Social factors underlying susceptibly to peer influence in
adolescents with mild-to-borderline intellectual disability
Student: Ayda Nakhostin Taghavi
Student number: 10988203
Study: Psychobiology, University of Amsterdam
Supervisor: Eline Wagemaker MSc
Assesor: Dr. Annemie Ploeger
Return date: 24-01-2020
Adolescents with Mild-to-Borderline Intellectual Disability (MBID) are more susceptible to peer influence compared to typically developing adolescents (Bexkens et al., 2019). This study aimed to find the underlying social factors that may contribute to the higher
susceptibility to peer influence in adolescents with MBID. Earlier research has found deficits in mentalizing in adolescents with MBID, also they seem to have higher interpretation bias compared to typically developing adolescents. Because these factors are crucial in their social information processing, this study investigates the mediating role of mentalizing and
interpretation bias in adolescents with MBID to peer influence. We studied 93 MBID and 39 control participants. The Balloon Analogue Risk-Task, in a solo and peer condition, was used to measure susceptibility to peer influence during risk-taking. The recognition task was used to measure interpretation bias and the Hinting Task was used to measure the ability to mentalize. The results showed that indeed adolescents with MBID had more deficits in mentalizing and that this was predictive for being more susceptible for peer influence. It was found that the ability to mentalize, mediates the susceptibility to peer influence in adolescents with MBID. However, it was not found that adolescents with MBID have a higher
interpretation bias than typically developing adolescents. Moreover, interpretation bias was not found to be predictive on risk-taking under peer influence. These findings suggest that training the ability to mentalize in adolescents with MBID might be an effective intervention to reduce susceptibility to peer influence.
Keywords: Adolescent risk-taking, Mentalizing, Interpretation bias, Mild-to-Borderline Intellectual disability, Peer influence
Peer relations are more prominent than ever in the period of adolescence: adolescents spend more time interacting with their peers, they appoint more priority to peer norms in behaviour and report to feel happier when they are with peers, compared with children or adults (Brown et al., 2009). In addition, it is found that adolescents are highly influenced by their peers in risk taking (Steinberg et al., 2008,). For example, aberrant peer relations are found to be the strongest predictor of delinquent behaviour for adolescents (Elliot &
Menerard, 1996). Delinquent behaviour seem to be more present in adolescents with learning disabilities, prevalence studies from the United Kingdom, Australia and the United States mainly found that these adolescents are overrepresented in the prison population
(Hayes et al, 2007). The focus of the present study is on adolescents with a mild-to-borderline intellectual disability (MBID). MBID is defined by an IQ between 50 and 85, significant limitations in adaptive functioning and particularly in social contexts (Schalock et al., 2007). Potentially, adolescents with MBID are more susceptible to peer influence compared to typically developing adolescents (Bexkens et al., 2018), which may explain their high crime rates. The current study will investigate deficits in adolescents with MBID that may increase this susceptibility to peer influence.
Earlier studies on susceptibility to peer influence in adolescents with MBID found higher risk taking under peer influence in adolescents with MBID compared to typically developing adolescents (Bexkens et al., 2018). This study investigated risk taking in male adolescents with MBID, adolescents with behavioural disorders and typically developing male adolescents. All adolescents were randomly assigned to either risk taking under peer influence or a solo risk-taking task. The results only showed higher risk taking under peer influence in adolescents with MBID compared to typically developing adolescents, and adolescents with behavioural disorders. Yet, the factors underlying this higher susceptibility to peer influence are not known. The current study aims to improve earlier research by
isolating two important social factors which may contribute to the higher susceptibility to peer influence in MBID: interpretation bias and their ability to mentalize.
The general peak in adolescence in susceptibility to peer influence may be attributed to the still developing ‘social brain’ (Blakemore & Mills, 2014). Brain areas that are associated with the ‘social brain’ are the anterior temporal cortex (ATC), posterior superior temporal sulcus (pSTS), prefrontal cortex (PFC), and the temporo-parietal junction (TPJ) (Fliessbach et al., 2007). These brain areas are still developing in adolescents. For example, the right ventrolateral PFC plays a critical role in reducing stress when people are coping with negative evaluations from peers. In general, it is found that adolescents experience more stress when they are excluded by their peers compared with adults or children (Gunther et al., 2010). Research shows that the right ventrolateral PFC is more active in adults compared with adolescents when they are socially excluded. Moreover, these adults show less physical distress during exclusion when using this brain area more actively (Sebestian et al., 2011). Another fMRI study found lower self-reports of distress by the participants when the ventrolateral PFC was used significantly more active during peer exclusion (Masten et al., 2009). Because this brain area is still developing in adolescence, adolescents may not be as effective at coping with distress during peer exclusion. The study of Falk et al., (2014) found that these neural responses to exclusion predict susceptibility to peer influence. From this neurological perspective it can be proposed that adolescents are more sensitive to peer influence by the potential larger immaturity of the social brain.
The social brain is involved in a lot of complex social cognitive skills. Social cognitive skills such as perspective taking and the interpretation of social situation are important factors to proper social information processing (SIP). Studies with typically developing children have shown that their SIP determine the extent in which adolescents show adaptive or problematic social behaviour (Dodge et al, 2003). SIP starts with encoding a social situation, it follows
with interpreting others and the situation. Then, own emotions are regulated, and it ends with generating a selected response. It is proposed that individual differences in the steps of SIP, will lead to different behavioural responses. Adolescents with MBID differ in their SIP compared with typically developing peers. It is suggested that this is because some parts of the SIP may require a certain level of intelligence (Van Giessen, 2016). Furthermore, it is known that important cognitive skills that predict SIP are found to be interpreting others and the ability to mentalize (Van Nieuwenhuijzen et al., 2004). However, it is not proven if these factors predict more susceptibility to peer influence under risk taking. In the following paragraphs, evidence for the link between negative interpretation bias, mentalizing, and susceptibility to peer influence is provided.
Negative interpretation bias is a social information-processing bias and is
characterized as the tendency to negatively interpret ambiguous stimuli, scenarios and social events are experienced as hostile or rejecting (Mathews & Macload, 2005). For example, in a situation in which one friend tells an adolescent he is too busy to meet, the adolescent can interpret this as their friend not liking him anymore. Adolescents with high interpretation bias have higher tendency to feeling rejected. As mentioned above, feeling excluded is highly correlated with susceptibility to peer influence (Falk et al., 2014). Moreover, sereval
longitudinal studies link peer rejection with increased risk-taking in adolescents in forms of substance use and deviant behavoir (Dishion et al., 1995, Prinstein and La Greca, 2004). So far, there is no study that have compared adolescents with and without MBID on their interpretation bias but suggest that adolescents with MBID have a higher interpretation bias. This is because social phobia is highly prevalent in adolescents with MBID and it is reported as one of the most common forms of social distress in this population (Emerson et al., 2003). Negative interpretation bias and social phobia are highly correlated too (Amin et al., 1998).
Therefore, it is expected that individuals with MBID might have higher interpretation bias and this may underlie their susceptibility to peer influence.
Mentalizing is the ability to reflect on mental states such as intentions, beliefs and desires of others (Wimmer & Perner, 1983). Given the relevance of being able to mentalize in social contexts, a study investigated this capacity in individuals with intellectual disability. This study found a strong correlation between IQ and mentalizing impairments (Schalock et al., 2011). Another study demonstrated that the ability to mentalize is significantlyimpaired in individuals with MBID (Benson et al, 1993; Baglio et al., 2016). Impairments in metalizing would make adolescents with MBID possibly more susceptible for peer influence, because they would be less able to read the intentions of their peers in risk contexts (Khemka et al. 2009). Therefore, it is expected that individuals with MBID have impaired mentalizing and this may underlie their susceptibility to peer influence.
The current study tries to improve earlier studies in three ways. First, the study of Bexkens (2019) had a between-subjects design: the adolescents were randomly assigned to a task that involved risk taking under peer influence or risk taking without peer influence. To control for preselected group differences, the current study will have a within-subjects design in which adolescents preform the risk-taking task with and without peer influence. Second, the study of Bexkens et al. (2019) only investigated susceptibility to peer influence in male adolescents with MBID. Noticeable is that research often indicates higher susceptibility to peer influence in boys compared to girls (Ngee Sim et al., 2003). But examples in the society such as traffickers, suggest that girls are suspectable to peer influence too. To better
generalize the results, the current study looks at both sexes and explores potential sex differences. Third, the study manipulated peer influence with audio fragments. This is not exemplary for real life peer influence. To make the manipulation more realistic, adolescents will interact with peers through a designed chat conversation. It is expected that the
adolescents experience more realistic peer influence through this manipulation. The measure for peer influence is therefore in this study more reliable.
The current study tests the following four hypotheses. First, it is expected that adolescents with MBID show higher susceptibility to peer influence compared to typically developing peers (Bexkens et al., 2019). Second, since evidence from earlier studies indicates higher prevalence of social phobia in adolescents with MBID (Emerson et al., 2003), it is expected that adolescents with MBID have a higher interpretation bias compared to typically developing adolescents. Third, lower mentalization is expected in adolescents with MBID compared to typically developing adolescents (Benson et al, 1993; Baglio et al., 2016). The fourth hypothesis is that higher interpretation bias and lower mentalization skills may predict higher susceptibility to peer influence in adolescents with MBID.
Materials and method Participants
A total of 133 adolescents with and without MBID between 14 and 19 years of age were tested. The inclusion criterion was an IQ between 50 and 85 for the MBID group, that were recruited from special education schools. Participants from the control group were recruited from regular schools, with an inclusion criterion of an IQ above 85. An informed consent was accessed from parents and the adolescents. The parents were called and asked for their consent a few weeks before testing. The participants were asked for their consent right before the test session began. It was explained to the participants that without any
consequences, they could withdraw from their participation whenever they wanted. The study was approved by the ethical review board of the University of Amsterdam.
Materials
IQ. To assess the IQ, the test session started with a shortened version of the Weschler Intelligence Scale for Children – V (WISC-V;Wechsler, 2018) or Weschler Adults
Intelligence Scale (WAIS-IV;Wechsler, 2014). Adolescents with ages between 6-16
performed the WISC-V test, and adolescents with an age above 17 performed the WAIS-IV test. Both tests were shortened by two subtests: performal IQ and verbal IQ. For the performal IQ, matrix-reasoning task was performed. In this subtest a matrix of abstract pictures was presented in which one picture was missing. Five possible options that could fit in the missing picture were presented. The participant had to choose the most fitting option, this tested the non-verbal problem solving ability of the participant. In total there were 32 items with 2 examples. In the examples the test leader told the participant the right answer with
explanation to make sure the adolescent understood the task. Verbal IQ was measured with the vocabulary subtest. This started with example pictures of objects. The participant had to
tell the test leader what was shown in the pictures. With a total of 4 examples and 29 verbal items for the WISC-V and a total of 30 words for the WAIS-IV test. The tests became increasingly more difficult. Therefore, adolescents with MBID started the task with the first item to keep the motivation high. The control group started at the entry-level which was item 9. All the tasks stopped when the participant answered wrong 3 times in a row. In the WISC-IV and WAIS-WISC-IV the dependent variable was the total IQ score. This was calculated by adjusting the raw score into a scaled score. The scaled score was adjusted through a script on SPSS into the IQ score. The validity of the shortened WISC-IV and WAIS-IV task is tested by Dutch Committee COTAN, both scored with a ‘good’. The reliability and validity were rated as ‘sufficient’ for both as well.
BART. To assess risk taking under peer influence the Balloon Analogue Risk Task (BART) (Lejuez et al., 2002) was used. Participants were sitting behind the screen of a notebook. Instructions were read out loud by the test leader and the participant could read along. Participants could earn money by pumping a balloon on the screen, with the risk of the balloon exploding and losing the earned money. The bigger the balloon got, the more money was earned. The task consisted of 2 blocks, with 15 balloons per block. Participants inflated the balloon by pressing the space bar. Each press was rewarded with one cent. The balloon could explode at every next click, this was randomized per balloon. If this happened, an explosion sound together with an exploded balloon was presented on the screen. All cents earned on that trial were then lost and the participant had to start with a new balloon. Participants could decide to sell the balloon by pressing the left Shift button when they wanted to. All points earned on that trial were then transferred to the bank, this came with a sound of a slot machine. There were two example blocks, before doing the actual task. In the first example block the participant was instructed to pump the balloon 3 times, each balloon exploded at different sizes. In the second example block the participant practise pumping and
selling the balloon. After the practice trials, the participant was asked if he had any questions. All the participants were able to perform the task.
The solo or peer version was counterbalanced. This was also done for the money that could be earned from one of the sessions. In the solo condition adolescents performed the BART as described above. In the peer condition the participant was told that another student from the same grade at another school was looking while he/she was performing the task. However, this was another researcher. A camera was placed behind the participant. This allowed the second researcher to watch. The participant was told that the camera was placed for the other adolescent to watch him or her perform the task. It was explained that the other adolescent had to guess, based on a conversation with the participant, how well he or she will perform the task. The participants first had to shortly introduce themselves through chat on a given research phone. The participant received a personal compliment first from the other adolescent. In addition, the writing style was adjusted to the style of a teenager. Moreover, a selfie from a collected database of adolescent selfies was sent to increase credibility. The second researcher had a chat protocol which mostly made every conversation the same. The names of the peers were randomly chosen from a standard name list and were matched with age, sex and a selfie. In each conversation the participant knew what age the peer was, what the peer enjoyed to do in their free time and what he or she looked like. The peer stimulated the participant two times through chat to pump more. First after the example trials, by saying: “You don’t pump enough. I would pump more if I were you”. And also after the first trial a message was sent to stimulate more pumping. Afterwards an questionnaire was taken to see if the participant believed the manipulation. Both conditions, the solo and peer, were done by the participant. To correct for the chat time, a neutral video through the same research phone was shown in the solo condition before the first trial and after one trial.
The average number of pumps on not-exploded balloon trials, was used as dependent variable (from now adjusted pumps). The score of the adjusted pumps was calculated by subtracting the BART peer from BART solo. This was used as a measure for susceptibility to peer influence. The validity of the BART task is proven by Hunt et al., (2005). This study found that higher self-reported psychopathy was a significant predictor of increased risk taking on the BART.
Hinting Task. The ability to mentalize was assed with the Hinting Task (Corcoran et al., 1995). The task is designed to test the ability of the adolescents to infer the real intentions behind indirect hints. There were 10 short stories, always involving two characters. All scenarios ended with one of the characters giving a hint to the other. The participant was asked what the character really meant when he/she said this. An appropriate response was given a score of two, and the next story was read out. If the participant fails to give the correct answer by, for example, simply paraphrasing the 'hint', another sentence is added to the story in the shape of a more obvious hint. The participant was then asked what the character wants the other to do. If the response was correct, a score of one was given. If the participant failed again to infer the intended meaning of the character, a score of zero was given for that item. All items on the task were read out loud to the participant. A sheet containing sketches of all the scenario was placed in front of the participant to read along. The test leader held a paper in front of the rest of the story and revealed the next sentence to make sure the participant could read along, and did not get distracted by the rest of the story. The task was adapted by
changing the names of the characters to more universal names. An example of a scenario is shown in the appendix (example 1).
The dependent variable was the total score obtained, with a maximum 20 points. Corcoran & Frith (2003) report that the hinting task has good face validity and is found to be sensitive to ToM deficits in a large number of studies (Corcoran et al., 1995).
IREC-T. The recognition task/IREC-T (Mathews & Mackintosh, 2000) was used to measure the interpretation bias. It consisted of seven social stories. The selected stories were highly correlated with social anxiety (Houtkamp et al., 2017). Each story was read out loud by the test leader, the participant could read along. All stories had a title and consisted of three lines. To make sure the participant understood the story, a control question was asked first which could only be answered with a yes or a no. If this was not answered correctly the test leader explained the story. Then the story was read out loud again. An example of such story is shown in the appendix (example 2). A maximum score of 28 could be obtained. The higher participants scored, the more biased the participant was. The dependent variable was the total score obtained. The IREC-T has repeatedly been used in adolescent samples (Micco et al., 2014; Salemink & Wiers, 2011). Moreover, internal consistency in the IREC-T has been found sufficient in earlier findings (Houtkamp et al. 2017; Salemink & Van den Hout 2010).
Procedure
Adolescents were brought to a separate test room at school by their teachers. First the information letter about the research was read. After reading this together with the research assistant, the participants gave their consent. The testing started with the shortened WISC-V or WAIS-IV that took around 15 minutes. The BART-task took 20 minutes. The first test session was finished with the IREC-T that took around 15 minutes. In the second test session the participant did the other version of the BART task. The session was finished with the hinting task. This task took about 10 minutes. More tasks were performed during these sessions. Because those tasks are not included in the current study, they are not mentioned. Both sessions lasted approximately 1.5 hours. Adolescents received 8 euros for participation and the money earned in one of the BART tasks.
Data analysis
Before testing the hypotheses, standardization checks were performed. To check if the MBID and control group were equally divided in sex, a Chi-square test was performed. Furthermore, it was checked whether the two groups were equally divided in age with an independent t-test. In case the standardization checks were violated, sex or age was added as covariates to the main analysis. The hypotheses were tested with two mediation models in SPSS with the PROCESS tool (Hayes, 2012).
Figure 1: The mediation model presented with the three components: MBID, mentalizing and peer influence.
In the first mediation model the independent variable was MBID or not, the mediator was the ability to mentalize, and the dependent variable was the difference score in the BART. The first hypothesis was that adolescents with MBID were more susceptible for peer influence than typically developing adolescents (see Figure 1). Therefore, a higher difference
score in adjusted pumps on the BART was expected for adolescents with MBID compared to typically developing adolescents (path c). Second, it was hypothesized that adolescents with MBID have a lower ability to mentalize compared to the typically developing adolescents. Therefore, lower scores on the hinting task for adolescents with MBID group were expected (path a). Third, it was assumed that a lower ability to mentalize was related to higher
susceptibility to peer influence. For this reason, it was expected that lower scores on the hinting task results in a higher difference score in adjusted pumps on the BART task (path b). Fourth, it was assumed that the ability to mentalize had a mediating role to susceptibly on peer influence in adolescents with MBID (path ab). It was assumed that adolescents with MBID are less able to mentalize, and are therefore more susceptible to peer influence.
Figure 2: The mediation model presented with the three components: MBID, interpretation bias and peer influence.
Furthermore, a second mediation model was performed in this study (see Figure 2). In this mediation model the independent variable was MBID or not, the mediator was the
interpretation bias, and the dependent variable was the difference score in the BART. The main hypothesis was that adolescents with MBID had a higher interpretation bias compared to the typically developing adolescents, and were therefore more susceptible to peer influence (path ab). The following was expected: First, higher scores on the recognition task for
adolescents with MBID compared to the typically developing adolescents (path a). Second, it was expected that higher scores on the recognition task result in a higher difference score in adjusted pumps on the BART task (path b), and thus higher susceptibility to peer influence. Lastly, it was assumed that adolescents with MBID were more susceptible to peer influence compared with typically developing adolescents (path c).
To test the significance of the size of the indirect (mediated) effect the t-values of path a and b were inverted in an online tool of the Baron and Kenny's (1986) modification
of Sobel's test (Preacher J. & Leonardelli J. 2010).
The sample size was calculated a-priori with the online MEDpower tool (Kenny, 2017). A power of 0.8, an alpha of 0.05 and an effect size of 0.30 beta for path a and path b that indicates a medium effect size (Leppink, O’sullivan, & Winston 2016), and an effect size of 0.10 for path c’ was inserted. This resulted in a necessary sample size of 113 participants.
All dependent variables were checked for outliers. Participants with scores that deviated three standard deviations or more from their group mean were excluded.
Lastly, potential sex differences in susceptibility to risk-taking will be exploratively investigated with an independent t-test.
Results
The average IQ of the MBID group was 74.66 with an SD of 0.74. The percentage of boys was 58% and the percentage of girls was 42% in this group. The average IQ of the control group was 101.76 with an SD of 1.98. The percentage of boys was 77% and the percentage of girls in this group was 23%. The MBID group had an average age of 15.81 with an SD of 0.87 and the control group had an average age of 17.01 with an SD of 1.12 The mean scores on the tasks for both groups are shown in table 1.
Table 1: An overview of the means (M) and standard deviations (SD) of the collected data Standardization checks
All dependent variables were checked for outliers. One participant was excluded (MBID) because of an extremely high BART difference score (a score of 31, which was 3 standard deviations above the group mean). This left a dataset of 93 MBID and 39 control participants.
From the chi-square it was found that the groups were not equally divided in sex χ2 (1) = 3.95, p < .05. The control group had significantly more boys than the MBID group. A small to medium effect of sex in the control group was found. Therefore, sex was took into the mediation analysis as a covariate. To compare the groups in age an independent t-test was
Hinting
score M SD IREC-T score M SD Adjusted pumps (Peer-Solo) M SD MBID (N = 93) 17.06 0.20 17.05 0.44 4.68 1.03 Control (N = 39) 19.16 0.16 16.97 0.62 3.25 1.11 Total ( N = 131) 17.69 0.17 17.02 0.36 4.25 0.79
done, this showed no significant difference between the control and the MBID group in age t(130)=6.33, p = 0.06.
Mediation model for mentalizing
To examine if the ability to mentalize mediates the effect of MBID on peer influence a mediation model was used. First, the MBID group did not show a higher difference score in the BART b=1.56 t(128)=0.91, p=0.36 (path c). Second, according to our expectations the MBID group showed a significant lower score on the Hinting Task than typically developing adolescents, which indicates that being in the MBID group is predictive for the ability to metalize, b=-1.96 t(128)=-5.84, p<.001 (path a). Third, contrary to our expectations, results showed that higher scores on the Hinting Task was a significant predictor for higher BART-difference scores, b=0.94 t(127)= 2.10, p< 0.05 (path b). The MBID group remained not to be a significant predictor for higher BART-difference scores after controlling for the mediator b=3.40 t(128)=1.78, p=0.08 (path c’). Approximately 4% of the variance in peer influence was estimated by the predictors (R² = .042). Fourth, the indirect effect (path ab) was examined by using a bootstrap estimation approach with 1000 samples (Shrout & Bolger, 2002). These results show that the indirect coefficient was: b=1.84, SE=0.86, 95% Cl[0.18, 3.58]. There was no 0 in the confidence interval which indicates a mediation effect. To make sure that there was a significant effect, the Sobel test was done, this gave a p < 0.05. Lastly, sex was not a significant covariate for hinting b=-0.22 t(128)=-0.70, p=0.49, nor to the BART-difference b=-0.40 t(127)=-0.25, p=0.81. These results reversely confirm our mediational hypothesis, see figure 3 for an overview of the results.
Figure 3: An overview of the b-values for all paths for the mediation analysis of the MBID group on peer influence, mediated by the ability to mentalize. The * indicates a p-value of < 0.05 and ** indicates a p-value of < 0.01.
Mediation model for interpretation bias
To explore the hypothesis if interpretation bias mediates the effect of MBID on peer influence again a mediation model was used with gender as a covariate. Contrary to our expectations, the MBID group did not show higher scores in the BART difference score b=1.32 t(122)=0.75, p=0.45 (path c). Also, not according to our expectations, the MBID group did not score significantly higher on the IREC-T b= -0.23 t(123)= -0.29, p= 0.77 (path a). Moreover, higher IREC-T scores were not predictive for higher BART-difference scores b=-0.04 t(122)=-0.18, p=0.86 (path b). In addition, the MBID group remained not to be a significant predictor on higher peer influence after controlling for the mediator, IREC-T, b= 1.33 t(123)=0.76, p=0.45 (path c’). Less than 1% of the variance in peer influence was accounted by the predictors (R² = .007). The indirect effect (path ab) was tested using a
bootstrap estimation approach with 1000 samples (Shrout & Bolger, 2002). These results showed that the indirect coefficient was not significant, b=0.01 SE=0.18, 95% Cl[-0.31, 0.49]. This was checked with the Sobel test, which also showed no significant effect, p=0.88. The covariate gender had a significant effect on interpretation bias b=-1.76 t(123)=-2.35, p<0.05. Higher IREC-T scores were found in girls compared to boys. But no effect between sex and BART-difference scores was found b=-0.68 t(122)=-0.39, p=0.69. The results do not support the hypothesis that MBID is associated with higher peer influence scores as mediated by interpretation bias, see figure 4 for an overview of the results of this mediation analysis.
Figure 4: An overview of the b-values for all paths for the mediation analysis of the MBID group on peer influence, mediated by interpretation bias. The * indicates a p-value of < 0.05.
Explorative research
To explore potential differences between boys and girls to susceptibility to peer influence, a independent t-test was done. This showed no significant difference between boys and girls in the BART difference score t(129)=-0.54, p = 0.72.
Discussion
This study investigated social factors that may underly the increased susceptibility to peer influence in adolescents with MBID. The aim of this study was twofold. First, we aimed to investigate if the ability to mentalize was a predictor in adolescents with MBID for higher susceptibility to peer influence. The second aim was if interpretation bias mediates higher susceptibility to peer influence in adolescents with MBID. The primary results was that adolescents with MBID were not more susceptible to peer influence compared with typically developing adolescents. In addition, no difference between boys and girls in susceptibility to peer influence was found. Moreover, it was not found that adolescents with MBID have a higher interpretation bias compared to typically developing adolescents. Also, interpretation bias was not found to be predictive on risk-taking under peer influence. Therefore, this study invalidates the hypothesis that interpretation bias has a mediating role in adolescents with MBID on the susceptibility to peer influence. In addition, it was found that girls had a higher interpretation bias compared with boys. Furthermore, the results showed that adolescents with MBID had more deficits in their ability to mentalize compared to typically developing
adolescents. Lastly, it was found that the ability to mentalize was predictive for being
susceptible for peer influence. But reversely to our expectations, the results show that a better developed ability to mentalize underlies the higher susceptibility to peer influence in
adolescents with MBID.
The study did not found that adolescents with MBID are more susceptible to peer influence than adolescents without MBID, this is not in line with previous research (Bexkens et al., 2019). This could be due to a too small sample size, especially in the control group. There were 93 adolescents tested with MBID and only 39 typically developing adolescents. The a-priori calculated amount participants needed was 113. This was not equally divided over the groups. This was because the main study investigates more factors, which means that
a larger sample size was needed. For this reason this resulted in a lower power of the collected data. A lower power increases the probability of making a type II error (Banerjee et al., 2009). More participants are needed in both groups for future research. Another explanation for not finding a difference in susceptibility to peer influence in adolescents with MBID compared to typically developing adolescents are the major differences between the groups in terms of gender. 23% of the participants in the control group were girls, for the MBID group this was 42%. Previous research often indicates higher susceptibility to peer influence in boys
compared to girls (Ngee Sim et al., 2003). Since there were more boys in the control group, this could have affected the results. However, in this study no difference in susceptibility to peer influence was found between the sexes. Moreover, sex was taking into the mediation analysis as a covariate. Nevertheless, it is important that in future research compares more equally divided groups.
The finding that adolescents with MBID have impairments in metalizing is in line with earlier research (Van Nieuwenhuijzen et al., 2004; Van Giessen, 2016; Benson et al, 1993; Baglio et al., 2016). However, it was found that the a better developed ability to mentalize has a mediating role in adolescents with MBID in susceptibility to peer influence. As mentioned before, the study of Khemka et al., (2009) hypothesized that impairments in metalizing would make adolescents with MBID potentially more susceptible to peer influence. Our results rejects this finding. The current study found that adolescents with MBID that are better in mentalizing, are more susceptible to peer influence. An explanation could be that the hints made during the BART by the peers were not considered by the adolescents with lower mentalizing skills. If the adolescent understands that it means he or she has to pump more when the other says ‘you are pumping too little’, they are more likely to do so. If it is not understood that the peer implies the adolescent do something, the chance of the adolescent doing it is less. Thus, it is possible that negative intentions of peers are less effective in people
who simply do not understand what the peer is implying the other to do. However, finding links between social understanding and social behaviour is very complex (Werner et al., 2006). Moreover, interpretations of the present findings are to be made with caution because of the implications in this study and the opposite finding in other research. These results provide more insight in differences in mentalizing in adolescents with MBID compared with typically developing adolescents.
The unexpected result that adolescents with MBID do not have a higher interpretation bias than adolescents without MBID is not in line with previous research. Earlier studies found that adolescents with MBID more often suffer from social phobia (Emerson et al., 2003), and social phobia is highly correlated with interpretation bias (Amin et al, 1998). An explanation could be that adolescents in general have a still developing social brain
(Blakemore & Mills, 2014). This is reflected in a higher sensitivity to rejection in peer contexts, which could make adolescents feel more distressed during peer exclusion, and put higher norms for behaviour in peer contexts (Sebestian et al., 201, Gunther et al., 2010). Therefore, it can be assumed that all adolescents are likely to have high interpretation bias. As could be seen in table 1, the means for both groups were approximately the same (17.05 for MBID and 16.97 for the control group). A maximum score of 28 could be obtained, the fact that the scores were higher than 14 indicate that on average both groups had an interpretation bias. Interpretation bias could still play a role in higher susceptibility to peer influence, but it is plausible that no medium to large effect of interpretation bias can be distinguished between the two groups. This study was the first that compared interpretation bias between these groups. Future research can be done to see if interpretation bias plays a role in peer influence by selecting adolescents without interpretation bias and compare them with adolescents with interpretation bias.
Another explanation for this unexpected finding could be that the IREC-T was the last task during the testing session. This might have resulted in adolescents having the tendency to do this task more quickly to finish the test session. Therefore, the test might not be valid enough. However, conceptual questions were asked to ensure that attention was paid. The test leader was also extra alert with errors. Still, it is found that adolescents have a lower attention span compared with adults (Chakrabarti et al., 2013). For both groups this attention span should not have been ignored, especially not with adolescents with MBID (Djuric et al., 2009). Therefore, follow-up research can better focus on one social factor instead of more, so this factor can be examined more validly over a shorter span of testing time.
The current study highlights that mentalizing in adolescents with MBID is limited compared with typically developing adolescents. Moreover, this was the first study that compared interpretation bias in adolescents with MBID to their typically developing peers. More research is needed to see if interpretation bias plays a role to susceptibility to peer influence in adolescents (with and without MBID). Finally, this study was the first that looked into potential social factors underlying the susceptibility to peer influence in adolescents with MBID. Knowing that adolescents with learning disabilities are overrepresented in the prison population (Hayes et al, 2007), mostly due to deviant peer relations (Elliot & Menerard 1996), interventions focusing in reducing risk-taking under peer influence in adolescents with MBID need to be explored more.
References
Amin, N. Edna, B. Foa, F. Meredith, E. (1998). Negative interpretation bias in social phobia, Behaviour Research and Therapy ISSN 0005-7967. 7967(98)00060-6.
Baglio, G. Blasi, V. Sangiuliano Intra, F. Castelli, I. Massaro, D. Baglio, F. (2016). Social
Competence in Children with Borderline Intellectual Functioning: Delayed Development of Theory of Mind Across All Complexity Levels. Frontiers in Psychology.
doi=10.3389/fpsyg.2016.01604
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173–1182. https://doi.org/10.1037/0022-3514.51.6.1173
Banerjee, A., Chitnis, U. B., Jadhav, S. L., Bhawalkar, J. S., & Chaudhury, S. (2009). Hypothesis testing, type I and type II errors. Industrial psychiatry journal, 18(2), 127–131.
doi:10.4103/0972-6748.62274
Benson, D.L., Wittrock, M.C., & Baur, M.E. (1993). Students’ preconceptions on the nature of gases. Journal of Research in Science Teaching, 28, 363-382.
Bexkens, A., Ruzzano, L., Collot d'Escury-Koenigs, A. M. L., Van der Molen, M. W., & Huizenga, H. M. (2014a). Inhibition deficits in individuals with intellectual disability: a meta-regression analysis. Journal of Intellectual Disability Research, 58(1), 3–16
Bexkens, A., Huizenga, H.M., Neville, D.A. et al. (2019). Peer-Influence on Risk-Taking in Male Adolescents with Mild to Borderline Intellectual Disabilities and/or Behavior Disorders. J
Abnorm Child Psychol 47, 543–555 doi:10.1007/s10802-018-0448-0
Calculation of the Sobel test. Retrieved from http://quantpsy.org/sobel/sobel.htm.
Blakemore S.J, Mills, K.L Is adolescence a sensitive period for sociocultural processing? Annu. Rev. Psychol., 65 (2014), pp. 187-207
Brown, B. B., Larson, J. (2009). Peer relationships in adolescents. In Steinberg, R. M. L. (Ed.), Handbook of adolescent psychology, Contextual influences on adolescent development (Vol. 2, 3rd ed., pp. 74–103). Hoboken, NJ: John Wiley & Sons.
Brown, B. B., Clasen, D. R., & Eicher, S. A. (1986). Perceptions of peer pressure, peer conformity dispositions, and self-reported behavior among adolescents. Developmental Psychology, 22(4), 521–530. https://doi.org/10.1037/0012-1649.22.4.521
Chakrabarti, Anwesha. (2013). Span of Attention across Stages of Intellectual Impairment: Does Affective Stimulation Matter?. Psychology. 04. 410-419.
Doi:10.4236/psych.2013.44058.
Corcoran R, Mercer G, Frith CD (1995). Schizophrenia, symptomatology and social inference: investigating theory of mind in people with schizophrenia. Schizophr Res.;17:5---13
Dishion T.J., Capaldi D.M., Spracklen K.M., Li F. (1995). Peer ecology of male adolescent drug use.
Dev. Psychopathol., 7 pp. 803-824
Djuric A., Japundza-Milisavljevic M., Macesic-Petrovic D. (2011). Arithmetic operations and attention in children with intellectual disabilities. Education and Training in Autism and
Developmental Disabilities, 46 (2), pp. 214-219
Dodge, K.A., Lansford, J.E., Burks, V.S., Bates, J.E., Pettit, G.S., Fontaine, R. and Price, J.M. (2003), Peer Rejection and Social Information Processing Factors in the Development of Aggressive ‐ Behavior Problems in Children. Child Development, 74: 374-393. doi:
10.1111/1467-8624.7402004
developmental patterns. Cambridge criminology series. Cambridge University Press; New
York: 1996. pp. 28–67.
Emerson, Eric. (2003). Prevalence of psychiatric disorders in children and adolescents with and without intellectual disability. Journal of intellectual disability research : JIDR. 47. 51-8. Doi: 10.1046/j.1365-2788.2003.00464.x.
Essau CA, Conradt J, Petermann F. Frequency of comorbidity of social phobia and social fears in adolescents. Behav Res Ther. 1999;37:831–843
Falk, E. B., Cascio, C. N., O'Donnell, M. B., Carp, J., Tinney, F. J., Jr, Bingham, C. R. Simons-Morton, B. G. (2014). Neural responses to exclusion predict susceptibility to social influence. The Journal of adolescent. doi:10.1016/j.jadohealth.2013.12.035
Fliessbach, B. Weber, P. Trautner, T. Dohmen, ,U. Sunde, C. E. Elger, A. Falk (2007). Social
Comparison Affects Reward-Related Brain Activity in the Human Ventral Striatum. Science. Doi: 10.1126/science.1145876
Golan, Ofer & Baron-Cohen, Simon & Hill, Jacqueline & Golan, Yael. (2006). The “Reading the Mind in Films” Task: Complex emotion recognition in adults with and without autism spectrum conditions. Social neuroscience. doi: 10.1080/17470910600980986.
Gunther Moor L., Van Leijenhorst, S.A.R.B. Rombouts, E.A. Crone, M.W. Van der MolenDo you like me? Neural correlates of social evaluation and developmental trajectories. Soc.
Neurosci., 5 (2010), pp. 461-482
Hayes S., Shackell P., Mottram P. & Lancaster R. (2007) The prevalence of intellectual disability in a major UK prison. British Journal of Learning Disabilities 35, 162–7
Houtkamp, E. S., van der Molen, M. J., Salemink, E., de Voogd, E. L., & Klein, A. M. (2017). Interpretation biases in socially anxious adolescents with a mild intellectual disability.
Houtkamp, Esther & Molen, Mariët & De Voogd, Esther & Salemink, Elske & Klein, Anke. (2017). The relation between social anxiety and biased interpretations in adolescents with mild intellectual disabilities. Research in Developmental Disabilities. Doi:
10.1016/j.ridd.2017.06.003.
Khemka, I., Hickson, L., Casella, M., Accetturi, N., & Rooney, M. E. (2009). Impact of coercive tactics on the decision-making of adolescents with intellectual disabilities. Journal of
Intellectual Disability Research, 53(4), 353–362
Klein, A. M., Houtkamp, E. O., Salemink, E., Baartmans, J. M. D., Rinck, M., & van der Molen, M. J. (2016). Social anxiety-related differences between self- and peer-rated likability in adolescents with mild intellectual disabilities. Manuscript submitted for publication.
Lejuez, C. W., Read, J. P., Kahler, C. W., Richards, J. B., Ramsey, S. E., Stuart, G. L., … Brown, R. A. (2002). Evaluation of a behavioral measure of risk taking: The balloon analogue risk task (BART). Journal of Experimental Psychology: Applied, 8(2), 75, 84.
Lejuez, C. W., Aklin, W. M., Zvolensky, M. J., & Pedulla, C. M. (2003). Evaluation of the balloon analogue risk task (BART) as a predictor of adolescent real-world risk-taking behaviours.
Journal of Adolescence, 26(4), 475–479.
Lejuez, C., Bornovalova, M. A., Reynolds, E. K., Daughters, S. B., & Curtin, J. J. (2007). Risk factors in the relationship between gender and crack/cocaine. Experimental and Clinical
Psychopharmacology, 15(2), 165–175.
Leppink, J., O'Sullivan, P., & Winston, K. (2016). Effect size - large, medium, and small. Perspectives
on medical education, 5(6), 347–349. doi:10.1007/s40037-016-0308-y
Mathews, A., & Mackintosh, B. (2000). Induced emotional interpretation bias and anxiety. Journal of
Mathews, Andrew & MacLeod, Colin. (2005). Cognitive Vulnerability to Emotional Disorders.
Annual review of clinical psychology. doi: 10.1146/annurev.clinpsy.1.102803.143916.
Masten, N.I. Eisenberger, L.A. Borofsky, J.H. Pfeifer, K. McNealy, Mazziotta, et al.(2009).Neural correlates of social exclusion during adolescence: understanding the distress of peer rejection.
Soc. Cogn. Affect. Neurosci., 4 (2009), pp. 143-157
McGregor, Evelyn & Whiten, Andrew & Blackburn, Pennie. (1998). Transfer of the Picture-in-the-Head Analogy to Natural Contexts to Aid False Belief Understanding in Autism. Autism. doi: 10.1177/1362361398024004.
MEDpower tool (Kenny, 2017) Power and N computation for Medation. Retrieved from
https://davidakenny.shinyapps.io/MedPower/.
Micco, J. A., Henin, A., & Hirshfeld-Becker, D. R. (2014). Efficacy of interpretation bias modification in depressed adolescents and young adults. Cognitive Therapy and Research.
doi:10.1007/s10608-013-9578-4
Ngee Sim, T. and Fen Koh, S. (2003), A Domain Conceptualization of Adolescent Susceptibility to Peer Pressure. Journal of Research on Adolescence. doi:10.1111/1532-7795.1301002
Nieuwenhuijzen, Maroesjka & Castro, Bram & Wijnroks, L. & Vermeer, Adri & Matthys, Walter. (2004). The relations between intellectual disabilities, social information processing, and behavior problems. European Journal of Developmental Psychology. Doi:
10.1080/17405620444000111.
Printstein J., La Greca A. M. (2004). Childhood peer rejection and aggression as predictors of adolescent girls’ externalizing and health risk behavoirs: a 6-year longitidunal study. J.
Consult. Clin. Psycho., 72, pp. 103-112
Salemink, E., & Wiers, R. W. (2011). Modifying threat-related interpretive bias in adolescents.
Salemink, E., van den Hout, M. A., & Kindt, M. (2010). Generalisation of modified interpretive bias across tasks and domains. Cognition and Emotion. doi:10.1080/ 02699930802692053
Schalock, R.A. Luckasson, K.A. Shogren, S. Borthwick-Duffy, V. Bradley, W.H. Buntinx (2007). The renaming of mental retardation: Understanding the change to the term intellectual disability.
Intellectual and Developmental Disabilities, 45 (2) (2007), pp. 116-124
Sebastian, G.C. Tan, J.P. Roiser, E. Viding, I. Dumontheil, S.J. Blakemore (2011). Developmental influences on the neural bases of responses to social rejection: implications of social neuroscience for education. NeuroImage, 57, pp. 686-694
Steinberg L. (2008). A social neuroscience perspective on adolescent risk-taking. Dev.
Rev., 28 pp. 78-106
Van Giessen, S. (2016). Peer-promoted prosocial behaviour in adolescents with Mild-to-Borderline Intellectual Disability. Master Thesis, Mild-to-Borderline Intellectual Disability, 1424165. Van Hoorn, J., Van Dijk, E., Crone, E. A., Stockmann, L., & Rieffe, C. (2017). Peers influence
prosocial behavior in adolescent males with autism Spectrum disorders. Journal of Autism
and Developmental Disorders, 47(7), 2225–2237
Werner, R.S., Cassidy, K.W. and Juliano, M. (2006), The role of social cognitive abilities in ‐ preschoolers' aggressive behaviour. British Journal of Developmental Psychology, 24: 775-799. doi:10.1348/026151005X78799
Wechsler, D. (2003). WISC-IV Technical and Interpretive Manual. San Antonio, Texas: Harcourt Associates Inc.
Wechsler, D. (2004). Wechsler Intelligence Scale for Children – Fourth UK Edition: Administrative and Scoring Manual. London: The Psychological Corporation. 188 References
San Antonio, Texas: Pearson
Wellman, Henry & Baron-Cohen, Simon & Caswell, Robert & Gomez, Juan & Swettenham, John & Toye, Eleanor & Lagattuta, Kristin. (2003). Thought-Bubbles Help Children with Autism Acquire an Alternative to a Theory of Mind. Autism : the international journal of research
and practice. doi: 63. 10.1177/1362361302006004003.
Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: Representation and constraining function of wrong beliefs in young children's understanding of deception. Cognition.
Appendix (see method section)
Example 1: Hinting Task Item
Paul has to go to an interview and he's running late. While he's cleaning his shoes he says to his wife, Jane: 'I want to wear that blue shirt, but it is very creased’. Question 1: What does Paul really mean when he says this?
Hint: Paul goes on to say: 'It is in the ironing basket.'
Example 2: IREC-T item
Your Birthday Today is your birthday.
But like every Monday morning, you have to go to school.
When you are about to leave your house, you don’t see any birthday cards on the doormat. Control question : Do you have to go to school on your birthday?
Question: How big is the chance that they forgot your birthday?
1. very low 2. Low 3. Big 4. very big
