Download Effects of Self-Instructional Training on Math and Printing in Retarded Children and more Exams Printing in PDF only on Docsity!
JOURNAL OF APPLIED BEHAVIOR ANALYSIS
A SELF-INSTRUCTIONAL PACKAGE FOR INCREASING ATTENDING BEHAVIOR IN EDUCABLE MENTALLY RETARDED (^) CHILDREN Louis D. BuRGIo, THOMAS L. WHITMAN, AND MOSES R. JOHNSON
UNIVERSITY OF NOTRE DAME
The purpose of this study was to develop a self-instructional package that would aid highly distractible retarded children in increasing their attending behavior in a training and two generalization (a one-to-one and a classroom) situations. Three untrained sub- jects were monitored for general comparison and^ social validation^ purposes. One^ of these control subjects was distractible and the other two (criterion comparison) were evaluated as not having attentional problems. A multiple baseline design was employed in which training was sequentially introduced across subjects. During training, the ex- perimental subjects were taught through self-instruction to focus their attention and to cope with two tasks, math and printing. After learning the self-instructions the subjects were systematically and sequentially exposed to photo-slides of distracting situations, to audio-distractors composed of noisy lunchroom^ verbal^ peer interactions, and^ to^ in vivo distractors provided by kindergarten children^ playing with^ wooden^ blocks^ in the^ training setting. The entire training procedure was handled in^ a^ game-like context to^ maintain subject interest and to facilitate generalization. The results suggested that the training package produced direct and generalized changes in^ self-instructional behavior. In addi- tion, a decrease in off-task behavior occurred during math, printing, and also during a phonics program in the one-to-one and classroom situations. However, reliable changes in academic task performance were not observed. Finally, no systematic changes on^ any of the dependent measures occurred for the three untrained subjects. DESCRIPTORS: self-instruction, attending behavior, academic behavior, generaliza- tion, retarded children
Within educational settings, teachers recog- nize that students must sustain attention to school-related materials and activities if learning
This paper is based on a thesis completed by the first author^ in^ partial fulfillment^ of^ the^ requirements for the master of arts (^) degree in (^) psychology at the University of^ Notre^ Dame.^ The^ research^ reported herein was supported by a National Research Train- ing Award #HD07184-01. Sections of this paper were presented at the annual convention of the Asso- ciation for the Advancement of Behavior Therapy, San Francisco, December, 1979. The authors thank Carol (^) Glass, Kathy Larsen, John Scibak, Sandy Levers, and Jim Little for their assistance in the project. Spe- cial thanks go to John Borkowski, Ellen Ryan, Denny Reid, and Tom Merluzzi for their helpful comments on (^) the manuscript. Thanks are also due to Sr. Ann Fox and the children of St. Bavos elementary school for their cooperation throughout the project. Reprints are available from Lou Burgio or Tom Whitman, De- partment of Psychology, University of Notre Dame, Notre Dame, Indiana 46556.
is to occur. Where attentional deficits exist, ini- tial emphasis is often placed on the teacher for
ameliorating this problem. Since the advent of
behavior modification a variety of techniques have been developed to modify distractible, non- attentive behaviors in normal and retarded popu- lations, including: positive reinforcement of in- compatible behaviors (Alabiso, 1975; Patterson, 1965; Patterson, Jones, Whittier, & Wright, 1965; Whitman, Caponigri, & Mercurio, 1971), time out procedures (Johnson, Whitman, & Bar- toon-Noble, 1978), and aversive stimulation (Forehand & Baumeister, 1970; Reardon & Bell, 1970). Recently, a number of behavioral therapists have emphasized the need to switch the locus of control for an individual's behavior from exter- nal agents (e.g., the teacher) to the individual himself (cf. Kazdin, 1975). They argue that 443
1980, 139 443-459^ NUMBER^3 (FALL 1980)
LOUIS D. BURGIO et (^) al.
teaching individuals to control their own behav- ior is educationally more efficient in that it frees the teacher from many (^) routine supervision re- sponsibilities and because once learned it allows children to effect positive changes more readily in their own behavior across situations. One ap- proach to teaching individuals to control their own (^) behavior has involved self-instructional training. The self-instructional approach evolved from early investigations of the effects of verbal
operants on motoric behavior (Bem, 1967; Lo-
vaas, 1964; Luria, 1961; Vygotsky, 1962). The results of these studies suggest that verbal in- structional procedures can facilitate the acquisi- tion (^) of new behaviors and increase appropriate responding in^ a^ variety of^ situations. Since these seminal investigations, self-in- structional procedures have been used success- fully in modifying a wide range of behaviors
(cf. Burron & Bucker, 1978; Meichenbaum,
1975), although researchers have confronted some difficulty in effecting change in academic
task behavior (cf. Robin, Armel, & O'Leary,
1975). With the exception of a study by Gural- nick (1976), research has typically employed adults (^) and children who (^) are of normal intelli- gence. Guralnick (1976) compared the effective- ness of feedback, modeling, and self-instruction techniques in developing problem-solving strate- gies for complex perceptual discriminations with a group of educable mentally retarded children. Results indicate that only the self-instructional approach significantly increases performance. In the present study the (^) utility of a self-in- structional program with highly distractible re- tarded children was examined. The use of a self- instructional procedure with hyperactive and highly distractible children was suggested by Luria (^) (1961). He stated that incorporating the child's own speech in a treatment program for hyperactivity would decrease the opportunity for disruption of goal-directed behavior and fa- cilitate the organization of the children's own activities. Guralnick (1976) has similarly con- tended that self-instructions channel an individ-
ual's attention skills in selecting the relevant
cues in a situation. In two earlier studies (Palkes,
Stewart, & Freedman, 1972; Palkes, Stewart, &
Kahana, 1968) the authors found that training
hyperactive children to use self-directed verbal commands improved posttest performance on
the Porteus Maze (a measure of impulsivity)
when compared with control subjects who sim- ply practiced the training exercises. However, until the present study no attempt to evaluate the effects of self-instructional training on dis- tractible retarded children has been conducted. The self-instructional (^) package used in (^) this study provided to the subjects: strategic inocula- tion and coping self-statements to assist them, respectively, in completing arithmetic and print- ing assignments, in^ ignoring distracting stimuli, and in dealing with task failure. In addition, be- cause behavior modification programs have not always been successful in^ effecting generalized behavior changes across situations (Stokes & Baer, 1977), several procedures for facilitating generalization into a one-to-one and classroom setting were introduced. Specifically, during
training a "storylike" instructional procedure
(Bornstein & Quevillon, 1976) employing class- room imagery was used to assist the children in
recognizing situations where self-instruction was
appropriate; distracting stimuli similar to those likely to be encountered in the classroom were
introduced; and finally, multiple exemplars
(Stokes & Baer, 1977) of academic tasks encoun-
tered in the classroom were employed. In order
to evaluate the direct and generalized effects of training, the subjects' use of self-instruction in the training situation, in a one-to-one and in a classroom situation, was assessed. In addition, changes in the subjects' (^) off-task behavior in the latter two generalization situations were moni- tored. The final goal of this study was to examine
whether correlated changes in academic behavior
occurred. In order to evaluate this question, per- manent-product data were collected during the math, printing, and phonics programs in the generalization settings.
444
LOUIS D. BURGIO et al.
events which (^) commonly occurred in the class- room (transfer II) setting.
Response (^) Definitions and Rating System
Self-instructional verbalizations were rated to assess whether the training program succeeded in teaching children to self-instruct and whether the children self-instructed in the transfer situa- tions. Off-task behavior was assessed in the trans- fer situations to see if there would be a change in the frequency of this response during self-in- structional training. These target behaviors were
defined as follows:
Self-instructional verbalization-Statements
made by the child pertaining to the appropriate
performance of a task. Specifically, a child was
rated as self-instructing when he or she made one of the (^) following statements: (1) asked a question [e.g., (^) "What does Sr. _ (the
teacher) want me to do?") (2) answered the
question (e.g., "She wants me to draw this word."); (3) provided direction on how to to do
the task (e.g., "First, I should look at both num-
bers," "I take away 4."); (4) reinforced himself
or (^) herself for completing the task (e.g., "I did
a good job," "I'm doing real good so far."); (5)
provided a cue to ignore distraction (e.g., "I hear
people talking but I'm not going to let them
bother me."); (6) specified how to cope with
task-failure (^) (e.g., "Oh, I was (^) messy in (^) printing
that word. That's okay, I'll^ be even more careful
on the next word."). In-seat, off-task-The child is sitting in his or her chair with buttocks touching the seat of the chair but is not performing the assigned task properly. For example, the student might be looking about the classroom or (^) talking to a class- mate but would not be (^) looking at the task material. Ratings were taken for each experimental stu-
dent three times a week on Monday, Wednesday,
and Friday mornings in the training and two transfer (^) settings. A fourth afternoon (^) rating ses- sion, scheduled randomly, was also^ taken^ on one of these (^) days each week. (^) During each session
behaviors were recorded during a 15-min train-
ing period, a 5-min transfer I period, and during a 1 5-min transfer II (classroom) period while the students performed the arithmetic, printing, and phonics (classroom only) tasks. In the training and transfer I, but not in the transfer II (classroom) settings, the six types of self-instruction (e.g., "question," "answer") were subcoded in an event fashion. Specific self-in- structional components were not recorded in the classroom because the noise level in the room made it difficult to (^) obtain a (^) reliable measure (^) of these behaviors in this setting. In the transfer I and II settings the incidence of self-instruction, irrespective of type, was rated in an interval fashion. An interval recording system was also used to rate off-task behavior in the transfer I. and transfer II (^) settings. Off-task behavior was not recorded in the training setting. When (^) rating self-instruction and off-task behavior with an in- terval system, these responses were scored in terms of their occurrence or non-occurrence within successive 10-sec intervals. Each specific behavior was recorded only once within an in- terval. A cassette tape recorder was used to record
self-instructions in the^ training setting. A^ video-
tape recorder was used to record off-task and
self-instructional behaviors in the transfer I pe-
riod and off-task behavior in the classroom set- ting. Thus, all ratings of the target responses
were made from these tapes except for self-in-
structional behavior in the classroom which was rated as it occurred. Performance measures-In addition to the preceding behavioral measures, measures of the
subjects' performance quality on the arithmetic,
printing, and phonics tasks were taken daily in the 15-min classroom situation. The overall
number of problems completed on the arithme-
tic task and the number of letters printed on the printing task were tallied for each session. In addition, a percent correct measure was calcu- lated for each task by counting the number of
correct answers given or letters printed and di-
viding them by the total number of problems completed or letters printed. The writing task
SELF-INSTRUCTIONAL PACKAGE FOR ATTENDING BEHAVIOR
was evaluated by a method derived from Hel- wig, Johns, Norman, and Cooper (1976) which utilized transparent overlays (^) to measure devia- tions of writing samples. A deviation of more than 2 mm from the standard was considered an incorrect response.
Children's distractibility rating form-In or-
der to assess the teacher's perceptions of how distractible the children in this study were, a dis- tractibility rating form composed of five ques- tions was administered to her prior to the initia- tion of the study and again at the conclusion of training. The questions on this form probed the teacher's views concerning the children's (^) ability to (^) maintain attention to their work (e.g., "Does this child (^) have trouble concentrating and keep- ing his mind on one thing?") and the extent of teacher prompting needed to assist the child in completing a task (e.g., "Does this child require prompting to finish his work?"). A 5-point scale indicating the extent to which the child displayed a particular (^) problem was used by the teacher in answering the questions.
Reliability Assessment For reliability purposes, at least twice in each condition a second observer rated the students' behaviors from randomly selected videotapes of the transfer situations and cassette recordings of the training sessions. Similarly, for the same pur- pose, a second observer was brought into the transfer II (^) (classroom) setting to assess self-in- structional (^) behavior. In calculating reliability coefficients for off-task behavioral and global self-instructional ratings in the transfer I and transfer II^ settings, individual rating records were compared on an interval-by-interval basis for each child. The reliability coefficients (^) were calculated by dividing the number of agreements of occurrence of the behaviors between the two observers by the total number of observations. For off-task (^) behavior, interobserver agreement
ranged from 53 to 100% with a mean of 83%.
Interobserver agreement for global self-instruc-
tion ranged from 73 to 100% with a mean (^) of
87%. All of the low reliability coefficients (<80%) obtained, occurred when the fre- quency of off-task and global self-instruction was extremely low (<4 occurrences). This occurred
three times. The reliability of the event recording
system used in measuring the occurrence of specific component self-instructions was assessed through the following formula: % agreement = (^) smaller number of occurrence/larger number of occurrence (^) X 100. The (^) percentage agreement across all reliability checks^ ranged from 81 to 100% with a mean of (^) 90%. That (^) is, in no in- stance did observer agreement, when a specific self-instructional component was rated, fall be- low 80%.
In order to assess the reliability of the rating
system used in evaluating the children's perfor- mance on the academic (^) tasks, each child's task performance was rated at least twice during each phase by dividing the number of agreements of
occurrence by the total number of agreements
and disagreements. Specific reliability coefficients
for the accuracy measure ranged across children
and tasks, from 97 to 100% with a mean of 99 %. Interobserver agreement for the rate mea- sure was (^) 100% at all times.
Design A multiple baseline design across children was employed. Training procedures were sequen- tially introduced to the two experimental chil- dren in the training setting. No intervention oc- curred in the transfer I or classroom setting. The behavior of the control and criterion comparison children, who did not receive (^) training, was moni- tored (^) throughout the (^) study in the (^) classroom.
Procedure Baseline. (^) For the two experimental children and one control (^) child, behavioral ratings were
taken in the training, transfer I, and classroom
situations. The two criterion comparison chil-
dren were observed only in the classroom setting.
No experimental manipulations were initiated
447
SELF-INSTRUCTIONAL PACKAGE FOR ATTENDING BEHAVIOR
another child who is trying to complete a task). The audio distractors were supplied via a tape recording of the children's peers interacting loudly in a lunchroom setting. Finally, to supply in vivo distraction, kindergarten children were introduced into the training setting and in- structed to play with wooden blocks and other noise-inducing (^) objects. During the initial presentations of the various distractors, the experimenter modeled the distrac- tion-ignoring self-statement by saying: "If I were in the classroom doing my work and saying the words and something like this happened (i.e., the visual, audio, or in vivo distraction), I would say something like this: 'I'm not gonna look, I'm (^) gonna keep doing my work.'" As each dis- tracting stimulus was (^) introduced the child was asked to identify it (e.g., "People are making noise."), and then asked to verbalize the distrac- tion-ignoring self-statement. When the subject could verbalize this appropriate self-statement along with the previously learned self-statements for a given set of distractors (^) (visual or audio) unprompted for three consecutive sessions, the next (^) type of distractors was introduced. The chil-
dren were^ continued^ in the in^ vivo^ distraction phase until training was terminated.
Maintenance Due to time limitations caused by the end of the academic year, only limited data were col- lected in this condition and only for one child. The first child was (^) gradually (over a 4-wk (^) pe- riod) (^) given fewer training sessions, at a rate of one less per week, until the last week of the study at which time training was phased out com- pletely.
RESULTS
Self-Instructional Behavior Training setting. Table 1 shows the mean unprompted frequencies of the various self-in- structional components during the training con- ditions. Although no self-instructions occurred prior to training, after training both experimen- tal children learned to verbalize each of the self- instructional (^) components without (^) prompting. The low mean frequencies of occurrence of the coping self-instruction in the math and printing
Table 1 Mean frequency of various types of self-instruction by the experimental children across tasks (math and printing) during the training condition in the training and transfer I settings.,
Self- Children Question Answer Task Coping Distraction reinforcement
TRAINING Math Judy 12.7 1.0 59.3 1.0 8.7 14. Angie 15.6 1.3 65.3 0.0 11.3 13. Printing Judy 1.6^ 1.2^ 161.4 4.1 14.3 14. Angie 2.5 1.4 199.1 0.3 11.4 14.
TRANSFER I Math Judy 8.4^ 0.9 32.1 0.3 0.0 9. Angie 5.4^ 0.7 33.2 0.0 0.2 4. Printing Judy 1.1 1.0 118.6 0.8 0.1 9. Angie 0.9 0.5 83.6 0.0 0.1 4. aSelf-instruction did not occur during the baseline condition.
449
LOUIS D. BURGIO et a.
tasks and the question self-instruction in the printing task were a function of the nature of the rating system and differential task demands. Although not presented in Table 1, the data also show that both children began producing some unprompted self-instructions as early as the first training session. Transfer I setting. (^) Figure 1 shows the per- centage of intervals in which general self-in- structional behavior (collapsing across the vari- ous types of self-instructions) occurred across tasks and conditions for the experimental chil- dren and for the (^) untrained child in (^) the transfer I setting. The two (^) children who received training showed a high frequency of self-instruction on both tasks in this generalization situation whereas neither of these children prior to train- ing nor the third child who never received train-
ing showed any self-instructional behavior. Both
of the trained children displayed a higher fre- quency of self-instruction on the printing task. Table 1 shows the mean frequencies of the vari- ous self-instructional components that were emitted in the transfer I setting by the two ex- perimental children on the math (^) and printing tasks. With the exception of the (^) coping and dis-
traction self-instructions, the other self-instruc-
tional components occurred on the average of
at least once and usually much more frequently
each session. The lower frequency of distraction-
ignoring self-instructions can be attributed to the lower number of distractors available in the transfer I situation. Transfer II (classroom) setting. Figure 2 shows the frequency of self-instruction by the experimental children and the untrained chil- dren during the math, printing, and phonics tasks in the classroom (^) setting. Although both experimental children showed self-instructional behavior in this generalization setting, they were quite variable. As Figure 2 shows, Angie dis- played a higher rate of self-instruction than Judy. Both children self-instructed more fre- quently during the printing than the math task.
With the exception of one of Angie's sessions,
neither (^) child showed self-instruction during the
phonics task. The third child who was not trained did not self-instruct nor did the criterion reference children whose data are not presented.
Off-Task Transfer I and II (classroom) settings. (^) Figures 3 and 4 show the frequency of off-task behavior displayed by the two experimental children and the untrained child in the transfer I and transfer II (classroom) setting. Data for the criterion comparison children are also presented in the classroom situation. Figure 3 shows that off-task behavior was (^) generally low across conditions (^) for all three (^) children in the transfer (^) I setting, al- though the data suggest that Angie was less off task after she received self-instructional training. The low rate of off-task behaviors in this situa- tion is probably related to the fact that there were no external distractors in the room. In (^) contrast, Figure 4 shows that (^) off-task be- havior was generally higher in (^) the transfer II (classroom) setting than in the transfer I setting. The two experimental children displayed a grad- ual but marked decrease in off-task behavior after training. Generally, this effect occurred across all tasks. Due to the nature and complex- ity of data presented in Figure 4, for these two
children a time-series statistical analysis (Glass,
Willson, & Gottman, 1974) was^ performed as
an adjunct to visual analysis.1 For Angie, the
time-series analysis showed a significant change
in both trend, t(41) =^ 5.37, p < 0.0005, one-
tailed, and level, t(41) =^ 1.72, p < 0.05, one-
tailed, of her off-task behavior during the math
task, a significant change in trend, t(41) -
1.97, p < 0.05, (^) one-tailed, during the printing
task, and a significant change in^ trend, t(41)
7.07, p < 0.0005, one-tailed, during the^ phonics
task. For Judy there was a significant post-inter-
vention change in the level of her off-task be-
havior during the printing, t(40) =^ 1.61, p <
'The authors performed the time-series analyses with the TMS (^) computer program developed by Cath- leen Bower, William Padia, and Gene Glass at the Laboratory of^ Educational Research, University of Colorado, October^ 1974.
LOUIS D. BURGIO et al.
BASELINE TREATMENT
HARVEY
0 6 12 la 24 30 37 43
SESSIONS Fig. 2. Percentage of intervals of self-instruction over sessions by the experimental children on the math, printing, and phonics tasks in the transfer II (classroom) setting. (An asterisk designates a 2-week^ school holiday when observations were not made. The arrow signifies the point at which^ training was faded out^ for Judy.)
452
0 Ca 0
(^0) CoCD
0
10
0
-J
LO z
LLU L)J CL
(^00)
(^0) Go
(^0) CD
0
0
0
(^00)
(^0) Co
0
0
(^0) ...........................&...
SELF-INSTRUCTIONAL PACKAGE FOR ATTENDING BEHAVIOR
BASEOLNE TREATMENT
JUDY
iAA X2 $
0 6 12 18 24 30 37 43
jw~
ANRGIE
HARVEY
(^0 6 12) 2a
'as
30 37 1aS
SESSIONS Fig. (^) 3. Percentage of (^) intervals of off-task behavior over sessions by the experimental children on the math and (^) printing tasks in the (^) transfer I (^) setting. (An asterisk designates a 2-week school holiday when observations were not (^) made. The arrow (^) signifies the point at which training was faded out for Judy.)
a
a"
aco
ax
no a
x
(n -J CR
C: LJ I-"z z
LU
cc
Gi
a co
co 'V
a
0
co as
a
Cy
a (^) mmm&.
SELF-INSTRUCTIONAL PACKAGE FOR ATTENDING BEHAVIOR
Table 2 Mean rate and accuracy of performance during math, printing, and (^) phonics tasks (^) for experimental and criterion comparison children (^) during baseline and (^) training conditions in the transfer II (classroom) setting.ab
Accuracy Rate Children Baseline Training (^) Baseline Training Math Judy (^) 71.6(41.5) 89.0(14.4) 4.0(2.1) (^) 13.1(6.2) Angie (^) 42.6(22.4) 56.3(16.9) 7.3(3.1) (^) 6.6(1.5) Donny 91.4(10.1) 87.3(13.8) (^) 10.2(5.3) 13.1(4.6) Kathy 47.7(28.4) (^) 35.2(25.2) 4.8(1.4) 4.0(1.6) Printing Judy (^) 92.6(3.4) 93.6(6.6) 43.8(12.4) (^) 43.5(13.2) Angie (^) 66.5(16.3) 67.8(11.3) 34.5(9.9) 26.4(5.7) Donny (^) 91.6(5.6) 91.5(10.1) 74.0(35.2) (^) 51.1(13.3) Kathy 87.7(6.8) 86.0(8.5) (^) 33.9(5.7) 30.6(4.9) Phonicsc Judy 82.4(18.7) (^) 83.5(18.7) Angie 53.3(33.7) 69.7(14.8) Donny 88.9(18.8) (^) 82.6(27.6) Kathy 67.6(29.4) (^) 80.0(24.5) aThe standard deviations appear in parentheses. bMeans and standard deviations for criterion (^) comparison children (Donny and Kathy) during baseline and training conditions were computed on (^) the task data for the first and last 23 sessions, (^) respectively. cRate data are not reported for (^) the phonics task due to the variable number of phonics (^) problems given in any one session.
Children's Distractibility Rating Form admin-
istered before and after training were compared.
The scale scores ranged from 0 to 4, with 4 in-
dicating high activity and 0 indicating a low
level of activity. Both experimental (^) children who received training were rated as (^) being less "hy- peractive" after (^) training than before training. Judy's and (^) Angie's prebaseline activity ratings were at a (^3) level, whereas their ratings after training were, respectively, 1.2 and 2.2. Both criterion reference children (Donny and Kathy) were rated as being slightly more active at the end of the study (1.6 and 1.4) than (^) before the study began (1.0 and 0.6) and were similar at this (^) point in their activity level to the experi- mental children. (^) At the end of the study the in- dividual seen by the (^) teacher as least "hyperac- tive" was Judy, an experimental (^) child.
DISCUSSION
From both (^) an experimental and clinical per- spective the (^) results of this study suggest that the
self-instructional training program employed can be used to decrease the off-task behavior of dis-
tractible retarded children in a classroom setting.
In contrast to most previous (^) research that has
used this approach, the present study not only
trained children to self-instruct but systemati-
cally monitored^ the extent to which the children
learned these self-instructions during training
and subsequently exhibited them in a nontrain- ing situation. The results indicated that the (^) ex- perimental children learned to (^) self-instruct in the training situation and to use these (^) self-in- structions extensively for the math and (^) printing tasks in the transfer I (^) situation where only an observer and the child were present. In (^) both the training and the transfer I situation it was evi- dent not only that both experimental children were self-instructing frequently (^) but also that they were using all of the (^) self-instructional com- ponents they had been taught. The results also indicate that both children self-instructed, albeit to a lesser extent, in the classroom situation. However, it may be that
455
LOUIS D. (^) BURGIO et (^) al.
more self-instructions occurred than were rated in this latter setting. Both children, in the early stages of treatment, spontaneously reported that they said (^) the words (self-instructions) to them- selves while they were doing their work in the classroom. Moreover, while the observer was re-
cording self-instructions in the classroom setting,
the children were often noticed mouthing, in- audibly, (^) what appeared to be (^) self-instructions. However, due to the absence of clear and definite identification, these apparent self-instructions were not recorded formally. Perhaps in future research a wireless microphone could detect these low volume verbalizations. In general, however, the fact that less self-instruction oc- curred in the classroom is consistent with past research. Meichenbaum (1977) has suggested that the (^) classroom setting may indeed have an inhibitory effect on overt self-instruction. One can easily understand how children may be re- luctant to self-instruct overtly in the presence of their peers for fear of drawing adverse attention to themselves. The self-instructional program used in (^) this study (^) differed in (^) several significant respects from those employed in past studies. First, it was more complex in that it taught the children self-in- structions for two very different tasks as well as procedures to cope with error and distraction, included within a game-like context. Second, the distraction (^) inoculation procedure was quite unique in that it not only taught the children what they should say when (^) distraction occurred, but (^) also, systematically, through the presenta- tions of a variety of visual, audio, and in vivo stimuli, showed the children the occasions when such statements would be appropriate. Third, because in pilot research with retarded children, modeling and prompting were not found suffi- cient to (^) produce good self-instructional behavior, the experimenter shaped, with praise as a (^) rein-
forcer, the chain of the various self-instructional
components to^ be used by the children. Although it^ took several sessions for the chil- dren to learn to self-instruct (^) without prompting, both children appeared to be (^) enjoying the train-
ing sessions and both reported that they enjoyed
playing the "game." One child even reported
that she played the "game" at home while doing her homework. These anecdotal data support the contention that by presenting the self-instruc- tional training as a "game that will (^) help you do your work better," children's attention will be better maintained during training (Meichen- baum, 1977; Bash & Camp, Note 1). This pro- cedure may be especially important with a re- tarded population that has marked attentional problems (House & Zeaman, 1963). As indicated in (^) Figure 4, there were marked decreases in the off-task behavior of the two ex- perimental (^) children in the classroom setting. Moreover, there is an obvious, although imper- fect, correlation between the self-instructional behavior of these children and their off-task be- havior during both the math and printing tasks (See Figures 2 and 4). (^) Conversely, the untrained child and the criterion comparison children (^) who did (^) not receive training did not show changes in either their (^) self-instructional or off-task be- havior. The direct monitoring of both of these subject behaviors allows stronger inferences to be made regarding whether the changes in off- task behavior were actually a function of the self-instructional program (Kazdin, (^) 1978). Al-
though similar reductions in off-task behavior
have been found by Meichenbaum and (^) Good- man (^) (1971), Bornstein and Quevillon (1976), and others, they did not monitor self-instruction directly. The results of this study suggest that the self- instructional (^) training package employed effected generalized changes in self-instructional and off- task behavior across situations, most importantly in the classroom, and across tasks. Prior to this
study only Bornstein and Quevillon (1976),
using a^ self-instructional program, obtained gen- eralization of behavioral (^) effects into a new set- ting, but (^) with a nonretarded population. One aspect of^ the^ results might, however, argue against the (^) conclusion that the self-instructional behavior and the (^) generalized effects (^) were really functionally related. That is, although decreases
458 LOUIS D. BURGIO et al.
common way of making this assessment is to ask "significant individuals" who are affected in im- portant ways by the child's behavior to assess whether the child is different after training. The problem with this type of assessment is that it often does not have established reliability and validity (O'Leary & Turkewitz, 1978). In the present research several methods of assessing the clinical significance of the decrease in the experi- mental children's off-task behavior were in- cluded. The teacher's perceptions of the chil-
dren's behavior, before and after training, were
solicited. In addition, the experimental children's behavior throughout the study was compared with that of two criterion comparison children who were seen as good students (frequently on
task) by the teacher. Both the teacher's ratings
and the interchild comparisons suggested that
after training the experimental children approxi-
mated and in^ some cases were off-task less than the criterion children in the classroom. The child who was most distractible in the classroom dur- ing baseline appeared after training more task oriented than the criterion comparison children. In summary, the results of this study indicate
that the self-instructional training package pro-
duced direct changes in^ self-instructional behav- ior in the training situation and that generalized change occurred in the self-instructional behav- ior of the experimental children in the transfer I and II (classroom) situations, and in their off- task behavior during three task situations in the classroom. It was established that retarded chil-
dren can^ be^ taught self-instructions^ even^ though
they are language deficient and commonly thought unable to gain control over their own behavior. Furthermore, this study suggests that self-instruction might be used to^ reduce hyper-
activity, a common^ problem among retarded
children. The results of this study are generally encouraging, but they do suggest that when a self-instructional approach is to be used to effect changes in academic performance, the instruc- tions employed may well need to be modified. In this regard and more generally, it is impera- tive that the^ self-instructional technology not be
viewed as a completed product, to be used but not refined. Although this approach should be compared ultimately with more direct behavior change procedures, the authors would argue that the self-instructional approach is still in its be- ginning stages of development, and that compar- isons of self-instruction with other techniques, such as token systems (cf. Friedling & O'Leary, 1979), is premature.
REFERENCE NOTES
- Bash, M., & Camp, B. Think aloud (^) program: Group manual.^ Unpublished^ manuscript.^ Univer- sity of Colorado Medical School, 1975.
- Wein, K. S., & Nelson, R. 0.^ The^ effects^ of^ self- instructional training on^ arithmetic^ problem^ solv- ing skills. Unpublished manuscript.^ University of North Carolina at Greensboro, 1978.
REFERENCES Alabiso, F. Operant control of attentive behavior: A treatment for hyperactivity. Behavior Therapy, 1975, 6, 39-42. Bem, S.^ L.^ Verbal^ self-control:^ The^ establishment of (^) effective self-instruction. Journal of Experi- mental (^) Psychology, 1967, 74, 4, 485-491. Bornstein, P. H., & Quevillon, R.^ P.^ The^ effects^ of a self-instructional package on overactive pre- school boys. Journal of Applied Behavior Anal- ysis, 1976, 9, 179-188. Burron, D., & Bucher, B. Self-instructions as dis- criminative cues for rule breaking or rule follow- ing. Journal of Experimental Child Psychology, 1978, 26, 46-57. Ferritor, D.^ E.,^ Buckholdt,^ D.,^ Hamblin,^ R.^ C.,^ & Smith, L.^ The^ noneffects^ of^ contingent^ rein- forcement for attending behavior on work accom- plished. Journal of Applied Behavior Analysis, 1972, 5, 7-17. Forehand, R., & Baumeister, A. A. Effects of^ varia- tion in^ auditory-visual stimulation on^ activity lev- els of^ severe mental^ retardates.^ American^ Journal of Mental^ Deficiency,^ 1970,^ 74,^ 470-474. Frielding, C., & O'Leary, S. G. Effects of self-in- structional training on second- and third-grade hyperactive children: A failure to replicate. Jour- nal of Applied Behavior Analysis, 1979, 12, 21 1-
Glass, G.^ V., Willson,^ V.^ L.,^ &^ Gottman, J. M.^ De- sign and Analysis of Time-series experiment. Boulder: Colorado Associated University Press,
SELF-INSTRUCTIONAL PACKAGE FOR ATTENDING BEHAVIOR 459
Guralnick, M. J. Solving complex perceptual dis- crimination problems: (^) Techniques for the (^) devel- opment of (^) problem-solving strategies. American Journal of Mental Deficiency, 1976, 81, 1, 118-
Helwig, J. J., Johns, J. C., Normal, J. E., and Cooper, J. O. The measurement of manuscript letter strokes. Journal of Applied Behavior Analysis, 1976, 9, 231-236. House, B. J., & Zeaman, D. The role of attention in retardate discrimination learning. In N. R. Ellis (Ed.), Handbook of mental (^) deficiency: Psycho- logical theory and research. New York: (^) McGraw- Hill, 1963. Johnson, M. R., (^) Whitman, T. (^) L., & Bartoon-Noble, R. A (^) home-based program for a preschool behavior- ally disturbed child with parents as therapists. Journal of Behavior (^) Therapy and (^) Experimental Psychiatry, 1978, 9, 65-70. Kazdin, A. E. Behavior modification in applied set- tings. Homewood, Ill.: The Dorsey Press, 1975. Kazdin, A. E. Methodological and interpretive prob- lems of single-case experimental designs. Journal of Consulting and Clinical Psychology, 1978, 46, 629-642. Lovaas, 0. I. Cue properties of (^) words: The control of operant responding by (^) rate and content of ver- bal operants. Child (^) Development, 1964, 35, 245-
Luria, A. R. The (^) role of speech in the regulation of normal and (^) abnormal behavior. New York: Live- right, 1961. Meichenbaum, D. H. Enhancing creativity by mod- ifying what subjects say to themselves. American Educational Research Journal, 1975, 12, 129-145. Meichenbaum, D. H. Cognitive-behavior Modifica- tion: (^) An integrative approach. New York: Plenum Press, 1977. Meichenbaum, D. H., & Goodman, J. Training im- pulsive children to talk to (^) themselves: A means of developing self-control. Journal (^) of Abnormal Psychology, 1971, 77, (^) 115-126. O'Leary, K.^ D., Becker, W. C., Evans, M. B., & Sau- dergas, R.^ A. A token reinforcement program in a (^) public school: A replication and systematic anal-
ysis. Journal of Applied Behavior Analysis, 1969, 2, 3-13. O'Leary, K.^ D., &^ Turkewitz, H.^ Methodological errors in marital and child treatment research. Journal of Consulting and Clinical Psychology, 1978, 46, 4, 747-758. Palkes, H., Stewart, M., & Freedman, J. Improve- ment in maze performance of hyperactive boys as a function of verbal-training procedures. The Journal of Special Education, 1972, 5, 4, 337-
Palkes, H., Stewart, M., & Kahana, B. Porteus maze performance of^ hyperactive boys after^ training in self-directed verbal commands. Child^ Develop- ment, 1968,39, 817-826. Patterson, G. R. An application of conditioning techniques to the control of a hyperactive child. In L. P. Ullman and L. Krasner (Eds.), Case stud- ies in behavior modification. New York: Holt, Rinehart & Winston, 1965. Patterson, G. R., Jones, R., Whittier, J., & Wright, M. A. A behavior modification (^) technique for the (^) hyperactive child. Behavior (^) Research and Therapy, 1965, 2, 217-226. Reardon, D.^ M., & Bell, G. Effects of sedative and stimulative music on activity levels of severely re- tarded boys. American Journal of Mental Defi- ciency, 1970, 75, 156-159. Robin, A. L., Armel, S., & O'Leary, K. D. The ef- fects of self-instruction on writing deficiencies. Behavior Therapy, 1975, 6, 178-187. Stokes, T. F., & Baer, D.^ M.^ An implicit technology of (^) generalization. Journal of Applied Behavior Analysis, 1977, 10, 349-367. Surratt, P. R., Ulrich, P. E., & Hawkins, R. P. An elementary student as a behavioral engineer. Jour- nal of Applied Behavior Analysis, 1969, 2, 85-92. Vygotsky, L. Thought and language. Cambridge, Mass.: MIT Press, 1962. Whitman, T. L., Caponigri, V., & Mercurio, J. Re- ducing hyperactive behavior in a severely retarded child. Mental Retardation, 1971, 9, (^) 17-19.
Received July 23, 1979 Final acceptance March 13, 1980
