This article was downloaded by: [FU Berlin] On: 15 May 2015, At: 02:04 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Clinical and Experimental Neuropsychology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ncen19
Verbal and nonverbal communication of events in learning-disability subtypes a
b
Katherine A. Loveland , Jack M. Fletcher & Vonda Bailey
c
a
University of Texas Medical School , Houston
b
University of Houston ,
c
Houston, Texas Published online: 04 Jan 2008.
To cite this article: Katherine A. Loveland , Jack M. Fletcher & Vonda Bailey (1990) Verbal and nonverbal communication of events in learning-disability subtypes, Journal of Clinical and Experimental Neuropsychology, 12:4, 433-447, DOI: 10.1080/01688639008400991 To link to this article: http://dx.doi.org/10.1080/01688639008400991
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-
Downloaded by [FU Berlin] at 02:04 15 May 2015
licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Clinical and Experimental Neuropsychology 1990, Vol. 12, No. 4, pp. 433447
0168-8634/90/1204-0433$3.00 0 Swets & Zcitlinger
Verbal and Nonverbal Communication of Events in Learning-Disability Subtypes* Katherine A. Loveland University of Texas Medical School - Houston Jack M. Fletcher University of Houston
Vonda Bailey Houston, Texas
Downloaded by [FU Berlin] at 02:04 15 May 2015
ABSTRACT This study compared a group of nondisabled children (ND) with groups of leaming-disabled children who were primarily impaired in reading and arithmetic skills (Reading-Arithmetic Disabled; RAD) and arithmetic but not reading (Arithmetic Disabled; AD) on a set of tasks involving comprehension and production of verbally and nonverbally presented events. Children viewed videotaped scenarios presented in verbal (narrative) and nonverbal (puppet actors) formats and were asked to describe or enact with puppets the events depicted in the stories. Rourke (1978, 1982) has shown that RAD children have problems with verbal skills, whereas AD children have problems with nonverbal skills. Consequently, it was hypothesized that children’s performance in comprehending and reproducing stories would be related to the type of learning disability. Results showed that RAD children made more errors than AD children with verbal presentations and describe-responses, whereas AD children made more errors than RAD children with nonverbal presentations and enact-responses. In addition, learning disabled children were more likely than controls to misinterpret affect and motivation depicted in the stones. These results show that learning disabled children have problems with social communication skills, but that the nature of these problems varies with the type of learning disability.
T h e learning-disabled (LD) child, in addition to having problems in school achievement, may be a poor communicator. I n referential communication tasks, learning-disabled children have been found to be less adept than nondisabled peers at providing appropriate and sufficient information to a listener (Bryan & Pflaum, 1978; Donahue, Pearl, & Bryan, 1980; Noel, 1980; Spekman, 1981) and
* The authors would like to thank Bella Jackson and David Breiger for their work in data collection and coding, and Mrs. Sally Woolrich of the St. Francis School for help in recruitment of children. Address for correspondence: Katherine A. Loveland, University of Texas Mental Sciences Institute, 1300 Moursund, Houston, TX 77030, USA. Accepted: April 28, 1989.
Downloaded by [FU Berlin] at 02:04 15 May 2015
434
KATHERINE A. LOVELAND ET AL.
at indicating their own need for more information when listening to an ambiguous message (Donahue et al., 1980). In conversational situations, LD children are less skilled than nondisabled peers at managing turntaking and initiating conversation and are more likely to defer to others (Donahue, 1983). Feagans and Short (1984) asked reading-disabled children to enact a narrative until it was fully comprehended and then to paraphrase the narrative. Even when comprehension was demonstrated, reading-disabled children of three age groups displayed deficiencies in oral expression on measures of linguistic complexity and content. Feagans and Short (1 986) found that 6-to 9-year-old disabled learners had more difficulty than nondisabled children comprehending information presented (“listener” role) as well as more difficulty explaining this information to others (“speaker” role) and in rephrasing their explanations when cued to do so. These studies raise the question of why the LD child, who by definition has difficulty in academic achievement, should also have difficulty in communicating with others. Although these difficulties are probably due in part to the well-documented language deficits experienced by many LD children (e.g., Wiig & Semel, 1976), they may also be related to difficulties in social perception or other nonverbal aspects of the interaction. In fact, several studies have characterized the LD child as less popular than nondisabled peers, less well-adjusted, more socially insensitive, and more prone to personality disorders (Bachara, 1977; Bryan, 1974, 1978; Denckla, 1972; Rourke & Fisk, 1981). Inability to infer emotions accurately and to grasp the meaning of gesture in interaction is frequently observed in LD children ( Bryan, 1974, 1978; Gerber & Zinkgraf, 1982; Saloner & Gettinger, 1985; Vogel, 1974, 1975; Wiig & Harris, 1974), supporting the hypothesis that problems in communication are not necessarily due only to deficits in language but may be due to deficits in social awareness in at least some disabled learners. It is also important to recognize that not all learning disabilities reflect the same underlying deficits (Rourke, 1978, 1987). Children with different underlying deficits may not exhibit the same difficulties in learning or in communication. Interestingly, problems with verbal and nonverbal aspects of communication have been differentially linked to neuropsychological deficits in disabled learners. 0201s and Rourke (1985) found that LD children with high visualspatial and low verbal skills performed better than children with low visualspatial and high verbal skills on tasks requiring comprehension of facial expression and gesture. However, when a verbal response was required, low visualspatial children performed better. These results suggest that only a subset of the LD population (those lower in visual-spatial skills and higher in verbal skills) may be experiencing difficulty in interpreting affect and interpersonal cues in conversation. Other children, those with specific language-based deficits, may experience difficulty mainly with linguistic aspects of communication. Different types of communication problems may further be linked to academic subtypes of learning disability. Reading-Arithmetic Disabled (RAD) children are predominately language impaired, whereas Arithmetic Disabled (AD)
Downloaded by [FU Berlin] at 02:04 15 May 2015
VERBAL AND NONVERBAL COMMUNICATION
435
children (those unimpaired in word identification) are predominantly impaired on non-language skills (Fletcher, 1985a; Rourke, 1978, 1982, 1987; Rourke & Strang, 1985; Siege1 & Linder, 1984). In addition to displaying poorer nonverbal than verbal skills, children with arithmetic disabilities have been hypothesized to represent a distinct subgroup in terms of social-emotional functioning. They are reported to have significant social problems, particularly in new or unstructured situations (Rourke, 1987) and are described as poor at interpreting others’ emotions, facial expressions, and gestures. This deficiency, which could result in inappropriate interpersonal behavior, could reflect poor visual-spatial perception of nonverbal cues. Consequently, although language problems may underlie the social-communication problems of RAD children, problems with nonlinguistic components may underlie those of AD children. It is important to ask how these underlying disabilities might affect the disabled learner’s ability to use language and understand events both in and outside the classroom. Finally, one might ask how verbal and nonverbal deficits affect the child’s performance as both speaker and listener in communicative interactions. Difficulty in interpreting nonverbal signals, affect, or the meaning of events, together with difficulty in language processing, can lead to deficits in comprehension or production or both (Bailey, 1984). Although it appears that disabled learners may have difficulty with both comprehension and production, it is not clear how the different demands of comprehending and producing information in communication are affected by the modality (verbal or nonverbal) in which a presentation or a response is made. It may be that the effects of modality on comprehension and production vary with the type of learning disability the child experiences. The present study was designed to address these issues by examining both comprehension and production of verbally and nonverbally presented events and comprehension of the affective/motivational content of these events by groups of disabled learners primarily impaired in reading (RAD) versus arithmetic (AD) and by a group of non-disabled control children (ND). We hypothesized the following: 1 . Learning disabled children’s performance in comprehending and reproducing stories would be related to the format (verbal or nonverbal) of the story and the type of learning disability present. In particular, a) Arithmetic disabled (AD) children were expected to make more errors than other children in both to nonverbal presentations of events and in nonverbal responses by the child. b) Reading-Arithmetic disabled (RAD) children were expected to make more errors than other children in both verbal presentations of events and in verbal responses. 2 . Children’s ability to interpret successfully the affective content of events was expected to be related to whether a learning disability is present and to the subtype of learning disability. In particular,
436
KATHERINE A. LOVELAND ET AL.
a) AD disabled children were expected to make a greater number of errors in interpreting affect than RAD or ND children (because of the relationships of visual-spatial proficiency and nonverbal cues in social-communicative interactions).
b) RAD children were expected to make fewer errors of interpreting affect than AD children but a greater number than ND children.
Downloaded by [FU Berlin] at 02:04 15 May 2015
METHODS Subjects Children for this study were recruited from area agencies and schools for leaming-disabled children and the local Association for Children with Learning Disabilities. ND children were friends of the LD children or were obtained through volunteer solicitation at local schools. In order to qualify for the study, all children were Caucasians with (a) Full Scale WISC-R or Stanford-Binet IQ above 85 (all children received the WISC-R except four controls with recent Stanford Binets); (b) age from 104-160 months; (c) no history or evidence of significant neurologic anomaly, emotional disturbance, or cultural deprivation; and (d) predominantly English-speaking background. Three groups of children were formed based on Wide Range Achievement Test (WRAT) scores (Jastak & Jastak, 1978). The ND children ( n = 14) had achievement centiles above 39 on the three WRAT subtests (reading recognition, spelling, arithmetic). Children with learning disabilities in arithmetic (AD; n = 12) had reading centile scores above 39, with arithmetic scores below the 31st centile and at least one standard deviation below the reading score. Spelling scores were not controlled because previous studies have not found AD children with and without spelling difficulties to differ on a variety of cognitive tasks (Fletcher, 1985b; Siege1 & Linder, 1984). Children impaired in reading and arithmetic (RAD; n = 13) had WRAT centile scores uniformly below 31 on both
Table 1: Means and Standard Deviations for Demographic and Selection Variables by Achievement Group WRAT Centiles Age (in months) FSIQ Percent Groupn Male M SD M SD
Reading
M
ND
14 57
131.8
13.7 109.3
6.2 78.2
AD
12 50
137.1
17.8 103.3
12.2 63.8
RAD 13 62
132.5
9.5
98.3
ND = Nondisabled AD = Arithmetic Disabled RAD = Reading-Arithmetic Disabled
7.6
13.3
SD
Spelling
Arithmetic
M
M
5.0 79.7
SD
SD
16.5
59.8
11.6
13.0 39.4 26.4
16.8
7.7
10.7
8.3
7.8
11.1
9.3
Downloaded by [FU Berlin] at 02:04 15 May 2015
VERBAL AND NONVERBAL COMMUNICATION
437
subtests with an average centile score below 26. Consequently, the two LD groups should be expected to differ in WRAT reading recognition scores, but not on arithmetic scores. Table 1 presents the sample size, gender (percent male), age, Full Scale IQ scores, and WRAT centile scores for the controls and two learning disability groups. Analysis of variance revealed no group differences in age or gender (p > .05). On FSIQ, the ANOVA was significant (F = 4.74; p < .05), with follow-up t tests showing that ND children had higher IQ scores than either learning-disability group, who were not significantly different. This difference between disabled and nondisabled learners is expected, given the sampling strategy, which controls for range of IQ. Greater control over mean IQ is not desirable, because the population of disabled learners has a lower IQ than the normal population (Yule & Rutter, 1975). Subject or group-based matching on IQ is inappropriate in situations such as this in which a quasi-experimental design is used because of potential selection bias and subsequent regression to the mean (Campbell & Erlebacher, 1970; Fletcher & Moms, 1986). As expected, the ND group scored significantly higher than the two LD subtypes on all three WRAT subtests. The AD group scored significantly higher than the RAD group on reading and spelling, but not on arithmetic. It should be noted that the AD group had reading centiles that are clearly in the average range. The differences in selection vanables are all expected, and they conform to patterns commonly found when ND learners are compared with AD and RAD children (Rourke, 1978). Because the primary comparisons of interest are between the two learning-disability subtypes, and because the magnitude of the IQ difference is relatively small, it is not likely that this difference interfered with interpretation of the results of this study.
Materials Eight videotapes of brief stones were made (Bailey, 1984). The stories involved simple, easily recognizable interactions among familiar characters (a bear, a dog, a chicken, etc.), and each contained some affectively charged situations (e.g.. a character becomes frightened and cries). Four different scripts were videotaped in both verbal and nonverbal presentation conditions. On the nonverbal tape, puppets act out a skit, and no talking occurs. On the verbal videotape, the face of a narrator telling the same story is shown, and the puppets do not appear. Care was taken to prepare story scripts that could be narrated and enacted with equal clarity and emphasis for all events in both versions. Two characters and one or two props were used in each story. All four stories were approximately equivalent in the number of events that occur and in the length of the videotapes, in both verbal and nonverbal format (about 1.5 min.). The same puppets and props that appeared i n the videotapes were used by the children to enact the stories. A sample story is given in the Appendix.
Procedure In one half-hour session, each subject was first shown two practice skits similar to the four later shown in the experimental trials, but involving different stories, characters, and props. One practice skit was nonverbal and the child was asked to describe it. The other was verbal, and the child was asked to act it out with the puppets. At this time, the procedure was explained and the child’s questions answered. If the child had any difficulty following the instructions (e.g., talked instead of acting out with the puppets, failed to complete the story, etc.), corrections and additional instructions were given. Each child was then shown the four experimental videotapes, one representing each condition (verbal-enact, verbal-describe, nonverbal-enact, nonverbal-describe). Each of these four tapes depicted a different story. Order of story presentation was fixed, but order of conditions for presentation and response was counterbalanced. The child viewed a tape, and then was asked either to enact or to describe the story. Children were not informed in advance which response would be required of them. Enact-
438
KATHERINE A. LOVELAND ET AL.
Downloaded by [FU Berlin] at 02:04 15 May 2015
ing the script entailed moving the puppets to make them do what they did in the story, without using words to substitute for actions. Lust, Loveland. and Komet (1980) used a similar procedure to test children’s comprehension of story sentences. Describing the script required telling, as accurately as possible, what occurred in the story, without using gestures to substitute for words. All responses were recorded on videotape for later coding. After a child responded to each story, a set of specific follow-up questions were asked; these dealt with the child’s interpretation of the emotions and motives of characters in the story. These responses were also videotaped for later coding. Coding Three coders separately coded the participants’ videotaped responses using a prepared format detailing in chronological order each of the events included in the story. As the child’s verbal or nonverbal response was viewed, it was coded for errors in the following areas: omission of an event, misinterpretation of an event, substitution of events, embellishments/intrusions, substituting or reversing characters, substituting or reversing props, wrong location, events out of sequence, implied information, use of nonspecific words (describe-response only), substituting gesture for words (describe-response only), and substituting words for gesture (enact-response only). Coding categories are explained in the Appendix. Several categories were eliminated because of very low frequency: substitution of events, substitution of word for gesture, substitution of gesture for word, location errors, and prop substitution or reversal. Responses to follow-up questions were also coded. Each response was first coded as either appropriate or inappropriate, according to whether it correctly (or reasonably) captured the emotion or motivation displayed by the character. Each response was then coded as primarily concrete or motivational in content, according to whether the focus of the response was on the characters’ actions or on feelings/motivations. For example, if asked, “Why did the dog growl at the rabbit?” a child might respond “Because he was angry” (motivational) or “Because the rabbit took the ball” (concrete). Note that both motivational and concrete responses could be either appropriate or inappropriate. Coders were initially trained to reach 90% agreement or greater across all error categories and responses to follow-up questions using videotapes of children not meeting subject-selection criteria. Agreements were counted by tabulating matching Occurzences of a particular error category for a particular line, event, or question in a story (see sample story in appendix) rather than for the story as a whole. Thereafter, periodic coding reliability checks were carried out to insure that interrater agreement remained high. When error categories were examined separately for all reliability re-coding performed (25% of data, randomly selected), percents of agreement ranged between 84% - 96%. Percents of agreement for responses to follow-up questions (appropriate/inappropriate, concrete/motivational) were 97% and 95%. respectively.
RESULTS
Omissions.Means and standard deviations for omission errors are presented in Table 2. A three-way ANOVA revealed no significant Group x Response x Presentation interaction. The two way interactions of Group x Response, F (2,36) = 6.0, p < .006, Group x Presentation, F (2,36) = 3.7, p c .03, and Presentation x Response, F (2,36)= 12.2, p < .001, were all significant. Follow-up tests using Tukey’s procedure (overall p = .05)showed that, when asked to describe, RAD children made
439
VERBAL AND NONVERBAL COMMUNICATION
Table 2: Means and Standard Deviations for Errors of Omission by Presentation and Response conditions Conditions Verbal
Downloaded by [FU Berlin] at 02:04 15 May 2015
Group
Enact
Nonverbal Describe
Enact
Describe
M
SD
M
SD
M
SD
M
SD
ND
2.1
1.3
4.8
1.9
2.4
1.7
7.7
2.8
AD
5.0
2.5
4.5
2.2
4.9
3.3
7.9
2.4
RAD
6.5
3.3
8.1
3.4
5.5
1.9
8.2
2.6
ND - Nondisabled AD - Arithmetic Disabled RAD - Reading-Arithmetic Disabled
more omissions than did AD or ND children, who did not differ from each other (Table 2). By contrast, in the enact response, both RAD and AD children made more omissions than did ND children. This seems to suggest that, when a verbal response is required, RAD children are at a disadvantage, making more omissions relative to AD children, but that when a nonverbal response is required, both RAD and AD children make more omissions than ND children. Similarly, with a verbal presentation, RAD children made more omissions than did AD and ND children; with a nonverbal presentation, AD children’s errors increased such they also differed from the ND children. For all children, describing a nonverbal presentation led to more omissions than did describing a verbal presentation; however, for the enact responses there were no differences between verbal and nonverbal presentations.
Embellishments. There were no group differences or interactions present for embellishments. However, there were significant main effects of Presentation, F (1,26) = 6.84, p < .01, and Response, F (1,36) = 5.0, p < .03. Tukey’s HSD procedure revealed that, for all children, verbal presentation led to significantly fewer embellishments than nonverbal presentations, and describe-responses led to more than enact-responses. This may suggest that, in general, verbal presentation constrains the adding of new material more than does nonverbal presentation; however, when extra material is inserted, it is more likely to be done verbally than nonverbally. Churucter reversal. Because of the low error rate, enact and describe response conditions were not examined for character reversals, and differences in presen-
440
KATHERINE A. LOVELAND ET AL.
Table 3: Means and Standard Deviations for Character Reversal Error by Presentation Conditions Group
Conditions
Downloaded by [FU Berlin] at 02:04 15 May 2015
Verbal
Nonverbal
M
SD
M
ND
.36
SO
.14
AD
.66
.99
SO
RAD
.54
.78
.15
SD .36 1.2 .38
ND - Nondisabled AD - Arithmetic Disabled RAD - Reading-Arithmetic Disabled
Table 4: Number of Children Misinterpreting Events by Group after Nonverbal Presentation Response Condition Enact
Group
Error
Describe Noerror
Error
No error
ND
12
0
14
AD
9
5
7
11
1
12
RAD ND - Nondisabled AD - Arithmetic Disabled RAD - Reading-Arithmetic Disabled
tation conditions only were examined. A significant interaction of Group x Presentation was found, F (2,36) = 3.9, p < .03. As Table 3 shows, A D children made more character reversals than did R A D and N D children in the nonverbal presentation. However, with the verbal presentation, the number of character reversals by R A D children increased such that there was no longer a difference between them and A D children. This result suggests that A D children were more likely than other children to confuse the roles of actors within events when the events
441
VERBAL AND NONVERBAL COMMUNICATION
were presented nonverbally, but that, when the events were presented verbally, RAD children made as many errors as did AD children. Misinterpretation of events. Table 4 presents the number of children making errors for the category of misinterpretation of events. In all groups this error was made exclusively with the nonverbal presentation. In addition, error rates were very low for all groups except AD. Consequently, separate Chi-square analyses were conducted comparing the number of children in each group who made errors for both the Nonverbal-Enact and Nonverbal-Describe conditions. For Nonverbal-Enact, Chi-square was not significant. However, the analysis was significant for Nonverbal-Describe, ( d f = 2) = 7.1 p c .05, and showed that the AD children were more likely to misinterpret nonverbal events than were either the RAD or ND children. Downloaded by [FU Berlin] at 02:04 15 May 2015
x2
Sequence errors. No group differences were present in sequence errors, but a significant main effect for Response was found, F (1,36) = 3 . 8 , < ~ .02. Followup comparisons using Tukey 's procedure were not significant.
Implied Information. Errors of implied information were coded only for the describe-response condition. ANOVA revealed a significant group difference, F (1,36) = 4.2, p < .02, with RAD children making more errors of implied information than did either of the other two groups (see Table 5 ) . This result indicates that RAD children more often than other groups failed to state explicitly information about story events.
Table 5: Means and Standard Deviations for Errors of Implied Information and Nonspecific Words for the Describe Response Conditions Error
Group Implied Information
M
SD
Nonspecific Word
M
SD
ND
.43
.85
.07
.27
AD
.25
.87
.07
.21
1.24
1.01
.46
.66
RAD
ND - Nondisabled AD - Arithmetic Disabled RAD - Reading-Arithmetic Disabled
442
KATHERINE A. LOVELAND ET AL.
Table 6: Means and Standard Deviations for Appropriate and Inappropriate Interpretations of Affect Responses by Category (Concrete, Motivational) Motivational Appropriate ~~~
Downloaded by [FU Berlin] at 02:04 15 May 2015
M
Concrete
Inappropriate ~
SD
Appropriate
Inappropriate
~
M
SD
M
SD
M
SD
ND
6.29 1.59
0.43
0.65
3.89 1.53
0.43
0.65
AD
4.50 1.98
1.33
1.16
4.50 1.24
0.58
1.24
RAD
4.80 1.80
.85
.90
4.23 1.24
1.08
1.38
ND - Nondisabled AD - Arithmetic Disabled RAD - Reading-ArithmeticDisabled
Nonspecific words, Use of nonspecific words was coded only for the describeresponse condition, Groups differed significantly, F (1,36) = 4.63, p < .02.Table 5 shows that RAD children made more of these errors than did the other two groups. RAD children were more likely than were other groups to substitute pronouns and other non-specific circumlocutions for specific story elements. Interpretation of affect. A significant main effect for Group, F (1,36) = 4.03, p < .02, was found for the category of appropriate/motivational responses with no effects for Presentation or Response (Table 6). Non-disabled children produced significantly more appropriate/motivational responses than did either of the learning disabled subtypes, indicating that they more often correctly identified a character’s motivation or emotion. RAD and AD children made more inappropriatekoncrete responses (errors of fact) and also more inappropriate/motivational responses (errors of misinterpreting affect) than did ND children. There were no significant differences between AD and RAD children. These results suggest that learning disabled children had more difficulty than ND children interpreting the affect-related events depicted in the and perhaps more difficulty in talking about these events. DISCUSSION These results largely support the first hypothesis of the study: i.e., that children’s performance in comprehending and reproducing stories would be related to the subtype of learning disability. The two subtypes of learning-disabled children differed from each other and from nondisabled children in the production and
Downloaded by [FU Berlin] at 02:04 15 May 2015
VERBAL AND NONVERBAL COMMUNICATION
443
comprehension of verbally and nonverbally presented stories. In general, AD children had more difficulty with nonverbal aspects of the story tasks, and RAD children had more difficulty with verbal aspects of the story tasks. Both subtypes of disabled learners committed more errors of omission and character reversal than did nondisabled children. Overall, RAD children made more omissions than did AD children when describing an event in either format, whereas AD children made more omissions than did RAD children when enacting nonverbal events. Similarly, when events were presented nonverbally, AD children reversed the role of actors more often than did the other groups; with verbal presentation, RAD children’s errors increased to the level of the AD group. The RAD children made more specifically verbal errors (such as implication errors and use of nonspecific words) than other children when describing events, whereas AD children performed similarly to controls with respect to verbal errors. Arithmetic disabled children were more likely than the other two groups to misinterpret events, particularly when describing events that were presented nonverbally. Overall, RAD children experienced more difficulty with verbal presentations and describe-responses, whereas AD children experienced more difficulty with nonverbal presentations and enact-responses. However, apparent problems with comprehension of verbal events by RAD children and nonverbal events by AD children seemed to lead to problems with production in both types of response format. This result suggests that, in the absence of adequate comprehension, production problems are likely to emerge regardless of the response format. There was less support for the second hypothesis: i.e., that AD children would have the most difficulty interpreting affect. Although both LD groups made more errors interpreting affect than did the controls, there were no significant differences between the two LD subtypes. However, in interpreting this result, it should be noted that there was a strong trend for the AD children to make more “inappropriate/motivational” responses than did RAD children, suggesting more misinterpretations of affect or motive. A larger sample might provide a more definitive test of this hypothesis. ln addition, all questions and responses about affect were given verbally, which may have placed RAD children at a disadvantage. It is possible that some errors made by RAD children were not problems of interpreting story affect but of talking about it, because the demands of oral expression in this task overtaxed the RAD child’s language system. Although Ozols and Rourke (1985) used slightly different criteria to form groups, they found that LD children with poor visual-spatial skills (all of whom were impaired on WRAT arithmetic but not necessarily reading) had more difficulty in pantomiming an appropriate affective response or in selecting an appropriate facial expression after a verbally presented story, whereas language impaired children performed like controls on these tasks. Consequently, requiring a verbal response in the present study may have masked comprehension of affect by the RAD group. These results support the hypothesis that the social-communicative problems
Downloaded by [FU Berlin] at 02:04 15 May 2015
444
KATHERINE A. LOVELAND ET AL.
of LD children vary according to the nature of the underlying cognitive impairment. In addition, they show that requirements for presentation and response interact with the nature of these impairments. Learning disabled children may, indeed, have special difficulty understanding and remembering narrative information or the actual sequences of events that they experience, both within and outside the classroom. They may have more difficulty than other children in describing what they know or in demonstrating a skill to someone else. Moreover, at least some children may have additional problems in interpreting and expressing emotions and social cues appropriately. These problems can have a negative impact not only on school performance but on peer acceptance and on the development of age-appropriate adaptive skills. Although it is likely that some LD children have social-emotional problems that result from their school failure and poor achievement, not all LD children exhibit such difficulties (Porter & Rourke, 1985; Rourke, 1988). The results of the current study add to the growing body of literature suggesting that the manifestations of socioemotional problems that are exhibited by LD children may also reflect differential patterns of cognitive assets and impairments in such children (Rourke, 1982, 1987, 1988a, 1988b). Future studies of social-communicative skills would do well to measure carefully the various processing deficits exhibited by LD children, as well as to outline more precisely the nature of selection criteria, particularly in terms of academic skills.
REFERENCES Bachara. G. (1977). Empathy in learning disabled children. Journal of Learning Disabilities, 10, 507-508. Bailey, V.E.(1984). Event description by two subtypes of learning disabled children. Unpublished master’s thesis, Rice University. Bryan, R.M. (1974). Peer popularity of learning disabled children. Journal of Learning Disabilities, 7, 66 1-625. Bryan, T. (1978). Social relationships and verbal interactions of learning disabled children. Journal of Learning Disabilities, 11, 107-115. Bryan, T.H. (1978). Verbal interactions and social relationships of learning disabled children in the classroom. Journal of Learning Disabilities, 11, 107-1 15. Bryan, T.S., & Pflaum, S. (1978). Social interactions of learning disabled children: A linguistic, social and cognitive analysis. Learning Disability Quarterly, 1,70-79. Campbell, D.T.. & Erlebacher, A. (1 970). How regression artifacts in quasi-experimental evaluations can mistakenly make compensatory education look harmful. In J. Helmuth (Ed.), Contemporary education: A national debate (p. 2-38). New York: BrunerMazel. Denckla, M.B. (1972). Clinical syndromes in learning disabilities: The case for “splitting’’ versus “lumping”. Journal of Learning Disabilities, 5,401 -406. Donahue, M. (1983). Learning disabled children as conversational partners. Topics in Language Disorders, 4 , 15-27. Donahue, M., Pearl, R., & Bryan, T. (1980). Learning disabled children’s conversational competence: Responses to inadequate messages. Applied Psycholinguistics, 1 , 387403.
Downloaded by [FU Berlin] at 02:04 15 May 2015
VERBAL AND NONVERBAL COMMUNICATION
445
Feagans, L., & Short, E.J. (1984). Developmental differences in the comprehension and production of narratives by reading disabled and normally achieving children. Child Development, 55, 1727-1736. Feagans, L., & Short, E.J. ( 1 986). Referential communication and reading performance in learning disabled children over a 3-year period. Developmental Psychology, 22, 177183. Fletcher, J.M. (1985a). External validation of learning disability typologies. In B. P. Rourke (Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis (pp.55-80). New York: Guilford Press. Fletcher, J. M. (1985b). Memory for verbal and nonverbal stimuli in learning disability subgroups. Analysis by selective reminding. Journal of Experimental Child Psychology, 40, 244-259. Fletcher, J. M., & Morris, R. (1986) Classification of disabled learners: Beyond exclusionary definitions. In S . Ceci (Ed.), Handbook of cognitive, social, and neuropsychological aspects of learning disabilities (Vol. 1) (pp 55-80). New York: Erlbaum. Gerber, P.J., & Zinkgraf, S.A. (1982). A comparative study of social-perceptual ability in learning disabled and nonhandicapped students. Learning Disability Quarterly, 5 , 374-378. Jastak, J.F., & Jastak, S.R. (1978). Wide Range Achievement Test. Wilmington, DE: Guidance Associates. Lust, B., Loveland, K., & Kornet, R. (1980). The development of anaphora in first language: Syntactic and pragmatic constraints. Linguistic Analysis, 6, 359-391. Noel, M.M. (1 980). Referential communication abilities of learning disabled children. Learning Disability Quarterly, 3, 70-75. Ozols, E.J., & Rourke, B.P. (1985). Dimensions of social sensitivity in two types of learning-disabled children. In B.P. Rourke, (Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis, (pp 281 -301). New York: Guilford Press. Porter, J.E., & Rourke, B.P. (1985). Identification of subtypes of learning disabled children: A multivariate analysis of patterns of personality functioning. In B.P. Rourke (Ed.), Learning disabilities in children: Advances in subtype analysis (pp 257-280). New York: Guilford Press. Rourke, B.P. (1 978). Reading, spelling, arithmetic disabilities: A neuropsychological perspective. In H.R. Myklebust (Ed.), Progress in learning disabilities (vol. 4 ) (pp 97120). New York:Grune and Stratton. Rourke, B.P. (1982). Central processing deficiencies in children: Toward a developmental neuropsychological model. Journal of Clinical Neuropsychology, 4, 1-18. Rourke, B.P. (1987) Syndrome of nonverbal learning disabilities: The final common pathway of white-matter diseaseldysfunction? The Clinical Neuropsychologist, 1,209234. Rourke, B.P. (1988a) Socioemotional disturbances of learning disabled children. Journal of Consulting and Clinical Psychology, 56, 801 -810. Rourke, B.P. (1 988b) The syndrome of nonverbal learning disabilities: Developmental manifestations in neurological disease, disorder, and dysfunction. The Clinical Neuropsychologist, 2, 293-330. Rourke, B.P., & Fisk, J.L. (1981). Socio-emotional disturbances of learning disabled children: The role of central processing deficits. Bulletin of the Orton Society, 31, 7788. Saloner, M.R., & Gettinger, M. (1985). Social inference skills in learning disabled and nondisabled children. Psychology in the Schools, 22, 201 -207. Siegel, L. S . , & Linder, A. (1984). Short-term memory processes in children with reading and arithmetic disabilities. Developmental Psychology, 20,200-207. Spekman, N.J. (1981). Dyadic verbal communication abilities of learning disabled and normally achieving fourth- and fifth-grade boys. Learning Disability Quarterly, 4 ,
Downloaded by [FU Berlin] at 02:04 15 May 2015
446
KATHERINE A. LOVELAND ET AL.
139-151. Strang, J.D.,& Rourke, B.P. (1985). Arithmetic disability subtypes: The neuropsychological significance of specific arithmetic impairment in childhood. In. B.P. Rourke (Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis (pp.167-182), New York: Guilford Press. Vogel, S.A. (1974). Syntactic abilities in normal and dyslexic children. Journal of Learning Disabilities, 7 , 103-109. Vogel, S.A. (1975). Syntactic abilities in normal and dyslexic children. Baltimore, MD: University Park Press. Wiig, E.H., & Hams, S.P. (1974). Perception and interpretation of nonverbally expressed emotions by adolescents with learning disabilities. Perceptual and Motor Skills, 38, 239-245. Wiig, E.H., & Semel, E.M. (1976). Language disabilities in children and adolescents. Columbus, OH: Charles C. Memll. Yule, W., & Rutter, M. (1976). The epidemology and social implications of specific reading retardation. In R. M. Knights & D. J. Bakker (Eds.). Neuropsychology of learning disorders (pp 25-40). Baltimore, MD: University Park Press.
APPENDIX Sample Story - “The Egg”
1. The chicken was sitting on an egg in her nest. 2. She looked around and yawned. 3. Suddenly, there was a loud noise outside. 4. The chicken jumped up and looked around. She ran out. 5. While she was gone, a fox sneaked in. 6. He sniffed and sniffed until he found the egg. 7. He looked at the egg and got excited. 8. While the fox was looking at the egg, the chicken came in and saw him there. 9. She was mad. 10. But she sneaked out before the fox noticed her. 11. The fox picked up the egg in his mouth. 12. He was about to eat it, when the chicken came back with a hammer in her beak. 13. She came up behind the fox and hit him on the head. 14. Startled, he turned around and dropped the egg back into the nest. 15. The chicken chased the fox out, hitting him on the head. 16. The chicken returned and put the hammer down by the nest. 17. Then she settled back on her egg.
VERBAL AND NONVERBAL COMMUNICATION
447
Sample affect question: How did the chicken feel when she sat back down on her egg?
Downloaded by [FU Berlin] at 02:04 15 May 2015
Coding Categories
Error Code Omission of event
Explanation An event is omitted from the child’s response. Responses did not have to be identical to the model as long as correct events were described or enacted. Where events were not omitted but distorted or altered, other categories were used (see below).
Misinterpretation of event
The meaning of an event is incorrectly interpreted by the child, e.g., in the original story the puppy growled at the rabbit making her cry, but in responding the child interprets the rabbit’s crying as laughter.
Substitution of event
An event in the original story is replaced in the child’s response by a completely different event, e.g., the puppy growled at the rabbit, but the child depicts the puppy as hitting the rabbit.
Embellishments/intrusions
In the child’s response, a completely new event is introduced (not one that takes the place of an original story event), e.g., the child invents a new character or location that did not appear.
Substitution or reversal of props
A different prop than used in the original is depicted in a particular event (ordinarily, a prop is imported from the previous story). A different character than used in the original is depicted as taking part in a particular event.
Substitution or reversal of characters Wrong location
A incorrect location is depicted for an event when a specific location was indicated in the original story.
Events out of sequence
Events are depicted in a different order than in the original story.
Implied information (describe - The child fails to state information that is assumed in response only) subsequent statements. Nonspecific words (describe - The child uses an ambiguous word or expression withresponse only) out specifying a referent, e.g., “And then he found that thing.” Substitution of gesture for words In the describe - response, the child uses gesture in(describe-response only) stead of describing an event, e.g., “And then the bear went like this (gesture).” Substitution of words for gesture’ (enact-response only)
In the enact-response, the child uses words to depict an aspect of an event instead of enacting it, e.g., “And now the bear is reading the note” (with no enactment).
Note: Each story was segmented into “events” for purposes of coding. A given category could only be coded once within a particular event. Unless otherwise noted, each error category could be scored for either an enact-response or a describe-response. Simultaneous narration of an enact-response was discouraged by reminding the child not to narrate, but was not penalized unless the child actually failed to enact an event narrated. Only the enact portion of such a response was coded.
Journal of Clinical and Experimental Neuropsychology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ncen19
Verbal and nonverbal communication of events in learning-disability subtypes a
b
Katherine A. Loveland , Jack M. Fletcher & Vonda Bailey
c
a
University of Texas Medical School , Houston
b
University of Houston ,
c
Houston, Texas Published online: 04 Jan 2008.
To cite this article: Katherine A. Loveland , Jack M. Fletcher & Vonda Bailey (1990) Verbal and nonverbal communication of events in learning-disability subtypes, Journal of Clinical and Experimental Neuropsychology, 12:4, 433-447, DOI: 10.1080/01688639008400991 To link to this article: http://dx.doi.org/10.1080/01688639008400991
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-
Downloaded by [FU Berlin] at 02:04 15 May 2015
licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Clinical and Experimental Neuropsychology 1990, Vol. 12, No. 4, pp. 433447
0168-8634/90/1204-0433$3.00 0 Swets & Zcitlinger
Verbal and Nonverbal Communication of Events in Learning-Disability Subtypes* Katherine A. Loveland University of Texas Medical School - Houston Jack M. Fletcher University of Houston
Vonda Bailey Houston, Texas
Downloaded by [FU Berlin] at 02:04 15 May 2015
ABSTRACT This study compared a group of nondisabled children (ND) with groups of leaming-disabled children who were primarily impaired in reading and arithmetic skills (Reading-Arithmetic Disabled; RAD) and arithmetic but not reading (Arithmetic Disabled; AD) on a set of tasks involving comprehension and production of verbally and nonverbally presented events. Children viewed videotaped scenarios presented in verbal (narrative) and nonverbal (puppet actors) formats and were asked to describe or enact with puppets the events depicted in the stories. Rourke (1978, 1982) has shown that RAD children have problems with verbal skills, whereas AD children have problems with nonverbal skills. Consequently, it was hypothesized that children’s performance in comprehending and reproducing stories would be related to the type of learning disability. Results showed that RAD children made more errors than AD children with verbal presentations and describe-responses, whereas AD children made more errors than RAD children with nonverbal presentations and enact-responses. In addition, learning disabled children were more likely than controls to misinterpret affect and motivation depicted in the stones. These results show that learning disabled children have problems with social communication skills, but that the nature of these problems varies with the type of learning disability.
T h e learning-disabled (LD) child, in addition to having problems in school achievement, may be a poor communicator. I n referential communication tasks, learning-disabled children have been found to be less adept than nondisabled peers at providing appropriate and sufficient information to a listener (Bryan & Pflaum, 1978; Donahue, Pearl, & Bryan, 1980; Noel, 1980; Spekman, 1981) and
* The authors would like to thank Bella Jackson and David Breiger for their work in data collection and coding, and Mrs. Sally Woolrich of the St. Francis School for help in recruitment of children. Address for correspondence: Katherine A. Loveland, University of Texas Mental Sciences Institute, 1300 Moursund, Houston, TX 77030, USA. Accepted: April 28, 1989.
Downloaded by [FU Berlin] at 02:04 15 May 2015
434
KATHERINE A. LOVELAND ET AL.
at indicating their own need for more information when listening to an ambiguous message (Donahue et al., 1980). In conversational situations, LD children are less skilled than nondisabled peers at managing turntaking and initiating conversation and are more likely to defer to others (Donahue, 1983). Feagans and Short (1984) asked reading-disabled children to enact a narrative until it was fully comprehended and then to paraphrase the narrative. Even when comprehension was demonstrated, reading-disabled children of three age groups displayed deficiencies in oral expression on measures of linguistic complexity and content. Feagans and Short (1 986) found that 6-to 9-year-old disabled learners had more difficulty than nondisabled children comprehending information presented (“listener” role) as well as more difficulty explaining this information to others (“speaker” role) and in rephrasing their explanations when cued to do so. These studies raise the question of why the LD child, who by definition has difficulty in academic achievement, should also have difficulty in communicating with others. Although these difficulties are probably due in part to the well-documented language deficits experienced by many LD children (e.g., Wiig & Semel, 1976), they may also be related to difficulties in social perception or other nonverbal aspects of the interaction. In fact, several studies have characterized the LD child as less popular than nondisabled peers, less well-adjusted, more socially insensitive, and more prone to personality disorders (Bachara, 1977; Bryan, 1974, 1978; Denckla, 1972; Rourke & Fisk, 1981). Inability to infer emotions accurately and to grasp the meaning of gesture in interaction is frequently observed in LD children ( Bryan, 1974, 1978; Gerber & Zinkgraf, 1982; Saloner & Gettinger, 1985; Vogel, 1974, 1975; Wiig & Harris, 1974), supporting the hypothesis that problems in communication are not necessarily due only to deficits in language but may be due to deficits in social awareness in at least some disabled learners. It is also important to recognize that not all learning disabilities reflect the same underlying deficits (Rourke, 1978, 1987). Children with different underlying deficits may not exhibit the same difficulties in learning or in communication. Interestingly, problems with verbal and nonverbal aspects of communication have been differentially linked to neuropsychological deficits in disabled learners. 0201s and Rourke (1985) found that LD children with high visualspatial and low verbal skills performed better than children with low visualspatial and high verbal skills on tasks requiring comprehension of facial expression and gesture. However, when a verbal response was required, low visualspatial children performed better. These results suggest that only a subset of the LD population (those lower in visual-spatial skills and higher in verbal skills) may be experiencing difficulty in interpreting affect and interpersonal cues in conversation. Other children, those with specific language-based deficits, may experience difficulty mainly with linguistic aspects of communication. Different types of communication problems may further be linked to academic subtypes of learning disability. Reading-Arithmetic Disabled (RAD) children are predominately language impaired, whereas Arithmetic Disabled (AD)
Downloaded by [FU Berlin] at 02:04 15 May 2015
VERBAL AND NONVERBAL COMMUNICATION
435
children (those unimpaired in word identification) are predominantly impaired on non-language skills (Fletcher, 1985a; Rourke, 1978, 1982, 1987; Rourke & Strang, 1985; Siege1 & Linder, 1984). In addition to displaying poorer nonverbal than verbal skills, children with arithmetic disabilities have been hypothesized to represent a distinct subgroup in terms of social-emotional functioning. They are reported to have significant social problems, particularly in new or unstructured situations (Rourke, 1987) and are described as poor at interpreting others’ emotions, facial expressions, and gestures. This deficiency, which could result in inappropriate interpersonal behavior, could reflect poor visual-spatial perception of nonverbal cues. Consequently, although language problems may underlie the social-communication problems of RAD children, problems with nonlinguistic components may underlie those of AD children. It is important to ask how these underlying disabilities might affect the disabled learner’s ability to use language and understand events both in and outside the classroom. Finally, one might ask how verbal and nonverbal deficits affect the child’s performance as both speaker and listener in communicative interactions. Difficulty in interpreting nonverbal signals, affect, or the meaning of events, together with difficulty in language processing, can lead to deficits in comprehension or production or both (Bailey, 1984). Although it appears that disabled learners may have difficulty with both comprehension and production, it is not clear how the different demands of comprehending and producing information in communication are affected by the modality (verbal or nonverbal) in which a presentation or a response is made. It may be that the effects of modality on comprehension and production vary with the type of learning disability the child experiences. The present study was designed to address these issues by examining both comprehension and production of verbally and nonverbally presented events and comprehension of the affective/motivational content of these events by groups of disabled learners primarily impaired in reading (RAD) versus arithmetic (AD) and by a group of non-disabled control children (ND). We hypothesized the following: 1 . Learning disabled children’s performance in comprehending and reproducing stories would be related to the format (verbal or nonverbal) of the story and the type of learning disability present. In particular, a) Arithmetic disabled (AD) children were expected to make more errors than other children in both to nonverbal presentations of events and in nonverbal responses by the child. b) Reading-Arithmetic disabled (RAD) children were expected to make more errors than other children in both verbal presentations of events and in verbal responses. 2 . Children’s ability to interpret successfully the affective content of events was expected to be related to whether a learning disability is present and to the subtype of learning disability. In particular,
436
KATHERINE A. LOVELAND ET AL.
a) AD disabled children were expected to make a greater number of errors in interpreting affect than RAD or ND children (because of the relationships of visual-spatial proficiency and nonverbal cues in social-communicative interactions).
b) RAD children were expected to make fewer errors of interpreting affect than AD children but a greater number than ND children.
Downloaded by [FU Berlin] at 02:04 15 May 2015
METHODS Subjects Children for this study were recruited from area agencies and schools for leaming-disabled children and the local Association for Children with Learning Disabilities. ND children were friends of the LD children or were obtained through volunteer solicitation at local schools. In order to qualify for the study, all children were Caucasians with (a) Full Scale WISC-R or Stanford-Binet IQ above 85 (all children received the WISC-R except four controls with recent Stanford Binets); (b) age from 104-160 months; (c) no history or evidence of significant neurologic anomaly, emotional disturbance, or cultural deprivation; and (d) predominantly English-speaking background. Three groups of children were formed based on Wide Range Achievement Test (WRAT) scores (Jastak & Jastak, 1978). The ND children ( n = 14) had achievement centiles above 39 on the three WRAT subtests (reading recognition, spelling, arithmetic). Children with learning disabilities in arithmetic (AD; n = 12) had reading centile scores above 39, with arithmetic scores below the 31st centile and at least one standard deviation below the reading score. Spelling scores were not controlled because previous studies have not found AD children with and without spelling difficulties to differ on a variety of cognitive tasks (Fletcher, 1985b; Siege1 & Linder, 1984). Children impaired in reading and arithmetic (RAD; n = 13) had WRAT centile scores uniformly below 31 on both
Table 1: Means and Standard Deviations for Demographic and Selection Variables by Achievement Group WRAT Centiles Age (in months) FSIQ Percent Groupn Male M SD M SD
Reading
M
ND
14 57
131.8
13.7 109.3
6.2 78.2
AD
12 50
137.1
17.8 103.3
12.2 63.8
RAD 13 62
132.5
9.5
98.3
ND = Nondisabled AD = Arithmetic Disabled RAD = Reading-Arithmetic Disabled
7.6
13.3
SD
Spelling
Arithmetic
M
M
5.0 79.7
SD
SD
16.5
59.8
11.6
13.0 39.4 26.4
16.8
7.7
10.7
8.3
7.8
11.1
9.3
Downloaded by [FU Berlin] at 02:04 15 May 2015
VERBAL AND NONVERBAL COMMUNICATION
437
subtests with an average centile score below 26. Consequently, the two LD groups should be expected to differ in WRAT reading recognition scores, but not on arithmetic scores. Table 1 presents the sample size, gender (percent male), age, Full Scale IQ scores, and WRAT centile scores for the controls and two learning disability groups. Analysis of variance revealed no group differences in age or gender (p > .05). On FSIQ, the ANOVA was significant (F = 4.74; p < .05), with follow-up t tests showing that ND children had higher IQ scores than either learning-disability group, who were not significantly different. This difference between disabled and nondisabled learners is expected, given the sampling strategy, which controls for range of IQ. Greater control over mean IQ is not desirable, because the population of disabled learners has a lower IQ than the normal population (Yule & Rutter, 1975). Subject or group-based matching on IQ is inappropriate in situations such as this in which a quasi-experimental design is used because of potential selection bias and subsequent regression to the mean (Campbell & Erlebacher, 1970; Fletcher & Moms, 1986). As expected, the ND group scored significantly higher than the two LD subtypes on all three WRAT subtests. The AD group scored significantly higher than the RAD group on reading and spelling, but not on arithmetic. It should be noted that the AD group had reading centiles that are clearly in the average range. The differences in selection vanables are all expected, and they conform to patterns commonly found when ND learners are compared with AD and RAD children (Rourke, 1978). Because the primary comparisons of interest are between the two learning-disability subtypes, and because the magnitude of the IQ difference is relatively small, it is not likely that this difference interfered with interpretation of the results of this study.
Materials Eight videotapes of brief stones were made (Bailey, 1984). The stories involved simple, easily recognizable interactions among familiar characters (a bear, a dog, a chicken, etc.), and each contained some affectively charged situations (e.g.. a character becomes frightened and cries). Four different scripts were videotaped in both verbal and nonverbal presentation conditions. On the nonverbal tape, puppets act out a skit, and no talking occurs. On the verbal videotape, the face of a narrator telling the same story is shown, and the puppets do not appear. Care was taken to prepare story scripts that could be narrated and enacted with equal clarity and emphasis for all events in both versions. Two characters and one or two props were used in each story. All four stories were approximately equivalent in the number of events that occur and in the length of the videotapes, in both verbal and nonverbal format (about 1.5 min.). The same puppets and props that appeared i n the videotapes were used by the children to enact the stories. A sample story is given in the Appendix.
Procedure In one half-hour session, each subject was first shown two practice skits similar to the four later shown in the experimental trials, but involving different stories, characters, and props. One practice skit was nonverbal and the child was asked to describe it. The other was verbal, and the child was asked to act it out with the puppets. At this time, the procedure was explained and the child’s questions answered. If the child had any difficulty following the instructions (e.g., talked instead of acting out with the puppets, failed to complete the story, etc.), corrections and additional instructions were given. Each child was then shown the four experimental videotapes, one representing each condition (verbal-enact, verbal-describe, nonverbal-enact, nonverbal-describe). Each of these four tapes depicted a different story. Order of story presentation was fixed, but order of conditions for presentation and response was counterbalanced. The child viewed a tape, and then was asked either to enact or to describe the story. Children were not informed in advance which response would be required of them. Enact-
438
KATHERINE A. LOVELAND ET AL.
Downloaded by [FU Berlin] at 02:04 15 May 2015
ing the script entailed moving the puppets to make them do what they did in the story, without using words to substitute for actions. Lust, Loveland. and Komet (1980) used a similar procedure to test children’s comprehension of story sentences. Describing the script required telling, as accurately as possible, what occurred in the story, without using gestures to substitute for words. All responses were recorded on videotape for later coding. After a child responded to each story, a set of specific follow-up questions were asked; these dealt with the child’s interpretation of the emotions and motives of characters in the story. These responses were also videotaped for later coding. Coding Three coders separately coded the participants’ videotaped responses using a prepared format detailing in chronological order each of the events included in the story. As the child’s verbal or nonverbal response was viewed, it was coded for errors in the following areas: omission of an event, misinterpretation of an event, substitution of events, embellishments/intrusions, substituting or reversing characters, substituting or reversing props, wrong location, events out of sequence, implied information, use of nonspecific words (describe-response only), substituting gesture for words (describe-response only), and substituting words for gesture (enact-response only). Coding categories are explained in the Appendix. Several categories were eliminated because of very low frequency: substitution of events, substitution of word for gesture, substitution of gesture for word, location errors, and prop substitution or reversal. Responses to follow-up questions were also coded. Each response was first coded as either appropriate or inappropriate, according to whether it correctly (or reasonably) captured the emotion or motivation displayed by the character. Each response was then coded as primarily concrete or motivational in content, according to whether the focus of the response was on the characters’ actions or on feelings/motivations. For example, if asked, “Why did the dog growl at the rabbit?” a child might respond “Because he was angry” (motivational) or “Because the rabbit took the ball” (concrete). Note that both motivational and concrete responses could be either appropriate or inappropriate. Coders were initially trained to reach 90% agreement or greater across all error categories and responses to follow-up questions using videotapes of children not meeting subject-selection criteria. Agreements were counted by tabulating matching Occurzences of a particular error category for a particular line, event, or question in a story (see sample story in appendix) rather than for the story as a whole. Thereafter, periodic coding reliability checks were carried out to insure that interrater agreement remained high. When error categories were examined separately for all reliability re-coding performed (25% of data, randomly selected), percents of agreement ranged between 84% - 96%. Percents of agreement for responses to follow-up questions (appropriate/inappropriate, concrete/motivational) were 97% and 95%. respectively.
RESULTS
Omissions.Means and standard deviations for omission errors are presented in Table 2. A three-way ANOVA revealed no significant Group x Response x Presentation interaction. The two way interactions of Group x Response, F (2,36) = 6.0, p < .006, Group x Presentation, F (2,36) = 3.7, p c .03, and Presentation x Response, F (2,36)= 12.2, p < .001, were all significant. Follow-up tests using Tukey’s procedure (overall p = .05)showed that, when asked to describe, RAD children made
439
VERBAL AND NONVERBAL COMMUNICATION
Table 2: Means and Standard Deviations for Errors of Omission by Presentation and Response conditions Conditions Verbal
Downloaded by [FU Berlin] at 02:04 15 May 2015
Group
Enact
Nonverbal Describe
Enact
Describe
M
SD
M
SD
M
SD
M
SD
ND
2.1
1.3
4.8
1.9
2.4
1.7
7.7
2.8
AD
5.0
2.5
4.5
2.2
4.9
3.3
7.9
2.4
RAD
6.5
3.3
8.1
3.4
5.5
1.9
8.2
2.6
ND - Nondisabled AD - Arithmetic Disabled RAD - Reading-Arithmetic Disabled
more omissions than did AD or ND children, who did not differ from each other (Table 2). By contrast, in the enact response, both RAD and AD children made more omissions than did ND children. This seems to suggest that, when a verbal response is required, RAD children are at a disadvantage, making more omissions relative to AD children, but that when a nonverbal response is required, both RAD and AD children make more omissions than ND children. Similarly, with a verbal presentation, RAD children made more omissions than did AD and ND children; with a nonverbal presentation, AD children’s errors increased such they also differed from the ND children. For all children, describing a nonverbal presentation led to more omissions than did describing a verbal presentation; however, for the enact responses there were no differences between verbal and nonverbal presentations.
Embellishments. There were no group differences or interactions present for embellishments. However, there were significant main effects of Presentation, F (1,26) = 6.84, p < .01, and Response, F (1,36) = 5.0, p < .03. Tukey’s HSD procedure revealed that, for all children, verbal presentation led to significantly fewer embellishments than nonverbal presentations, and describe-responses led to more than enact-responses. This may suggest that, in general, verbal presentation constrains the adding of new material more than does nonverbal presentation; however, when extra material is inserted, it is more likely to be done verbally than nonverbally. Churucter reversal. Because of the low error rate, enact and describe response conditions were not examined for character reversals, and differences in presen-
440
KATHERINE A. LOVELAND ET AL.
Table 3: Means and Standard Deviations for Character Reversal Error by Presentation Conditions Group
Conditions
Downloaded by [FU Berlin] at 02:04 15 May 2015
Verbal
Nonverbal
M
SD
M
ND
.36
SO
.14
AD
.66
.99
SO
RAD
.54
.78
.15
SD .36 1.2 .38
ND - Nondisabled AD - Arithmetic Disabled RAD - Reading-Arithmetic Disabled
Table 4: Number of Children Misinterpreting Events by Group after Nonverbal Presentation Response Condition Enact
Group
Error
Describe Noerror
Error
No error
ND
12
0
14
AD
9
5
7
11
1
12
RAD ND - Nondisabled AD - Arithmetic Disabled RAD - Reading-Arithmetic Disabled
tation conditions only were examined. A significant interaction of Group x Presentation was found, F (2,36) = 3.9, p < .03. As Table 3 shows, A D children made more character reversals than did R A D and N D children in the nonverbal presentation. However, with the verbal presentation, the number of character reversals by R A D children increased such that there was no longer a difference between them and A D children. This result suggests that A D children were more likely than other children to confuse the roles of actors within events when the events
441
VERBAL AND NONVERBAL COMMUNICATION
were presented nonverbally, but that, when the events were presented verbally, RAD children made as many errors as did AD children. Misinterpretation of events. Table 4 presents the number of children making errors for the category of misinterpretation of events. In all groups this error was made exclusively with the nonverbal presentation. In addition, error rates were very low for all groups except AD. Consequently, separate Chi-square analyses were conducted comparing the number of children in each group who made errors for both the Nonverbal-Enact and Nonverbal-Describe conditions. For Nonverbal-Enact, Chi-square was not significant. However, the analysis was significant for Nonverbal-Describe, ( d f = 2) = 7.1 p c .05, and showed that the AD children were more likely to misinterpret nonverbal events than were either the RAD or ND children. Downloaded by [FU Berlin] at 02:04 15 May 2015
x2
Sequence errors. No group differences were present in sequence errors, but a significant main effect for Response was found, F (1,36) = 3 . 8 , < ~ .02. Followup comparisons using Tukey 's procedure were not significant.
Implied Information. Errors of implied information were coded only for the describe-response condition. ANOVA revealed a significant group difference, F (1,36) = 4.2, p < .02, with RAD children making more errors of implied information than did either of the other two groups (see Table 5 ) . This result indicates that RAD children more often than other groups failed to state explicitly information about story events.
Table 5: Means and Standard Deviations for Errors of Implied Information and Nonspecific Words for the Describe Response Conditions Error
Group Implied Information
M
SD
Nonspecific Word
M
SD
ND
.43
.85
.07
.27
AD
.25
.87
.07
.21
1.24
1.01
.46
.66
RAD
ND - Nondisabled AD - Arithmetic Disabled RAD - Reading-Arithmetic Disabled
442
KATHERINE A. LOVELAND ET AL.
Table 6: Means and Standard Deviations for Appropriate and Inappropriate Interpretations of Affect Responses by Category (Concrete, Motivational) Motivational Appropriate ~~~
Downloaded by [FU Berlin] at 02:04 15 May 2015
M
Concrete
Inappropriate ~
SD
Appropriate
Inappropriate
~
M
SD
M
SD
M
SD
ND
6.29 1.59
0.43
0.65
3.89 1.53
0.43
0.65
AD
4.50 1.98
1.33
1.16
4.50 1.24
0.58
1.24
RAD
4.80 1.80
.85
.90
4.23 1.24
1.08
1.38
ND - Nondisabled AD - Arithmetic Disabled RAD - Reading-ArithmeticDisabled
Nonspecific words, Use of nonspecific words was coded only for the describeresponse condition, Groups differed significantly, F (1,36) = 4.63, p < .02.Table 5 shows that RAD children made more of these errors than did the other two groups. RAD children were more likely than were other groups to substitute pronouns and other non-specific circumlocutions for specific story elements. Interpretation of affect. A significant main effect for Group, F (1,36) = 4.03, p < .02, was found for the category of appropriate/motivational responses with no effects for Presentation or Response (Table 6). Non-disabled children produced significantly more appropriate/motivational responses than did either of the learning disabled subtypes, indicating that they more often correctly identified a character’s motivation or emotion. RAD and AD children made more inappropriatekoncrete responses (errors of fact) and also more inappropriate/motivational responses (errors of misinterpreting affect) than did ND children. There were no significant differences between AD and RAD children. These results suggest that learning disabled children had more difficulty than ND children interpreting the affect-related events depicted in the and perhaps more difficulty in talking about these events. DISCUSSION These results largely support the first hypothesis of the study: i.e., that children’s performance in comprehending and reproducing stories would be related to the subtype of learning disability. The two subtypes of learning-disabled children differed from each other and from nondisabled children in the production and
Downloaded by [FU Berlin] at 02:04 15 May 2015
VERBAL AND NONVERBAL COMMUNICATION
443
comprehension of verbally and nonverbally presented stories. In general, AD children had more difficulty with nonverbal aspects of the story tasks, and RAD children had more difficulty with verbal aspects of the story tasks. Both subtypes of disabled learners committed more errors of omission and character reversal than did nondisabled children. Overall, RAD children made more omissions than did AD children when describing an event in either format, whereas AD children made more omissions than did RAD children when enacting nonverbal events. Similarly, when events were presented nonverbally, AD children reversed the role of actors more often than did the other groups; with verbal presentation, RAD children’s errors increased to the level of the AD group. The RAD children made more specifically verbal errors (such as implication errors and use of nonspecific words) than other children when describing events, whereas AD children performed similarly to controls with respect to verbal errors. Arithmetic disabled children were more likely than the other two groups to misinterpret events, particularly when describing events that were presented nonverbally. Overall, RAD children experienced more difficulty with verbal presentations and describe-responses, whereas AD children experienced more difficulty with nonverbal presentations and enact-responses. However, apparent problems with comprehension of verbal events by RAD children and nonverbal events by AD children seemed to lead to problems with production in both types of response format. This result suggests that, in the absence of adequate comprehension, production problems are likely to emerge regardless of the response format. There was less support for the second hypothesis: i.e., that AD children would have the most difficulty interpreting affect. Although both LD groups made more errors interpreting affect than did the controls, there were no significant differences between the two LD subtypes. However, in interpreting this result, it should be noted that there was a strong trend for the AD children to make more “inappropriate/motivational” responses than did RAD children, suggesting more misinterpretations of affect or motive. A larger sample might provide a more definitive test of this hypothesis. ln addition, all questions and responses about affect were given verbally, which may have placed RAD children at a disadvantage. It is possible that some errors made by RAD children were not problems of interpreting story affect but of talking about it, because the demands of oral expression in this task overtaxed the RAD child’s language system. Although Ozols and Rourke (1985) used slightly different criteria to form groups, they found that LD children with poor visual-spatial skills (all of whom were impaired on WRAT arithmetic but not necessarily reading) had more difficulty in pantomiming an appropriate affective response or in selecting an appropriate facial expression after a verbally presented story, whereas language impaired children performed like controls on these tasks. Consequently, requiring a verbal response in the present study may have masked comprehension of affect by the RAD group. These results support the hypothesis that the social-communicative problems
Downloaded by [FU Berlin] at 02:04 15 May 2015
444
KATHERINE A. LOVELAND ET AL.
of LD children vary according to the nature of the underlying cognitive impairment. In addition, they show that requirements for presentation and response interact with the nature of these impairments. Learning disabled children may, indeed, have special difficulty understanding and remembering narrative information or the actual sequences of events that they experience, both within and outside the classroom. They may have more difficulty than other children in describing what they know or in demonstrating a skill to someone else. Moreover, at least some children may have additional problems in interpreting and expressing emotions and social cues appropriately. These problems can have a negative impact not only on school performance but on peer acceptance and on the development of age-appropriate adaptive skills. Although it is likely that some LD children have social-emotional problems that result from their school failure and poor achievement, not all LD children exhibit such difficulties (Porter & Rourke, 1985; Rourke, 1988). The results of the current study add to the growing body of literature suggesting that the manifestations of socioemotional problems that are exhibited by LD children may also reflect differential patterns of cognitive assets and impairments in such children (Rourke, 1982, 1987, 1988a, 1988b). Future studies of social-communicative skills would do well to measure carefully the various processing deficits exhibited by LD children, as well as to outline more precisely the nature of selection criteria, particularly in terms of academic skills.
REFERENCES Bachara. G. (1977). Empathy in learning disabled children. Journal of Learning Disabilities, 10, 507-508. Bailey, V.E.(1984). Event description by two subtypes of learning disabled children. Unpublished master’s thesis, Rice University. Bryan, R.M. (1974). Peer popularity of learning disabled children. Journal of Learning Disabilities, 7, 66 1-625. Bryan, T. (1978). Social relationships and verbal interactions of learning disabled children. Journal of Learning Disabilities, 11, 107-115. Bryan, T.H. (1978). Verbal interactions and social relationships of learning disabled children in the classroom. Journal of Learning Disabilities, 11, 107-1 15. Bryan, T.S., & Pflaum, S. (1978). Social interactions of learning disabled children: A linguistic, social and cognitive analysis. Learning Disability Quarterly, 1,70-79. Campbell, D.T.. & Erlebacher, A. (1 970). How regression artifacts in quasi-experimental evaluations can mistakenly make compensatory education look harmful. In J. Helmuth (Ed.), Contemporary education: A national debate (p. 2-38). New York: BrunerMazel. Denckla, M.B. (1972). Clinical syndromes in learning disabilities: The case for “splitting’’ versus “lumping”. Journal of Learning Disabilities, 5,401 -406. Donahue, M. (1983). Learning disabled children as conversational partners. Topics in Language Disorders, 4 , 15-27. Donahue, M., Pearl, R., & Bryan, T. (1980). Learning disabled children’s conversational competence: Responses to inadequate messages. Applied Psycholinguistics, 1 , 387403.
Downloaded by [FU Berlin] at 02:04 15 May 2015
VERBAL AND NONVERBAL COMMUNICATION
445
Feagans, L., & Short, E.J. (1984). Developmental differences in the comprehension and production of narratives by reading disabled and normally achieving children. Child Development, 55, 1727-1736. Feagans, L., & Short, E.J. ( 1 986). Referential communication and reading performance in learning disabled children over a 3-year period. Developmental Psychology, 22, 177183. Fletcher, J.M. (1985a). External validation of learning disability typologies. In B. P. Rourke (Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis (pp.55-80). New York: Guilford Press. Fletcher, J. M. (1985b). Memory for verbal and nonverbal stimuli in learning disability subgroups. Analysis by selective reminding. Journal of Experimental Child Psychology, 40, 244-259. Fletcher, J. M., & Morris, R. (1986) Classification of disabled learners: Beyond exclusionary definitions. In S . Ceci (Ed.), Handbook of cognitive, social, and neuropsychological aspects of learning disabilities (Vol. 1) (pp 55-80). New York: Erlbaum. Gerber, P.J., & Zinkgraf, S.A. (1982). A comparative study of social-perceptual ability in learning disabled and nonhandicapped students. Learning Disability Quarterly, 5 , 374-378. Jastak, J.F., & Jastak, S.R. (1978). Wide Range Achievement Test. Wilmington, DE: Guidance Associates. Lust, B., Loveland, K., & Kornet, R. (1980). The development of anaphora in first language: Syntactic and pragmatic constraints. Linguistic Analysis, 6, 359-391. Noel, M.M. (1 980). Referential communication abilities of learning disabled children. Learning Disability Quarterly, 3, 70-75. Ozols, E.J., & Rourke, B.P. (1985). Dimensions of social sensitivity in two types of learning-disabled children. In B.P. Rourke, (Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis, (pp 281 -301). New York: Guilford Press. Porter, J.E., & Rourke, B.P. (1985). Identification of subtypes of learning disabled children: A multivariate analysis of patterns of personality functioning. In B.P. Rourke (Ed.), Learning disabilities in children: Advances in subtype analysis (pp 257-280). New York: Guilford Press. Rourke, B.P. (1 978). Reading, spelling, arithmetic disabilities: A neuropsychological perspective. In H.R. Myklebust (Ed.), Progress in learning disabilities (vol. 4 ) (pp 97120). New York:Grune and Stratton. Rourke, B.P. (1982). Central processing deficiencies in children: Toward a developmental neuropsychological model. Journal of Clinical Neuropsychology, 4, 1-18. Rourke, B.P. (1987) Syndrome of nonverbal learning disabilities: The final common pathway of white-matter diseaseldysfunction? The Clinical Neuropsychologist, 1,209234. Rourke, B.P. (1988a) Socioemotional disturbances of learning disabled children. Journal of Consulting and Clinical Psychology, 56, 801 -810. Rourke, B.P. (1 988b) The syndrome of nonverbal learning disabilities: Developmental manifestations in neurological disease, disorder, and dysfunction. The Clinical Neuropsychologist, 2, 293-330. Rourke, B.P., & Fisk, J.L. (1981). Socio-emotional disturbances of learning disabled children: The role of central processing deficits. Bulletin of the Orton Society, 31, 7788. Saloner, M.R., & Gettinger, M. (1985). Social inference skills in learning disabled and nondisabled children. Psychology in the Schools, 22, 201 -207. Siegel, L. S . , & Linder, A. (1984). Short-term memory processes in children with reading and arithmetic disabilities. Developmental Psychology, 20,200-207. Spekman, N.J. (1981). Dyadic verbal communication abilities of learning disabled and normally achieving fourth- and fifth-grade boys. Learning Disability Quarterly, 4 ,
Downloaded by [FU Berlin] at 02:04 15 May 2015
446
KATHERINE A. LOVELAND ET AL.
139-151. Strang, J.D.,& Rourke, B.P. (1985). Arithmetic disability subtypes: The neuropsychological significance of specific arithmetic impairment in childhood. In. B.P. Rourke (Ed.), Neuropsychology of learning disabilities: Essentials of subtype analysis (pp.167-182), New York: Guilford Press. Vogel, S.A. (1974). Syntactic abilities in normal and dyslexic children. Journal of Learning Disabilities, 7 , 103-109. Vogel, S.A. (1975). Syntactic abilities in normal and dyslexic children. Baltimore, MD: University Park Press. Wiig, E.H., & Hams, S.P. (1974). Perception and interpretation of nonverbally expressed emotions by adolescents with learning disabilities. Perceptual and Motor Skills, 38, 239-245. Wiig, E.H., & Semel, E.M. (1976). Language disabilities in children and adolescents. Columbus, OH: Charles C. Memll. Yule, W., & Rutter, M. (1976). The epidemology and social implications of specific reading retardation. In R. M. Knights & D. J. Bakker (Eds.). Neuropsychology of learning disorders (pp 25-40). Baltimore, MD: University Park Press.
APPENDIX Sample Story - “The Egg”
1. The chicken was sitting on an egg in her nest. 2. She looked around and yawned. 3. Suddenly, there was a loud noise outside. 4. The chicken jumped up and looked around. She ran out. 5. While she was gone, a fox sneaked in. 6. He sniffed and sniffed until he found the egg. 7. He looked at the egg and got excited. 8. While the fox was looking at the egg, the chicken came in and saw him there. 9. She was mad. 10. But she sneaked out before the fox noticed her. 11. The fox picked up the egg in his mouth. 12. He was about to eat it, when the chicken came back with a hammer in her beak. 13. She came up behind the fox and hit him on the head. 14. Startled, he turned around and dropped the egg back into the nest. 15. The chicken chased the fox out, hitting him on the head. 16. The chicken returned and put the hammer down by the nest. 17. Then she settled back on her egg.
VERBAL AND NONVERBAL COMMUNICATION
447
Sample affect question: How did the chicken feel when she sat back down on her egg?
Downloaded by [FU Berlin] at 02:04 15 May 2015
Coding Categories
Error Code Omission of event
Explanation An event is omitted from the child’s response. Responses did not have to be identical to the model as long as correct events were described or enacted. Where events were not omitted but distorted or altered, other categories were used (see below).
Misinterpretation of event
The meaning of an event is incorrectly interpreted by the child, e.g., in the original story the puppy growled at the rabbit making her cry, but in responding the child interprets the rabbit’s crying as laughter.
Substitution of event
An event in the original story is replaced in the child’s response by a completely different event, e.g., the puppy growled at the rabbit, but the child depicts the puppy as hitting the rabbit.
Embellishments/intrusions
In the child’s response, a completely new event is introduced (not one that takes the place of an original story event), e.g., the child invents a new character or location that did not appear.
Substitution or reversal of props
A different prop than used in the original is depicted in a particular event (ordinarily, a prop is imported from the previous story). A different character than used in the original is depicted as taking part in a particular event.
Substitution or reversal of characters Wrong location
A incorrect location is depicted for an event when a specific location was indicated in the original story.
Events out of sequence
Events are depicted in a different order than in the original story.
Implied information (describe - The child fails to state information that is assumed in response only) subsequent statements. Nonspecific words (describe - The child uses an ambiguous word or expression withresponse only) out specifying a referent, e.g., “And then he found that thing.” Substitution of gesture for words In the describe - response, the child uses gesture in(describe-response only) stead of describing an event, e.g., “And then the bear went like this (gesture).” Substitution of words for gesture’ (enact-response only)
In the enact-response, the child uses words to depict an aspect of an event instead of enacting it, e.g., “And now the bear is reading the note” (with no enactment).
Note: Each story was segmented into “events” for purposes of coding. A given category could only be coded once within a particular event. Unless otherwise noted, each error category could be scored for either an enact-response or a describe-response. Simultaneous narration of an enact-response was discouraged by reminding the child not to narrate, but was not penalized unless the child actually failed to enact an event narrated. Only the enact portion of such a response was coded.
