LOGO as an Information Prosthetic for Communication and Control

LOCO AS AN INFORMATION PROSTHETIC FOR COMMUNICATION AND CONTROL

Sylvia Weir LOGO Group Ma.<~achu~ett~ IMtitute of Technoloay Cambridae. Ma 02139

risk in regard to the development of a dqiCit ill spolia/ Ctlmpttt/ICt as a result of the restricted manipulatory experience available to them. An exposure to LOGO has provided a source of spatial mafliplllatary acti·itits for the~e individuals. The LOGO theme of the "learner a~ m.odel builder" takes on a poignant significance, since lines on the graphics screen can become model~ of objects (AS-IF object~). and producing l!n effect on such graphics objects can become highly significant actions for a severely handicapped person. This ha.< had a dramatic effect on the cerebral palsy victims with whom we have worke.d, both on their academic progress and on the level of self confidence they demonstrate <­ section VI).

ABSTRACT

When u~ed with physically handicapped and learning di~abled children, computer·b~ed LOGO activities can serve diagno~tic and therapeutic purposes, as well a.< provide insights into theories of knowledge repre~entation and cognitive functionina in general. The impact of LOGO nows from the active role taken by the Mudent~ in initiating and controlling their activities; from the ~pecifically spatial geometric emphasis of the!oe activitie~: from the ready availability of problem-~lution pat~ for scrutiny by the. ~tudent and by the teacher; and from the prosthetic value of the computer in providing a means for the upres~ion of otherwise trapped intellisence in this population.

I. INTRODUCTION

The LOGO learning environment (I] provides a ver~atile tool for diagnostic, in~tructional and remedial u~e with children who have special educational needs (II 14 16 17 18l Weir (18] has described the ~ucce~sful use of LOGO with severely phy~ically handicapped but mentall(alert cerebral pal~yl adolescents at the Cotting School for Handicapped Children. The educational impact has been ~ub~tantial. It is our belief that the navor of the work ha.< been innuenced by the A.l. ~etting which spawned it. Themes which arise include an emphasi~ on proces~ rather than product; a concern for the varjety of type~ of knowledge involved in solving a problem; the learner as model builder; and the specificity/generality issue -- how specific are the computational mechanisms underlying cognitive proce~~?

Observing children with di~abilitie~ offers a magnified and slowed down view of familiar proce~~es. Although we do not know how many unusual and deviant . mechani~ms have been invoked in the~e cl)ildren, it is instructive to explore their behavior as a source of evidence for pouiblt cognitive mechanisms

II. MODELS AND PROCESSES

To the extent that sen~ri-motor activity contributes to the full development of the cognitive structures underlying spatial understanding. individuals with severe motoric restrictions suffer from a reduced opportunity for such learnina. Hence they are at Performance te~ts are notoriou~ly difficult to administer to this aroup. The very motor handicap that lead~ u~ to su~pect a deficit in spatial cognition, renders the ascertainment of that deficit very difficult. Mea~ure~ of vi~ual perception in Cerebral Palsy yield progres~iveiy lower score~ a~ the motor component of the test u~ed increa.<e~ (20~ Cerebral pal~y children who fail on the bloek-design reproduction task of the Wec~ler Intelligence Scala for Children, can ~ucceed, most of the time, on a ta.<lt which require~ them to choose the one correct design out of three pre~ented block d.e~igris (2~ However, turning assessment tasks into · a m11ltiple choice format in this way gi.ves a ~mewhat impoveri~hed ~etting, since all information on haw the task is performed is lo~t .

I. Cerebral palsy is a di~rder of movement and posture resulting from a permanent, non-proaressive defect or lesion in the brain oecurrina before the end of the fi"t year of life.

In contrast, exploiting the pos~ibilities of an interactive graphics situation can generate a significantly novel way of as~e~~ina motorically re~tricted per~ons. We have developed an exten~ion of the stand~rd LOGO sy~tem which allow~ the use of scrn11 rc>rsiafiJ· af assnsmrflt taslrJ, thus reducing the motor component of the ta.·lt to a minimum. \Ising these, we have found deficits in spatial rea.<oning in ~me cerebral pal~ied subjects when they are asked to perform ta.<h involving ·ordering by length, shape recognition, ~patiai localization and mental rQtation (7 14.18l

The computer place~ con~traints on the execution of a ta.\k that empha.<ize~ the intermediate steps of the ta~k. This re~ults in a wealth of detail in the protocol~ of both handicapped and non-handicapped controls. We recommend a more widespread use of this approach as an adjunct to more standard recognition, torting and copying ta~h. where end results tend to predominate over evidence of the problem-solving process. It is of ~oune po~sible to emphasize proces~ without a computer, but the computer does facilitate thi~. and provides ail kinds of other perb like automatic records of the ~c:tions performed and automatic tim ina.

Ill. SPECIFICITY OF COMPUTATIONAL MECHANISMS

There i~ a theoretical a.· well a.· a practical intere~t in exploring the pattern~ of diHuption di~played by per~on~ with the~e unu~ual experience~. Are 1here dissociations ·· a pattern of ~pared and impaired competencie~ ~uch ~ ~patial excellence coupled with marked language deficit~? Thi~ would lend ~upport to the view, gaining ground among cognitive Kienti~t~. that ~orne computational mechani~m~ at lea~t are quite ~pecific to the domain and level of proce·~ing involved in the particular problem being JOived.

It is difficult to obtain good data on subjects with gro~s limitation of e11pres~ive power ·· restricted or absent speech, and/or InabilitY to write or oerform O\her manioulation~. Text·editin£ fac:ilitie~ provide a way for motorically re~tricted intellectuals to keep notrs in tht form of romputcr, filrs. and to engage in problem·solving at a high level without having to do it all in the head, by using the compuur as a scratchpad

Combining informatioo from a variety of LOGO activities provid., a powerful way of making fine·grain analyses of performance in spat-ial, numerical and writing domain~. Very clear diuociations ar~ exhibited.

DiSJ«iation in an fourtrrn ytar old quadriple-gic

Ja's favorite project was the production of a new~paper, in which he showed hi~ - ability to plan his activity. He first generated the Nc:tion heading~. and then went bacl; to fill in detail:., taking each Nc:tion heading and expanding it, beginning with the e~ie~t and not with the first ·· an exfmple of top·down planning. In contrast, his ~patial ability was very poor. He programmed a simple drawing of a car step-by·step and painfull)' slowly. He failed on a mental rotation task (7 9) that five )'ear·old~ can do. He could not seriate tilt ~tick~ .. neither using real ~ticks [12 14~ nor in the graphics tcrHn version or the l'iagetian task (18l He could seriate four tcrHn "sticks" but not four real ~tich. He could not insert new tticb into an ordered row of four sticks, graphic or real.

His performance on Piaget's village (topographical) t~k [8] was revealing. In this t~k. the child and the experimenter each have klenticel drawings of a Kene. :rhe experimenter places a doll successively at different locations in the drawing, and, for each plac:ement or the doll, the child is required to place his doll in the flme positions on his own dra'll·ing. The first time round, the drawings of experimenter and subject are in the same north·south orientation. Then the experimenter's drawing is rotated 180 degree~. while the child's drawing is left at the same upright orientation. In our version of the ta~k. each of these parh is done fint with physical modeh of the Kene, and then with the child's tcene drewn on the Kreen, using the cursor a.· doll. Ja's strategy in the "reel" task wa.· to copy the physical movement of the e11perimenter, thus getting the wrong placement~. (Kore 3/'1). In the computer ta.~k he began drscribing tht position of tht doll ill "'"~($ and achiev.ed a score of 8/9 correct placements. He succeeded in turning the computer version of the ta~k into a tinguhtic ta.·k which he could do. When he performed the "real" tuk, the underdev~IDped state of his place·findins skill led him to U!le a le~~ ~ucce~~ful copying ~trategy.

Ja.'~ work with number~ wa~ far below age·level, and idio~yncratic. Th~re are many place~ in hi~ work with Turtle �eC'Im~try where he Marted with 14, which is hi~ age, a~ the argument for a turtle command, and then continued along the number ~equ~nc. e regardle~~ of whether he w~ in~tructing the turtle to turn right, to go forward or .to go back, and regardle~s of the n~ed~ of the problem.

The contra~! between Ja'~ ~kill in the lingui~tic domain and hi~ weakne~~ in ~patial and number·handling areas leads u~ to reject the propo~al that he ha.· a "general sequencing" problem, ~ sugge~ted by his cliniciam, since he can order letters very skillfully. lmtead we hypothe~ize a domoin-sprcific problrm ini"OivinR inadrquatr dcwlopnrrnt of appropriatt nrttlfal rrprruntations for spatial knowlt~lgt (data~tructure~ and operation~ on them; ~patial metaknowledge).

IV. VARIETIES OF KNOWLEDGE

Many of our ~ubject~ give the impre~sion of having small local packet~ of knowledge, without the appropriate procedural metaknowledge recarding how to fit the~e into any particular ~olution path. Thi~ i~ illumated by their behavior in the mhsile game. To give Ja. and ·other ~tudent~ a sen~e of a number ~ I quantity, in thh ca~e a~ something which doe~ a particular amount of work, we set up a target game, in which the student was required to e~timate the number needed to move the mi~~ile to the target .. a hit wa~ rewarded with an explo~ion. A ~econd cerebral pahy ~ubjecl, K~t. the ~arne age and with ~imilarly poor ~patial ability and good verbal skill~ a~ Ja. ~howed an even Mronger tendency to ~tep through the number ~equence when playing this target game. She went ( 12 II I 0 Q 8 7) to hit the fir~t target at 7; (6 5 4 3 2) to get the next target at 2, and so on. Her use of the $Ucce~~or function wa~ without error, but inappropriately inflexible for a fourteen·\'ear old.

li~r problem would appear to be not so much that she did not have a given ~chema, but that ~he did not know when to activate it or how to choose the mo~t u~eful of several po~sible Khemata .. she did not have the flexibility which accompanies efficient metaknowledge. Thi~ is illu~trated further by her performance when required to face the turt·te in particular direction, and drive it to a given de~tination .. She alway~ ~topped short, and made no midcour~e correction even though the wrong direction she had taken took her way pa~t the locality involved. The experimenter found that if he took the driver's seat and a.'ked the appropriate que~lion~ after each step ·· "h the turtle facing the right wayr .. thi~ wa.· ~ufficienl, mo~t of the time, for her to "solve the problem. When the appropriate invocation of ~chemata wa~ facilitated by embedding the game in a linguistic context .. driving the turtle eround the ~creen wa~ expres~ed as succe~sive visit~ to depicted characters in a story .. her aim became more accurate, and she readily mad~ appropriate mid·cour~e direction corrections without any prompting. When the problem took the form of a linguistic ta~k. she showed appropriate invocation of ~chemata.

A dl'1our: Spatial Giftmnm

Di~~C'Iciation between the development of spatial and lingui~tic ability can go the other way .. spatial cxulltncr in tht prrstncc of linguistic deficits. Evidence i~ accumulating for the exi~tence of a group of children who in fact are spatially gifted. Such children turn up in the spC'C'ific lrarning disability category of ~chool-children who~e reading ~kills are below expected level . They have difficulty with language u~. and ~ince ~chool~ do not provide an opportunity for them to di~play their ~uperior ~patial ~kill in any formal way, they lo~e out. We have begun working with ~uch children and find the predicted excellence at LOGO [17 19~ For example, one eleven-year-old dy~lexic boy at a local ~chool for the learning <lo~abled demonMrated more ~kill at LOUO Turtle Geometry in the first half hour than any other child of that age we have encountered; and it wa~ easy to detect the clever heuriMic~ he adopted ~pontaneou~ly, e.g .. con~tructing his own "unit" and "yard~tick" appropriate to hi~ need~. Our explanation i~ that hi~ ~uccess rPsl' on highly developed metaknowledge in the ~patial domain. Storie~ about such children have accumulated in the LOGO community [6 17 19~ and there is independPnt neurop~ychological evidence for the existence of such a category among learning disabled children [S~

The questi<'n arises a~ t<' whether one can do anything with this finding in practical terms, i.e., how it can affect decisiom about the e-ducat ion of such children. ResParch on individual differpnces in aptitude in relation to instructional variables has shown that aptitude-treatment interactions do occur, but that these relate more to general mental ability rather than to the contra~! between verbal and n<'n·verbal abilit)' (4). Inconsistent results are reported on the use of pictorial materials to enhance the learning or tho~oe ~tudent~ who score high on the visualization dimension. However, the figural enhancements of imtructional materials used up till now have taken the form of graphs and diagrams ·· a rather static form. Nolice the a.·sumption that learning consists of the acquisition of facts ·· a diagram i~ a perfectly appropriate visual enhancement of factual acquisition. This approach does not take account of the need to know when and how to deploy a fact in a problem-solving situation ·· crucial procedural components of metaknowledge.

U~ing a computer, 11 i~ now possible to introduce a more dynamic approa~h by making a substantial proportion of the formal ~chool subjects available in spatial mode on a graphics screen. Most importantly from a pedagogic point of view, we can arrange for computrr-basrd spatia/learning to br procrdural and constructi~t in nature, as i~ the case with LOGO, with all the advantages which now from this.

Providing an <'pportunity for spatial problem-solving in a formal srtting, in addition to the more usual place~ like the playing fields and the arti~t's drawing board is es~ential to the healthy academic growth of some people ·· future engineers, computer programmers ·· allowin~t them to exploit their ~patial facility to acauire academic skills.

The excellence we have been talking about has been shown to have a physiol<'gical counterpart, in that 11 is related to maturati<'nal rate [1~]. Neuro-anatomical issues ·rise when Inte-rpreting the significance of the finding of a spatial deficit in cerebral palsy subjects. We return to the main topic of phy~ically handicapped children and their ~patial deficit~.

V. AN AMBIGUITY: BRAIN DAMA�E~ LACK OF EXPERIENCE

While there has undoubtedly been a re)triction in ruotor activity, there ha' al~o undoulitedly been brain damage in these subjects. Although mot<'r dy~function i~ the predominant, defining characteristic of thi~ group, the variou~ etiological agents of cerebral pal~y do n<'l a~ a rulr rrcognize functional boundarie~; a pure motor di~order i~ rare. In particular, the original in~ult, be it vascular, mechanical, infectious or whatever, can affect those area~ in the brain which recent re~earch ha) shown to be concerned with spatial undentanding (reviewed in [S]). Thus, in any particular case, the ~patial deficit in que~tion may have re~ulted either from dirC'C't damage to the neural ~ub~trate which ~upport~ ~uch activity in the undamaged brain, or indirC'C'tly from damage to motor area~. the dysfunctional comequences of which, namely, re~tricted manipulatory experience, have led to the ~patial defect under con~ideration.

We hope to ca~t ~ome light on the matter by observing the effect ,,f intemive spatial problem-solving on the degree of ~patial deficit shown by our subject~. To the extent that we can produce an improvement in performance ~o would we be inclined to attribute the deficit to the lack of exoerience.

An Examplr from tht' Li1111uistic Domain

Disambiguating brain damage and lack of experience ari~e~ in the area of language production. We have not yet worked with .non-vocal individuals, but some of our ~tudent~ have never before written anything what~oever, becau~e their handicap i$ ~evere Pnough to prevent them from holding and using a pencil. The standard pr<'cedure here i~ for an adult to act a.' ~cribe, and in the proce~$, · in~ert all those "minor", "trivial" marken ·etc., whoJe pres~nce or ab~ence is of such intere~t.

If ~everely handicapped individual~ are to hold down a job, the quality <'f their writing is import~nt. For many of these ~tudents, thi~ i~ the fint opportunity to generate their own text, without the intervention of an adult. Their <'Utput can be u~ed in diagnostic mode, to establish patterns of expressive language di~ability, and, where required, a~ a rrmt'dialtool for language re-education.

We show ~uch a sample of writin1; produced by Ma., a quadriplegic cerebral palsied ~tudent. He was 17 years old when he started LOGO, and hi~ programming and math achievements have confirmed his high intelligence. He ha~ now been admitted to the Univer~ity of Ma~sachusetts Bo~ton campu~ as a computer Kience $tUdent. Thi~ is the fir~! piece or writing he produced.

"I ment Dr. Sileva Where, Jo~e Valente and Gary Drescher on October S, I 978 at 9 ; 32 ; 47 AM. which the compuer I Wti ~o excized it like being it a waitting & maternace room at a hospiltal whiting to fine if oot'~ abo)' or a grail.

We had a and we whont you to do it fist for us I am LOGO number "J" ginny pig. When the)' get a new idse they say to hel, michalel we had a and we whont you to do it fi~t for u~

like a nice guy I do if I wont to or not. I do and then I give my por and con on the id~e .. l tell them why I came up for a allturative. Why you mite a~k? Beco~e I know how the per~n on the arther end feel. Becose I am the middle man betwee M.I.T. and handcap people.

I also teach five Mundt~. No tow lean at rate. Some tik then I know they are intered in leaning about logo. Then the one that take for granted. I can tell .

When I teach I lean form my ~Iandt. ti well or bedter then form a book. I call that on the job training. My fi~t and every day e.xperreance with the compuer when 11 cash and it lo~t but it keep on losting all that I have tort it but keep no teaching it overy and overy 1gian when I bring back

to live."

This loob like ~poken Engli~h l''hich h~ been written down. There if much phonetic ~pelling. omi~~ion of word~ and letter~. ·~ well u letter rever~··~ and a curiou~ con~truction for pre~ent ten~· verb~. How much i~ thi~ due to lack of experience and how much to damage to the relrvant brain area~ Are the errors random, re~ulting from inaccurate typing, thinking fa~ter than he is able to type, lack of ~pelling experience, and the like? Or i~ there evidence for particular pattern~ ~een in patient~ with known le~ion~. e.g. the t)'pe of di~turbance a.-.-ociated with a le~ion in the anterior portion of the dominant hemi~phere (Broca\ expre~~ive apha.·ia). In thi .- type of le~ion, inflectional marker~ for plural$, p<n~e.-~ive~. and third per~on ~ingular h/ are affected differentially [3~

We can ~ee errou with teme marker~ in the above p~~age:

"feel" {feel~); "handicap" {handicapped); "take" {take~); "ca.·h" {cruhed); "keep" {kept).

However, the~e error~ could conceivably be due to his mi~perception of the .-poken word~. upon which he i~ ba~ing his ·pelling {notice 'fine" in~tead of "find"). Mi~perception could cover the in~tance~ of letter omi~~ion~ which occur, e~pecially "r" before 1 con.-onant, which i~ not ~ounded:

"fist" {fir~t); "lean" {learn).

Either ha.·te and/or lack of experience, or brain le~ion, could conceivabl)l account for the numerou~ omi~~ion of word~:

It [wa.·] like being in; we. had a (idea~ No tow lean at (~ame] rate.

and the revenal of letter ordP.r:

·arail" {girl): "id~e" {idea); "por" {pro); "Mundt~· (~tudent~); "tow" {two); "form" (from): "no" (on).

In order to inve~tigate and treat thi~ problem we have ~et up 1 focal rrmcdial program. All hi~ Engli~h le~~on~ take place in the computer room, u~ing the text editor, in~tead of a ~cribe who~e activitie~ were ma~king the true ~ituation . He write~ hi~ own book review~. e~~ap and ·the like, on the computer; in addition, and, in collaboration with hi~ Engli~h teacher, we have in-'tituted a ~rie~ of remedial writing -'e~~ion~ inV!llving ~pecific exerci~e-' ba~d on weakne-'~e~ ~h!lwn in hi~ writing, in order to 8-'~e-'~ how much improvement we can achieve. Early ind ication~ are that a great deal of improvement can be achieved, a very nece~my ~tep for his college career. He recently wrote thi~ letter in re-'pon~e to a query about ·LOGO from another victim of cerebral pal~y .

I 07 Second Street Medford, Ma~s. 02155 February 6, 1981

Mr. NeviUe Andenon Bu-'h Farm 2400 Acacia Rd Green.field, N.C. 27405

Dear Nick,

My name i~ Michael Murphy. I am the pcr~on whom your m<>ther .-aw on "I'M Magazine". I attend the Cotting School in BoMon ma~~- I have been work with the compuer~ for about two and 1 half year~. The n1me of the ~y~tem is "LOGO". It h~ open many new door~ for me. Now I can draw picture on 1 cheen ane write letter likl' thi~ one.

I wa~ the fir.'t C.l'. per~on in the U.S. to tired the "LOGO" sy~tem. Enclo~e I will ~end ~ome of my work.

Sincerely Yours, Michael J. Murphy Jr.

P.S. I would be greatful if you could ~nd me a piece of your art.

VI. SOCIO-AFFECTIVE 'ISSUES: EFFECT OF I'ARTJCII'ATION IN THE PROCESS

Severely handicapped per~on~ cannot engage in the motor activitie~ which ordinarily ~upport the acqui~ition of cognitive structure~. What i~ mi~.'ing from their live~ b the explicit C!lnfrontation between the actiom they perform and the con.'equence.' of .'Uch actiom in the phy~ical world, an experience taken for granted in normal development.

Thi.' LOGO experience i~ often the fint in which the~e ~tudent-' have had an opportunity to participate in the proce~ of tackling problem~ where they are required to initiate ~olution~. try them out, decide when tc:> change track and when to per~i~t. re~pond to feedback -- all tho~e th i ng~ which tend not to happen in the U$UII dependent, adult-dominated ~itllations which characterize their lives. .

It is characteri.'tic of the LOGO ~y~tem to seek to promote a clo.'e a~$ociation between a child'~ cognitive development and an environment where I~ or ~he take~ the active role in his or her learning. Thi~ a~pect become~ e~pecially important in work with $pecial need~ children, e.g., with an auti~tic child (16~ with the learning-di$abled child [17 I q~ and, mo~t poignantly, with the cerebral pal~y child, where a particularly debilitating cOn$equence of the grM.' phy~ic. al handicap i~ the dependent, pa.·~ive role it impo$e~ on it~ victim~. A profound le$$On ha~ to be learned, one which i~ taken for granted in "normal" development. A handicapp(t/ indil-idual has to lrarn to bdirvl' that an action ht talceJ can have an cff«t on his e-nvironment. The uncompromi~ing way in which L~�O place~ initiative and contr!>l in the hand~ .Df the._ u~_er and gtve-' the U$er a chance to know what ~he know-', means that -'he may now engage in ~u~tained, independent activity and be her own critic. The re~ulting enjoyment and ~en~e of per~onal fulfillment can greatly enhance their general ~n~e of personal WMth and competence, and could have important effects on the member~ of thi~ di~advantaged ~ection of ~ociety, not the le~t of which i~ the creation of a high degree of motivation which can then form the ba~i~ for ~ignificant learning in formal $Ubject-'.

Th.e~e i~ much evidenc.e of incrt'ast'tl motivation among the parttctpant~. a~ reflected in an inten~e competition for the u~ of machines -- C<'ming in an hour before Khool and ~tipping half the lunch break to work on the computer, .. and the frequency of 1nimated di~CU$~ion about variou~ computer project~ repre~ent-' a new experience for the school. The ~chool authoritie~ have been ~o impres~ed by all thi~ that they have rai~ed $27,000 to equip a computer center. Computer progra.mming ha~ become an integral part of the -'Chool curriculum and -41 out of ~0 ~nior students have opted to take computing. All 5 of the ~nior~ involved in our project have changed their career plans to computer programming. Three of the~e have been accepted into the computer ~cience cour~e at Univer~ity of Ma.·~hu~tt~ Bo~ton campu~. lncrea~ed. vocational ecce~~ -- the po~ibility of independent living -- i~ tlrt great advantage accruing from the computer activitie~.

The ver~atility of the ~)·~tern ~upport~ a wide range of cognitive ~lyle~ and leveh of ability [I 0~ allowing a ~tudent to carve out an area in which he feeb comfMtable -- a microworld -- in which to work. The teacher can aim for a balance between working with strength~ and exercizing weakne~~e~. Individual ca~ ~tudies reveal the range of reaction~ to the introdvction to LOGO, the preferred activitie~ of different ~tudent~. difference~ in Myle~ of debugging, ~lyle~ of working, amount of ri~k·taking, re~pon~e to the anxiety of a ne-w ~ubject and ~o on.

A~ an illu~trative example, we quote the subject who handled the devastation of failure by denial . Ja. wa.< alway~ ready to change hi~ purpose to fit hi~ latest error. It was a great step forward when he learned to tolerate making an error and allowed himself to benefit from the proce~~ of debugging that error. Another ~ubject, aged ~eventeen, was so uncertain, ~hy and lacking in ~elf-confidtonce that ~he had the reputation of making no intellectual progre~~ what~oever. The virtue~ of LOGO applied particularly to her, and 5he ha5 moved from a totally dependent situation, where she made no move on her own, and looked for rea.<surance after each command she l)'ped, to one in which she worh on her own for long period~ of time and initiates her own quite ~ophisticated projects.

We believe that a major element in the improvement we see in motivation and in ·the po~itive affect generated by LOGO activity hi$ to do with issue~ of who is in control of the aetivity. We have only scratched the ~urface of what is pos~ible.

ACKNOWLEDGEMENTS

The work reported here was ~upported by the Bureau of Education of the Handicapped under grant number CFDA 13-443. It was carried out with the cooperation and 5Upport of ~ymour l'apert, the director of the LOGO laboratory, by the member~ of the Cerebral l'al~y project: Syvia Weir, J~ Valente, Gary Dresche.r, and linda Laatsch.