i‘. ‘ III 1II \4 . 4.4 'u \‘U’gg I. {I .11. .. .. 1 8.; L.I.._ x» u M... u . .._. an? Mgr. 1.. _ n J. . , . .. 1 . I q . \ Bl‘ihflll‘hli1i ..-I~4.ll.i 4 57."...1'.‘ 1'13... ‘I‘IIOHJIH, inn-Hui « .1 ‘50... at .up".. will I" . v . 1 . _ 4 . Q Ii in...- ‘ I . nah vu.‘ *n: . . . - . “2.... s .o 9”,, AdSTRACT AI-TALYSIS OF A I'JLE‘HCD OE T.\’AINI“‘TG "Ht; IVENTAI LY EHLTARDILT) by Fsrjorie Ann Yascolt This non-comparative study was undertaken to explore the practibility and possibility of training the severely retarded on prepositional and abstract concepts in response to verbal commends. Fourteen institutionalized severely re- tarded children and young adults served as subjects. Eight were trained on a Nultiple Differential Response And Feedback Apparatus (NUDRAFA) using a methodology which specified the deficits of the retarded subject and applied learning principles to overcome them. It seemed that knowing a high initial number of concepts at pretest led to faster subsequent learning, but this made no appreciable difference in retention. Subjects k“owing a low initial number of concepts retained as well as those with a high initial number. That subjects could learn to use verbal mediation is evidenced by the fact that some non-verbal subjects were trained to read meaningfully. It was concluded that PUDRAFA was effective in bringing the behavior of the retardate under the control of abstract verbal commands. Specific recommendations concerning the use of NUDRAFA and how to interact with the child were given. Implications of this study for further learning research were also discussed. Approved_, 6 Z 2 T‘J ANALYSIS 09 A Lesion OF THAINING TnE LnflTAIIX RETARDLD by Marjorie Ann Yascolt A THESIS submitted to Michigan State University in partial fulfillment of the requirements for the degree of FASTER OF ARTS Department of Psycholopy 1966 to Stanley and Felicia ii ACK ‘IO .JIZJD CI -‘LZ‘JTS The writer wishes to thank all those who made this study possible. Special appreciation is due Dr. R. Ray Denny, Chairman, for his inspiration and encouragement. Dr. William Stellwagen and Dr. Donald Burke, members of the committee, deserve thanks for their support and helpful suggestions. The aid of undergraduate experimenters in observation of the subjects and their insights into individual behaviors is deeply appreciated. I am deeply indebted to Mr. Stanley G. Smith for his dedication and perseverance throughout this Stlldy. To Dr. Harion J. Skowronski, H.D., Howell State Hospital, Howell, Michigan, go the writer's special thanks for his cooperation and enthusiasm in this study. Ho H. p. m ggLI-m 7'1 n "'mmvm'j 4A-; 4.11 Of VO.-4'.._‘1-1.L~J Chapter I . ITTTQfiiG. UCTILIT II . I‘LLTLIOD ) .0 III 0 R;QII—’T IV. 31553173530: V . SUI '1 if BY 7 fli‘]¥i‘_‘); _‘I.‘$C'_‘I(;I l{.4? ..'1. L._‘.- -. -4») APPEFTUICIES m r-J V W. \o -\J \o l-‘ o w \xl C\ o: co LIST 03 HIGURLS Figure 1 Concept Acquisition and Retention of Experimental Subject number 1 and Control Subject number 11 2 same but: E $2 and G #g 3 E #3 and C #12 h E fh and C #10 5 V #5 and c #13 6 E #6 and C #12 7 E #7 and C #11 3 E }3 and C 51h M A R) i) Appendix rage A. Concept Knowledge Test Form and Instructions........... 33 J. Sex, Chronological Age, Kentsl Ape, Ion, and Type of Retardation for Experimental and Control SU'IFDjex'C-bSOOOOOOCOOOOOOOOOOOOOOOOOOC [4'5 C. Concept knowledge Scores for Each Subject on Pretest and Retest.... 44 O c c n n a d. b n . \ o . A. u . . I . . a u n v . 9 D . s a 4 a o n O ChAPTLfi I “-ITRCDL CT ION This study attempted to explore the effectiveness of a method for training the mentally retarded. The broad aim is to facilitate more independent living on the part of the retardate by bringing his behavior under better control of verbal stimuli. This means training him to respond appropriately to a large variety of verbal commands from the experimenter or parent. Since this was an exploratory study, no hypotheses were generated. Our basic premise is that the severely mentally retarded can learn and retain,well, but that they suffer certain deficits in the cond'tions of learning. Je have attempted to identify these deficits and to employ optimal conditions and principles of learning to overcome them. Benoit (1957) considers attention to be the primary factor in learning. For the mentally retarded, the focus of attention is largely determined by the stimuli around them. When learning is incomplete, a complex situation leads to mass activation which results in disorganized behavior. At the beginning of training he recommends simplifying the con- text by diminishing he over-a l stimulation so that a response will not be disrupted while it is in the process of l being reinforced in a repetitional series. Later, after the response is fairly well strengthened, these precautions may be eliminated. ‘ Zeaman k House (1961) in a study of two-choice spatial and delayed responses attributed the absence of a delay- 1earning-set in the mentally retarded to position habits and perseveration errors. These habits and errors were ascribed to a deficit in the attention function. In 1963, they related the inhibition deficit to the poor attention or distractibility to account for the mentally retarded subject's ignoring the relevant stimuli. Denny (196M) suggests that the inhibition deficit may be more basic than the attention deficit, that is, the subject's failure to inhibit the effects of extraneous stimuli accounts for his inability to attend to the task. Since the subject cannot inhibit responses to other stimuli, he cannot attend to a set of stimuli for any length of time. In any event, the retarded subject lacks the self-initiated sets which allow for consistent and continued responding. For Zeaman 3 House (1961), the delayed response deficit is also related to the lack of ability to maintain a self- initiated set. Because the mentally retarded are more stimulus bound than normal (Benoit, 1957; Denny, 196h) and respond more to irrelevant stimuli and engage in task irrelevant behavior (Cruse, 1961; Terdal, 1965), his may lead to an incidental 2 learning deficit in the retarded. In incidental learning an individual must maintain sets which are not established for him as they are in intentional learning. Because the men- tally retarded are distractable and show poor ability to establish sets, they do not encounter consistent pairings of stimulus and response. When the mentally retarded are sufficiently well instructed or guided, as under intentional learning conditions, they often do not show a learning deficit (Denny, l96h). For example, Stevenson (1960) in two experiments on object and pattern discriminations found no difference between mentally retarded and normal subjects when matched on MA. Singer (1963) has found that the men- tally retarded are poor incidental learners. On a passive incidental learning task in which there was no response involved, the mentally retarded showed a marked deficit. A misdirected task in which the subject was to respond with the color of an object revealed a less severe deficit. There was no significant difference between the mentally retarded and normal subjects on an intentional learning task. Baumeister (1963) found an incidental learning deficit in the mentally retarded on an immediate recall test, but the normals and mentally retarded were equal on a recognition test of incidental material MS hours later. These data indicate the need for the mentally retarded to be directed and guided in every step of the learning process. Presumably, an incidental learning deficit makes it difficult to perform extended behavior sequences, especially 3 with respect to language behavior. The mentally retarded may know familiar objects such as chair, table, or food well, but prepositional and relational concepts such as behind, throurh, and over are known only very poorly and are extremely difficult to learn because 'he subject must attend to the relevant stimulus over a time span to learn its meaning. Such a deficit could account in part for the lack of verbal behavior in the mentally retarded. Griffith, Spitz, and Lipman (1959) in an abstraction task found that the mentally retarded seem to be about three years behind their MA in language development. Luria and Vinogradovia (1959) point out the dissociation between the verbal and motor systems in the mentally retarded. They state that the language function has not developed sufficiently to regulate the behavior of the subject; inhibition is impaired and differential conditioning is poor. The subject fails to inhibit because of a lack of verbal control. He does not I attend to the subtle verbal cues because of his relative inability to inhibit other responses. Thus, the mentally retarded suffer from deficits in learning which seem to stem from deficits in duration of attention, incidental learning, inhibition, and verbal mediation and behavior. One set of principles which have been found useful in r»: traininx the mentally retarded are presented below (Denny, F 1966) and are placed in the framework of Elicitation Theory (Denny and Adelman, 1955). According to this position the it nain parameters of learning constitute ways in which the situation can be arranged so that the organism makes or continues to make the designated response. These principles attack the deficits listed above by a methodology which employs optimal conditions in the intentional learning of abstract concepts, prepositional phrases, and relational concepts. The main principle is the elicitation or specification of the response. According to Elicitation theory learning occurs only when a response is consistently elicited in a given stimulus situation each time the stimulus is present. In order to insure elicitation of the correct response without evokind competing responses, barriers are used so that the subject can choose only the correct alternative. The use of barriers is particularly important because the mentally retarded tend to perseverate and follow position responses (House and Zeaman, 1958; Kaufman, 1959) and make little use of negative cues (douse, Orlando, and Zeaman, 1957; Oatley, 1965). Once the response is strengthened, the barriers are opened up. Crutch cues, manual assistance, and coaching are also liberally used at first to specify the response anl are later dropped out. It seems possible that the mentally retarded fail to form the associations necessary for adequate adaptive be- havior partially because the ambiguous and inconsistent reinforcement supplied by the normal environment does not lead to adequate discrimination between the relevant and 5 irrelevant cues in the multiplicity of stimuli impinging upon them. A light, social-verbal reinforcement, tangible and token reinforcement provide immediate knowledge of results necessary to specify the correct reaponse and form these associations. This is particularly important in the later stages of learning when other alternatives are avail-l able to the subject. The correct response is immediately followed by the light and incentive the incorrect response is not reinforced and a different response must be made. The role of kinesthetic feedback is also emphasized since the occurrence of motor responses to relevant cues facilitates learning (Smith and heans, 1961). The method outlined below allows the subject to perform the response, to see his response, etc., so that the subject provides himself with a discriminative stimulus as well as having someone else provide it with a light and social-verbal reinforcement. Such cross modality learning has been found to facilitate learning (O'Connor and Hermelin, 1962; hayden, 1966). Differential visual, auditory, kinesthetic and/or taetual feedback also identify whether or not the associated response was correct. Cole and Shaefer (1961) found that generalization of learned response occurs in the severely mentally retarded in much the same fashion as in monkeys and man. In order to promote stimulus generalization and positive transfer, we train for the meaning of each concept in a variety of different contexts and use different versions of the same 6 stimulus quality or object. In concept training the general- ized response is inhibitory in nature. Thus relevant cues continue to elicit the appropriate response while nonessential contextual cues tend to remain irrelevant. In order that the concept be the only aspect of the total stimulus situation which is a consistently relevant cue, we must randomize the contextual stimuli. All other possible cues such as position, shape or size of slot, etc. are randomized so that the correct response is consistently associated with the same cue in a series of like commands. Lance (1965) found that the mentally retarded benefit more from overlearning than normals, and are identical to normals on savings scores under conditions of overlearning. This repetition of learning yields best results when it is distributed within a session and across sessions. The repeated use introduces a variety of contexts and thus increases generalization and transfer. Elicitation theory emphasizes the importance of incentives and the removal of incentives as consistent eliciters of the to be learned response. That is, there is no learning of a differential response without the presence of an incentive in he goal which elicits approach and conditions avoidance of wrong alternatives. Tne rein- forcement gets the organism to make the response which the experimenter has designated to be learned. The mentally retarded subject must be interested and remain interested in the task in order to make appropriate responses and to 7 maintain these responses without making alternative responses. High success (Butterfield, 1963; Kass and Stevenson, 1961) and switching of incentives (Aldrich and Doll, 1931) has been noted to increase performance of the mentally retarded. Therefore we select the agent most reinforcing to the sub- ject to maintain tis attention and interest in the task. Also a variety of incentives, a variety of commands, and a whistle, are used to maintain attention. An obstruction to getting the behavior of the mentally retarded under verbal control is the dissociation between heir verbal and motor systems. O'Connor and Hermelin (1962) found that the advantage of a cross modality situation is lost when verbal encoding is prevented. Therefore we train for specific use of the language by having the subject verbalize the concept and by building in verbal mediators for guiding and directing his own behavior. The more complex responses are obtained through shaping techniques which involve reinforcement of behaviors success- ively approximating the criterion behavior. Through chaining of responses by shaping techniques, the abstract concepts which are difficult for the mentally retarded to learn by other less explicit methods are developed and made resistant to extinction. In this sequential building we take advantage of positive transfer by shaping in easy steps on what has I previously been learned. CHAPTER II METHOD Subjects Fourteen institutionalized mentally retarded children and young adults, ranging in age from E to 33 years, and recorded EA from 7 months to h years, 3 months, were selected from howell State HOSpital. Six of these subjects served as controls and were roughly matched with the experi- mental subjects on the basis of number of initial concepts known, MA, CA, and ward. T16 subjects were preselected over a 10 week interaction and observation period on the oasis of H H overall responsiveness and physical capacity. Overs responsiveness included attending to clapping hands, ver- balization of the subject's name, following hand movement with eyes, and behavior on the ward. Only subjects with extreme spasticity (i.e., could not grasp and hold onto an object) were eliminated. Hospital records were then in— vestigated to determine CA, NA, and IQ. This study was to have included only subjects with an MA of at least two years, however, two younger subjects with recorded IA of 7 months were retained. Since the purpose of this study was to explore the effectiveness of a method for training the severely retarded, a representative sample was con- sidercd to be important and therefore selection was not made on the basis of type of retardation. 9 Faterials The subjects were taken individually from the hospital to an unused building in which the office and experi— mental room were located. The experimental room was carpeted, free of distracting stimuli, and contained two chairs (for experimenter and subject) and RbDRaFA. Stimuli for the individual subject, primary and secondary reinforcers, and Sticky-Gum for attaching stimuli were kept inside KUDRAFA. Stimuli and crutch cues were cutouts from Sears catalogues and children's magazines, and were pasted on heavy paper (1% x 1% in.). Categories of stimuli included: human (parts of body, faces, people jumping, people sitting,...); animal (cats, dogs, aobits,...); clothing (shoes, coats, dresses,..); household furnishings (plates, televisions, beds,...); toys (trucks, stuffed animals, bicycles,...); colors (any and all of the above, paint chips, colored pipe cleaners, colored tissue); and forms (3 and 2 dimensional plastic, wooden, and paper objects). Varied reinforcers included paper money, marshmallows, baseball cards, Christmas cards, trinkets, toys, magazines, and 2" x 2" ceramic tiles. Tally sheets (entered daily for each subject) listed experimenter, sub- ject, length of session, number of concepts trained for, number of slots open, number correct, number wrong, percentage reinforcement, reinforcers used, and comments. Apparatus The training apparatus, a Multiple Differential Response And Feedback Apparatus (KUDRAFA) is essentially the same as the hultiple Operant Problem Solving Aparatus (ROPSA) lO which was developed as a therapeutic procedure for hronic schizophrenics (King, Armitage, and Tilton, 1960). DUDRAFA is a 2' x 2' X 2' wooden cube, open on one side, with a sloping front panel. Independent of the rest of the front panel is a 16" x 16" panel containing a cross like aperture which can be rotated and thus oriented in four directions. Rotation of the board prevents a particular slot in the aperture from becoming a cue for a response. The M slots are each of different lengths and widths. The varying lengths and widths of the slots allow differential kinesthetic feedback for each response made. The slots can be closed off by means of transparent plastic barriers, thus allowing specification of a particular response with the gradual opening of the other slots. Projecting from the center of the crosslike opening is a metal lever wita a translucent plastic handle in which a light is mounted. This liaht is operated by the experimenter for a correct response. A small white pointer may be attached to the lever so that the subject can go in front of and behind stimuli. This lever is maintained in neutral position at the center by tension springs. The stimuli or crutch cues can be placed at any position on the panel with Sticky-Gum so that the subject can "move to the apple," "go next to the cat," "50 over the eyes..." Procedure Phase I - interaction, observation, and 0 selection: This pha~e consisted of observation of the subjects I on the wards, selection of children, and filling out the 11 general information page of the Concept Knowlejge Test. The subject was played with, taken outside, and familiarized with the experimental room. Also during; this phase the agent most reinforcing to the child was determined. Pretest was conducted in the hospital playroom and during interaction with the child on the hospital premises. Scores on pretest and retests were number of concepts known. Criterion for knowledge was performance of all four tasks under a concept. Phase IIE (Experimental Subjects) - training: After the subject had been femiliarized with the experimental room, shaping of a response to approach tie machine and to reach and hold the lever began. Guidance in the form of manually assisted responses and crutch cues were liberally used at the beginning of training and dropped out later» Subjects were trained from concrete to more abstract con- cepts (e.g., first "go to (object familiar to subject)," then "go to red," and finally ( "go through red"). Time in the experimental situation was gradually increased from h to us minutes, always stoppin before the subject became tired or snowed signs of the situation becoming aversive“ A retest followed training. Retest for group I experimentals followed 10 weeks of training, after which they were trained for 8 more weeks, were retested, and then went to phase III. Group II experimentals were retested after 3 weeks of train- ing, and then went directly to phase 111. Phase IIC (Control Subjects) - interaction, imitation, play actiV1ty: Subjects were engaged in play activities 12 on a 2 or M year level which encouraged muscle coordination, eye-hand coordination, and attending to the activity.* Again, the subject was returned to the ward before he became tired. In imitation activity the subject Was told "do this" in building up a sequence of tenavior such as patting head, extending arm, clappin hands, picking up crayon,... These 01? subjects were also shown pictures and read to from pre-school children‘s books. They were reinforced with candy at the end of each play session. Subjects were retested at the end of 10 weeks. Phase III - No contact period: During this time there was no interaction with the subject by the experimenter. At the end of 5 weeks the subject was retested to determine retention of the concepts over the rest interval. Training sessions were held 5 days a week. Each session the machine was checked over, stimuli and reinforcers selected, and tally sheet prepared. The subject was then taken from the heapital directly to the distraction free experimental room, using concepts along the way (e.g., push the elevator button, open the door, look at the red car,...) At first no more than 2 or 3 concepts were covered per session, with each session lasting no longer than h or 5 minutes. The situation was to remain a pleasant one for the subject and he was always removed before he became bored or fatigued. If the subject's attention wandered during the —-——- —..._..———-~_—-——-.———. .._. '———- U U of Physical Education, Lichigan S‘: 13 *Activities pladned by hiss Ca w, Department L; o experiment, the experimenter clapped his hands, blew a ‘\ whistle, or banged the front of the machine. A wide variety f7 and number of concepts in a time span of us to 60 minutes was gradually built up as the subject became more attentive. Concepts first used were such as up, down, riyht, and left. After these were mastered, more abstract and preposi- tional concepts were trained for. Three more advanced sub- jects were then trained intensively on the concept of numerousity by presenting l, 2, and 3 objects on the board with the symbols under them, then addition and subtraction of these objects. They were later trained on the symbols and words for numbers up to ten. Letters making up simple words, words in conjunction with pictures, and phrases using these words were then taught. A picture with the word were presented together at the end of one slot, the letters making up the word were presented at the other three slots. The subject was trained to go to one letter, opening up the slots gradually. Then he was trained on each of the other letters. He was then trained to go to the letters in a sequence which spelled the word, and finally to the word itself. After, complex words and phrases which could be used on the wards were taught these subjects. During the initial training of a concept the subject was assisted and given manual guidance, if necessary, as well as verbal instructions by the experimenter. Concepts and commands were repeated within a session so the child would not acquire a set to do something different each time. 1h Concepts were repeated across sessions to provide overlearning and give the subject the opportunity to perform a concept he already knows. Responses were shaped by reinforcing approximating behavior. Crutcn cues in the form of arrows and known t m H. }-Jo muli were liberally used at f rst and then drOpped out as a concept was better learned. Since the retarded subject tends to perseverate, all but the appropriate response was blocked off by means of the transparent plastic barriers. As the subject came to respond with shorter latency, the barriers were opened up one at a time. If the subject made an error he was told "no" or "wrong.' i If the error persisted »arriers were closed and gradually reopened. In order for generalization (i.e., transfer into daily life) of a concept to occur, we maximized the stimulus context in which it occurred. To accomplish tnis with KUDRAFA, we used a variety of commands and presentation of stimuli. We varied the position and orientation of stimuli by different placements and by rotating the face plate. Commands were varied by different intonation and phrasing, as well as mixing the commands. Secondary reinforcers (ceramic tiles) were built up on a percentage reinforcement schedule. For each response, the subject received a tile token, and for an increasing (over sessions) number of tokens he received a primary reinforcer. This type of schedule was introduced to prevent 15 extinction of the response and to decrease the rate of satiation by increasingly sparing use of the primary reinforcer. lo During the first week or two, leasniig proceeded slowly and there was poor retention from session to session. After this initial period of adjustment, the experimental group showed a fairly rapid learning rate. Subjects 1, 2, 3, and h (experimental group I) respectively acquired 53, ll, 33, and 19 concepts. Subjects 5, e, 7, and 3 (experimental group II) respectively acquired 3, 17, A \ 3h, and 3 concepts. In the control group, subjects ; and 13 lost one concept, subject 12 lost 3 concepts, subject 11 gained 5, and subjects 10 and 1M remained the same. sub- Flo ject number ll did show a slight increase. Th 3 subject would not pick out colors during pretesting. During play activity, it was discovered that he used colors approp- .L. riately and at retest he responded b0 naming of the colors. '1 ‘ 1 This subject was particularly shy and it is very ‘ lil ‘( ely that the increase is due to his feeling at -ter case 7"» H (3“) 9'1 ter se sion. A t—test U) with the xperimenter during the l (1’: based on the mean number of concepts acquired per week for both experimental groups as compared with the control group was sirnificant at the 0.05 level (t=l.73, df=12). C... ‘ Figures 1 to 3 incicate that etperimental subjects in L‘ niti 1 number of concepts ’i—Jo both groups I and II with a nigh t 17 tend to have a steeper acquisition slope (i.e., they learned more concepts within a certain time span than subjects with low initial number of concepts, However, there was no difference in retention. Eoth suijects witn high and low umber of initial concepts known showed good retention over a 5 week period. The only exception is experimental sub- ject number 3, for whom it was difficult to determine a reinforcing agent. For this subject the experimental situation became increasingly aversive and he was later excluded from the study. gu alia tive results The two subjects with Ra's of 7 month were retained in this study since the experimenter concluded that a 10 ng observation period was a more valid estimate of ability than a report which ended with the statement "the psychologist was kicked thoroughly through- out the testing." Subjects differed in the mount of time (‘11 that they could be kept in the experimental room and the rate at which tokens could be introduced and increased. One highly distractible subject would not attend to the machine but rather approached the experimenter. Another subject was then put on the machine and was reinforced while he watched. After two sessions of the other subject responding and being reinforced, the di stractit 1e subject voca lized and approached the machine while pointing to his mou‘h. This behavior was rein; orced and further training proceec ded quite well. a. Iiniature mars imallows were finally usec as the only 13 candy reinforcer since other candies required chewing and caused srbjects to choke, thus distracting the subject from the experimental situation. The light in the lever was found to be particularly reinforcing as is indicated by the gesturing and vocalizing of the subjects when the light had gone out one day. The experimental session had to be discontinued as a result. The tokens (ceramic tiles) seemed to acquire a great deal of reinforcing value to the subject. The sub- jects smiled and seemed to obtain pleasure from stacking up a number of tokens and turning them in. One subject even stole a few tokens to take back to the ward rather than turn them in. The tokens produced less reinforcer intrusive- ness than other tangibles (candy, trinkets) which were often investigated by the subject before putting them down or eating. Decreased latency for accepting the tokens was found to be a good indicator of when it was feasible to increase percentage reinforcement. Reading (letters, words, and sentences) were found to be highly motivating. The subjects to whom reading had been introduced (numbers 1, 3 and 7) were taken off tangible reinforcers and tokens and received a marshmallow and trinket only at the end of the session. Only the light and social- verbal reinforcement were Liven during the session. Haver- theless, reading was well learned and retained from session to session. All subjects rapidly learned their own name. Subject number 3, a semi-verbal mongoloid, responded well to animal pictures and household articles. Subjects number 1 and 19 7, non-verbal cerebral palsied spastics most quickly learned phrases that involved them and that could be used on tn= wards as, for example, "I am tired,” ”I an hungry,” ”Lleee give Rayford a drink of water." Tnese subjects were later given a set of 3" X 5" cards on wgicn tnese pureses were printed to use in tne hospital. A good understandin‘ of f >- \4 «I the sentences is demonstrated by the subjects' use of the sentences in a variety of contexts. Eor example, when asked "Jhat should I have said when I came in?" the subject ! H went to 'Good morning, hatny, the only correct alternative for the situation. Thus, verca mediation and its use had been introduced in both verbal and non-verbal subjects. T V) T l; 3 T 3 3 T. Q“ 3 P R . o ‘~~' '( V‘-v JJTI7 .L -- m --, - -n nvn~‘ -54.} 1‘. " £1.14 — A. .LLJJL lCO’ _ .3 Er. 20~ LTIOI T .4 7:11 1WD 3 UL: ‘... -L bITI O _r"r'~.‘r’-‘ V v.“,'J.L A u .351. H fi-r COX FIGURE 1. ll fiv-q-n, DOJJT 4"” O u COITRCL iTAL SUBJECT l AID r1 - .11- EXP ERIE-1 Uli-BtdtIJO'E’JOO ’IJO WLIJtU'E‘Td'Z—fi' 50‘ 4O 3O 20 10 p A RET ELITIC-LI RETE'ST - :- l l 5 10 15 20 Figure 2. COl-ICEET ACQUISITIOIT AID RETZI'TIOLI ELQERILEXML §_ 2 AID 001:1:ch §_ 9. (QCOli-CON L)‘ C) J hJ 12%;: E 2-« .l 02ewm020n *IJO PJbedEZGZ 7O 60 .333333 ©RETE3T AXREthrzzr RETEST NON-CONTACT: E ‘3 H13 L l ‘ I ——3 5 10 15 2o 25 WEEKS Figure 3. 0013033? ACQUISITIOEF AZTD 333331110}: CF mewMOZOQ mo mmwzdz 4O 30 2O 10 A RETEZTTIOIT D 17 77‘ ." "1 .L‘LLJT 4..) J J. pron-cozeice: E lo 4’23 I I l L . -lL 5 10 15 20 25 Figure 4, COECEPT ACQUI3I3103 AID RETEITIOE OF exessixzrelt s 4-AHD COETROL s 10. 24 NUMBER OF CONCEPTS .0 5 L l I «D vi ll .5 f T ml 3 3 H 5 Ma... . u «L T.“ mg ml T. ml 1 E E E n. ma “nan . G A. F p p O O O O 3 2 1 ._~ ry" " LIV) f _s_ 13. 'TROL fifi‘ ‘UVL- D mm _s_ 5 A: farm “IE-fl" r] - , ‘ .LJLL Ai-‘Jw' . V £14 50” NUMBER — O .4 V I 5 l 2) l s C E I I 3 T T T M4.“ V3 . V) O. n; 7.4 D. 7; .fii E T m.-. T C on. E T... _;._ in P R Du. .3.“ n... M... O . O A O L l mu _ Thu .|. 31‘ 8 l 5 . his 0 O O O 3 2 1 OF CONCWAPTS *1 ‘fi‘u‘f "I .242..le '( ii. r“I ‘~D \“.‘" :VI“ ILL.‘ ‘kAJ;-‘J...a-l- A ‘.V' "‘T I III.“ -‘ .LI :1} ‘9, I“ r‘TV \J COITCEET A 6. U. I‘ G *1'-.. Bldg CL i 12. "-1 aH—n'flw « - 1 , lgqngL O A ' g“ I 'r‘ ‘— . .L ‘9' .5. .n ”fl ‘ P I ‘1‘" ' ht; . I ‘4 :3 “‘"n‘ -Lg- TU‘ \Ix. nu HBER OF CONCEPTS E *3 3. av 20 10 7"?" 1" 4.1.4le c: F .11". v—‘t‘f’v‘ I [N I? .‘HHJ— DJ». 1. LA. "N "7 J.) Q ‘0 TI -9 .L’\ ‘- ISIT' A mnfiu it‘d-‘lv . '3; i" W --—~ \J ‘4 H CO *Ire L A F? l ( wmmmdv JCDHOC) '"IJO r“ . (new: 2d» NONaCONTACT: E A RETEHTION RP'T‘F‘C‘“ . L M‘A—dk—J ] o l I 5 10 'r_ —\ v‘u'rrw II ‘ . I ..J..-:-‘..J ’1'-.. 'r’" "1fl""‘(""““"1 anrv' *w ~1 “a" 1.-.. 1‘" .ure U. VUI.‘J.L.L’.L Au ,‘IJII 1w“: 231.2 *_'_J T‘V'T.‘ -“ -<‘ '7' . ‘ '1 'pl ,' - h‘ (I) ‘0‘ - r'V‘ 'X. . ' I‘m 9 1'.‘T m ! J‘J‘X; g-L.L....- .a.. Pool} ‘3 U ‘1‘.!1J ‘JI‘J. .L-L\I-L-I a 1+. A". ‘04 :13 15 «am 7» ”WT/W" A...‘ -41. .. ‘.._\¢‘.L. LL} CEAIYHER IV DISCUSSION The examples described above lend support to our premise that the severely retarded can learn, given the anpropriate conditions. LUDRAFA and the methodology contained herein provide these conditions and bring the behavior of the subject under control of verbal commands in both verbal and non-verbal subjects. Thus, as in other studies (Ellis, Barnett, and Fryer, 1960; O‘Connor and Hermelin, l§59) we find that the retarded subject can learn to use verbal mediators when trained to do so. The results indicate that there is great intersubject variability in the concepts being mastered. Subjects with a high number of initial concepts known tended to snow a steeper acquisition slope than those with a low number of erence in reten- H) t 1 concepts. However, there is no Qii }.J¢ 9': H. D H. tion of concepts once learned. Tiis is similar to studies which compared normal and retarded subjects and found no evidence for a retention deficit (Ellis, Barnett, and fryer, 1960; Nischner, Braun, and Patton, 1962)- Good learnini and retention of reading is probably explained by the fact that this was a highly reinforcing behavior for the subjects. The mongoloid boy had previously been shown msrazines with large colored pictures and had been 2 / given magazines as reinforcers. The two non-verbal cerebral palsied spastics had been read to from children's books and had previously been told to lift their left hand when asked "Are you tired?". dhen reading was introduced they wer told they would now be able to tell about themselves by reading a card and showing it to the experimenter. An important factor in learning was the sequential building of words and phrases which could be used by these subjects on the wards. Although retest and retention scores for further reading training are not available because of a quarantine in the hospital, inapection of daily records indicates that the acquisition and retention from day to day was even better than for other concepts. The example of the highly distractible subject implies the possibility of increased learning by using a model in training. Specific recommendations which may be made from the results of this study are: 1. Use a distraction free room so that attention is directed to the machine. Introduce new stimuli often to attract the subject's attention. 2. Although training is best approached as a game, it appears to be more reinforcing to call it going to school. The subject is often reinforced on the wards by the aides smiling and saying "Oh, Johnny's going to school now” or H 0‘3 "flhat did you learn in school today, Diane 3. Be kind but firm. Lee a positive approach. Keep 3O the situation pleasant but insure that the subject recognizes the concept as a command to which he is to respond. h. Se an active participant yourself in training, make wide use of manual guidance and crutch cues. Later be an active observer-reinforcer. S. Attempt to keep the subject active, do not allow him to get bored. Offer a variety of stimulus presentation and vocal commands. 6. Remember the characteristics of each child and con- sider his individual abilities and attention span. A given duration of experimental session may be appropriate for one child but not another, just as a certain reinforcer may be appropriate for one and not another. 7. Introduce new concepts before the subject becomes fatigued, remove the subject from the experimental room be- fore he becomes fatigued, 8. Make available powerful and lasting reinforcers to elicit the behavior wanted.. Since the retarded subject tends to perseverate, it is important to specify and elicit the correct response. This is done by giving the reinforcer (tangible and social-verbal), a light for a correct response, ' for errors, and prevention of incorrect saying "wrong' responses by the use of barriers. 9. Make wide use of social-verbal reinforcement but do not use so much as to caLse reinforcer intrusiveness. A H g. simple statement such as yood Lirl" with appropriate intonation will suffice. 31 10. Progress slowly, offering familiar activities and using repetition across sessions and within sessions. This gives the subject an opportunity to receive reinforcement in performing something he knows well. Use repetition to build concepts on what the subject already knows. ll. Introduce the token system as soon as possible. This produced very little reinforcer intrusiveness and seems to specify the response as correct better than tangible reinforcers. The subject seems to anticipate receiving a reward for a stack of tokens and works without distraction to accumulate them. 12. Naximize generalization and transfer by varying the cont xt in which the concept occurs. In the experimental session, vary the commands and their intonation. Rotate the board and change the stimuli and their position. Use the I concepts outside of the experimental room as '30 through the door." 13. Randomize the stimuli 0 that only one aspect U) (the concept) is consistently relevant over a series of like commands and all other possible cues such as position, shape, or size of slots do not act as relevant cuesm 1h. Talk to the subject even if he is non—verbal. Tell him in simple language throughout training what you are going to do. For example, in removing tokens you might say ”Kathy you are so good we‘ll just have me and the liLht tell you when you're right or wron:. You can have a candy and ribbon when we're through." 32 1: Have the subject verbalize. Ask him cuestions. Shape his verbalization by reinforcing sounds at first and later reinforcing only progressively more suitable responses. 1 .0 Future Directions A new fithAPA Cl basically the same design, but more compact and with a mechanized rotatinp front plate is now being constructed. The compact rachine will be portable and thus can be taken to schools, homes, te 9‘.- and bed patients. The automatically rotating face pl is circular and produces less distraction since the esperi- m nter will no longer have to interrupt training to change orientation of the slots. This new machine will be used with the subjects who are presently control subjects in this study. They will serve as their own controls since a baseline for number of concepts known has been established over time in interaction activity. This new study will also determine the effective- ness of training the subject to follow the activity of the experimenter and training to respond to the command "do Ho this." It s assumed that imitation activity will facilitate learning and avoid the slow initial learning and short re- tention period. Also, a study is presently being des'gned to use KUDRAFA with culturally retarded children. The children will be trained on basic geometric designs, numerousity, and reading. The control group of culturally retarded children will be matched on CA, EA, and pretest number of concepts. They will be trained individually by conven- 33 tional classroom methods. A retest on concepts and IQ test will determine the effectiveness of the two methods. . “fin.“ u.’ ' d;‘.~u-t w.. \J- i 'P. V---“-- r a» .. .M» 3 .r e éI‘B .. V' I non-c OI". 'lS ‘ 4“ (”‘i - LL .«A 1. Cu I1 “ded c I , reta .4; l5 2 .~ 1 Lu 2“ .3 .1 S .. . WA .2 .3 Cu 39 s .«u a. Q» E n. ‘ o. i C C mg. 5 a) “V‘! H. ”—1 . ”4 « wrvl , -_\.../-.l&- I \ C“, 5’1““ T“"‘ ‘vfl---..--.‘- 1; d 9" \V i!“ V -. \..-- cine t to over 'ples flCl pri - V.“ C‘l-l' :1 5-.- e. ; ‘1f_C|-. _‘ je \. five-‘ 3 by pretest led to fa (C b. .3 a— 5... ‘ 001108: of it'er. , er I 'l 1‘. “(Aid A v hjects c 1 at 81 FT). .1 :3. .0 y i .U the non-verbal I), :1" v e an .‘ t «Lu-Lu " t ,4 J. s conclude we It ringing tie b 1 (— r ‘.Lf.«v "1"" .. ... \ . l stract verbal co at c. of «II. d nb nD my C; 9W4 \ srnin" tudy for further le Q L.— his t AL) .qufiiJ r‘flJi:r U133 Baumeister, A. A. A comparison of normals and retardates with respect to incidental and intentional learnin5, Am. J. Lent. Defic., 19 63, 68’ LLOLL’bOU O Benoit, E. P. Relevance of Hebb's theory of the organization of belwaiior to educational research on the mentally retarded. Am. J. Hent. Defic., OS 61 ‘c “1’ / 7’ 3 “)7-DUOO Denny, H. R. in Intwna1onal review of mental retar ation. Vol. II, Ellis, H. wed.), new York, Academic Press, 1966. Denny, H. R. Research in learnin5 and performance. In Stevens, H. and Haber, R. (Eds.), Lental retardation: .A review of research. University of Chica5o Press, l96h. Ellis, N. R. and Pr;.rer, R. 8. Primary vs. secondary reinforcement in simple discrimination learning of the mentally deficient. PSYChO-io (.313. 1959’ b" 67‘700 Griffith, B. C., Spitz, H. H., and Iipnan, H. B. Verbal mediation and concept formation in retarded and normal subjects. J. Bro. Psychol., 1959, 58, 2&7-500 Hayden, F. J. Physical fitness for the mentally retarded. April 1966, J. Health, Phys. Ed.,and Rec. House, B. J. and anman, D., Effects of tractice in the delayed response of retardates. J. Come . Physiol. Ps5rchol., 1961, En, 233-66. King, G. E., Armitage, S. G., and Tilton, J. R. A therapeutic ap\roach to schizophrenics of extreme path0105y: An opersnt—interpe‘sonal method. J. Ab.. Soc. Psychol., 1960, 61, 276-286. Luria, A. R. and Vino 5radova , O. S. 1959. An objective i vesti5aticn of the dynamics of semen tic systems. :rit. J. Pc*chol., 1959, 50, 89-105 0 O'Conner, N. and Hm muslin, 3., Learnin5 and recognition in imbeciles. In Richards, B. J., (Ed.) Proceidin5s of the London Con1erencc on the scient ic study of mental deficiency, J. I. Dage 1am, England, Mary & Baker Ltd., 1962. Singer, R. V., Incidental learning in rater rdedi normal children Ph.D. Dissertation, 1963, Hichi5an State University. 36 r!‘ 1 ‘1 i ‘3' Smith, M. P. and Heans, J. 3., Effects of type of stimulus pretrainir on discrimina ion learnin5“ in the mentally retarded. m. J. Ment. Defic., 1961, 66, 239-265. Stevenson, H. W. Learning of complex troolexs by normal and re ta :ded SUbjeCtSO AIPLGI‘. Jo 38H‘.D~;191C., 19C/O , 61; ,1021-2‘6. Terdal, Leif G. Stimulus sa.tiation and.mental retardation. Ph.D. Dissertation, 1965, HichL 5an State University. Hischner, G. J., Braun, H., and Patton, 3., 1960. Pre5rer s reports 1earnin5 set projects. University of Pi tsbur5h, U. S. Puelie Hea1+h%rviee. Zeaman, D. and House, B. J., The role of attention in retardates discrimination learning. In Ellis, w. R. Ed.) Handbook of iental deficiency: psycholo5 ical theory and resecr x. New lork: neu1aJ-H111, 1963. Zigler, E. F., Hodgden, L., and Stevenson, H.'H. The effect of support and non-supLort on the performance 01 normal and feebleminded children. J. Pe:s., 1956, 26, 106—22. K}! .4 p. .. C.) /C1.$C.m(f CD Ace/195:5? fit“, “9((“ja Subje at "3 Name 5 EKDIII “rn‘L H 3T0! Pre-test/r6—test Birth Date Recorded I. Q. Sex Recorded CA Recorded MA .4 I ‘1 ”W "' {up I Date started Date finished Ward mm In Q0 1:83th Date institutionalized Physical Condition Spastic Amb uJatory/nonwambula for; they in (Wee Verbal/Non-Verbal Right/Left Handed HBdication Given _5 Type In Speech Therapy (If nonmembulatony what are ) GENERAL160fiHRHTS (i.e. what nurse has to say about patient, or anything you notice that is pertinent to stufiy) 'Pnn +0. *1. r .1. I I "5 ‘1. It. . .u, < q 1 q. . 0.. n. ... 3.1. "7"”. ' I (Ar! p: I 3 9. ‘._4 .J '13 E .1 . 5 ., I I... rv . ,T 1. r . an us... . t. ”.06. ‘. . ... .t .1. 0......) p ..u s o . 3“ 7" 'I "f I; i ( ‘I'J V‘I 3“ r L! .u x. ‘ .. LA \~ 5 , V'I '. .1 / q _ L ' W ,1 39 ‘1“) 53 T171 1} “ram - ' ' my hand the toy the chair the truck my hand the toy Pull the chair the truck Reach for the dog (other is cat) Reach for the ball Reach for the block Reach for my head . POINT~-(Have child use pointer from Machine) Point to the light 1. PUSH a. Elfih b0 Mb c. Push do Push 2o PULL a. Pull 5. Pull Co d. Full 3- REACH a. b. C. do h a. ho C. d. Point to the dog (have 2 pictures on tablcu-other is cat) Point to the ball Point to yourself 5. TOUCH a. Touch my hand b. Touch the table c. Touch your eye d. Touch the tqy 6. STOP 3. Have walking subject step b. Stop rotating wheel c. Stop rolling ball d. Have subject pushing truck and tell him stop 7. GO a. Have.8topped subject go b. hhkn wheel go c. Make ball go do rhka true}: go a. DON’T/DO nor a. Have subject playing with doll and then say don‘t play b. Have subject.walking and then tell him don't walk c. Have subject eating marshmallows and than say "don"t cat" do Have subject scribbling on paper and then tell him “dealt write" 90 UP 30 Look up b. Stand up c. Put hand up d. Put this tqy up 10. Bil-m a. Look down .- b. Sit down Co Put hand dCHn d. Put toy down 11. Rough 1.2. SMOOTH a. a. Pick out difference be Ween rough and smooth sandpaper b. b. Brietly pipe cleaner versus plain pipe cleaner c. 0. Glass versus sandpaper d. d. Rough block/smooth block 130 BIG 11.. MEDIUM 1:3. LITTIE a. a. a. Pick out hig/ m::iium/little pencil ”—‘B. “—1). “I'b. Big! radium llittle bottles 0. c. c. Which box is big flmdium/littls d. d. d. Which block is big /mdium/littlc 16. SHORT 1?. LONG 11. a. Pick out the short/long pencil b. “flab. Short/long wire ' . c. Which piece of paper is short/long d. d. Which pipe cleaner is short/long l8. THICK 19. THIN a. a. Pick out the thick/thin pencil b. b. Pick out the thick/thin block c. c. Pick out the thick/thin lines (on paper) (1. “Ind. Thick versus thin cardboard strips 20. STRAIGHT 21. CHOOKED . a. a. Pick out straight pipe cleaner from crock-ed pipe cleamr+ (making sure pipe cleaner is crooked not curved) b. b. Straight versus crooked 11:..."t‘m'fier c. c. Straight versus crooked cut out pieces of paper . (1. Straight versus crooked wire 22. I? 23. OUT 3. :1. Have subject plawd so that the Opening of a box is away from him. Tell subject to place object 1n/or to take object out of the box. 555k). b. Go out/in room c. c. Put object ill/take object out of was tobaslcet d. d. Put pencil in my hand-”take it out of W hand 21.... cums 25. STRAIGHT a. a. Pick out straight pipe cleaner from curved pipe cleaner (making sure pipe cleaner is curved. not crooked} _b. b. Straight versus cunrod lines 011353.} c. c. Straight versus curved cutnout pieces of paper (1. (1. Straight wire versus curved wine 41 28:, *a0 b. Co d. 30., ac .mba C o m:dn BE ‘5 I If?) Putt .3; uaci. Hrs-:31“ tic-3 b: Pia ace your h’ and 0411” :32‘ My 1:311:31 Co Put t3" Mei/om the tulle do P511; “013.0"1'1/13333:the (”131? L) ',,r,t "“‘au 29 o I}? F‘s". 13‘s} ‘1‘ OF 3. Put toy in front ref/behind '33:): CO in from; of nag/behimj r15:- Put your hand in front of/tphihi yo. Put your "33;! in 1 :0 front of finhind Ctn rt??? THROUG. Poke your finger through this piece of page? P’xt your timer throu3h t‘te hole Come t rough the door way Ptu’sh the m’u’ble {ah ro 1,1311 tine tn. 3]., EE’A".2”?1’3N a. 31:. your hand be. tween my 1121' 33 .b0 Push the my be Queen tho-’38 taco clocks -~ 0. Fax? {363311-30 I t. :3 Chair and the this :*‘.-. Walk be’?’ wee-:1 me and t‘. 3 245311 32317311? 1‘0 ea Stand next to tire halal; ”‘13:. Go next to ”he table 0-.- Plea-n your mm next to 22);; 1’3.2’..--- . . ..‘, :, .,.'.. s. I J . ”n . .. c. c. 11.. 5.: ~05 mi... ow, u'-.';v—"U:.' « e n h? Y- «:2. ”at“ an...“ ' ' .' .’ _‘ "' ‘ ‘ .. -,~.-- -. \ - -. , ,v - . —_ - n "w {a o my; \‘ mauajum we a? a». (u. we ”1“) 11:, 3D s - C :5 (3.4} '5’th La :41 O u—‘snu. " a q -..C v; tun-m ‘ ‘4' ' a 8.4.3.30 .marMu—T'u ’ cAdo I—Ittm \ J i .‘ J (a . no N Na“? " ’0. am N _., K. . - s' Au Lgbu mm“. - : :5 ‘. YA"! vi,l L'.;v ]‘ ' 1 ,;P(_‘)‘ r. f .- 11‘. L.,' -V_~1 :1” TI. nn. 2‘7 «1 m,- 41 . "a v-D a 'or or” "hm Pan 5." '.. -.u-‘ “in. ('1 _ u. .\ I 6" 8.1”? ') ‘-’.:~31.‘Q S7. A mun-r. CL-““IH TU‘IH mv‘ .1 V 1 ‘5! Tu n1 ‘I‘u'm ‘ P1 ~ L ‘3371 ! '. '.' s": '4 L -3 \ .1”: 1H. .J aflb- f cgd, ZYNI.‘ 731...?) aroma}. t0: crank fins (T‘ .. 60¢ Emmi -J mm C -..~ W”; . O from? wry: left/rt" {H3 ti‘sy C; 74- 45 ‘1 ‘I I. m a . .. .-. _ . 9‘ "14"; uu' . v - \ . . a“ ‘1'. a". ’f‘ 1A 1“"!- r .4 ' 31:11.; 1; s m; r?" ‘\ .'~ ‘ 1 .Liix. ‘3. .1 L -" *m ~ 'mlh‘ ‘Q~“h2.:f‘l" - u—-~_‘u. at to fio each of d :2: 21—17 kg ~--;.. u: ”e .w-_ ..'“L‘ ‘W ”HM-‘9‘ Oil") 7.3 10'de ‘13 {5} Subject EXperime 1 f0 NOU‘IJ-‘rw Control: 9 10 11 12 13 14 up Sex, Chronological Age, Lentel Age, IQ, and Etiolory of Retardation H.) For Exterimentel and C ntrol Subjects in RA in are & Years & the Konths 13 Etiolorv ._____111 --te.l: F 33-0 3-7 11 congenital cerebral palsy X 5-6 0-7 11 encephalopathy I 13-2 3-0 15 mongolism K 12-1 2-3 19 enox e at birth F 13-2 2-5 18 develOpacatal cranial anegoly H 29-0 4-3 14 congenital cerebral pclsy f r:— r 2 '- / r‘ M 1" ' r: 7, 1-“ ‘n‘ - n (V -,r h ~-J J—; 20 coibeeitcl Cereoiel peled H 13~3 2-5 20 Longolis; H 6-3 -8 ll enceyhelopathy “j H 7“ .q 1}) O H F" I" O\ hydrocephalus, arrested x 19—4 3—o H U] encephalooethy due to syphillitic (prenatal) infection H 17-5 2-6 undifferentiated 1...: U! I CO 1—: U] ’d H 7“ KL) erythoblaetosis fetelis H \fl I O ['0 m R) H O\ r. "r . J'- '1 - ~ . vo- conoeliual cereore p; s; 45 Jud - ~-., ‘1. rw .- .- *1 T‘ “r, '. 7.. -- J. '2‘ n-¥-« 1 ti 4- -t ,. | .\’f.' f. . ’3 N 3‘} 5 — ’3:- N 3,“ r‘ ‘1 - ‘ '. .\ ‘ 1' b“ f f) C '3 --C/ (.3 . t 5.---KJIO‘1'3 'nL .C '3 (J O i L U r L’ -- “L" V -‘~ Q ~‘ ~ {I [-4 ‘v‘ U U -- 4 I v U ‘u i.) t L. “‘f a 3 n -.a H) *5. . ° 1 4. r1 ., experinentel wrong I: Eon- Retention Pretest Tre ning Retest Training Retest Contact Retest S biect Score Sgpn Scor Sren Score Sign Score 1 45 10 wks. 61 8 mks. E3 5 Uns. 97 " fl '1 O N H PO H }. F4 [U 10 I" N R) V.) U] u: C\ [U C\ e- " H 41‘ U! 1..) \D R) \1) '\ \fl 0) W KO 3xceri.citel Grout II: ron- Retention Pretest Training Retest Contact Retest DJD‘eCt Score Stan Score Eben Score 5 3 0 wks. 6 9 wks. 6 6 26 u n 43 n n 49 L C C C3 C3 CO H O CO I. 13 N I! Control Group: Pretest Interaction Retest gubiect Score Seen Score 9 l 10 WIS. O 10 21 n n 21 11 4. O p- Ui 12 23 n u 23 13 2 u u l 14 7 u n 7 usutumtwgtguuiyujulfllmlmymlmm