Understanding Social Adaptation in Children with Mental Retardation: A Social-Cognitive Perspective.
Clearly, if children with mental retardation (MR) are to successfully meet the challenges of everyday social adaptation, they must do more than simply acquire and rehearse behavioral social skills. As the social skills training literature has consistently recognized, helping children to learn specific social behaviors, such as taking turns, sharing possessions, and asking for and offering help, is only part of the solution. According to this literature, social skills instruction confronts educators with the equally important challenge of facilitating the transfer of newly-acquired behaviors to real-life situations (Stokes & Baer, 1977), or, in Gresham's (1986) terminology, addressing children's "performance deficits" as well as their "skill deficits." We believe that in order to address the key question of why children with MR often have difficulty demonstrating the right social behavior at the appropriate time--and, consequently, what can be done to help these children to become more skillful in real life social situations--we need to move beyond an exclusively behavioral paradigm to adopt a social-cognitive perspective.
The social-cognitive perspective leads, us to shift our focus to a more process-oriented conception of social skills; specifically, away from an exclusive focus on discrete, observable behaviors and toward an examination of underlying social-cognitive processes which guide children's performance of socially adaptive behavior. The study of children's social cognition is not new (see Shantz, 1983, for an earlier review). Its history dates back to initial attempts in the 1950s and 60s to explore the implications for social development of Piaget's theories concerning children's development of internal schemas for understanding the nonsocial world (Feffer, 1959, Kohlberg, 1969), as well as investigate children's ideas for solving interpersonal problems (Goldfried & D'Zurilla, 1969; Spivack & Shure, 1974). In recent years, the emergence of new theoretical models of social-cognitive processing (e.g., the Social Information Processing model of Crick & Dodge, 1994) has stimulated renewed interest in the examination of social-cognitive processes and their role in children's social adaptation.
In our view, educators who are seeking to understand the nature of the difficulties in social functioning that children with MR experience can benefit from examining these children's social cognitive processes for two reasons. First, social-cognitive research illuminates specific linkages between the cognitive limitations of children with MR, which are a defining characteristic of their disability, and these children's problems in social adaptation. Second, the social-cognitive perspective, which has informed social skills training programs for other populations of children (e.g., Elias & Tobias, 1996; Kusche & Greenberg, 1994), may also contribute to the development of innovative social skills remediation strategies for children with MR.
This study investigated the social-cognitive processing skills of children with MR by focusing on two key processes, social perception and strategy generation, that have been found to be of special importance for meeting the social challenges of the classroom (Crick & Dodge, 1994; Wong, Day, Maxwell, & Meara, 1995). The first of these processes, social perception, refers to an individual's ability to interpret or "read" relevant social messages from others (Maheady, Harper, & Sainato, 1987). These messages, known as social cues, consist of verbal and nonverbal stimuli in the environment. Social cues can include physical actions, words, facial expressions, tone of voice, and body language that tell about others' behaviors, feelings, and intentions. One can further differentiate two logically distinct but related competencies involving the perception of social cues (Crick & Dodge; Wong, et al.): the encoding of cues (focusing one's attention on particular social cues in a situation and representing these cues in working memory) and the accurate interpretation of cues (drawing conclusions about the meaning of these cues), for example, in making inferences about others' intentions. This distinction between the encoding and interpretation of social cues is valuable for accurately diagnosing a child's specific problems in social perception. Faulty encoding of the available social cues may deprive a child of information needed to arrive at an accurate interpretation of a social situation. Alternatively, the child may have successfully encoded the available social cues, but nonetheless made an error in interpreting these cues because of faulty interpretation skills.
The two social perception processes of encoding and interpretation are challenging for children with MR because the social cues that confront these children during everyday social problems occur quickly and are continuously in flux. At one moment, for example, a child may be jostled in a way that appears to have been by accident; at another time, the same child may be deliberately shoved, perhaps even by the same classmate. Somehow, the child has to quickly differentiate between these two events. Furthermore, in order to make plausible inferences, for example, regarding a classmate's intentions, children must coordinate and integrate information from multiple social cues, which can include potentially conflicting information. For example, a child must integrate the fact of being jostled by a classmate with equally important cues such as the classmate's tone of voice or eye contact as he says, "I'm sorry," or that the hall was narrow and crowded. For these reasons, social perception is a social-cognitive process which is both critical and challenging for all children, but particularly for children with MR, who have limitations in their cognitive processing abilities.
The second key social-cognitive process, strategy generation, is also challenging because it involves the ability to think of solutions for resolving social problems that are age-appropriate and that fit the immediate situation. For example, when a child wants to join an ongoing activity, asking may be a suitable strategy. However, waiting and watching may be a better strategy if the child's peers are very intensely involved in a game. Similarly, when a child's books are knocked off his desk by a peer, complaining to the teacher may be an appropriate strategy, but not if social cues clearly indicate that this happened by accident. Moreover, complaining to the teacher about being provoked or rebuffed by a peer may be an appropriate strategy for a second-grader but not for a fifth-grader. Thus, strategy generation skills, as well as social perception skills, are critical to the child's capacity to respond in a developmentally appropriate way to the dynamic social environment of the classroom.
Having established the importance of these underlying processes for successful social adaptation, we are now ready to examine how successful children with MR are at performing these key social-cognitive skills. We believe that children with MR are at risk for experiencing problems in social perception and strategy generation, in light of preliminary evidence indicating that not only do these children experience difficulties in performing both of these social-cognitive processes (Leffert & Siperstein, 1996), but also that their ability to perform these processes is related to maladaptive and adaptive patterns of social behavior and whether these children are socially accepted or rejected by classmates (Siperstein & Leffert, 1997).
Furthermore, the recent evidence of limitations in social perception and strategy generation among children with MR is consistent with earlier research findings. For example, Maheady, Maitland, and Sainato (1984) found that children and adolescents with MR were less successful at accurately interpreting social cues than children without disabilities, children with learning disabilities, and children with social/emotional difficulties. Also, Smith (1986) found that children with MR in the sixth grade generated strategies for resolving social problems that were less mature than those generated by other sixth-graders and, in fact, resembled those generated by third-graders without disabilities.
The present study continues this earlier line of research by examining more closely the ability of children with MR to perform these two processes, social perception and strategy generation. In addition, the present study examined these processes in a larger and more heterogeneous sample, including children with and without MR and in Grades 1 through 5, than had been employed in previous studies (Leffert & Siperstein, 1996; Siperstein and Leffert, 1997).
Specifically, in the current investigation we assessed the social perception and strategy generation skills of children with and without MR in relation to the dynamic social environment of the classroom by presenting them with contrasting social cues and different types of social conflicts. By doing so, we were able to assess how effective these children were at interpreting another child's intentions, particularly when the social environment contained conflicting social cues involving outcome and intentions, and at generating social strategies in response to varied social conflicts. We also examined the extent to which children's interpretation of a peer's intentions ("being mean" or "not being mean") influenced their selection of strategies, thereby examining the ability of children with MR to coordinate the two social-cognitive processes. Through this investigation of social perception and strategy generation processes, we hoped to better understand the capacity of children with MR to meet the challenges of successful social adaptation in their school and classroom.
The present study was part of a larger investigation of the social functioning of children with MR. Subjects were 59 elementary school children with mild MR (31 girls and 28 boys), and 58 comparison children without MR (30 girls and 28 boys) who were between the ages of 7.0 and 11.9 years. The children with MR were drawn from self-contained special education classrooms in eight urban elementary schools. Their breakdown by grade was as follows: 4 first-graders, 10 second-graders, 18 third-graders, 21 fourth-graders, and 6 fifth-graders. Children were included if they were identified by school personnel as receiving special education services and a recent standardized IQ test indicated that they were functioning within the range of mild MR (IQ score between 50 and 75). The mean IQ score of this group was 65.2 (SD = 6.1). Eighty-seven percent of the children were from ethnic minority backgrounds.
Children without MR were randomly selected from general education classrooms within the same schools as the children with MR and were similar with regard to age, gender, and ethnicity. Eighty-three percent were from ethnic minority backgrounds. Their breakdown by grade was as follows: 4 first-graders, 15 second-graders, 13 third-graders, 15 fourth-graders, and 11 fifth-graders.
Assessment of Social Cognitive Processes
To assess the ability of children with and without mild MR to perform the social-cognitive processes of social perception and strategy generation, we conducted individual structured interviews accompanied by videotaped stimuli. The video materials and interview format have been used successfully with children in the primary elementary grades (Dodge, Bates, & Pettit, 1990; Weiss, Dodge, Bates, & Pettit, 1992) and with children with mild MR in the upper elementary grades (Leffert & Siperstein, 1996).
The stimuli consisted of 18 vignettes, (approximately 30 s each) representing two types of conflict situations that children typically experience: peer entry (9 stories) and peer provocation (9 stories). The peer entry vignettes depicted a negative event in which the child protagonist's attempt to initiate play is rebuffed by a peer (e.g., the child protagonist is told he cannot join the lunch group). The peer provocation vignettes depicted a negative event, in which the child's activity is disrupted by a peer (e.g., a peer knocks over the child protagonist's blocks). Each type of story also systematically varied according to the type of social cues present in the vignette portraying the peer's intentions when the negative event was enacted. Hostile intention vignettes (6) depicted situations in which the accompanying social cues signaled clear hostile intentions (e.g., another child says, "We don't want you here" in a sneering tone of voice). Benign intention vignettes (6) depicted situations in which the accompanying social cues signaled benign intentions even though a negative event occurred. For example, another child says that she needs to fix a mistake in her writing on the blackboard. While erasing her writing, she erases part of the protagonist's writing, and then says "Oops." Ambiguous intention vignettes(6) depicted situations in which the accompanying social cues were neutral, and therefore, left the intentions of the peer unclear. For example, a child shrugs when another child asks if she can join their computer group. Vignettes were enacted by Caucasian and African-American child actors of both genders who were in the elementary school age range.
Each child was presented with all 18 video vignettes and asked to imagine being the protagonist (i.e., the child with the numbered shirt). After the child watched each vignette, the interviewer asked a series of questions designed to assess social perception and strategy generation processes as follows.
Encoding. To assess encoding, the interviewer asked the child, "What happened in the story?" If his or her response provided incomplete information, the follow-up question, "Can you tell me anything else that happened?" was added. Responses were scored for evidence of whether the child encoded the social conflict (i.e., the negative event involving peer entry or peer provocation). In the six benign intention stories, the child's responses were also scored as to whether the child encoded benign intention cues. These were verbal and nonverbal social cues indicating that the child actor who caused the negative event to happen had benign intentions, despite the negative outcome. Therefore, these cues presented a contrast with the negative event in the story. Examples of benign intention cues were statements such as, "Oops, I didn't see you there!," nonhostile body language, or tone of voice. Responses were coded for evidence of encoding at least one benign intention cue. Coders utilized a list of benign intention cues compiled by Leffert and Siperstein (1996).
In the present study, we wanted to maximize children's opportunity to encode social cues regarding the nature of the social conflict and other child's intentions so that we could subsequently assess their ability to interpret social cues and generate social strategies. Consequently, those children who demonstrated encoding difficulties after viewing the vignette once were given the opportunity to view the vignette a second time. They were then queried once again regarding what happened. Whether the child demonstrated improved ability to encode the story conflict or benign intention cues after a second viewing was noted.
Cue Interpretation. To assess cue interpretation, we then asked the child, "Was the other kid in the story being mean or not being mean?" Responses were coded for hostile or benign interpretation of intentions. The child was also asked a follow-up question, "How can you tell that the kid was being mean/not mean?" to assess his or her reasoning for attributing "mean" versus "not mean" intentions. Finally, in the benign intention vignettes, responses were coded for spontaneous evidence of "discounting" benign cues. These were statements such as "He said, `Oops!,' but he didn't really mean it," which indicated that the child saw and understood the benign cue(s), but didn't believe them.
Strategy Generation. To assess strategy generation, we asked the child a final question, "what would you do if this happened to you?" Responses were coded according to one of five content categories: (1) aggressive, (2) nonaggressive assertive, (3) accommodating, (4) appeal to authority, and (5) avoidant, plus "other." Only the first response that the child generated was scored. This coding system was adapted from those previously used by Carlson (1985) and Leffert and Siperstein (1996).
For children with MR, we were able to utilize available data consisting of children's composite scores from individual administration of the Test of Auditory Comprehension of Language--Revised (TACL-R; Carrow-Woodfolk, 1985) to examine the relationship between social-cognitive processing abilities that involved language comprehension and these children's language comprehension abilities.
Children with and without MR were seen individually in a private setting outside of the classroom for two sessions, each lasting approximately 25 min. Children were presented with the 18 vignettes on a color television monitor, and responded to the interviewer's questions. The interviews were tape-recorded and responses were coded from the audio-recordings.
Children's audiotaped responses were coded by two coders, who were blind as to the age and educational status of the child. To assess intercoder agreement, kappa coefficients were calculated separately for each social-cognitive variable, based on 31 interview protocols. Kappa coefficients ranged from .72 to .94 with an average of .85.
Out of the 14 children with MR in Grades 1 and 2 who participated in the social-cognitive assessment, 3 children, all 7 years of age, appeared to have difficulty performing the task, (i.e., attending to the video vignettes as well as responding to the questions that followed each vignette). Specifically, they required numerous prompts to redirect them to the task and responded to the structured interview with comments that were unrelated to the video vignette they were shown. Therefore, we decided not to include their responses in the analysis of data.
In our analysis, we compared the social perception and strategy generation abilities of children with and without MR and, in particular, their ability to take into account the social conflict type (peer entry or peer provocation) and the intention cues (hostile, ambiguous, or benign) presented in the vignettes. The comparisons involved a series of ANOVAs in which group (children with MR, children without MR) and grade (three levels consisting of Grades 1 and 2, Grade 3, and Grades 4 and 5) were between-subjects variables. (We combined Grade 1 with Grade 2 and Grade 4 with Grade 5 because of the smaller number of participants from Grades 1 and 5.) Depending on the specific analysis, Intention Cue (hostile, ambiguous, or benign) or Social Conflict Type (peer entry, peer provocation) served as within-subjects variables.
Analysis of Social Perception
Our analysis of children's performance of social perception processes covered both major aspects of social perception, encoding, and cue interpretation. For encoding, we examined children's ability to encode the social conflict depicted in each vignette as well as their ability to encode the benign intention cues that were present in six of the vignettes. For cue interpretation, we examined children's interpretation ("mean" or "not mean") of the hostile, ambiguous, or benign intention social cues that were present in the vignettes.
Encoding of the Story Conflict. For the analysis of children's encoding of the story conflict, we conducted a Group (2) x Grade (13) x Social Conflict Type (2) analysis of variance. The results revealed a main effect for Group, F (1, 108) = 13.29, p [is less than] .001, which indicated that children with MR were less successful (91% frequency) at encoding the social conflict depicted in the story than children without MR (96% frequency). It is important to note that when given the opportunity to see the video a second time, the percentage of children in the two groups who correctly encoded the conflict in the story rose to 98% for children with MR and 100% for children without MR.
Since a few children sometimes failed to encode the social conflict, in the subsequent analyses we only utilized children's responses to stories in which the social conflict was correctly encoded. Thus, for each of the remaining measures, a child's score consisted of a percentage based on the total number of stories depicting conflicts that the child correctly encoded. We did this so as not to include, for example, a child's interpretation of intent or a child's generation of a social strategy given in response to a vignette in which the child failed to encode the basic story conflict.
Encoding of Benign Intention Social Cues. In examining children's encoding abilities, we next focused on their ability to encode benign intention cues. In this analysis, we focused only on the six stories in which benign intention cues were present because these stories presented children with a contrast between a negative event (i.e., a peer rebuffs a child's attempt to enter an ongoing activity or a peer's action disrupts a child's activity) and cues which indicated that the peer had benign intentions. Similar to the previous analysis, we again employed a Group (2) x Grade (3) x Social Conflict Type (2) analysis of variance. The proportion of times children correctly encoded the benign intention cues was the dependent variable.
The analysis showed a main effect for Group, F (1, 104) = 23.93, p [is less than] .001. Children with MR had greater difficulty encoding benign intention cues than children without MR (mean percentages = 65% and 85%, respectively). In addition, there was a main effect for Social Conflict Type, F (1,104) = 7.30, p [is less than] .01, which showed that children, regardless of group, had greater difficulty encoding benign intention cues in peer entry conflicts (mean frequency = 70%) than in peer provocation conflicts (mean frequency = 79%).
For the children with MR, the ability to encode the conflict and to encode benign intention cues were not significantly related to their IQ, r = .10 and .14, respectively, or language comprehension ability (i.e., TACL-R composite score), r = .26 and .20, respectively.
Overall, children with MR had difficulty encoding the benign intention cues that were embedded in the vignettes. Further, children in both groups were less likely to successfully encode benign intention cues when the conflicts involved peer entry than when the conflicts involved peer provocation.
Interpretation of Intention Social Cues. Next, we analyzed children's interpretation ("mean" vs. "not mean") of the benign intention social cues embedded in the vignettes in contrast to their interpretation of ambiguous and hostile intention social cues. In this analysis, unlike the previous analyses, we performed a Group (2) x Grade (3) x Social Conflict Type (2) x Intention Cue (3) analysis of variance, in which we included Intention Cue (hostile, ambiguous, or benign) as a second within-subjects variable. The number of times the child arrived at a "mean" interpretation of intent in the vignettes served as the dependent measure. (Note: Analysis of "not mean" interpretations would have provided us with the mirror image of the following results.)
Overall, there was a main effect for Intention Cue, F (2, 102) = 148.90, p [is less than] .001. As expected, children in both groups attributed "mean" intentions to the other child most often in the vignettes which contained hostile intention cues (mean frequency = 98%), somewhat less often in the vignettes which contained ambiguous intention cues (mean frequency = 64%), and least often in the vignettes which contained benign intention cues (mean frequency = 46%). Thus, both groups were able to distinguish hostile from benign cues in interpreting the other child's intent.
It is important to point out that while children with and without MR were able to distinguish among the different types of intention cues, they still incorrectly interpreted benign intention cues as "mean" about half of the time. To help explain this finding, we found a significant Group x Social Conflict Type x Intention Cue interaction, F (2, 102) = 4.87, p [is less than] .01, that showed that it was children with MR who interpreted benign intention cues as "mean" about half the time in the peer entry conflicts, while children without MR interpreted benign intention cues as "mean" only about a third of the time. In the peer provocation conflicts, however, both groups interpreted benign intentions as "mean" about half of the time (see Table 1).
TABLE 1 Children's "Mean" Interpretation of Benign Intention Cues by Group
Type of Social Conflict Group Peer Entry Peer Provocation Children with MR 48 47 Children without MR 37 53
Note: Numbers represent mean percentage frequencies.
While children with MR misinterpreted benign intention cues in both the peer entry and the peer provocation conflicts, children without MR misinterpreted benign intention cues mostly in the peer provocation conflicts. The reason for this was that a number of children without MR, when asked to interpret benign intention social cues, spontaneously reported that they perceived the benign intention cues in the peer provocation conflicts but did not believe them. Consequently, they gave a "mean" interpretation of the other child's intent. This "discounting" of benign intention cues was not evident among children with MR. Specifically, children without MR showed evidence of discounting benign cues 25% of the time while children with MR discounted the benign cues only 7% of the time.
Thus, in the peer entry context, children with MR arrived at "mean" interpretations of intent in the presence of benign intention cues more frequently than children without MR. In the peer provocation context (in which the children without MR often discounted benign intention cues), both groups had a similar frequency of "mean" interpretations.
In addition to the interaction involving differences between children with and without MR, the analysis of interpretation of cues indicated a significant Grade x Social Conflict Type interaction, F (2,103) = 8.12, p [is less than] .001, which was, as would be expected, further qualified by a three-way interaction involving Grade x Social Conflict Type x Intention Cue, F (4, 206) = 3.86, p [is less than] .01. This three-way interaction indicated that in peer entry conflicts, younger children with and without MR (Grades 1 and 2) were significantly more likely to interpret benign intention cues as "mean" than older children, in Grades 3, 4 and 5 (see Table 2).
TABLE 2 Children's "Mean" Interpretation of Benign Intention Cues by Grade
Type of Social Conflict Grade Level Peer Entry Peer Provocation Grades 1 & 2 64 46 Grades 3 42 55 Grades 4 & 5 31 49
Note: Numbers represent mean percentage frequencies.
For children with MR, the frequency with which children interpreted the other child's intentions as "mean" was unrelated to IQ or to language comprehension ability, as indicated by separate correlational analyses carried out for the vignettes which contained hostile, ambiguous, and benign intention cues. Pearson's r values ranged from -.03 to -.18, ns, for the relationship with IQ and from -.06 to -.21, ns for the relationship with language comprehension ability.
Overall, both children with and without MR correctly interpreted hostile intention cues as showing "mean" intent. However, when benign intention cues were present, children with MR were more apt to still arrive at a mean interpretation, particularly in peer entry conflicts. Lastly, younger children were more apt to misinterpret benign intention cues as "mean" than older children in peer entry conflicts.
Analysis of Generation of Strategies
For the analysis of the social-cognitive process of generation of strategies, we conducted a Group (2) x Grade (3) x Story Conflict Type (2) multivariate analysis of variance in which the, five strategy types (aggressive, assertive, accommodating, appeal to authority, and avoidant) were the dependent variables. The results indicated significant global main effects for Group, F (5, 104) = 10.33, p [is less than] .001, and Grade, F (10, 210) = 2.83, p [is less than] .01. Univariate tests for the Group effect indicated that regardless of the story conflict type, children with MR generated more aggressive strategies and appeal to authority strategies and fewer assertive, accommodating, and avoidant strategies than children without MR (see Table 3). [F (1, 108) values ranged from 5.08, p [is less than] .03 to 19.97, [is less than] p .001.] Univariate tests for the Grade effect indicated that children in Grades 1 and 2 generated fewer assertive strategies (mean frequency = 11%) than children in Grades 3 (mean frequency = 26%) and 4 and 5 (mean frequency = 21%), F (2, 108) = 8.63, p [is less than] .001.]
TABLE 3 Children's Generation of Social Strategies
Type of Strategy Group Aggressive Assertive Accommodating Children with MR 17 13 13 Children without MR 9 25 17 Type of Strategy Appeal to Avoidant Group Authority Children with MR 27 23 Children without MR 14 32
Note: Numbers represent percentage of total strategies generated.
The Group and Grade global main effects were qualified by a significant Group x Grade interaction, F (2, 108) = 5.09 p [is less than] .01. Univariate tests indicated that there was no difference in the use of the assertive strategies among younger children in either group (mean frequency = 8% and 12% for children with and without MR, respectively). However, among older children in Grades 3, 4, and 5, children without MR generated a higher frequency of assertive strategies (32%) than children with MR (15%), F (2, 108) = 5.09, p [is less than] .01. Thus, children without MR are primarily responsible for the higher frequency of assertive strategies which occurred among older children.
The analysis of children's strategy generation abilities also showed a significant main effect for Social Conflict Type, F (5, 104) = 41.79, p [is less than] .001, which was qualified by a significant interaction between Group and Social Conflict Type, F (5, 104) = 2.74. p [is less than] .02. This showed that children with MR had difficulty varying their strategy when presented with different story conflicts. For example, with regard to assertive strategies (see Table 4), children without MR significantly increased their use of this strategy type when the social conflict shifted from peer entry to peer provocation. By contrast, children with MR did not employ assertive strategies in either the peer entry or peer provocation conflicts, F (1, 108) = 10.24, p [is less than] .01. With regard to avoidant strategies (see Table 5), children without MR significantly increased their use of this strategy type when the social conflict shifted from peer provocation to peer entry, while children with MR increased their use of avoidant strategies only slightly, F (1, 108) = 10.24, p [is less than] .01.
TABLE 4 Children's Generation of Assertive Strategies
Type of Social Conflict Group Peer Entry Peer Provocation Children with MR 6 18 Children without MR 18 34
Note: Numbers represent percentage of total strategies generated.
TABLE 5 Children's Generation of Avoidant Strategies
Type of Social Conflict Group Peer Entry Peer Provocation Children with MR 33 12 Children without MR 49 15
Note: Numbers represent percentage of total strategies generated.
Overall, children with MR generated a higher rate of appeal to authority and aggressive strategies and a lower rate of avoidant, assertive, and accommodating strategies than children without MR. Furthermore, children with MR showed less ability than children without MR to differentiate between peer entry and peer provocation conflicts in the strategies they generated.
Relationship Between Interpretation of Intention Social Cues and Generation of Strategies
After analyzing children's social perception and strategy generation abilities, we next examined the connection between these two processes. Specifically, this analysis focused on the question of whether children's interpretation of the intentions of the peer in the story influenced the type of strategy they generated. To address this question, we compared the strategies children generated when they perceived "mean" intent with the strategies they generated when they perceived "not mean" intent. In a previous analysis, one of our independent variables was the type of intention cues presented in the different vignettes. In this analysis, our emphasis was not on portrayed intentions, but rather on the children's perceived intentions. Therefore, we limited this analysis to the six stories which contained ambiguous intention cues.
We performed a Group (2) x Grade (3) x Perception of Intent ("mean," "not mean") multivariate analysis of variance in which the five strategy types were the dependent variables. As expected from the previous analysis, there was a significant main effect for Group, F (5, 84) = 5.62, p [is less than] .001. More importantly, there was a significant main effect for Perception of Intent, F (5, 84) = 28.90, p [is less than] .001, qualified by significant interactions between Group and Perception of Intent, F (5, 84) = 2.43. p [is less than] .05, Grade and Perception of Intent, F (10, 170) = 2.35. p [is less than] .01, and, further, by a significant three-way interaction between Group, Grade, and Perception of Intent, F (10, 170) = 2.56, p [is less than] .01.
Univariate tests for the important Perception of Intent main effect indicated that children both with and without MR were more apt to generate aggressive, appeal to authority, and avoidant strategies when they interpreted "mean" intent and accommodation strategies when they interpreted "not mean" intent. [F (1, 88) values ranged from 17.66 to 103.12, p [is greater than] .001]. The difference in the generation of avoidant strategies was further qualified by a Group x Perception of Intent interaction indicating that children without MR were more apt to generate avoidant strategies when they interpreted "mean" intent (33%) than when they interpreted "not mean" intent (7%), F (1, 88) = 4.69, p [is less than] .01. This pattern of use of avoidant strategies was less obvious among children with MR (26% when they interpreted "mean" intent as opposed to 17% when they interpreted "not mean" intent). The univariate tests for the Group x Grade x Perception of Intent interaction, which was significant for assertive strategies, F (2, 88) = 3.72. p [is less than] .05, were not able to be interpreted given the complexity of the mean differences.
Overall, children with and without MR generated different strategies based on whether they interpreted the intentions of the peer in the vignette as "mean" or "not mean." In particular, they generated appeal to authority and avoidant strategies when they interpreted "mean" intentions and accommodation strategies when they interpreted "not mean" intentions.
In the present study, we investigated the capacity of children with MR to meet the challenges of successful social adaptation from a social-cognitive perspective. Specifically, we examined the ability of these children, in light of their cognitive processing difficulties, to perform two critical social-cognitive processes--social perception and strategy generation. The results indicated that children with MR have difficulties in social perception, particularly in recognizing and interpreting social cues about classmates' intentions. They were also less likely than children without MR to alter their social strategy in response to different social problems. Instead, they were more likely to look to the teacher for help in resolving the conflict. By further documenting the nature of the social-cognitive processing difficulties experienced by children with MR, the results focus attention on the role that cognition plays in these children's ability to socially adapt to their surroundings. Below, we will highlight the key findings and place them with the context of existing cognitive and social-cognitive theories that might help to explain why children with MR are having these difficulties.
Children with MR had social perception difficulties when a negative event (e.g., a peer says that they cannot join a game or erases part of their writing on the blackboard) was accompanied by social cues that signaled benign intentions. These situations, which are common in school settings, presented the child with a discrepancy between the outcome of the event and the other child's intentions. The response of children with MR to this discrepant information is that they frequently arrived at a "mean" interpretation, even though benign intention cues were present.
How do we explain the difficulties that children with MR had interpreting benign intentions? One possible explanation is that children with MR lacked understanding of the meaning of benign intention cues. Another possible explanation is that the children with MR may have understood the meaning of social cues indicating benign intentions, but failed, in the context of a negative event, to utilize these cues in making a decision about another person's intentions. There is some suggestive evidence in the present study for this second explanation in that the children with MR were nearly always aware of the negative event, but often failed to report noticing the benign intention cues. By overfocusing on the negative event to the exclusion of the social cues indicating benign intentions, children with MR missed key information about the social problem which could have led them to correctly conclude that another child had "not mean" intentions.
The failure of children with MR to utilize benign social cues in interpreting intentions was present in both the peer entry and peer provocation social conflicts. In contrast to the peer entry conflicts, when the conflict involved peer provocation, children with and without MR were equally likely to conclude that the other child was "being mean" despite the presence of benign intention social cues. However, unlike the children without MR, who reported the benign social cues but "discounted" them, saying that they doubted the other child's sincerity, the children with MR often failed to report the benign intention cues and usually explained their negative interpretation of intentions by citing the negative event. Overall, unlike children without MR, who reconciled conflicting information derived from multiple cues when interpreting intentions, children with MR overfocused on the negative event, without taking into consideration the available benign social cues. In interpreting the other child's motivations, they did not seem to recognize that a child's mental state, (i.e., intentions) could be distinct from the outcome. Rather, they seemed to draw their conclusions about intentions primarily from the negative event.
We suggest that this problem in the social perception abilities of children with MR reflects a developmental immaturity in these children's cognitive processing abilities, which Feffer (1959; Feffer & Suchotliff, 1966) described 40 years ago as problems in simultaneous cognitive decentering. Feffer extended Piaget's work by elaborating on the implications of Piaget's theories for the study of children's social development. By applying the concept of simultaneous cognitive decentering to MR, we build a bridge between the processes by which children develop understanding of the physical world (Piaget, 1950) to their understanding of the social world.
According to Piaget (1950), cognitive processes reflect greater maturity and adaptiveness to the degree that immediate sense impressions are subordinated to thought in organizing experience. A sign of growing cognitive maturity is the emergence of the ability to "decenter," (i.e., to focus on multiple aspects of the perceptual field as opposed to "centering" on only one). The development of the ability to ¢er simultaneously is particularly important because it enables the individual to consider a number of aspects of a situation in relation to each other at the same time, thus reducing or eliminating the perceptual distortions that result from focusing exclusively on any one aspect.
It is precisely this ability to decenter cognitively during social encounters that we believe is difficult for children with MR. In focusing exclusively on the negative effect of the other child's actions and ignoring cues regarding intentions, these children show a "one-sidedness," in which they focus on one type of information to the exclusion of the other.
Furthermore, it would appear to be particularly challenging for children to decenter when trying to understand their social world (as opposed to the physical world) because of an imbalance that exists between the two conflicting dimensions, outcome and intentions, that children are called upon to integrate. Children with MR are likely to focus on the negative outcome (e.g., the rebuff of an entry attempt or disruption of their activity), because it is highly concrete in its impact on the child. At the same time, they are not as likely to focus on social cues suggesting benign intentions, which are less concrete in their impact while demanding more from the child in terms of having to draw inferences from them. The imbalance between these conflicting dimensions of the social situation increases the likelihood that children with MR will remain "centered" on the negative outcome of another child's behavior and not take into consideration social cues that reveal the child's intentions. The resulting misinterpretation of the situation will make it difficult for the child to generate effective social strategies.
Interestingly, we found that in those relatively few instances in which children with MR did conclude that another child was "not being mean," despite the occurrence of a negative event, they generated a much higher rate of accommodation strategies and a lower rate of appeal to authority strategies than when they reached the opposite conclusion. This finding is consistent with Feffer's insight that more socially mature strategies arise when a child's social perception processes integrate multiple dimensions of the interpersonal field. It is also an intriguing finding because it provides suggestive evidence that children with MR can adjust their strategies in accordance with their interpretation of intentions.
In addition to their limitations in social perception, we found that children with MR had problems generating social strategies which fit the different social conflicts. Unlike children without MR, who recognized that different social problems call for different strategies, children with MR did not show a clear-cut preferred strategy for either the peer entry or peer provocation conflict. Instead, they appeared to rely on more "general strategy preferences," regardless of the nature of the conflict.
As was the case with social perception, we believe that difficulties in strategy generation among children with MR reflect specific problems in their cognitive processing, in this instance in the performance of executive functions, (i.e., processes that help children to decide the nature of the problem and determine which strategy from their existing repertoire is most appropriate [Gallagher, 1984; Sternberg, 1987]). Executive functions have been considered a critical aspect of Sternberg's triarchical theory of intelligence and are represented in his concept of "metacomponents" which include "higher order executive processes which are used to plan what one is going to do" (Sternberg, p. 141).
When we look at the strategies that children generated for resolving social conflicts, the strategy of appealing to authority stands out, for two reasons. First, while appealing to authority was seldom mentioned by children without mental retardation, it was the strategy that was most often suggested by children with MR, regardless of the type of social conflict. Second, one reason that this strategy may have been so popular among children with MR is that it has the benefit of allowing children with MR to compensate for their social-cognitive limitations by relying on the adult's superior social perception and strategy generation skills in "reading" the social situation and guiding them to an appropriate response. From a social-motivational point of view, appealing to authority may also be attractive to children with MR because these children feel a lack of power to influence others to respect their wishes in social conflicts unless they can enlist a more powerful ally to intervene on their behalf.
Although at times the strategy of appealing to authority may pay off for the child with MR, this strategy nonetheless has significant drawbacks. First and foremost, children with MR may not come away from the experience with improved skills, in social perception or in matching strategies to situations, which they can apply when future conflicts arise. Second, there is the risk that children with MR will overgeneralize their use of this strategy by appealing to authority even when another strategy better fits the particular social context. Finally, there is reason to suspect that relying on appeals to the teacher as a preferred strategy for solving problems may place children with MR at a social disadvantage, particularly given the expectation within general education settings that children, as they get older, demonstrate increasing independence in working out solutions to social problems with peers.
An interesting finding of the present study was that for children with as well as without MR, the interpretation of a child's intentions appeared to influence their selection of strategies. This result provides suggestive and, indeed, intriguing evidence that children with MR can adjust their strategies to fit their interpretation of intentions. It is important, however, to interpret these results cautiously, because they represent a relatively small number of children's responses, i.e., to the six ambiguous vignettes. Moreover, only about one-third of these responses involved a "not mean" interpretation of intentions.
While the results of the present study highlight problems in social-cognitive processing in children with MR that, we believe, impact on these children's social adaptation, these results also point to the need for future investigations. Specifically, we need to go beyond the present findings, which indicate that children with MR have difficulty performing social-cognitive processes, to examine further the linkages between these children's social-cognitive limitations and their actual social adjustment, (i.e., social behavior and relationships). While there is already suggestive evidence from past studies linking social perception and strategy generation difficulties among children with MR to maladaptive patterns of social behavior (Leffert & Siperstein, 1996) and the generation of developmentally appropriate social strategies by these children to social acceptance by the peers in general education classrooms (Siperstein & Leffert, 1997), further investigation of these connections is needed. In particular, we need to examine whether those children with MR who demonstrate greater proficiency in social-cognitive processes have better social outcomes.
IMPLICATIONS FOR PRACTICE
The present study's findings highlight the importance of social-cognitive processes in enabling children with MR to meet the challenges of successful social adaptation in their school and classroom. Although the participation of children with MR in general education settings is on the rise, we contend that these children are not likely to improve their social-cognitive processing abilities solely as a consequence of inclusion in general educational settings. Rather, as is the case with other skill deficits, we believe that specialized instruction needs to be provided to children with MR in the area of social-cognitive processes. This belief is supported by studies which confirm that when children with MR participate in the general education classroom, they experience significant problems of social rejection by peers (Marwell, 1990; Sale & Carey, 1995) and show different patterns of behavior than their classmates (Roberts, Pratt, & Leach, 1991; Taylor, Asher, & Williams, 1987).
What kinds of intervention and instruction are called for? To engage in successful social adaptation, children with MR must continuously "read" social cues in real-time so that they can make sense of their complex, ever-shifting social environment. Moreover, they must utilize the knowledge that they gain from these social cues to generate social strategies that best fit a particular situation. Both general and special education teachers can help equip children with MR to meet the social challenges of daily life at school by providing instruction that focuses not only on discrete observable social behaviors, as many social skills curricula do, but also on stimulating the development of underlying social-cognitive skills.
Teachers can help students develop social perception skills. That makes it possible for a student to evaluate, for example, whether a classmate deliberately or accidently knocked his book off the desk or if classmates would not let him join their activity out of spite or because they were already in the middle of an activity. In teaching social perception skills, teachers need to identify and target those aspects of social perception that a given child needs help with.
Teaching students to recognize and interpret social cues regarding intentions is a case in point. First, students with MR may need instruction in recognizing social cues that signal different intentions. These can include facial expressions, tone of voice, body posture, verbal statements, physical actions, and other visual details. For example, the student may not notice that after the classmate knocked the book off the desk, he said "Oops!," that the classmate did not have a hostile facial expression, or that the edge of the desk where the book had been resting was situated right by an open doorway.
Second, in teaching students to interpret social cues, instruction should focus on helping students to learn the basic distinction that exists between a person's intentions--what he ox she wants to happen--and the effect of the person's action. This will allow students to interpret social conflict situations by utilizing available information about intentions as well as information about outcomes. If students are to arrive at accurate interpretations, they need to learn to search for and synthesize information from all available social cues, rather than focusing exclusively on a single piece of information. An important part of the process of teaching students to distinguish between outcomes and intentions is providing them with instruction and practice in accurately applying concepts such as "mean" and "not mean" and "accident" and "on purpose" when reasoning about intentions. For example, children need to learn that when things happen by accident, no one was being mean (although someone may have been careless).
In addition to teaching social perception skills, teachers can also promote the socially adaptive behavior in children with MR by working with them on the social-cognitive process of strategy generation. In particular, many students with MR need to improve the process by which they select strategies. For example, students need to learn that one strategy is not the "right" strategy for all situations. Teachers can help students generate situationally-appropriate strategies by teaching them to utilize available social cues, for example, regarding others' intentions and the nature of the social conflict. Teachers can help stimulate this process by directing the student's attention to relevant social cues when working with them on generating strategies, and by encouraging them to think out loud about the likely consequences within the specific social conflict situation for the different social strategies that they generate.
Overall, the present study highlights the challenges that day-to-day interpersonal interactions pose for children with MR. Moreover, the results point to the need for instruction in social-cognitive skills such as social perception and strategy generation to prepare these students to successfully adapt to a dynamic and increasingly diverse social environment. Because of their cognitive limitations, some students with MR may continue to require a high degree of on-the-spot direction from teachers or peers regarding how to interpret or respond to specific social conflicts. We believe, however, that many students can benefit from instruction that provides them with guided practice in social-cognitive skills. By focusing instruction on processes such as social perception and strategy generation, and in doing so, addressing the link between these children's cognitive limitations and their social functioning, teachers may help children with MR to adapt more successfully to the varied social situations that they will encounter.
Carlson, C. (1985). Social interaction goals and strategies of children with learning disabilities. Journal of Learning Disabilities, 20, 306-311.
Carrow-Woodfolk, E. (1985). TACLl-R examiner's manual: Revised edition, Allen, TX: DLM Teaching Resources.(*)
Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social information processing mechanisms in children's social adjustment. Psychological Bulletin, 115, 74-101.
Dodge, K., Bates, J., & Pettit, G. (1990). Mechanisms in the cycle of violence. American Association for the Advancement of Science, 250, 1678-1683.
Elias, M. J., & Tobias, S. E. (1996). Social problem-solving: Interventions in the Schools. New York: Guilford.(*)
Feffer, M. (1959). The cognitive implications of role-taking behavior. Journal of Personality, 27, 152-168.
Feffer, M., & Suchotliff, L. (1966). Decentering implications of social interaction. Journal of Personality and Social Psychology, 4, 415-422.
Gallagher, J. (1984). The search for developmental models for understanding mental retardation. In P. H. Brooks, R. Sperber, & C. McCauley (Eds.), Learning and cognition in the mentally retarded (pp. 39-53). Hillsdale, NJ: Erlbaum.(*)
Goldfried, M. R., & D'Zurilla, T. J. (1969). A behavior-analytic model for assessing competence. In C.D. Spielberger (Ed.), Current topics in clinical and community psychology (pp. 151-196). New York: Academic Press.(*)
Gresham, F. M. (1986). Strategies for enhancing the social outcomes of mainstreaming: A necessary ingredient for success. In C. J. Meisel (Ed.), Mainstreaming handicapped children: Outcomes, controversies and new directions (pp. 193-218). Hillsdale, NJ: Erlbaum.(*)
Hughes, C. (1999). Identifying critical social interaction behaviors among high school students with and without disabilities. Behavior Modification, 23(1), 41-60.
Kohlberg, L. (1969). Stage and sequence: The cognitive-developmental approach to socialization. In D. A. Goslin (Ed.), Handbook of socialization theory and research (pp. 347-480). New York: Rand McNally.(*)
Kusche, C. A. & Greenberg, M. T. (1994). The PATHS Curriculum. Seattle: Developmental Research and Programs.(*)
Leffert, J. S., & Siperstein, G. N. (1996). Assessment of social cognitive processes in children with mental retardation. American Journal on Mental Retardation, 100, 441-445.
Maheady, L., Harper, G. F., & Sainato, D. M. (1987). Social perception deficits: Have we misinterpreted the data? Journal of Research and Development in Education, 20(3), 27-35.
Maheady, L., Maitland, G., & Sainato, D. M. (1984). The interpretation of social interactions by mildly handicapped and nondisabled children. The Journal of Special Education, 18, 151-159.
Marwell, B. E. (1990, November). Integration of students with mental retardation: Summary evaluation report. Report issued by the Madison, Wisconsin Metropolitan School District. (ERIC Document Reproduction Service No. ED 355 727).
Piaget, J. (1950). The psychology of intelligence. New York: Harcourt, Brace.(*)
Roberts, C., Pratt, C., & Leach, D. (1991). Classroom and playground interaction of students with and without disabilities. Exceptional Children, 57, 212-224.
Sale, P., & Carey, D. M. (1995). The sociometric status of students with disabilities in a full-inclusion school. Exceptional Children, 62, 6-19.
Shantz, C. U. (1983). Social cognition. In J. H. Flavell & E. M. Markham (Eds.), Handbook of child psychology: Vol 3. Cognitive development (pp. 629-649). New York: Wiley.(*)
Siperstein, G. N., & Leffert, J. S. (1997). Comparison of socially accepted and rejected children with mental retardation. American Journal on Mental Retardation, 101, 339-351.
Smith, D. C. (1986). Interpersonal problem-solving skills of retarded and non-retarded children. Applied Research in Mental Retardation, 7, 431-442.
Spivack, G., & Shure, M. B. (1974). Social adjustment of young children: A cognitive approach to solving real-life problems. San Francisco: Jossey-Bass.(*)
Sternberg, R. J. (1987). A unified theory of intellectual exceptionality. In J. D. Day & J. G. Borkowski (Eds.), Intelligence and Exceptionality: New directions for theory, assessment, and instructional practices. Norwood, NJ: Ablex Publishing Corporation.(*)
Stokes, T. F., & Baer, D. M. (1977). An implicit technology of generalization. Journal of Applied Behavior Analysis, 10, 349-367.
Taylor, A. R., Asher, S. R., & Williams, G. A. (1987). The social adaptation of mainstreamed mildly retarded children. Child Development, 58, 1321-1334.
Weiss, B., Dodge, K., Bates, J., & Pettit, G. (1992). Some consequences of early harsh discipline: Child aggression and a maladaptive social information processing style. Child Development, 63, 1321-1335.
Wong, C-M. T., Day, J. D., Maxwell, S. E., & Meara, N. M. (1995). A multitrait-multimethod study of academic and social intelligence in college students. Journal of Educational Psychology, 87(1), 117-133.
(*) To order a book marked by an asterisk (*), please call 24 hrs/365 days: 1-800-BOOKS-NOW (266-5766) or (732) 728-1040; or visit them on the Web at http://www.BooksNow. com/ExceptionalChildren.htm. Use VISA, M/C, AMEX, Discover, or money order + $4.95 S&H ($2.50 each add'l item) to: Clicksmart, 400 Morris Avenue, Long Branch, NJ 07740; (732) 728-1040 or FAX (732) 728-7080.
This research was supported by a grant from the U.S. Department of Education (H023A40051).
Manuscript received September 1999; accepted March 2000.
JAMES S. LEFFERT GARY N. SIPERSTEIN The University of Massachusetts Boston
EMILY MILLIKAN University of Denver
JAMES S. LEFFERT, Senior Research Associate; and GARY N. SIPERSTEIN (MA Federation), Director, Center for Social Development and Education, The University of Massachusetts Boston. EMILY MILLIKAN, Doctoral Student, Department of Psychology, University of Denver, Colorado.
Correspondence should be sent to James Leffert at the Center for Social Development and Education, University of Massachusetts Boston, 100 Morrissey Boulevard, Boston, MA 02125-3393. E-mail: firstname.lastname@example.org
|Printer friendly Cite/link Email Feedback|
|Author:||LEFFERT, JAMES S.; SIPERSTEIN, GARY N.; MILLIKAN, EMILY|
|Date:||Jun 22, 2000|
|Previous Article:||Improving Graduation and Employment Outcomes of Students with Disabilities: Predictive Factors and Student Perspectives.|
|Next Article:||Mentoring Beginning Special Education Teachers and the Relationship to Attrition.|