIIIIIIIII I I I I I IIIIIIIIIIIIIIIIIIIIIIIIIIII IIIII I II I I II IIIIIIIIIIIIIIIIIIIII I I I I I I I I IIIIIIIIIIIIIIIIIIII I

I I I I IIIIIIIIIIIIIIIIIIII I IIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIII I II I I

I I I I I I I I I I I I II I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII II I I I I I I I I II I I II I I I I I I I I I I I I IIIIIIIII II I I I I I I I I II I II II I I I I I I I I I I I II I I I III I II IIIIIIIIIIII IIIIIIIII II III II II I I I I

II I IIIIIIIIIIIIIIII I I I II III IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII I I II I II III IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII I I I I I II IIIIIIIIIIIII II I II II IIIII II II I I I I II I I I I I I I I I I II I I I I

17 trials

II

I

II

II

II

II

II

II

II

II

II

IIIIII

II I

IIIIIIIIIIIIIIIII

II

II

II

II

II

II

II

II I

II

II

II

II

II

II

II

II I

II

III

II

II

II

III

II I

II

II

II

II

II

II

II

III

II

II

II

II

II

II

II

II

II

I

II

II

II

II

I

II

I

II

II

II

II

I II II I I I I IIII IIIIIIIIIIIIIIIIIII II IIIIIII I I I II I III I II I I II II I II IIII II III I II

I I I I IIIIII IIIIIIIIIIIIIIIIIIIIIIIIII IIII I I I I I I I I I I I I I I I I I I II I I I I III I I I II I I I I I I I I

II I II II IIII II I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII I II I I I I I I I I II I I I I IIIII IIIIIIIIIIIIIII IIIIIIIII III I

IIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIII I I I I I I I I IIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIII

III IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII II II IIIII I IIIIIIIIII I I I III IIIIIIIIIIIIIIIII II I I II I I I I I I I I

I IIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII II I I II I I I I I I II I I I II III III II II I I I I IIIII

I I I IIIIIIIIIIIIIIIIIIII IIIIIIIII II I I I I II I I II I III IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII II II I II I I I

I I I I I I I I III I I I IIIIII IIIIIIIIIIIIIIIII IIII III I I I I I II I I I I I I I I II I I I I IIII I I I I I I I I I I III I I I I I I I

III II I I I I I I I I I II I II II IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII I III I I I I I I II III III IIIIII IIII IIIIIIIIIIIIIII III I I II I II I II IIIIII IIII IIIIIIIIIIIIIII IIIIIIIIII II I I I I I I I I

I I I IIIII IIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIII II III I I I I I I I I I I I I II I I II II II II I IIII I I I I I I I I I IIII II III I I I I I I I I I I I I II I I II IIIIIII IIII I I I I I I I I I I I I II I I II IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIII

I III I IIIIIIIIIIIIIIIIIIIIIIIIIIIII IIII I I II II III I II I I II I II I I I I I II I I II III I I II II I I I I III I I I I I I IIIIII I I II III II II IIIIIII II II I I III I I I II II I I I I I I III I I I I I II I II II I II II II I IIIIIIIIIIIIIIIIIIIIIIIIIII I II I IIII II IIII II I II II IIIIII IIIIII I IIIIIIIIIIIIIIIIIIIIIIIIIII II I I IIII II I I I I I I I I II II I I I III I I

I I I II IIIIIIIIIIIII IIIIIII III I I I I I I II IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII III I I

II III I I I II I IIIIIIIIIIIIIIIIIIIIIIIIIII II I I I I I I I I I II I I III I

26 trials

II

I

III

II

II

II

II

II

II

III

II

II

II

II

II

IIII

IIIIIIIIIIIII

II

II

III

III

II

II

II

I

II

II

I III

II

II

II

II

I

II

II

I III

II

II

II

II

I

II

II

I II

II

II

II

II

II

II

II

I II

II

II

II

III

I

II

I

I I

II

II

II

II

I

II

I

I I

II

I

II

II

I

I IIIIIIIIIIIIIIIIIIIII I I IIIIIIIIIIII I I II IIIIIIIIIIIIIIII IIII II I I I I II I I I I I II II I IIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIII II

I I I I IIIIIIIIIIII I I III II I I I I I I I I I III I I III I I I I III I I II I I I II I I IIIIIIIIIIIIIIIIII II I I I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII IIIIII I

III IIIIII I I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII II I I I I IIIIIIIIIIIIIIIIIIII I I

I I I III II IIIII I II III IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIII I I IIIIIIIIIIII I

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII II IIII I I II I II III I I I I I I I I I III

I I II I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIII III IIIII I I I III I II I II I

IIIIII IIII II IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII II I I I I I IIIII II I I I I I

I II I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIII I I III I II I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII I I II I I I I I I I I

I I II I I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIII IIIIII I I I III II IIIII I

I I I II I II IIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII II II I I I I I I I I IIIIIIIIIII I III I I III IIII IIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII II I I II II I I II I II

I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII II I I II I I III I III III I I I I I IIII I II I III I I I I I I I I IIIIIII I I II I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII III

I II I I I I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII I I I I III I I II I IIII IIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIII I I I I III I I II I IIII IIIIII IIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII II I I I I

II II I I I I I IIIII II I I I I II I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIII I IIII I

35 trials

II

I

I

II

I

II

II

I II

II

IIIII

II

I

IIII

IIIIIIIIIIIII

II

II

I II

II

II

II

II

II

IIII

I II

II

II

III

II

I

III

I

I II

II

II

III

II

I

II

II

III

II

II

II

II

II

II

II

III

II

II

II

II

II

II

II

III

II

II

I

II

II

II

I

I II

II

I

I

I

II IIII I I IIII II I I I I I I III IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIII II II III II IIII III II II I I I I I I I I I I I I I I I I I I I I II I II III IIIIIIIIIIII

I II I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII II I I I I II II I I I II I I III I I I I I I I I I I I I II I II III III II IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII III

IIII IIIIIII I I III I II I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII I I II I I I I I I I I I I I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII III II I III IIII II III I I I I I I I III IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIII II III IIIII

II I I I II IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIII IIIIIIIIII IIII II IIIIIIIIIIIIIIIIIIIIIIIIIIII II III III I I I I I I I I I I I I III I I I I III I I I II I I IIIIIIIIII I I II IIII IIIIIIIIIIIIIIII IIIIIIIIIIIII IIIII IIIII I I IIIIII IIIIIIIIIIIIIIIIIII III II III III IIII II II II I II I I I I II

I II I I II II IIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIII II I I II I I IIII I II I IIIIIIIIIIIIIIII IIII

II II IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII II III I I II II I I I I I I I IIII III IIII III II III I IIII I I I I I I II IIIIIIII IIIIIIIII III I IIII I I I I I I I I I II I I I I I I II III I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIII I II

I I I II IIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIII I III II III III I I II II I I I I

I I I II III IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII III III I I II I I III I I I I I I I I I I I I I I I I I I I I IIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII I I I I I IIII I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIII II I II II IIII I I I I I III I I II II II I I IIIIIII IIIIIIIIIIIIIIIIIIIIIIIII III II II I

III I III IIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII III II II II I IIII I IIII I I III I I II I IIIIIIIIIIIIIIIIIII I I I I I I

I I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIII II I II I I II I I IIIIIII IIII II III I I I I I I I I I I I II I I I I I I I I I I I I III II IIIIIIII III II IIII II I II I II I I I I I I I II I I I II I I I I I III III IIIIIIIIIIIIIIIIIIIIIIIIIIIIII I I I I I I I I I I I I I

I III I I II IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII I I I II IIII I IIII III IIIIIIIIIIII IIII IIIII IIIII III I

I I I I I I II I I II II III IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII II II III II II II II II I I I I I I I I I I I I I I II I I II I I

28 trials

l l l l l l l l l l

l l l l l l l l l l

l l l l l l l l l l

l l l l l l l l l l

Spikespersec

Spikespersec

Spikespersec

Spikespersec

Approach

Contact

Grasp

Previous

Lift

Lower

Relax

Release

Next

Hold

Unit H17094-10-2.4

FIG. 5. Rasters of the 1st 17 trials of each knob aligned to hand contact and spike density functions averaging all knob trials for the neuron shown in Figs.2 and 4. Colored markers on the raster show the onset times of the task stages; markers above the spike density graphs show the mean onset times of the stagesaveraged across the trials. Although all 4 knobs evoked the same peak firing rates at contact, the spike trains were of longer duration for the round knobs,persisting through the hold stage. Spike bursts near the right margin represent responses on the next trial (gold marker).




8/24

speed of acquisition, and the previous history of stimulation.The longest duration spike trains in these records occurredwhen the large round knob was tested (red AC). Responses to

successive trials of the same object decreased in peak rateregardless of shape, particularly if the animal lifted the objectwithout prior relaxation of the grip (red E, green F, J).

0

25

50

75

100

125

150

-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000

Spikespersec

Left Hemisphere, Right hand

N = 180 trials

0

20

40

60

80

100

-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000

Spikespersec

small round

left rectanglelarge round

right rectangle

N = 226 trials

0

20

40

60

80

100

0 1 2 3 4 5 6 7

Spikespersec

left rectangle

small round

large round

right rectangle

0

25

50

75

100

125

150

-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000

Spikespersec

0

25

50

75

100

125

150

0 1 2 3 4 5 6 7

Spikespersec

H17094 70-4.2

N = 79 trialsP = 0.042small round

left rectangle

large round

right rectangle

0

25

50

75

100

125

150

-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000

Spikespersec

Time re contact (ms)

0

20

40

60

80

100

0 1 2 3 4 5 6 7

Spikespersec

Task stage

H17094 131-3.1N = 217 trials,P = 0.0055

small round

left rectangle

large round

right rectangle

H17094-10-1.4

H17094129-2.1 P = 0.279

P = 0.0001

l l l l l l l l l

l l l l l l l l l

A

B

C

D l l l l l l l l l

0

25

50

75

100

125

0 1 2 3 4 5 6 7

Spikespersec

l l l l l l l l l

FIG. 6. Superimposed spike density functions (left) and average firing rates (SE) per task stage (right) for 4 neurons illustrating the range of selectivity for specificobjects observed in monkey H17094. Stage 0 indicates the pretrial interval. Knob preferences are generally highest in stages 24; in these examples, the round knobsevoked the highest firing rates. P values in the right panels indicate the selectivity for knobs as determined by repeated-measures ANOVA analyses. The order of listingof the knobs in the keys denotes their left-to-right locations on the shape box. Images of the hand postures used to grasp each of the knobs are shown in Fig. 2.




9/24


10/24

2. All of the neurons analyzed responded to the four testobjects, but the amplitude and/or duration of the spike trainvaried between neurons and objects. Distinctions in firing ratesbetween objects, when present, usually occurred after contactas the hand was moved over the object surface and duringgrasp and lift actions.

The neuron in Fig. 6A was one of the most highly selective

for shape in the population analyzed. It was recorded on thesame track as the neuron in Figs. 4 and 5; it shared a similarstrong preference for the large round knob after contact andresponded least to the left rectangle. Although the firing ratesevoked by each object were nearly indistinguishable duringapproach, they diverged after contact. The greatest spread infiring rates occurred during lift, when view of the knob waspartially obscured by the animals hand. The differences inmean firing rates between knobs across the seven task stageswere highly significant [F(3,176) 11.9, P 0.0001], as wereinteractions between firing rate per stage and knobs [ F(7,21)4.81, P 0.0001].

The spike trains of the neuron in Fig. 6B were modulated by

the size, shape, and location of the knobs during the grasp andlift stages but not during other task actions. It responded mostvigorously to the small round knob during the preferred stagesand least to the large round knob. The difference in mean firingrates was significant [F(3,73) 2.87, P 0.042]. The smallsphere was the lightest object and was grasped with the mostflexed hand posture (Fig. 2B). It was placed at the most mediallocation on the shape box (knob 1) and was furthest from thetest hand; the large sphere was located at the most lateralposition (knob 4). Responses to the two rectangular knobsduring grasping were intermediate between the round knobs,reflecting their central locations on the shape box.

The neurons in Fig. 6, Cand D, were typical of the majorityof cells, showing only subtle distinctions in firing patterns

between knobs. Representation by the spike train of handactions during the task was stronger than the distinctionsbetween grasped objects. Responses to three of the four knobsoverlapped and did not differ significantly (P 0.05). Theremaining knob evoked weaker responses than the other three;in these examples, knob 4 placed lateral to the animals handproduced the lowest firing rates (Fig. 2, C and D). The spreadin firing rates was clearest during stages 3 and 4 and wereminimal during approach before the objects were touched.

Weak object selectivity was observed in both animals, andwas independent of the laterality of the recording site in thebrain. Figure 7 shows a 33-s epoch of recordings made simul-taneously on three electrodes placed in the right hemisphere of

monkey N18588, which was trained to use its left hand. Likethe continuous responses shown in Fig. 4, the four neuronsillustrated began their responses at the onset of approach andfired maximally at contact and during grasping of each knob.Their firing rates returned to baseline at the start of lift.Synchronized responses occurred across all of the recordingsites during the initial task stages but not in later stages norduring other movements such as licking the juice tube onreward delivery (magenta trace). The neurons distinguishedwhether acquisition was completed (A, C, E, G, and J) or if thetrial was aborted after contact (B). Similarly, lift of a knobwithout prior relaxation of grasp (F, H) evoked weaker re-sponses than trials in which all of the task actions wereperformed (E, G).

Superimposed spike density functions and average firing rategraphs are shown in Fig. 8, A and B, for two of these neurons,together with similar records from another pair of cells from adifferent recording session (Fig. 8, C and D). Like the neuronsfrom the left hemisphere illustrated in Fig. 6, the spike densityplots evoked by all four objects were similar in time coursewith peak activity at contact or during grasping, and a precip-

itous drop in firing during lift. The most striking feature of thedata are the overlap in spike trains evoked by the four knobs,particularly during approach. Only one of these neuronsshowed significant differences in firing rates (Fig. 8 D, P 0.05); it responded best to the large sphere and least to thesmall one. The weakest responses in three of the cells occurredon trials testing the left rectangle that was placed at the mostlateral position in the workspace; the fourth neuron respondedleast to the small round knob. The varying neuronal responsesare particularly striking because they were obtained during theexact same trials on different electrodes.

The average firing rates per stage diverged more sharplythan the PSTHs apparently because of differences in stage

duration between objects (Table 2). The small round knobevoked high firing rates in stages 1 and 2; firing rates duringstatic grasp and lift were generally lower than for the otherobjects. The small sphere required the longest acquisitiontimes; the mean duration of stages 24 was 120 ms longer forthis knob than for the other three objects. The large round knobevoked the highest firing rates during lift and hold actions butwas retained in the hold posture for the shortest interval. Thehold stage for the large sphere was 120 ms shorter than that ofthe neighboring left rectangle and 300 ms less than the holdstage of the small sphere and the medial rectangle.

Images of the grasp posture used by this animal suggest thatthe timing differences may have resulted from the hand kine-matics used to grasp these objects (Fig. 9). This monkey did

not fully enclose the large knobs in his hand. Instead he placedhis fingers below the two rectangles and the large sphere andscooped them up with the wrist supinated. In contrast, the smallsphere was usually clasped in a semi-precision grip betweenthe thumb and digits 2 and 3 with hand postures similar tothose used by monkey H17094. The formation of the precisiongrip resulted in longer duration contact, grasp, and hold stagesfor the small round knob (Table 2).

We also recorded responses to prehension in the left hemi-sphere of this animal with similar results (Fig. 10, A and B).Only one of six neurons analyzed in area 5 of the left hemi-sphere showed significant knob preferences; the small roundknob was the preferred object. Neurons in adjacent regions of

S-I and area AIP/7b showed greater object selectivity thanthose in area 5; examples are provided in Fig. 10, Cand F, forcompleteness. However, the small number of neurons sampledprecludes definitive conclusions concerning the prevalence ofobject selectivity in these regions (see Table 1).

Population distribution of object selectivity in area 5

The trends observed in the data from individual neurons inarea 5 were clearly evident when responses of all cells studiedwere quantified statistically. Mean firing rates per stage werecomputed for each neuron on a trial-by-trial basis, and ana-lyzed using repeated-measures ANOVA protocols with knobtype as a between subjects factor. One-third of the neurons




11/24

tested (28/83) showed statistically significant differences infiring rates (P 0.05) as a function of knob tested during theseven task stages (Table 1). Thirty-six percent of neuronsshowed significant interactions between firing rate per stageand knob, indicating preferences for particular objects duringsome actions but not during others. Object selectivity washigher in monkey H17094 than in N18588.

Despite differences in grasp style, the neurons recorded inboth animals showed similar preferences for particular actions,objects, and places within the workspace during task perfor-mance (Fig. 11). Peak responses in the population occurredmost frequently at contact and during grasp (stages 2 and 3)when the hand first engaged directly with the knobs (Fig. 11C).The small round knob was the most preferred object in both

animals, evoking maximum responses in 43% of the popula-tion (Fig. 11A: H17094 20/46 neurons; N18588 13/31neurons). This knob was placed at the midline in 13/13 cells inthe right hemisphere of monkey N18588, in 8/20 cells inmonkey H17094, and at the most medial location in the re-maining 12 cells in H17094. Of the 28 neurons that showedstatistically significant effects of object on firing rates, 13(46%) preferred the small sphere, 6 (21%) preferred the largesphere, 6 preferred the medial rectangle, and 3 favored thelateral rectangle.

The medial rectangle was the next most preferred object in bothanimals. The left rectangle was more likely to evoke peak activitywhen tested with the right hand, and the right rectangle wasfavored in trials testing the left hand (Fig. 11A). The object placed

0 5 10 15 20 25 30Time in clip (sec)

0 5 10 15 20 25 30

0

25

50

75

100

0

25

50

75

100

Spikespers

ec

Right Hemisphere, Left Hand N18588 Track 313 Clip 7

A

B

C

D

0

25

50

75

100

0

25

50

75

100

Spikespersec

0

25

50

75

100

0

25

50

75

100

Spikespersec

0 5 10 15 20 25 30

0 5 10 15 20 25 30

0

50

100

150

0

50

100

150

Spikespersec

Time in clip (sec)

Knob

Task stage

Burst

Burst threshold

Knob

Task stage

Burst

Burst threshold

Tongue

Tongue

Small round Right rectangle Left rectangle

CA B D E G H JF

IIII II IIIIIIIIIIII III IIII II IIIIIIIIIIIIIIIIIIIII I III IIIIIIIIIIIIIIII III I IIIII IIIIIII IIIIIIIIIII I I I IIII IIIIII II I IIII IIIIIIIIIIIIIIIIIII II I I I I IIIII I III IIIIIIIIIIIIIIIIIIIIIIII I IIII IIIIII IIIIIIIIIIIIIIIIIIIII IIIIIIII IIIIIIIIIIII III I I II IIIIIIIII IIIIII II IIIIIIIIIIIIIIIIIIIIIIIIII I

IIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIII I II IIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIII IIIII IIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIII IIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIII IIIIIIIIIII I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIII IIII IIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIII IIIIIII I IIIIII IIIIIIIII I IIIIIIIIIIIIIIIIIIIIIII IIIIIII II IIIII III IIIIII IIIII IIIIIIIIIIIIIIII IIIIII IIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII II IIII IIIIIIIIIIIIIIIIIIIIII IIIIII

FIG. 7. Burst analysis records from 4 simultaneously recorded neurons in monkey 18588. Same format for burst traces as in Fig. 5; the colored spike rastersat the top correspond to the matching colored histograms in the burst traces. The neurons show coincident firing during object acquisition but not during otheractions. The magenta tongue trace shows actions of the tongue as the animal licked the juice tube for reward. The 1st upward deflection indicates mouth openingand tongue protrusion, the 2nd upward deflection indicates full extension of the tongue; downward deflection marks retraction of the tongue into the mouth. Onlythe neuron in D responded to tongue movements. A and B: units 313_12.1 (A) and 313_1-2.4 (B) recorded on electrode 1. C: unit 313_2-2.2 recorded onelectrode 2. D: unit 313_4-2.3 recorded on electrode 4. Images of the hand actions in these records are provided in Fig. 9, A and B.




12/24

at the most lateral position was the least likely to evoke maximumfiring (Fig. 11B: knob 4 in H17094 and knob 1 in N18588). As aresult, the large round knob and the lateral rectangle were the least

preferred objects in terms of peak firing rates, in part because theywere located in front of or lateral to the test arm, and were easiestto regrasp on subsequent trials.

0

20

40

60

80

-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000

Spikespersec

Time re contact (ms)

0

25

50

75

100

125

150

-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000

Spikespersec

0

20

40

60

80

-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000

Spikespersec

0

20

40

60

80

-1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000

Spikespersec

Right Hemisphere, Left hand

Task stage

small round

left rectangle

large round

right rectangle

N18588 307_2-1.2

N18588 313_1-2.1

N18588 313_2-2.2

small round

left rectangle

large round

right rectangle

small round

left rectangle

large round

right rectangle

small round

left rectangle

large round

right rectangle

l l l ll l l l l

l l l l l l l l lA

B

C

D

l l l l l l l l l

0

15

30

45

60

0 1 2 3 4 5 6 7

Spikespersec

N = 89 trialsP = 0.74

0

20

40

60

80

0 1 2 3 4 5 6 7

Spikespersec

N = 89 trials

P = 0.078

0

25

50

75

100

125

150

0 1 2 3 4 5 6 7

Spikespersec

N = 90 trials,P = 0.173

l l l ll l l l lN18588 307_2-1.3

0

20

40

60

80

0 1 2 3 4 5 6 7

Spikespersec

N = 90 trialsP = 0.0092

FIG. 8. Superimposed spike density functions (left) and average firing rate graphs (right) illustrating the range of object selectivity observed in monkeyN18588; same format as Fig. 6. A and B: simultaneously recorded neurons from 2 different electrodes on track 313; burst analysis records for these cells areshown in Fig. 7, A and C. C and D: neurons recorded simultaneously by the same electrode on track 307. Images captured from the digital video recordings forthese cells are shown in Fig. 9, C and D. Only the neuron in D showed a significant knob preference, responding best to the large sphere.




13/24

Selectivity among the four objects during each stage wasalso assessed using the PI and dPI as described in METHODS.These metrics do not make assumptions about the specificcharacteristics of objects that yield maximum and minimumresponses; they have been used in other studies to assess objectselectivity of neurons recorded in areas AIP, F5, F2vr, and M1during similar prehension tasks (Murata et al. 2000; Raos et al.2004, 2006; Umilta et al. 2007). Figure 12 plots PI and dPIvalues for the neurons illustrated in Figs. 6 and 8; both indices

yielded nearly identical results. Although the firing rates ofthese neurons were highest in stages 13, the index values werelowest during these acquisition stages. The highest values of PIand dPI were recorded in stages 4 and 5 (lift and hold) whenfiring rates dropped back to baseline. PI and dPI values some-times exceeded 50 in stage 5 when neurons fired at low rates tothe most preferred knob, and were nearly silent for the leasteffective one (Fig. 12, A and E). These findings suggest thatmetrics such as the PI and dPI exaggerate differences in firingrates when neural responses are weak and minimize them whenneurons are most responsive.

We used similar techniques to measure the relative influenceof object size, shape, and location on mean firing rates perstage. To analyze the effect of object size, we compared

responses to the small and large spheres. Twenty-one percentof area 5 neurons (17/83) showed significant effects of objectsize on task responses in repeated-measures ANOVA analyses(Table 1); size selectivity was higher in monkey H17094.

The size index allowed us to compare which knob waspreferred during specific task stages. The small round knobwas generally favored during object acquisition but rarelyduring lift and hold actions (black curves, Fig. 13). The largeround knob was the most preferred object in stage 5 in both

animals (H17094 16/49 cells; N18588 23/31 cells), andthe small round knob was the least preferred in the holdstage.

The knob location in the workspace had weaker effects onfiring rates than the object size, particularly in monkey

H17094. Sixteen percent of neurons recorded in area 5(13/83) showed significant differences in firing rates be-tween the left and right rectangles (Table 1). Likewise, theplace index comparing the two rectangles (gray curves, Fig.13) was smaller than the size index, particularly duringobject manipulation (stages 35).

The indices obtained for the individual neurons shown inFigs. 12 and 13 are typical of the population studied. Meanvalues for the PI, dPI, size and place indices are plotted for

38:06:29 37:59:1638:35:23 38:52:17

38:18:29 40:28:02 38:31:16 39:57:23

Right rectangle Small round Large round Left rectangle

A

B

C

D

22:07:03

22:31:26

22:03:03

22:49:13

22:53:09 22:14:26

23:40:03 22:27:20

C E

J

A

G

FIG. 9. Digital video images of the grasp postures used for the 4 knobs by monkey N18588. Spike density plots and average firing rate graphs for these neuronsare shown in Fig. 8. A and B: image captures from clip 7 of track 313. Letters designate the corresponding trials in the burst analysis records in Fig. 7. C and

D: image captures from clip 5 of track 307. Note the similarity in hand postures across days and sessions. Only the small round knob is fully grasped in the hand;the others are held with the digits extended.




14/24

the entire population of 83 neurons in Fig. 14, B and C. As inthe individual examples, the population PI and dPI values werelowest during approach and contact when firing rates werehigh. Indeed the difference in firing rates between the mostand least preferred objects during hand preshaping andinitial contact was slightly lower than in the pretrial interval.Mean preference indices increased progressively during thegrasp, lift, and hold stages as the hand interacted directlywith the object. These findings suggest that tactile feedbackfrom the hand about intrinsic object properties modulatedthe firing rates of area 5 neurons during manipulatoryactions.

The distribution of PI values in the population during eachof the task stages is illustrated in Fig. 14A by their respec-tive cumulative PIs. The smallest distinction between ob-jects occurred during the approach and contact stages, withmedian PI values 20; 90% of the neurons tested had PIvalues 40 in these stages. As in the mean PI graphs, themedian PI increased progressively from grasp through lift,hold, and lower actions to 30; the 90th percentile in thelower stage yielded PI values 50.

The population size index reversed as the task progressed. Thesmall round knob was generally favored by higher firing ratesduring acquisition stages when the hand was preshaped and the

Task stage

Left rectangle

Large round

Small round

Right rectangle

Task stage

Left rectangle

Large round

Small round

Right rectangle

Left rectangle

Large round

Small round

Right rectangle

0

10

20

30

40

50

0 1 2 3 4 5 6 7

Spikespersec

N18588 94-7.2N = 119 trialsP = 0.26

PPC neurons S-I neurons

0

10

20

30

40

50

0 1 2 3 4 5 6 7

Spikespersec

N18588-32-4.3N = 59 trialsP = 0.22

0

10

20

30

40

0 1 2 3 4 5 6 7

Spikespersec

N18588-36-1.4N = 209 trialsP = 0.0049

Left rectangle

Large round

Small round

Right rectangle

0

10

20

30

40

50

60

0 1 2 3 4 5 6 7

Spikespersec

H17094-137-4.2N = 97 trialsP = 0.065

Left rectangle

Large round

Small round

Right rectangle

A D

N18588-25-10.3N = 120 trialsP < 0.0001

0

10

20

30

40

0 1 2 3 4 5 6 7

Spikespersec

0

10

20

30

40

50

60

0 1 2 3 4 5 6 7

Spikespersec

H17094 55-2N = 97 trialsP = 0.68

Left rectangle

Large round

Small round

Right rectangle

B E

C F

FIG. 10. Average firing rate graphs illustrating the range of object selectivity of neurons recorded in area 5 (A and B), area 7b/AIP (C), and S-I (F) of theright hemisphere in monkey N18588; same format as Fig. 6. The neurons in C and F showed significant preferences for the left (medial) rectangle; the area 5

neurons did not distinguish between knobs. Two neurons recorded in S-I in monkey H17094 are included for comparison (D and E).




15/24

object grasped. Object preferences shifted during lift and holdwhen the large round knob evoked stronger responses. The selec-tivity for the large sphere was most prominent in monkey N18588,which used distinctive grasp postures for the large and smallspheres.

The population place index indicates that the knob location inthe workspace had relatively weak effects on population firingrates. Although the medial rectangle evoked slightly higher firingrates during approach and contact, there was little difference inlater stages when firing rates evoked by both objects were low.

Population representation of grasped objects in area 5

To quantify population responses to individual objects,we compiled normalized response profiles for each knob.

Four sets of normalized firing rate graphs were computed foreach neuron by dividing the mean firing rates per stage toeach knob by the peak mean response of the neuron aver-aged across all trials during the session. The same valueswere used to normalize responses across neurons in aprevious report (Gardner et al. 2007a). Normalized firingrates for each knob were averaged across the population andplotted as a function of task stage in Fig. 15, left. The mostsalient feature of the population curves is the overlap infiring rates evoked by the four knobs during the task. Theirtemporal profiles of activity rose and fell in parallel, yield-ing maximum responses at contact. Differences in firingrates between knobs were small in comparison with thechanges in mean activity across the task stages.

Left rectangle Large Round Small round Right Rectangle0

10

20

30

40

50

%T

otalneurons

Left rectangle Small Round Large round Right Rectangle0

10

20

30

40

50

%T

otalneurons

H 17094: Right handA

B

Best Knob

N 18588: Left hand

Best task stage0 1 2 3 4 5 6 7

0

15

30

45

60

%T

otalneu

rons

Knob 1

Knob 2

Knob 3

Knob 4C

Best Knob

0 1 2 3 4 5 6 70

15

30

45

60

%T

otalneu

rons

Knob 1 Knob 2 Knob 3 Knob 40

10

20

30

40

50

Best Place

%T

otalne

urons

Knob 1

Knob 2

Knob 3

Knob 4

Best task stage

Left rectangle

Small round

Large Round

Right Rectangle

Knob 1 Knob 2 Knob 3 Knob 40

10

20

30

40

50

%T

otalneurons

Best Place

N = 31 neuronsN = 46 neurons

FIG. 11. Population distribution of the objects (A), locations (B), and task stages (C) that yielded the maximum firing rates per stage for the neurons studied.The knobs were shuffled in different recording sessions in H17094 when testing the right hand; they remained at the indicated places in the recordings from

N18588. The small sphere was the most preferred object in both animals; it was most likely to evoke peak responses in stage 2 (contact) and at the most mediallocations on the shape box.




16/24

Although the two animals used different hand postures toperform the task, their population response profiles displayed

similar contours (Fig. 15, B and C). The small round knobevoked the highest firing rates during object acquisition (stages13), followed by the medial rectangle (left rectangle whentesting the right hand and right rectangle when testing the lefthand). The medial rectangle evoked stronger responses thanthe lateral one during acquisition, but the responses to bothrectangles were indistinguishable during lift, hold, and lowerstages. The large round knob evoked weaker responses than thesmall one during acquisition, particularly in monkey N18588but produced the highest firing rates during lift and hold in thisanimal.

We also used a second normalization procedure to assessobject selectivity modeled on protocols used by other investi-gators studying prehension tasks in monkeys (Murata et al.

2000; Raos et al. 2004, 2006; Umilta et al. 2007). Here themean firing rate per stage for each object was divided by the

best responsethe highest firing rate obtained across allseven task stages and all four test objects (Fig. 11)andmultiplied by 100. The four objects were then ranked frombest to worst based on their relative firing rates during thebest stage. The normalized mean responses for each rankordered set were then averaged to compute the populationnormalized mean rate for the best, second best, third best,and worst knobs.

The rank ordered response profiles in the right column ofFig. 15 show similar selectivity among objects and theirassociated grasps in area 5 as has been previously demon-strated for neurons in areas F5, F2vr, and M1 (Raos et al.2004, 2006; Umilta et al. 2007). The spread in responses tothe four objects emerged during approach, as the hand was

B

Object Preference Index

H17094-70-4.2N = 77 trialsP = 0.042

H17094-10-2.4N = 105 trialsP = 0.0003

H17094-129-2.1N = 226 trialsP = 0.278

H17094-131-3.1N = 220 trialsP = 0.0055

N18588 313_1-2.1N = 89 trialsP = 0.743

N18588 313_2-2.2N = 88 trialsP = 0.078

N18588 307_2-1.2N = 89 trialsP = 0.173

N18588 307_2-1.3N = 95 trialsP = 0.0092

0

20

40

60

80

100

0 1 2 3 4 5 6 7

Prefere

nceIndex

PI

dPI

0

20

40

60

80

100

0 1 2 3 4 5 6 7

PreferenceIndex

0

20

40

60

80

100

0 1 2 3 4 5 6 7

PreferenceIndex

0

20

40

60

80

100

0 1 2 3 4 5 6 7

PreferenceIndex

Task stage

0

20

40

60

80

100

0 1 2 3 4 5 6 7

PreferenceIndex

0

20

40

60

80

100

0 1 2 3 4 5 6 7

Prefere

nceIndex

0

20

40

60

80

100

0 1 2 3 4 5 6 7

PreferenceIndex

Task stage

0

20

40

60

80

100

0 1 2 3 4 5 6 7

PreferenceIndex

A

C

DH

G

E

F

FIG. 12. Place index (PI) and differential place

index (dPI) values per stage for the neurons shownin Figs. 6 (AD) and 8 (EH). The indices arelowest in stage 2 when firing rates are maximal andrise in the late stages when firing rates are low. Pvalues indicate the selectivity for knobs as deter-mined by repeated-measures ANOVA analyses.




17/24

preshaped, and peaked at contact, when direct interactionbetween the hand and object first occurred. Distinctionsbetween objects persisted as grasp was secured but wereprogressively reduced during lift, hold, and lower stages.Although responses to the best knob remained distinctivethrough holding, the firing rates evoked by the other threeobjects merged during and after lift (stage 4). Note that thepeak responses in stage 2 were less than the maximumpossible rate because the best responses occurred in differ-ent task stages in individual neurons (Fig. 11C).

The distribution of rank object preferences provided in Table 3closely follows the distribution of mean firing rates illustrated in Fig.15, left. The representation of individual objects and their associated

grasps is distributed nonuniformly across the ensemble, with specificgroupings of neurons signaling distinctive objects and actions. Thesmall sphere was the best knob in 42% of the population, followed bythe medial rectangle (27%). These two objects each comprised thesecond best choice for 30% of area 5 cells. The lateral rectanglewas the least preferred object (36%). These findings indicate thatneurons in area 5 have distinctive object preferences that are influ-enced by the object size, shape, and location in the workspace.

D I S C U S S I O N

Hand manipulation neurons that respond to object graspingwere first identified in PPC areas 5 and 7 by Mountcastle and

FIG. 13. Size index (black) and place index

(gray) per stage for the neurons shown in Figs. 6(AD) and 8 (EH). Positive values indicate prefer-ence for the small sphere (size) and right rectangle(place). P(s) values indicate the selectivity for sizeas determined by repeated-measures ANOVA anal-yses for trials of the large and small spheres. P(p)values indicate the selectivity for place as deter-mined by repeated-measures ANOVA analyses fortrials of the left and right rectangles.




18/24

co-workers (1975). These cells have been postulated to formpart of a how system in which action-relevant parameters ofobjects are encoded to enable performance of skilled tasks(Jeannerod et al. 1995; Milner and Goodale 1995). Recog-nition of the size, shape, and orientation of objects is

important for hand preshaping prior to acquisition and forselection of appropriate grasp postures and hand contactpoints to accomplish task goals. The relevant informationcan be provided by visual pathways and/or by tactile inputsfrom the hand (Jenmalm and Johansson 1997; Jenmalmet al. 1998, 2000, 2003; Johansson et al. 2001). Slowlyadapting SA1 fibers innervating Merkel cells in the fingersare the most likely receptors mediating tactile object recog-nition. They respond parametrically to the curvature ofspheres pressed or scanned on the skin with the highestfiring rates evoked by objects with the smallest radii (Good-win and Wheat 2004; Goodwin et al. 1995, 1997; Jenmalmet al. 2003; Johnson and Hsiao 1992; Khalsa et al. 1998;LaMotte and Srinivasan 1987; LaMotte et al. 1998; Srini-

vasan and LaMotte 1987). The surface curvature of objectsgrasped in the hand is therefore likely to produce similarmodulation of cortical firing rates.

Recordings from neurons in the anterior wall of the intrapa-rietal sulcus (IPS) by Iwamura and co-workers (Iwamura and

Tanaka 1978, 1996; Iwamura et al. 1985, 1995) providedevidence that area 5 might serve as a higher-order somatosen-sory area for shape selectivity. They postulated that area 5neurons integrated features such as surface curvature and edgesto distinguish round and rectangular objects. However, theirfindings were based on tests of naturalistic stimuli and didnot use behavioral tasks to quantify neural responses.

The experiments described in this report were designed totest for the first time the effects of object properties on firingpatterns of hand manipulation neurons in area 5 when objectswere grasped and manipulated in repeated, quantifiable pat-terns. The main finding of these studies was that the object sizeand shape had weak modulatory effects on firing rates duringtask performance. Neural responses rose and fell in parallel

H17094 right hand

N = 46 neurons

N18588 left and right hands

N = 37 neurons

A

C

B

M

L

Area 5: all neurons

N = 83 neurons

0

20

40

60

80

100

0 10 20 30 40 50 60 70 80 90 100Cumulative Preference Index

%t

ota

lneurons

0

20

40

60

80

100


%t

ota

lneurons

0

20

40

60

80

100


%t

ota

lneurons

-40

-20

0

20

40

0 1 2 3 4 5 6 7

PreferenceIndex

Task stage

Size

Place

-40

-20

0

20

40

0 1 2 3 4 5 6 7

PreferenceIndex

Task stage

-40

-20

0

20

40

0 1 2 3 4 5 6 7

PreferenceIndex

Task stage

0

10

20

30

40

50

60

0 1 2 3 4 5 6 7

PreferenceIndex

Task stage

0

10

20

30

40

50

60

0 1 2 3 4 5 6 7

PreferenceIndex

Task stage

PI

dPI

0

10

20

30

40

50

60

0 1 2 3 4 5 6 7

PreferenceIndex

Task stage

Approach

Contact

Grasp

LiftHold

Lower

FIG. 14. Population object selectivity in the 2 animals studied; error bars show SE per task stage. A: cumulative PI values per task stage for the 83 neuronsanalyzed in area 5. The distribution of PI values in the population was concentrated at low values during approach and contact stages, and rose duringmanipulatory actions (grasp through lower). B: mean PI and dPI averaged across the population paralleled the modal values in A. Object preferences were lowestin stage 2 (contact). C: mean size and place indices in the 2 animals. The size index was greater than the place index in both animals. The small sphere was favoredduring acquisition (stages 13) and the large 1 during holding (stage 5). The medial rectangle evoked slightly stronger responses during acquisition; the rectangleswere indistinguishable in stages 35.




19/24

with the hand actions as the task progressed from acquisition tomanipulation and release of grasp. However, only 34% of thepopulation tested showed statistically significant modulation offiring rates related to the objects shape, size, or location.Object-linked differences in firing rates were small in compar-ison with the changes in mean activity across task stages.Selectivity for particular objects, when present, was character-ized by amplitude modulation of firing rates following hand

contact and differences in response duration as the object wasmanipulated. Objects that were grasped with the same handpostures evoked similar responses in area 5 neurons, regardlessof whether their surface was curved or flat or where they werelocated in the workspace.

Unlike visuomotor neurons in area AIP (Murata et al. 1996,2000), none of the cells tested in area 5 responded selectivelyto only one object class nor did we find cells that were

Left rectangle

Large round

Small round

Right rectangle

H17094 right hand

N = 46 neurons

0

20

40

60

80

100

0 1 2 3 4 5 6 7

Normalizedfiringrate

Left rectangleLarge round

Small round

Right rectangle

N18588 left hand

N = 37 neurons

A

C

B

0

20

40

60

80

100

0 1 2 3 4 5 6 7


Medial rectangle

Large round

Small round

Lateral rectangle

N = 83 neurons

0

20

40

60

80

100

0 1 2 3 4 5 6 7

NormalizedMaximum

Rate

Task stage

Best Knob

Second Best Knob

Third Best Knob

Worst Knob

Area 5: all neurons

0

20

40

60

80

100

0 1 2 3 4 5 6 7


Task stage

0

20

40

60

80

100

0 1 2 3 4 5 6 7

NormalizedMaxim

um

Rate

0

20

40

60

80

100

0 1 2 3 4 5 6 7

NormalizedMaximum

Rate

FIG. 15. Population object selectivity in the 2 animals studied; error bars show SE per task stage. Left: firing rates of each neuron were normalized by thepeak mean response across all trials and plotted as a function of the knob tested. Right: responses normalized by the maximum response evoked across all knobsand stages and grouped as a function of rank object preference in the best stage. A: mean normalized firing rate per task stage for all of the neurons tested inthis study. Object selectivity was less prevalent in monkey H17094, but the small sphere evoked the highest rates in both animals. B and C: mean normalizedfiring rates in the left hemisphere of monkey H17094 (B) and the right hemisphere of monkey N18588 (C).




20/24

unresponsive to grasp of one object type alone. However,Murata and co-workers tested a wider variety of objects thatrequired greater diversity of hand postures to be grasped. It istherefore possible that object selectivity might occur morefrequently in area 5 if tested with objects other than spheres orrectangular blocks.

Rank-ordered preferences for specific objects emerged in thepopulation during approach as the hand was preshaped forgrasping. The greatest spread in firing rates between the most

and least preferred objects occurred at contact when the objectwas secured in the hand and during static grasp. Responsesconverged in the later task stages when the objects were liftedand held. Individual neurons often showed distinct preferencesfor particular objects, but the selectivity was distributed non-uniformly in the population. Forty-three percent of the neuronsstudied responded most vigorously to the small sphere. Thisobject was held in a semi-precision grip between the thumb anddigits 2 and 3 or with a tightly flexed whole hand grasp. Thethree larger objects were held in a power grasp by the entirehand. The digits were extended along the flat face(s) ofthe rectangular blocks or flexed around the circumference ofthe large sphere. Acquisition responses were stronger whenobjects were placed at medial locations in the workspace,

especially sites near the midline of the animals body, than atpositions lateral to the shoulder. The distinctions based onworkspace location disappeared in later stages. During the liftand hold stages, the most preferred object was the large sphere,apparently because it was the heaviest object tested, andrequired greater load forces to accomplish the task goals.

Our findings differ somewhat from earlier reports byIwamura and co-workers that neurons in area 5 respondedselectively to one class of objects but not to others (Iwamuraand Tanaka 1978, 1996; Iwamura et al. 1985, 1995). Webelieve that the discrepancy between our results and theirs maybe due to methodological factors. The test objects used in ourstudy differed in size, shape, weight, and location in the

workspace but were all made of metal and therefore had thesame compliance. They were all simultaneously visible orblocked from view and therefore had no distinguishing saliencebeyond their relevance to reward on particular trials. The studiesby Iwamura and co-workers were based on single trial observa-tions when a monkeys hand was placed on a variety of commonobjects such as fruits, rulers, wooden blocks, or brushes. Theanimals were not trained to do anything in particular with theobjects, but one can imagine that these items might have verydifferent behavioral significance to them. It is possible thatthe selectivity for shape described in their studies mayactually reflect the animals intentions or interest in the testobjects. It is clear from Mountcastles original observations(Mountcastle et al. 1975) that the intentions motivating

grasp of food are different from those directed towardinedible wooden objects. Alternatively, the shape selectivityreported in the Iwamura and Tanaka studies may reflectdifferences in hand postures and distinctive muscle activa-tion patterns used by their animals to grasp the various testobjects (Brochier et al. 2004).

The task goals in our paradigm were to manipulate objects ina particular fashion (lift them). The animals were not requiredto discriminate shape, although they may have used shape cues

during approach to distinguish objects when they did not lookdirectly at them. In fact the cues delivered on each trialsignaled only the location of the rewarded object not its shape(see Fig. 1A). Therefore it is possible that greater objectselectivity might be revealed in area 5 if object shape and/orsize cues were relevant features of the task, requiring theanimal to locate a matching object.

Interaction of object features with hand kinematicsduring prehension

The orientation of the hand during approach and the selec-tion of specific grasp points by the fingers are flexible and canbe readily adapted to task requirements. Objects can be grasped

in a variety of ways depending on their physical properties,such as size and shape, the context in which they are acquired,and the goal of the grasping action (Ansuini et al. 2006;Castiello 2005; Mason et al. 2001, 2004; Roy et al. 2002;Santello et al. 1998, 2002). A pen may be grasped differentlyif the goal is writing or passing it to someone else. Conversely,different objects can be grasped in the same manner when t

Documents

Gardner, Babu, Ghosh, Sherwood, Chen 2007 - Neurophysiology of Pre Hens Ion. III. Representation of Object Features in Posterior Parietal Cortex of the Macaque Monkey