bs_bs_banner
Journal of Intellectual Disability Research 845
doi: 10.1111/jir.12183
volume 59 part 9 pp 845–859 septemBer 2015
Social cognition dysfunction in adolescents with 22q11.2 deletion syndrome (velo-cardio-facial syndrome): relationship with executive functioning and social competence/functioning L. E. Campbell,1,2,3 K. L. McCabe,1,4,6 J. L. Melville,3 P. A. Strutt1,3 & U. Schall1,2,5,6 1 Centre for Translational Neuroscience and Mental Health Research, University of Newcastle, McAuley Centre, Waratah, NSW 2298, Australia 2 Hunter Medical Research Institute, Locked Bag 1, Hunter Region Mail Centre NSW 2310, Australia 3 School of Psychology, University of Newcastle, Ourimbah, NSW 2258, Australia 4 Brain & Mind Research Institute, University of Sydney, Camperdown, NSW 2050, Australia 5 School of Medicine and Public Health, University of Newcastle, Callaghan, NSW 2308, Australia 6 Schizophrenia Research Institute, 405 Liverpool St Darlinghurst NSW 2010, Australia
Abstract Background Social difficulties are often noted among people with intellectual disabilities. Children and adults with 22q.11.2 deletion syndrome (22q11DS) often have poorer social competence as well as poorer performance on measures of executive and social-cognitive skills compared with typically developing young people. However, the relationship between social functioning and more basic processes of social cognition and executive functioning are not well understood in 22q11DS. The present study examined the relationship between social-cognitive measures of emotion attribution and theory of mind with executive functioning and their contribution to social competence in 22q11DS.
Correspondence: Dr Linda Campbell, School of Psychology, University of Newcastle, Science Offices, Ourimbah, NSW 2258, Australia (e-mail: [email protected]).
Method The present cross-sectional study measured social cognition and executive performance of 24 adolescents with 22q11DS compared with 27 age-matched typically developing controls. Social cognition was tested using the emotion attribution task (EAT) and a picture sequencing task (PST), which tested mentalising (false-belief), sequencing, cause and effect, and inhibition. Executive functioning was assessed using computerised versions of the Tower of London task and working memory measures of spatial and non-spatial ability. Social competence was also assessed using the parent-reported Strengths and Difficulties Questionnaire. Results Adolescents with 22q11DS showed impaired false-belief, emotion attribution and executive functioning compared with typically developing control participants. Poorer performance was reported on all story types in the PST, although, patterns of errors and response times across story types were similar in both groups. General sequencing ability was the strongest
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 846 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
predictor of false-belief, and performance on the false-belief task predicted emotion attribution accuracy. Intellectual functioning, rather than theory of mind or executive functioning, predicted social competence in 22q11DS. Conclusions Performance on social-cognitive tasks of theory of mind indicate evidence of a general underlying dysfunction in 22q11DS that includes executive ability to understand cause and effect, to logically reason about social scenarios and also to inhibit responses to salient, but misleading cues. However, general intellectual ability is closely related to actual social competence suggesting that a generalised intellectual deficit coupled with more specific executive impairments may best explain poor social cognition in 22q11DS. Keywords emotion attribution/recognition, executive function, intellectual disability, social cognition, theory of mind, velo-cardio-facial syndrome
Introduction Healthy social relationships are integral to a good quality of life. The ability to initiate, form and maintain social relationships is dependent on the successful integration of social-cognitive factors including skills in understanding, reasoning and predicting other people’s emotions and behaviours. Social difficulties and isolation are often noted among people with intellectual disabilities (IDs) (Emerson & Hatton 2007). Poor social functioning severely limits opportunities for a full and productive life leading to great social and economic burden not only to the individual, but also for families and society at large (Emerson & Hatton 2008). In addition, social dysfunction is one of the many risk factors for behavioural and psychiatric problems that young people with IDs experience (Dykens 2000). Indeed, persons with IDs are estimated to be three to four times more likely than those in the general population to experience an emotional, behavioural or psychiatric disorder (Emerson 2003; Emerson & Hatton 2007). 22q11.2 deletion syndrome (22q11DS; also known as velo-cardio-facial syndrome) results from a de novo or familial deletion of 30–50 genes on the long arm of chromosome 22 (Driscoll et al. 1992; Scambler et al. 1992). The syndrome is the most
common microdeletion syndrome in humans (Botto et al. 2003) and has been estimated to occur as often as between 1 in 1600 and 1 in 2000 live births (Shprintzen 2008), and although very variable, is associated with physical anomalies such as cardiac abnormalities, velopharyngeal insufficiency, hypoparathyroidism and immune deficiency (for a review see e.g., Shprintzen 2008). In addition, the vast majority of people with 22q11DS have cognitive impairments and IDs (Swillen et al. 1997), and similar to other developmental disabilities, often experience behavioural and psychiatric disorders including anxiety, attention-deficit/hyperactivity disorder and autism spectrum disorders (Feinstein et al. 2002). Moreover, 22q11DS is one of the highest known risk factors for psychosis in adolescence and early adulthood (Murphy et al. 1999). Over the past two decades, it has been well documented that children and adults with 22q11DS often have poorer social competence including, for example, mood lability, shyness, and difficulties in initiating and maintaining social relationships compared not only with their typically developing (TD) peers (Golding-Kushner et al. 1985; Fuerst et al. 1995; Swillen et al. 2001; Woodin et al. 2001), but also with other children with chronic illnesses (Looman et al. 2010). While early retrospective parental reports do not identify delays in achieving social developmental milestones (Roizen et al. 2007), later in childhood, parents are often concerned by the problems their children have with peers (Campbell et al. 2011), and as teenagers and young adults, parents often report increasing social withdrawal and isolation (Swillen et al. 1997, 2001). Social problems can be caused by, and be associated with, a range of biopsychosocial factors including speech problems, anxiety and bullying (Shashi et al. 2012); however, they can also be caused by deficits in ‘social cognition’; that is, the set of mental processes that govern how people think about themselves, others, social situations and social interaction (Green & Leitman 2008). Niklasson et al. (2002) reported that children and adults with 22q11DS may have deficits in the social-cognitive skill known as ‘theory of mind’ (ToM). ToM tasks test the ability to go beyond observed behaviour to explain actions in terms of inferred mental states. While Niklasson et al. (2002) measured ToM using classical false-belief tasks (such as the Smarties and
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 847 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
Sally-Anne tasks; (Baron-Cohen et al. 1985); Strange Stories (Happé 1994); and Non-mental and Mental Cartoons (Nyden et al. 2000), they did not report any supporting statistics for ToM deficits in 22q11DS compared with typically developed controls, other than to indicate that in most 22q11DS participants (n = 15/20), poor performance on measures of false-belief were observed. Subsequently, our group has investigated ToM skills among children with 22q11DS compared with age-matched siblings and age- and intelligence quotient (IQ)matched children with Williams syndrome (Campbell et al. 2009, 2011). We did not identify specific ToM performance deficits in children with 22q11DS compared with children with Williams syndrome on false-belief tasks, and although the children with 22q11DS performed poorer on the Strange Stories task, these difficulties may have been influenced by low comprehension as suggested by low performance on non-mentalising stories as well. In addition, while the younger children with 22q11DS in that study performed poorer, primarily on second-order false-belief tasks (which require a person to make inferences about someone’s belief about another’s belief) compared with TD siblings, we concluded that the findings were likely indicative of delayed development of ToM rather than a deficit. On reflection, tasks that are reliant on verbal comprehension may not be optimal measures of ToM for people who report difficulties within both of these domains. Instead, using implicit tasks may better delineate how well people with 22q11DS comprehend mentalising skills. Adopting this approach, Ho et al. (2012), using the Animations task (Abell et al. 2000; Castelli et al. 2000), found that participants with 22q11DS scored lower than TD controls when required to explain purposeful behaviour and when asked to describe vignettes requiring mentalising (designed to imply complex mental states), but not on the random control vignettes. Mentalising ability was positively correlated with a functional measure of reciprocal social behaviour (Ho et al. 2012). In addition, Jalbrzikowski et al. (2012) utilised videotaped vignettes of everyday social interactions to investigate the ToM ability of participants to understand the intentions of others, with a particular focus on white lies and sarcasm. As expected, the 22q11DS
group performed poorer than healthy controls, and it was also reported that performance on the ToM task was a strong predictor of psychosis positive symptom severity (Jalbrzikowski et al. 2012). This is important, as Chow et al. (2006) reported that adults with 22q11DS and schizophrenia (compared with people with 22q11DS, but no diagnosis of schizophrenia) performed worse on a ToM task in which the participants were required to make social inferences; indicating that ToM ability may possess endophenic marker characteristics of psychosis proneness. Interestingly, it has been proposed that the developmental progression of ToM and executive functions (EF) follow similar trajectories in early childhood and that these two cognitive skills are closely related to one another (Miller & Marcovitch 2012). Executive processes underpin control of thought and action, and involve three main components: the ability to retain and to manipulate information in the mind (working memory), to suppress inappropriate responses (inhibition) and to be able to change behaviour in response to new information (set-shifting; Miyake et al. 2000). Whether working memory is a component of EF or a separate set of processes remains an ongoing source of debate. For the purposes of the present study, working memory is considered a component of EF. Specifically, the working memory processes needed to hold relevant information in mind to guide lower-level processes towards executing a goal (Miller & Marcovitch 2012). EFs such as working memory and setshifting are impaired in 22q11DS (Swillen et al. 1999; Woodin et al. 2001; Henry et al. 2002; Sobin et al. 2005; Lewandowski et al. 2007; Campbell et al. 2010b). Furthermore, children with 22q11DS perform poorer at non-verbal EF tasks that require sequencing, integration of patterns, and an understanding of spatial relationships (Sobin et al. 2005). It is likely that social problems are related to a range of cognitive mechanisms among young people with 22q11Ds; indeed, using parent-rating scales, Kiley-Brabeck & Sobin (2006) report a link between executive dysfunction and social skills among children with 22q11DS. More specifically, the skills of self-starting tasks or problem solving on one’s own, and the ability to monitor own work or behaviour, significantly predicted social skills.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 848 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
Over the last few years there has been an increasing focus to design remediation strategies to assist in improving social skills among young people with 22q11DS; one example of such a programme is visá-vis (Glaser et al. 2012). This programme aimed to improve face processing skills among children with IDs. However, in order to design similar programmes with a wider scope of social skills it is important to investigate if ToM should be considered as the primary focus for intervention. Therefore, we must first ascertain whether ToM can predict social competence in the context of other known predictors, such as EFs. Earlier studies conducted by members of our group (Campbell et al. 2010b) indicate strong positive correlations among Tower of London (ToL) and set-shifting tasks with impaired performance observed on both tasks by people with 22q11DS. Despite set-shifting being a core component of EF, set-shifting was not assessed in this study. Specifically, the present study aimed to determine what the strength of the relationship is between ToM and EF in 22q11DS. On the basis of existing literature, we hypothesised that: (1) adolescents with 22q11DS would show poorer performance on measures of ToM [indexed by picture sequencing (false-belief) and emotion attribution tasks (EAT)] and EF (as indexed by working memory, the ToL task and general sequencing ability); (2) EF would predict performance on ToM tasks; and (3) both ToM ability and EF would significantly predict social competence (as indexed by parent-rated peer competence from the Strengths and Difficulties Questionnaire) in 22q11DS.
Methods Participants A total of 24 adolescents with 22q11DS (10 males, 14 females) and 27 TD adolescent controls (13 males, 14 females) were recruited in New South Wales, Australia. Participants with 22q11DS were aged 12–21 years [mean age = 16.75 years; standard deviation (SD) = 3.14] and had a mean fullscale IQ of 75.88 (SD = 14.93). The TD control group were aged 8–22 years (mean age = 16.26 years; SD = 3.65) and had a mean full-scale IQ of 108.48 (SD = 14.21). TD control participants were
recruited from the local community (n = 16) or were siblings (n = 11) of the 22q11DS participants. Written informed consent was given by all parents/ guardians, and participants provided assent when applicable. This study was granted ethical approval by the University of Newcastle’s Human Research Ethics Committee. Participants completed structured diagnostic interviews to assess for common mental health diagnoses [Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders – Fourth Edition (Text Revision) Axis I Disorders (SCID; First et al. 2002) and Schedule for Affective Disorders and Schizophrenia for School-Age Children – Present and Lifetime Version (K-SADS-PL; Kaufman et al. 1997) ]. Of the 22q11DS participants, the following diagnoses were reported: mild ID (n = 15); anxiety disorders (n = 9; i.e., obsessive– compulsive disorder, generalised anxiety disorder, agoraphobia with/out panic disorder, panic disorder without agoraphobia, separation anxiety, paroxysmal anxiety, other anxiety symptoms); attentiondeficit hyperactivity disorder (n = 3); oppositional defiant disorder (n = 3); dysthymic disorder (n = 3); prodromal symptoms (n = 2); insomnia (n = 2); schizotypal disorder (n = 1); depressive symptoms (n = 1); schizophrenia-catatonic (n = 1); trichotillomania (n = 1); and epilepsy (n = 1). Six participants reported no psychological diagnoses. Furthermore, of the 22q11DS participants, the following medications were being used at the time of testing: methylphenidate (n = 4), selective serotonin reuptake inhibitor (n = 3), risperidone (n = 2), valproate (n = 1), amitriptyline (n = 1), clonidine (n = 1), cogentin (n = 1), unspecified antipsychotics with mood stabilisers (n = 1) and antiepileptic medication (n = 1). Fifteen participants were receiving no current medication at the time of testing. The inclusion and exclusion criteria for the study have been previously reported for the 22q11DS group (Campbell et al. 2010a; McCabe et al. 2011). In brief, TD control participants were excluded from the study if they had an ID/developmental delay or any history of neurological disorders or Axis 1 psychiatric disorders. All participants underwent a structured diagnostic interview based on the SCID (research version; First et al. 2002) and the K-SADS-PL (Kaufman et al. 1997) conducted by a psychiatrist (Author: US) at the time of testing.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 849 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
Experimental tasks and procedure Assessment of intellectual functioning The Wechsler Abbreviated Scale of Intelligence (WASI; Wechsler 1999) was used to assess participants’ general intellectual functioning. Social competence and emotional well-being Parental ratings of the Strengths and Difficulties Questionnaire (SDQ; Goodman et al. 2000) were collected for each participant. The SDQ is a brief (25 items) behavioural screening questionnaire that uses a 3-point Likert scale (‘not true’, ‘somewhat true’ or ‘certainly true’) to assess: emotional symptoms, conduct problems, hyperactivity/inattention, peer relationship problems and prosocial behaviour. The SDQ is a valid and reliable measure of adjustment and psychopathology in children and adolescents (Goodman et al. 2000). Social cognition and theory of mind Emotion attribution task. Emotion attribution ability was assessed using a shortened version of the EAT reported previously by Langdon et al. (2006) Participants were shown five cartoon strips (as in Langdon et al. 2006) presented in a fixed order of increasing length and difficulty level, in which the characters’ faces were blank. Placed underneath these, in predetermined incorrect order, were a series of facial-affect cards (depicting happiness, sadness, anger and shock). Participants were asked to rearrange the facial-affect cards according to which emotions they thought matched the characters in the cartoon strips. Correctly and incorrectly positioned facial-affect cards were given a score of 1 or 0, respectively. On completion of this task, participants were asked to name the emotion shown in each of the facial-affect cards. The mean number of correctly identified emotions was calculated for each participant. Picture sequencing task. Picture sequencing task (PST) (see Langdon et al. 2006, for a more detailed description) is used to assess selective ToM impairments. It is comprised of four story types: one ToM/mentalising story type (false-belief) and three control story types (social-script, mechanical and capture). False-belief stories assess the ability to
understand that a story character’s actions were due to misinformation about a situation. Social-script stories control for ability to logically sequence commonly occurring social events/behaviour. Mechanical stories control for ability to understand physical cause-and-effect relationships. Capture stories control for ability to inhibit misleading information. As noted by Langdon et al. (2006) capture stories are an important control for two reasons: (1) they require a similar level of processing to the false-belief stories and (2) they are of higher difficulty so as to prevent any group differences in false-belief scores being explained purely by task difficulty. As well as providing information about selective ToM impairments individually, the three control story types, when taken together, provide information about participants’ general sequencing ability, which is an important control to ensure that any group differences in false-belief scores are not better accounted for by general EF differences. Participants completed two practice trials followed by four experimental trials (i.e., four versions of each story type), with four picture cards per story. In a fixed pseudo-random incorrect order, the cards were placed face-down and participants were asked to turn each card and rearrange them to form a logical sequence of events. For each sequence, two points were given for correctly positioning the first card, two points for the last card, and one point for the second and third cards (total six points available). Card position scores and response times (RTs) were averaged across each story type to give a mean position score and RT per story type. Executive functioning Planning task. We designed a computerised planning task based on the ToL task to assess participants’ planning ability. Briefly, participants viewed a start configuration of three pegs of differing heights and three balls of differing colours, and were instructed to rearrange the balls into a goal configuration using a touch screen. The start and goal configurations were shown simultaneously in the upper and lower half of the screen and there was no time limit for completion. The balls could only be moved one at a time and if there was no other ball on top. Three balls could be placed on the longest peg, two on the medium peg, and one on the shortest peg.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 850 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
The number of moves necessary for completion ranged from two to seven. Variables recorded were moves above minimum, initial thinking time and subsequent thinking/move time. Participants were given feedback for passing a trial (smiley face) or for failing the trial (sad face). Additionally, all participants completed a motor control task in order to control for motor movement time; these trials were not included in the data analysis. Working memory. Participants performed one visuospatial (location) and two non-spatial working memory tasks on a touch screen. The visuo-spatial task consisted of a sequence of two to nine stimuli (shapes and faces with a neutral expression; size 100 × 128 pixels) that appeared sequentially at random, non-symmetrical locations for 2 s, with no inter-stimulus interval. Empty blue boxes then appeared in the same location as the preceding stimuli and participants were asked to touch the boxes in the same order as the stimuli had appeared. In both non-spatial tasks, participants were presented with one image (either complex shapes or faces) at a time, centrally on the computer screen. The stimuli were presented in a random order at the bottom of the computer screen and participants were asked to touch the stimuli in the order that they had appeared in. If successful in three same-length sequences, one more stimulus was introduced up to a maximum of nine stimuli/ levels. Stimulus size and delivery rate were identical in all three tasks and working memory accuracy and span (defined as the maximum number of levels completed; i.e., two to nine) were recorded for each task.
Analysis Demographic information and intellectual functioning (WASI) were examined using independent samples t-tests and chi-square analysis. A (2 × 4) mixed between-within subjects analysis of variance (ANOVA) was conducted to analyse mean scores from the PST. The between-factor had two levels for participant group (22q11DS vs. TD control) and the within-factor had four levels for story type (social-script vs. mechanical vs. capture vs. false-belief). An index of general sequencing ability was created by averaging the scores for social-script, mechanical
and capture stories. A one-way analysis of covariance (ANCOVA) was performed to assess possible effects of general sequencing ability on false-belief scores. EAT scores were analysed using a one-way ANOVA.1 A one-way ANCOVA was performed to assess the effect of emotion recognition accuracy on EAT scores. A one-way ANOVA was also used to assess scores on the planning task and working memory task. Multiple regression analysis was used to assess (1) the ability of general sequencing ability, emotion attribution ability, planning, working memory2 and full-scale IQ to predict performance on false-belief across groups; and (2) the ability of general sequencing ability, planning, false-belief, working memory, and full-scale IQ to predict performance on emotion attribution. The threshold set for significance was P < 0.05.
Results Demographic data The 22q11DS and TD control groups did not differ significantly on gender distribution or age (P’s > 0.05). The 22q11DS group reported significantly lower IQ scores than the TD group t(49) = −7.99, P < 0.001; IQ differences were greater than two SDs between the TD control and 22q11DS groups and were therefore considered a group-defining characteristic. Table 1 describes means, SDs and patterns of significance.
Strengths and Difficulties Questionnaire When assessing SDQ scores, a significant main effect was found for group, F(1, 47) = 28.59, P < 0.001, partial eta squared = 0.38; and for subscale type F(4, 44) = 34.63, P = 0.001, partial eta squared = 0.76. Further, a significant group by subscale interaction effect was found, F(4, 44) = 11.16, P < 0.001, partial eta squared = 0.5. Compared with the TD control group, the 22q11DS group scored particularly high for emo1 Response times for the EAT were not assessed because of faulty timing data for a large proportion of the participants. 2 The working memory sub-tests were highly correlated with one another; therefore, where working memory was entered as a covariate in analysis, we used only the face accuracy sub-scale of this task.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 851 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
22q11DS
Gender (% male) Age (years) Intelligence quotient (WASI)** SDQ Emotional symptoms** Conduct problems Hyperactivity** Peer problems** Prosocial behaviour†
TD control
Mean (SD)
Min–Max
Mean (SD)
Min–Max
41.7 16.75 (3.14) 75.88 (14.93)
N/A 12–21 56–115
48.1 16.26 (3.65) 108.48 (14.21)
N/A 8–22 83–141
5.33 (2.96) 2.04 (1.88) 4.63 (2.06) 4.54 (2.45) 6.79 (2.78)
1–10 0–6 0–8 1–10 0–10
1.64 (2.27) 1.36 (1.63) 2.36 (2.08) 1.40 (1.94) 7.76 (2.07)
0–8 0–5 0–7 0–7 2–10
Table 1 Means and SDs of demographic data for 22q11DS and control groups
* P < 0.05, ** P < 0.001. † Reversed scored. N/A, not applicable; SD, standard deviation; SDQ, Strengths and Difficulties Questionnaire; TD, typically developing; WASI, Wechsler Abbreviated Scale of Intelligence.
tional symptoms, hyperactivity/inattention, and peer relationship problems (Ps < 0.001), but scored comparably on conduct problems and prosocial behaviour (Ps > 0.1). See Table 1 for means and SDs.
Picture sequencing task Position scores When assessing mean position scores, there was a significant main effect of group, F(1,49) = 23.51, P < 0.001, partial eta squared = 0.32; and story type, F(3, 47) = 37.88, P < 0.001, partial eta squared = 0.71; however, there was no significant group by story type interaction, F(3, 47) = 0.71, P = 0.55, partial eta squared = 0.04. Participants with 22q11DS made significantly more positioning errors than TD controls across all story types (Ps ≤ 0.005). All participants made the most positioning errors for capture, followed by false-belief and social-script, and made the least errors for mechanical. Furthermore, across all participants, positioning scores differed significantly between almost all story types (Ps ≤ 0.001 for social-script/false-belief, capture/ false-belief, social-script/capture, and capture/ mechanical; and P ≤ 0.01 for false-belief/mechanical) with the exception of scores between social-script and mechanical stories (P = 0.13). See Table 2 for means, SDs and patterns of statistical significance.
An ANCOVA was undertaken to assess the effect of general sequencing ability on the false-belief scores of 22q11DS and TD control participants. This revealed that general sequencing ability was a significant predictor of false-belief scores, F(1, 48) = 21.94, P < 0.001, and after correcting for general sequencing ability, the difference in falsebelief scores between 22q11DS participants and TD controls was no longer significant (F < 1).
Response times To assess mean RTs for correctly sequenced trials, a one-way ANOVA was performed due to missing data for 12 participants.3 Participants with 22q11DS were significantly slower than TD controls at sequencing the false-belief F(1, 45) = 5.91, P < 0.05, social-script F(1, 48) = 10.85, P < 0.01, and mechanical F(1, 47) = 8.17, P < 0.01 stories; however, there was no difference between the two groups in RTs to
3 Twelve participants (11× 22q11DS and 1× TD control) had RTs excluded for at least one story type because of having no correctly sequenced trials for that story type. The control participant had their capture RTs excluded, and for the 22q11DS participants, RTs were excluded for 1× social-script, 2× mechanical, 5× false-belief and 10× capture stories.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 852 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
Table 2 Group means and standard deviations for the picture sequencing task and emotion attribution task
22q11DS Mean (SD) Picture sequencing task Position scores False-belief** Social-script*** Mechanical*** Capture*** Response times (seconds) False-belief* Social-script** Mechanical** Capture Emotion attribution task Emotion identification accuracy** Overall emotion attribution accuracy
TD control Min–Max
Mean (SD)
Min–Max
16.67 (6.63) 19.79 (4.43) 18.54 (5.49) 10.83 (3.89)
1–24 6–24 3–24 0–17
20.96 (3.32) 23.04 (1.83) 22.52 (2.68) 16.26 (5.24)
12–24 18–24 14–24 8–24
26.31 (8.04) 18.56 (4.69) 23.16 (10.14) 29.89 (10.84)
15.95–45.71 13.22–29.99 11.29–49.35 11.75–54.63
19.89 (9.56) 14.10 (4.84) 16.44 (6.16) 29.33 (12.51)
8.95–61.60 7.16–28.45 6.75–39.75 13.65–61
23.54 (2.06) 2.54 (1.69)
18–27 0–5
25.77 (3.05) 3.96 (1.09)
16–29 1–5
* P < 0.05, ** P < 0.01, *** P ≤ 0.001. SD, standard deviation; TD, typically developing.
sequence the capture stories (P = 0.89; see Table 2 for means, SDs and range of RT scores). Across all participants, RTs were significantly different among all four story types (Ps < 0.001 for social-script/ mechanical, social-script/false-belief, social-script/capture and false-belief/mechanical, and P ≤ 0.05 for capture/ false-belief and capture/mechanical). All participants were quickest to correctly sequence social-script stories followed by mechanical stories, then falsebelief stories and were slowest to correctly sequence capture stories.
Emotion attribution task Emotion identification accuracy Participants with 22q11DS made significantly more errors in identifying the facial affect of the cartoon characters compared with TD control participants, F(1, 48) = 8.99, P < 0.01. See Table 2 for means and SDs.4
4 Emotion identification accuracy score is missing for one TD control participant.
Emotion attribution accuracy Participants with 22q11DS were significantly less accurate on emotion attribution than the TD control participants, F(1,49) = 12.98, P = 0.001. See Table 2 for means and SDs. Emotion identification accuracy was included in an ANCOVA analysis to assess whether emotion identification accuracy had an effect on the ability to correctly attribute emotion. The results showed that emotion identification accuracy was a significant predictor of emotion attribution scores, F(1, 47) = 5.78, P < 0.05, and after correcting for emotion identification ability, the difference in emotion attribution scores between 22q11DS participants and TD controls lowered, but remained statistically significant, F(1, 47) = 6.51, P = 0.01. General sequencing ability was then included as a covariate to examine its effect on emotion attribution accuracy scores between the two groups. The results showed that general sequencing ability was a significant predictor of emotion attribution scores, F(1, 48) = 5.87, P < 0.05. After adjusting for general sequencing ability, the difference in emotion attribution scores between 22q11DS participants and TD controls were no longer significant, F(1, 48) = 2.63, P = 0.11.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 853 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
Table 3 Group means and standard deviations for the planning task and working memory task
22q11DS
Planning task Pre-planning time for correct trials (ms)** Move time for correct trials (ms)* Number of moves above minimum*** Working memory task Spatial accuracy*** Object accuracy*** Facial accuracy*** Spatial span*** Object span*** Facial span***
TD control
Mean (SD)
Min–Max
Mean (SD)
Min–Max
7539.35 (2639.76) 34518.87 (45969.07) 33.95 (14.89)
4718–13109.67 3116.5–200352.25 15–74
12186.22 (7752.6) 13483.69 (1478.89) 14.12 (8.28)
3834.27–35248.13 1518.93–37986.15 1–36
13.88 (3.84) 11.44 (3.42) 11.24 (3.03) 6.04 (1.4) 5.16 (1.25) 11.24 (3.03)
1–21 4–17 5–16 1–8 3–7 5–16
9.27 (4.2) 6.64 (3.43) 5.91 (2.65) 4.41 (1.47) 3.5 (1.26) 5.91 (2.65)
1–16 3–16 1–9 2–7 2–7 1–9
* P < 0.05, ** P ≤ 0.01, *** P < 0.001. SD, standard deviation; TD, typically developing.
Working memory variable
1
2
3
4
5
1 2 3 4 5 6
– 0.983*** 0.634*** 0.663*** 0.541*** 0.564***
– 0.646*** 0.674*** 0.561*** 0.586***
– 0.982*** 0.621*** 0.659***
– 0.602*** 0.635***
– 0.978***
Spatial accuracy Spatial span Object accuracy Object span Facial accuracy Facial span
Table 4 Pearson’s r correlations between all variables of the working memory task, n = 47
*** P < 0.001.
Executive functioning Planning task A one-way ANOVA revealed significant group differences on all three levels of the planning task (see Table 3 for means and SDs). Compared with TD controls, 22q11DS participants spent significantly less time planning their moves F(1, 46) = 7.18, P ≤ 0.01, but took a longer time F(1, 46) = 5.27, P < 0.5 and made a greater number of moves F(1, 46) = 33.87, P < 0.001 in completing the task.
trols on all types of the working memory task (spatial = F(1, 45) = 15.42, P < 0.001; object = F(1, 45) = 23.05, P < 0.001; facial = F(1, 45) = 40.62, P < 0.001). Consequently, 22q11DS participants had significantly shorter working memory span than TD controls across working memory tasks (spatial = F(1, 45) = 15.18, P < 0.001; object = F(1, 45) = 20.48, P < 0.001; facial = F(1, 45) = 35.68, P < 0.001). See Table 3 for means and SDs. A Pearson’s r analysis revealed that all accuracy and span scores were positively correlated with one another (see Table 4).
Working memory A one-way ANOVA revealed that 22q11DS participants were significantly less accurate in remembering the location of stimuli than TD con-
Multiple regression analysis Preliminary analyses were conducted to ensure no violation of the assumptions of normality, linearity,
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 854 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
multicollinearity and homoscedasticity. Multiple regression was used, firstly to assess the ability of general sequencing, emotion attribution, planning, working memory and full-scale IQ to predict performance on false-belief across groups. The total variance explained by the model as a whole was 56.5%, F(6, 45) = 8.45, P < 0.001. The variable with the strongest significant unique contribution to explaining performance of false-belief was general sequencing ability beta = 0.44, P < 0.01 while emotion attribution ability was slightly lower beta = 0.28, P = 0.06. Secondly, the ability of general sequencing ability, planning, false-belief, working memory and fullscale IQ to predict performance on emotion attribution ability were investigated. The total variance explained by the model as a whole was 52.1%, F(5, 45) = 8.69, P < 0.001. The variable with the strongest significant unique contribution to explaining differences in emotion attribution was false-belief beta = −33, P < 0.05. Thirdly, across groups, the variables mentioned earlier were included in a model to investigate the predictive ability of peer problems. The variable with the strongest significant unique contribution to explaining differences in peer relationships was intellectual functioning beta = −0.72, P < 0.005, while working memory was the second largest beta = 0.35, P = 0.09. Follow-up within group multiple regression analyses revealed that the best significant predictor of false-belief in the 22q11DS group was emotion attribution beta = 0.39, P < 0.05. The total variance explained by the model was 65.6%, F(4, 23) = 9.06, P < 0.001. Meanwhile, the best predictor of emotion attribution in the 22q11DS group was false-belief beta = 0.42, P = 0.12. The total variance explained by the model was 65.6%, F(5, 20) = 3.67, P < 0.05. Subsequent analyses revealed that the best predictors of peer relationships in the 22q11DS group were working memory (indexed using the face accuracy subscale; beta = 0.55, P < 0.02, intellectual functioning beta = −0.94, P < 0.005, and emotion attribution beta = 0.69, P < 0.05. The total variance explained by the model was 79.9%, F(7, 20) = 8.24, P < 0.05. The multiple regression analysis for the TD group did not reach statistical significance (P > 0.2).
Discussion 22q11DS is associated with a range of cognitive and social cognition difficulties (Henry et al. 2002; Niklasson et al. 2002). The present study’s findings are consistent with previous research and support our first hypothesis by showing impairment of social cognition, as measured by false-belief and EATs, and EF, as measured by planning and working memory tasks, by individuals with 22q11DS compared with TD control participants. However, the measures employed in the current study were more sensitive and allowed for overall performance on the social cognitive tasks to be broken down into components. Our findings indicate that the socialcognitive impairments observed in 22q11DS are not specific to false-belief or indeed emotion attribution dysfunction. In the current study, we confirmed the findings of previous research that people with 22q11DS display impaired performance on a range of social cognition (Niklasson et al. 2002; Campbell et al. 2011; Ho et al. 2012; Jalbrzikowski et al. 2012) and EF (Swillen et al. 1999; Woodin et al. 2001; Henry et al. 2002; Sobin et al. 2005; Lewandowski et al. 2007; Campbell et al. 2010b) tasks. More specifically, the data revealed that the young people with 22q11DS performed poorer on each of the four story types in the PST indicating a general underlying dysfunction, including problems with understanding cause and effect, the ability to logically reason about social scenarios (without needing to understand false-beliefs), and also to inhibit responses to salient, but misleading cues. Although the 22q11DS group consistently performed at lower levels compared with TD controls, the errors and RTs across story types were similar in both groups. These findings suggest that the 22q11DS group displayed a generalised deficit coupled with slower responses. Thus, a methodological strength of the present study is the sensitivity of the tasks selected to examine ToM and EF. The suitability of EF tasks to groups with mild intellectual deficits is a necessary consideration for 22q11DS researchers. Langdon et al.’s (2006) PST is well suited to ID groups because it is designed in such a way that participants, to some degree, act as their own control. Specifically, the capture stories are incorporated to account for differential
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 855 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
sensitivity to increasing task difficulty, while in combination with social-script (logical sequencing) and mechanical (cause/effect) these components of the task act as an index of EF (general sequencing ability). The inclusion of these components permitted the examination of patterns and hierarchies of performance on these tasks showing that our 22q11DS group responded in the same way to our HC group, only slower and with less accuracy. Similarly, on the EAT, it was found that although the TD group were better at completing the task, the performance of the young people with 22q11DS was partially contingent on their ability to recognise emotions. Further analysis revealed that the strongest predictor of performance on the social cognition task examining ToM measures of false-belief (as indexed by the PST) was general sequencing ability, an approximate measure of EF, while the ability to understand false-belief, in turn, predicted the accuracy of emotion attribution. These findings confirmed our second hypothesis that deficits in social cognition in 22q11DS would be accounted for by impaired EF. Finally, contrary to our predictions, we found only partial support for our third hypothesis that both ToM ability and EF would predict social competence, with none of the measures of social cognition predicting social competence (as indexed by peer relationship problems in the SDQ). Instead, intellectual ability and working memory were the strongest unique predictors of peer relationships in 22q11DS. Performance by participants with 22q11DS on the social cognition tasks reported here indicates evidence of a general underlying dysfunction. Difficulty grasping cause and effect, interpreting social scenarios and inhibiting responses to salient, but misleading cues demand further exploration. Disturbance of lower-order sensory and cognitive functions may be associated with higher-order social cognition processing deficits reported in 22q11DS. For instance, Simon (2008) suggests that a ‘hypergranularity’ of attention processes in 22q11DS may best explain performance impairments on measures of attention, which they identify as associated with higher-order processes such as magnitude and numerical processing (Simon 2008). Similarly, studies in schizophrenia groups
describe relationships between sensory and cognitive impairment, which show associations between basic sensory processes such as pitch perception and emotion recognition (Leitman et al. 2010; Gold et al. 2012), as well as spatial frequency with object recognition (Calderone et al. 2013). Although the present findings did not focus on basic sensory processing, we did find support for the notion of generalised impairments, whereby impaired executive control may best explain impairment on tests of social cognition, rather than specific difficulties with social stimuli per se. Future research will benefit from the continued exploration of the relationship between basic cognitive and sensory processes and higher-order cognitions. Our finding that working memory was a significant predictor of social functioning problems (second to intellectual functioning) was somewhat unexpected. Although this finding is supported by earlier studies that report working memory as one of the strongest predictors of social skills in 22q11DS, but not sibling controls (Kiley-Brabeck & Sobin 2006), in the current study, we expected that ToM measures would be more likely to explain differences in peer relationship problems (our measure of social functioning). From a cognitive perspective, social functioning includes a complex network of processes that are reliant on working memory features such as the timely retrieval and integration of information under conditions that include competing responses and that are likely to be vulnerable to distraction. The limited capacity of people with 22q11DS to hold and manipulate information in working memory would likely impair their ability to react and respond to complex social cues in an efficient manner.
Clinical implications The findings from the present study suggest that conclusions of generalised impairment in 22q11DS are likely far more nuanced. The patterns of similarity between 22q11DS and TD control participants on the SDQ indicate that generalised impairment does not extend to some aspects of social functioning with similar rates of conduct problems and prosocial behaviour reported by 22q11DS participants as their age-matched peers. Coupled with
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 856 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
evidence of similar although poorer patterns of accuracy and slower reaction times on measures of emotion attribution, EF and ToM, these findings also raise important questions regarding whether EF and social cognition should be considered as single or separate constructs for the purpose of designing effective interventions in this population. As a neurodevelopmental disorder, 22q11DS is associated with a wide range of atypical cognitive functions and it seems unlikely that models of cognition, developed on the basis of TD people or indeed people with very specific cognitive difficulties such as autism, will offer an adequate explanation for the social problems among young people with 22q11DS. Indeed, while social perception, social knowledge, attribution bias and emotion processing are associated with ToM, they do depend on other cognitive mechanisms such as EF, all of which start to develop in very early childhood. Hence, in order to get a full picture of the social difficulties among children with 22q11DS, it will also be important to investigate cognitive and perceptual precursors to the higher-order social cognition deficits reported in 22q11DS.
Limitations As noted in previous studies conducted by our group (Campbell et al. 2010a; McCabe et al. 2013), significant differences in IQ between the 22q11DS and TD control groups were reported. 22q11DS is associated with significant intellectual impairment and is considered a group-defining characteristic. Thus, it was not controlled for statistically in our analysis. It is also necessary to consider the potential influences of medication on the measures used in this study. Several of the medications with potential sedative side effects (e.g., risperedone, valproate, amitriptyline) may have impaired performance. In addition, the possibility of different subtypes is worth considering when doing research into 22q11DS as this may affect our ability to reliably identify behavioural characteristics. This is particularly interesting as Sinderberry et al. (2013) identified that one 22q11DS subtype was primarily characterised by EF deficits. In the future, larger sample sizes and more sophisticated statistical analysis should be
carried out to identify specific behavioural phenotypes within the syndrome.
Conclusions The present study would seem to suggest support for a domain-general rather than domain-specific interpretation of social cognition deficits in 22q11DS because, while we have reported ToM deficits, when we accounted for the contribution of EF deficits, these were no longer significant. However, when this relates to measures of functioning (social competence indexed using a measure of peer problems), other domains (intellectual ability, working memory) best explain dysfunction.
Acknowledgements We would like to thank Dr Kathryn Leadbeater for her assistance in recruitment and testing for the current study and A/Prof Robyn Langdon for her advice during the preparation of this paper. Funding for the current study was received from the Hunter Medical Research Institute (HMRI) in the form of a Port-Waratah Coal Services post-doctoral fellowship, and from the National Health and Medical Research Council (NMMRC) in the form of an Australian Training Fellowship (455624). Infrastructure support was also obtained from the Schizophrenia Research Institute. The funding bodies had no input in the conduct of the study or in the analysis and reporting of the data. We thank our participants, with and without 22q11.2 deletion syndrome and their families, for their generosity of time and information, and support for our research.
References Abell F., Happe F. & Frith U. (2000) Do triangles play tricks? Attribution of mental states to animated shapes in normal and abnormal development. Cognitive Development 15, 1–16. Baron-Cohen S., Leslie A. M. & Frith U. (1985) Does the autistic child have a ‘theory of mind. Cognition 21, 37–46. Botto L., May K., Fernhoff P., Correa A., Coleman K., Rasmussen S. et al. (2003) A population-based study of
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 857 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
the 22q11.2 deletion: phenotype, incidence, and contribution to major birth defects in the population. Pediatrics 112, 101–7.
rated health of English men and women with mild and moderate intellectual disabilities: cross sectional survey. European Journal of Public Health 18, 31–7.
Calderone D. J., Hoptman M. J., Martinez A., Nair-Collins S., Mauro C. J., Bar M. et al. (2013) Contributions of low and high spatial frequency processing to impaired object recognition circuitry in schizophrenia. Cerebral Cortex 23, 1849–58.
Feinstein C., Eliez S., Blasey C. & Reiss A. (2002) Psychiatric disorders and behavioral problems in children with velocardiofacial syndrome: usefulness as phenotypic indicators of schizophrenia risk. Biological Psychiatry 51, 312–18.
Campbell L., McCabe K., Leadbeater K., Schall U., Loughland C. M. & Rich D. (2010a) Visual scanning of faces in 22q11.2 deletion syndrome: attention to the mouth or the eyes? Psychiatry Research 177, 211– 15.
First M. B., Spitzer R. L., Gibbon M. & Williams J. B. W. (2002) Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Research Version, Patient Edition With Psychotic Screen (SCID-I/P W/ PSY SCREEN). Biometrics Research, New York.
Campbell L. E., Stevens A., Daly E., Toal F., Azuma R., Karmiloff-Smith A. et al. (2009) A comparative study of cognition and brain anatomy between two neurodevelopmental disorders: 22q11.2 deletion syndrome and Williams syndrome. Neuropsychologia 47, 1034–44.
Fuerst K. B., Dool C. B. & Rourke B. P. (1995) Velocardiofacial syndrome. In: Syndrome of Nonverbal Learning Disabilities; Neurodevelopmental Manifestations (ed. B. Rourke), pp. 119–37. Guilford Press, New York.
Campbell L. E., Azuma R., Ambery F., Stevens A., Smith A., Morris R. G. et al. (2010b) Executive functions and memory abilities in children with 22q11.2 deletion syndrome. The Australian and New Zealand Journal of Psychiatry 44, 364–71.
Glaser B., Lothe A., Chabloz M., Dukes D., Pasca C., Redoute J. et al. (2012) Candidate socioemotional remediation program for individuals with intellectual disability. American Journal on Intellectual and Developmental Disabilities 117, 368–83.
Campbell L. E., Stevens A. F., McCabe K., Cruickshank L., Morris R. G., Murphy D. G. et al. (2011) Is theory of mind related to social dysfunction and emotional problems in 22q11.2 deletion syndrome (velo-cardiofacial syndrome). Journal of Neurodevelopmental Disorders 3, 152–61.
Gold R., Butler P., Revheim N., Leitman D. I., Hansen J. A., Gur R. C. et al. (2012) Auditory emotion recognition impairments in schizophrenia: relationship to acoustic features and cognition. American Journal of Psychiatry 169, 424–32.
Castelli F., Happe F., Frith U. & Frith C. (2000) Movement and mind: a functional imaging study of perception and interpretation of complex intentional movement patterns. Neuroimage 12, 314–25. Chow E. W., Watson M., Young D. A. & Bassett A. S. (2006) Neurocognitive profile in 22q11 deletion syndrome and schizophrenia. Schizophrenia Research 87, 270–8.
Golding-Kushner K., Weller G. & Shprintzen R. (1985) Velo-cardio-facial syndrome: language and psychological profiles. Journal of Craniofacial Genetics and Developmental Biology 5, 259–66. Goodman R., Ford T., Simmons H., Gatward R. & Meltzer H. (2000) Using the Strengths and Difficulties Questionnaire (SDQ) to screen for child psychiatric disorders in a community sample. British Journal of Psychiatry 177, 534–9.
Driscoll D. A., Spinner N. B., Budarf M. L., McDonald-McGinn D. M., Zackai E. H., Goldberg R. B. et al. (1992) Deletions and microdeletions of 22q11 in velo-cardio-facial syndrome. American Journal of Medical Genetics 44, 261–8.
Green M. F. & Leitman D. I. (2008) Social cognition in schizophrenia. Schizophrenia Bulletin 34, 670–2.
Dykens E. M. (2000) Annotation: psychopathology in children with intellectual disability. Journal of Child Psychology and Psychiatry 41, 407–17.
Henry J. C., Van Amelsvoort T., Morris R. G., Owen M. J., Murphy D. G. & Murphy K. C. (2002) An investigation of the neuropsychological profile in adults with velo-cardio-facial syndrome (VCFS). Neuropsychologia 40, 471–8.
Emerson E. (2003) Prevalence of psychiatric disorders in children and adolescents with and without intellectual disability. Journal of Intellectual Disability Research 47, 51–8.
Happé F. (1994) Autism, An Introduction to Psychological Theory. UCL Press, Ltd., London.
Emerson E. & Hatton C. (2007) Mental health of children and adolescents with intellectual disabilities in Britain. The British Journal of Psychiatry 191, 493–9.
Ho J. S., Radoeva P. D., Jalbrzikowski M., Chow C., Hopkins J., Tran W. C. et al. (2012) Deficits in mental state attributions in individuals with 22q11.2 deletion syndrome (velo-cardio-facial syndrome). Autism Research 5, 407–18.
Emerson E. & Hatton C. (2008) Socioeconomic disadvantage, social participation and networks and the self-
Jalbrzikowski M., Carter C., Senturk D., Chow C., Hopkins J. M., Green M. F. et al. (2012) Social
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 858 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
cognition in 22q11.2 microdeletion syndrome: relevance to psychosis? Schizophrenia Research 142, 99–107. Kaufman J., Birmaher B., Brent D., Rao U., Flynn C., Moreci P. et al. (1997) Schedule for affective disorders and schizophrenia for school-age children-present and lifetime version (K-SADS-PL): initial reliability and validity data. Journal of the American Academy of Child and Adolescent Psychiatry 36, 980–8. Kiley-Brabeck K. & Sobin C. (2006) Social skills and executive function deficits in children with the 22q11 deletion syndrome. Applied Neuropsychology 13, 258–68. Langdon R., Coltheart M. & Ward P. (2006) Empathetic perspective-taking is impaired in schizophrenia: evidence from a study of emotion attribution and theory of mind. Cognitive Neuropsychiatry 11, 133–55. Leitman D. I., Laukka P., Juslin P. N., Saccente E., Butler P. & Javitt D. C. (2010) Getting the cue: sensory contributions to auditory emotion recognition impairments in schizophrenia. Schizophrenia Bulletin 36, 545– 56. Lewandowski K., Shashi V., Berry P. & Kwapil T. (2007) Schizophrenic-like neurocognitive deficits in children and adolescents with 22q11 deletion syndrome. American Journal of Medical Genetics. Part B, Neuropsychiatric Genetics 144, 27–36. Looman W. S., Thurmes A. K. & O’Conner-Von S. K. (2010) Quality of life among children with velocardiofacial syndrome. Cleft Palate Craniofacial Journal 47, 273–83. McCabe K., Rich D., Loughland C. M., Schall U. & Campbell L. E. (2011) Visual scanpath abnormalities in 22q11.2 deletion syndrome: is this a face specific deficit? Psychiatry Research 189, 292–8. McCabe K. L., Melville J. L., Rich D., Strutt P. A., Cooper G., Loughland C. M. et al. (2013) Divergent patterns of social cognition performance in autism and 22q11.2 deletion syndrome (22q11DS). Journal of Autism and Developmental Disorders 43, 1926–34. Miller S. J. & Marcovitch S. (2012) How theory of mind and executive function co-develop. Review of Philosophical Psychology 3, 597–625. Miyake A., Friedman N., Emerson M., Witzki A., Howerter A. & Wagner T. (2000) The unity and diversity of executive functions and their contributions to complex ‘frontal lobe’ tasks: a latent variable analysis. Cognitive Psychology 41, 49–100. Murphy K., Jones L. & Owen M. (1999) High rates of schizophrenia in adults with velo-cardio-facial syndrome. Archives of General Psychiatry 56, 940–5. Niklasson L., Rasmussen P., Oskarsdottir S. & Gillberg C. (2002) Chromosome 22q11 deletion syndrome (CATCH 22): neuropsychiatric and neuropsychological
aspects. Developmental Medicine and Child Neurology 44, 44–50. Nyden A., Hjelmquist E. & Gillberg C. (2000) Autism spectrum and attention-deficit disorders in girls. Some neuropsychological aspects. European Child & Adolescent Psychiatry 9, 180–5. Roizen N., Antshel K., Fremont W., Abdulsabur N., Higgins A. M., Shprintzen R. J. et al. (2007) 22q11.2DS deletion syndrome: developmental milestones in infants and toddlers. Journal of Developmental & Behavioral Pediatrics 28, 119–24. Scambler P. J., Kelly D., Lindsay E., Williamson R., Goldberg R., Shprintzen R. et al. (1992) Velo-cardiofacial syndrome associated with chromosome 22 deletions encompassing the DiGeorge locus. Lancet 339, 1138–9. Shashi V., Veerapandiyan A., Schoch K., Kwapil T. R., Keshavan M. S., Ip E. et al. (2012) Social skills and associated psychopathology in children with chromosome 22q11.2 deletion syndrome: implications for interventions. Journal of Intellectual Disability Research 56, 865–78. Shprintzen R. J. (2008) Velo-cardio-facial syndrome: 30 years of study. Developmental Disabilities Research Reviews 14, 3–10. Simon T. J. (2008) A new account of the neurocognitive foundations of impairments in space, time and number processing in children with chromosome 22q11.2 deletion syndrome. Developmental Disabilities Research Reviews 14, 52–8. Sinderberry B., Brown S., Hammond P., Stevens A. F., Schall U., Murphy D. G. et al. (2013) Subtypes in 22q11.2 deletion syndrome associated with behaviour and neurofacial morphology. Research in Developmental Disabilities 34, 116–25. Sobin C., Kiley-Brabeck K., Daniels S., Khuri J., Taylor L., Blundell M. et al. (2005) Neuropsychological characteristics of children with the 22q11 deletion syndrome: a descriptive analysis. Child Neuropsychology: A Journal on Normal and Abnormal Development in Childhood and Adolescence 11, 39–53. Swillen A., Devriendt K., Legius E., Eyskens B., Dumoulin M., Gewillig M. et al. (1997) Intelligence and psychosocial adjustment in velocardiofacial syndrome: a study of 37 children and adolescents with VCFS. Journal of Medical Genetics 34, 453–8. Swillen A., Vandeputte L., Cracco J., Maes B., Ghesquiere P., Devriendt K. et al. (1999) Neuropsychological, learning and psychosocial profile of primary school aged children with the velocardio-facial syndrome (22q11 deletion): evidence for a nonverbal learning disability? Child Neuropsychology 5, 230–41. Swillen A., Devriendt K., Ghesquiere P. & Fryns J. (2001) Children with a 22q11 deletion versus children with a
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 859 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
speech-language impairment and learning disability: behavior during primary school age. Genetic Counseling 12, 309–17. Wechsler D. (1999) Wechsler Abbreviated Scale of Intelligence. The Psychological Corporation, San Antonio, TX.
Woodin M., Wang P., Aleman D., McDonald-McGinn D., Zackai E. & Moss E. (2001) Neuropsychological profile of children and adolescents with the 22q11.2 microdeletion. Genetics in Medicine 3, 34–9.
Accepted 30 November 2014.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research 845
doi: 10.1111/jir.12183
volume 59 part 9 pp 845–859 septemBer 2015
Social cognition dysfunction in adolescents with 22q11.2 deletion syndrome (velo-cardio-facial syndrome): relationship with executive functioning and social competence/functioning L. E. Campbell,1,2,3 K. L. McCabe,1,4,6 J. L. Melville,3 P. A. Strutt1,3 & U. Schall1,2,5,6 1 Centre for Translational Neuroscience and Mental Health Research, University of Newcastle, McAuley Centre, Waratah, NSW 2298, Australia 2 Hunter Medical Research Institute, Locked Bag 1, Hunter Region Mail Centre NSW 2310, Australia 3 School of Psychology, University of Newcastle, Ourimbah, NSW 2258, Australia 4 Brain & Mind Research Institute, University of Sydney, Camperdown, NSW 2050, Australia 5 School of Medicine and Public Health, University of Newcastle, Callaghan, NSW 2308, Australia 6 Schizophrenia Research Institute, 405 Liverpool St Darlinghurst NSW 2010, Australia
Abstract Background Social difficulties are often noted among people with intellectual disabilities. Children and adults with 22q.11.2 deletion syndrome (22q11DS) often have poorer social competence as well as poorer performance on measures of executive and social-cognitive skills compared with typically developing young people. However, the relationship between social functioning and more basic processes of social cognition and executive functioning are not well understood in 22q11DS. The present study examined the relationship between social-cognitive measures of emotion attribution and theory of mind with executive functioning and their contribution to social competence in 22q11DS.
Correspondence: Dr Linda Campbell, School of Psychology, University of Newcastle, Science Offices, Ourimbah, NSW 2258, Australia (e-mail: [email protected]).
Method The present cross-sectional study measured social cognition and executive performance of 24 adolescents with 22q11DS compared with 27 age-matched typically developing controls. Social cognition was tested using the emotion attribution task (EAT) and a picture sequencing task (PST), which tested mentalising (false-belief), sequencing, cause and effect, and inhibition. Executive functioning was assessed using computerised versions of the Tower of London task and working memory measures of spatial and non-spatial ability. Social competence was also assessed using the parent-reported Strengths and Difficulties Questionnaire. Results Adolescents with 22q11DS showed impaired false-belief, emotion attribution and executive functioning compared with typically developing control participants. Poorer performance was reported on all story types in the PST, although, patterns of errors and response times across story types were similar in both groups. General sequencing ability was the strongest
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 846 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
predictor of false-belief, and performance on the false-belief task predicted emotion attribution accuracy. Intellectual functioning, rather than theory of mind or executive functioning, predicted social competence in 22q11DS. Conclusions Performance on social-cognitive tasks of theory of mind indicate evidence of a general underlying dysfunction in 22q11DS that includes executive ability to understand cause and effect, to logically reason about social scenarios and also to inhibit responses to salient, but misleading cues. However, general intellectual ability is closely related to actual social competence suggesting that a generalised intellectual deficit coupled with more specific executive impairments may best explain poor social cognition in 22q11DS. Keywords emotion attribution/recognition, executive function, intellectual disability, social cognition, theory of mind, velo-cardio-facial syndrome
Introduction Healthy social relationships are integral to a good quality of life. The ability to initiate, form and maintain social relationships is dependent on the successful integration of social-cognitive factors including skills in understanding, reasoning and predicting other people’s emotions and behaviours. Social difficulties and isolation are often noted among people with intellectual disabilities (IDs) (Emerson & Hatton 2007). Poor social functioning severely limits opportunities for a full and productive life leading to great social and economic burden not only to the individual, but also for families and society at large (Emerson & Hatton 2008). In addition, social dysfunction is one of the many risk factors for behavioural and psychiatric problems that young people with IDs experience (Dykens 2000). Indeed, persons with IDs are estimated to be three to four times more likely than those in the general population to experience an emotional, behavioural or psychiatric disorder (Emerson 2003; Emerson & Hatton 2007). 22q11.2 deletion syndrome (22q11DS; also known as velo-cardio-facial syndrome) results from a de novo or familial deletion of 30–50 genes on the long arm of chromosome 22 (Driscoll et al. 1992; Scambler et al. 1992). The syndrome is the most
common microdeletion syndrome in humans (Botto et al. 2003) and has been estimated to occur as often as between 1 in 1600 and 1 in 2000 live births (Shprintzen 2008), and although very variable, is associated with physical anomalies such as cardiac abnormalities, velopharyngeal insufficiency, hypoparathyroidism and immune deficiency (for a review see e.g., Shprintzen 2008). In addition, the vast majority of people with 22q11DS have cognitive impairments and IDs (Swillen et al. 1997), and similar to other developmental disabilities, often experience behavioural and psychiatric disorders including anxiety, attention-deficit/hyperactivity disorder and autism spectrum disorders (Feinstein et al. 2002). Moreover, 22q11DS is one of the highest known risk factors for psychosis in adolescence and early adulthood (Murphy et al. 1999). Over the past two decades, it has been well documented that children and adults with 22q11DS often have poorer social competence including, for example, mood lability, shyness, and difficulties in initiating and maintaining social relationships compared not only with their typically developing (TD) peers (Golding-Kushner et al. 1985; Fuerst et al. 1995; Swillen et al. 2001; Woodin et al. 2001), but also with other children with chronic illnesses (Looman et al. 2010). While early retrospective parental reports do not identify delays in achieving social developmental milestones (Roizen et al. 2007), later in childhood, parents are often concerned by the problems their children have with peers (Campbell et al. 2011), and as teenagers and young adults, parents often report increasing social withdrawal and isolation (Swillen et al. 1997, 2001). Social problems can be caused by, and be associated with, a range of biopsychosocial factors including speech problems, anxiety and bullying (Shashi et al. 2012); however, they can also be caused by deficits in ‘social cognition’; that is, the set of mental processes that govern how people think about themselves, others, social situations and social interaction (Green & Leitman 2008). Niklasson et al. (2002) reported that children and adults with 22q11DS may have deficits in the social-cognitive skill known as ‘theory of mind’ (ToM). ToM tasks test the ability to go beyond observed behaviour to explain actions in terms of inferred mental states. While Niklasson et al. (2002) measured ToM using classical false-belief tasks (such as the Smarties and
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 847 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
Sally-Anne tasks; (Baron-Cohen et al. 1985); Strange Stories (Happé 1994); and Non-mental and Mental Cartoons (Nyden et al. 2000), they did not report any supporting statistics for ToM deficits in 22q11DS compared with typically developed controls, other than to indicate that in most 22q11DS participants (n = 15/20), poor performance on measures of false-belief were observed. Subsequently, our group has investigated ToM skills among children with 22q11DS compared with age-matched siblings and age- and intelligence quotient (IQ)matched children with Williams syndrome (Campbell et al. 2009, 2011). We did not identify specific ToM performance deficits in children with 22q11DS compared with children with Williams syndrome on false-belief tasks, and although the children with 22q11DS performed poorer on the Strange Stories task, these difficulties may have been influenced by low comprehension as suggested by low performance on non-mentalising stories as well. In addition, while the younger children with 22q11DS in that study performed poorer, primarily on second-order false-belief tasks (which require a person to make inferences about someone’s belief about another’s belief) compared with TD siblings, we concluded that the findings were likely indicative of delayed development of ToM rather than a deficit. On reflection, tasks that are reliant on verbal comprehension may not be optimal measures of ToM for people who report difficulties within both of these domains. Instead, using implicit tasks may better delineate how well people with 22q11DS comprehend mentalising skills. Adopting this approach, Ho et al. (2012), using the Animations task (Abell et al. 2000; Castelli et al. 2000), found that participants with 22q11DS scored lower than TD controls when required to explain purposeful behaviour and when asked to describe vignettes requiring mentalising (designed to imply complex mental states), but not on the random control vignettes. Mentalising ability was positively correlated with a functional measure of reciprocal social behaviour (Ho et al. 2012). In addition, Jalbrzikowski et al. (2012) utilised videotaped vignettes of everyday social interactions to investigate the ToM ability of participants to understand the intentions of others, with a particular focus on white lies and sarcasm. As expected, the 22q11DS
group performed poorer than healthy controls, and it was also reported that performance on the ToM task was a strong predictor of psychosis positive symptom severity (Jalbrzikowski et al. 2012). This is important, as Chow et al. (2006) reported that adults with 22q11DS and schizophrenia (compared with people with 22q11DS, but no diagnosis of schizophrenia) performed worse on a ToM task in which the participants were required to make social inferences; indicating that ToM ability may possess endophenic marker characteristics of psychosis proneness. Interestingly, it has been proposed that the developmental progression of ToM and executive functions (EF) follow similar trajectories in early childhood and that these two cognitive skills are closely related to one another (Miller & Marcovitch 2012). Executive processes underpin control of thought and action, and involve three main components: the ability to retain and to manipulate information in the mind (working memory), to suppress inappropriate responses (inhibition) and to be able to change behaviour in response to new information (set-shifting; Miyake et al. 2000). Whether working memory is a component of EF or a separate set of processes remains an ongoing source of debate. For the purposes of the present study, working memory is considered a component of EF. Specifically, the working memory processes needed to hold relevant information in mind to guide lower-level processes towards executing a goal (Miller & Marcovitch 2012). EFs such as working memory and setshifting are impaired in 22q11DS (Swillen et al. 1999; Woodin et al. 2001; Henry et al. 2002; Sobin et al. 2005; Lewandowski et al. 2007; Campbell et al. 2010b). Furthermore, children with 22q11DS perform poorer at non-verbal EF tasks that require sequencing, integration of patterns, and an understanding of spatial relationships (Sobin et al. 2005). It is likely that social problems are related to a range of cognitive mechanisms among young people with 22q11Ds; indeed, using parent-rating scales, Kiley-Brabeck & Sobin (2006) report a link between executive dysfunction and social skills among children with 22q11DS. More specifically, the skills of self-starting tasks or problem solving on one’s own, and the ability to monitor own work or behaviour, significantly predicted social skills.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 848 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
Over the last few years there has been an increasing focus to design remediation strategies to assist in improving social skills among young people with 22q11DS; one example of such a programme is visá-vis (Glaser et al. 2012). This programme aimed to improve face processing skills among children with IDs. However, in order to design similar programmes with a wider scope of social skills it is important to investigate if ToM should be considered as the primary focus for intervention. Therefore, we must first ascertain whether ToM can predict social competence in the context of other known predictors, such as EFs. Earlier studies conducted by members of our group (Campbell et al. 2010b) indicate strong positive correlations among Tower of London (ToL) and set-shifting tasks with impaired performance observed on both tasks by people with 22q11DS. Despite set-shifting being a core component of EF, set-shifting was not assessed in this study. Specifically, the present study aimed to determine what the strength of the relationship is between ToM and EF in 22q11DS. On the basis of existing literature, we hypothesised that: (1) adolescents with 22q11DS would show poorer performance on measures of ToM [indexed by picture sequencing (false-belief) and emotion attribution tasks (EAT)] and EF (as indexed by working memory, the ToL task and general sequencing ability); (2) EF would predict performance on ToM tasks; and (3) both ToM ability and EF would significantly predict social competence (as indexed by parent-rated peer competence from the Strengths and Difficulties Questionnaire) in 22q11DS.
Methods Participants A total of 24 adolescents with 22q11DS (10 males, 14 females) and 27 TD adolescent controls (13 males, 14 females) were recruited in New South Wales, Australia. Participants with 22q11DS were aged 12–21 years [mean age = 16.75 years; standard deviation (SD) = 3.14] and had a mean fullscale IQ of 75.88 (SD = 14.93). The TD control group were aged 8–22 years (mean age = 16.26 years; SD = 3.65) and had a mean full-scale IQ of 108.48 (SD = 14.21). TD control participants were
recruited from the local community (n = 16) or were siblings (n = 11) of the 22q11DS participants. Written informed consent was given by all parents/ guardians, and participants provided assent when applicable. This study was granted ethical approval by the University of Newcastle’s Human Research Ethics Committee. Participants completed structured diagnostic interviews to assess for common mental health diagnoses [Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders – Fourth Edition (Text Revision) Axis I Disorders (SCID; First et al. 2002) and Schedule for Affective Disorders and Schizophrenia for School-Age Children – Present and Lifetime Version (K-SADS-PL; Kaufman et al. 1997) ]. Of the 22q11DS participants, the following diagnoses were reported: mild ID (n = 15); anxiety disorders (n = 9; i.e., obsessive– compulsive disorder, generalised anxiety disorder, agoraphobia with/out panic disorder, panic disorder without agoraphobia, separation anxiety, paroxysmal anxiety, other anxiety symptoms); attentiondeficit hyperactivity disorder (n = 3); oppositional defiant disorder (n = 3); dysthymic disorder (n = 3); prodromal symptoms (n = 2); insomnia (n = 2); schizotypal disorder (n = 1); depressive symptoms (n = 1); schizophrenia-catatonic (n = 1); trichotillomania (n = 1); and epilepsy (n = 1). Six participants reported no psychological diagnoses. Furthermore, of the 22q11DS participants, the following medications were being used at the time of testing: methylphenidate (n = 4), selective serotonin reuptake inhibitor (n = 3), risperidone (n = 2), valproate (n = 1), amitriptyline (n = 1), clonidine (n = 1), cogentin (n = 1), unspecified antipsychotics with mood stabilisers (n = 1) and antiepileptic medication (n = 1). Fifteen participants were receiving no current medication at the time of testing. The inclusion and exclusion criteria for the study have been previously reported for the 22q11DS group (Campbell et al. 2010a; McCabe et al. 2011). In brief, TD control participants were excluded from the study if they had an ID/developmental delay or any history of neurological disorders or Axis 1 psychiatric disorders. All participants underwent a structured diagnostic interview based on the SCID (research version; First et al. 2002) and the K-SADS-PL (Kaufman et al. 1997) conducted by a psychiatrist (Author: US) at the time of testing.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 849 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
Experimental tasks and procedure Assessment of intellectual functioning The Wechsler Abbreviated Scale of Intelligence (WASI; Wechsler 1999) was used to assess participants’ general intellectual functioning. Social competence and emotional well-being Parental ratings of the Strengths and Difficulties Questionnaire (SDQ; Goodman et al. 2000) were collected for each participant. The SDQ is a brief (25 items) behavioural screening questionnaire that uses a 3-point Likert scale (‘not true’, ‘somewhat true’ or ‘certainly true’) to assess: emotional symptoms, conduct problems, hyperactivity/inattention, peer relationship problems and prosocial behaviour. The SDQ is a valid and reliable measure of adjustment and psychopathology in children and adolescents (Goodman et al. 2000). Social cognition and theory of mind Emotion attribution task. Emotion attribution ability was assessed using a shortened version of the EAT reported previously by Langdon et al. (2006) Participants were shown five cartoon strips (as in Langdon et al. 2006) presented in a fixed order of increasing length and difficulty level, in which the characters’ faces were blank. Placed underneath these, in predetermined incorrect order, were a series of facial-affect cards (depicting happiness, sadness, anger and shock). Participants were asked to rearrange the facial-affect cards according to which emotions they thought matched the characters in the cartoon strips. Correctly and incorrectly positioned facial-affect cards were given a score of 1 or 0, respectively. On completion of this task, participants were asked to name the emotion shown in each of the facial-affect cards. The mean number of correctly identified emotions was calculated for each participant. Picture sequencing task. Picture sequencing task (PST) (see Langdon et al. 2006, for a more detailed description) is used to assess selective ToM impairments. It is comprised of four story types: one ToM/mentalising story type (false-belief) and three control story types (social-script, mechanical and capture). False-belief stories assess the ability to
understand that a story character’s actions were due to misinformation about a situation. Social-script stories control for ability to logically sequence commonly occurring social events/behaviour. Mechanical stories control for ability to understand physical cause-and-effect relationships. Capture stories control for ability to inhibit misleading information. As noted by Langdon et al. (2006) capture stories are an important control for two reasons: (1) they require a similar level of processing to the false-belief stories and (2) they are of higher difficulty so as to prevent any group differences in false-belief scores being explained purely by task difficulty. As well as providing information about selective ToM impairments individually, the three control story types, when taken together, provide information about participants’ general sequencing ability, which is an important control to ensure that any group differences in false-belief scores are not better accounted for by general EF differences. Participants completed two practice trials followed by four experimental trials (i.e., four versions of each story type), with four picture cards per story. In a fixed pseudo-random incorrect order, the cards were placed face-down and participants were asked to turn each card and rearrange them to form a logical sequence of events. For each sequence, two points were given for correctly positioning the first card, two points for the last card, and one point for the second and third cards (total six points available). Card position scores and response times (RTs) were averaged across each story type to give a mean position score and RT per story type. Executive functioning Planning task. We designed a computerised planning task based on the ToL task to assess participants’ planning ability. Briefly, participants viewed a start configuration of three pegs of differing heights and three balls of differing colours, and were instructed to rearrange the balls into a goal configuration using a touch screen. The start and goal configurations were shown simultaneously in the upper and lower half of the screen and there was no time limit for completion. The balls could only be moved one at a time and if there was no other ball on top. Three balls could be placed on the longest peg, two on the medium peg, and one on the shortest peg.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 850 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
The number of moves necessary for completion ranged from two to seven. Variables recorded were moves above minimum, initial thinking time and subsequent thinking/move time. Participants were given feedback for passing a trial (smiley face) or for failing the trial (sad face). Additionally, all participants completed a motor control task in order to control for motor movement time; these trials were not included in the data analysis. Working memory. Participants performed one visuospatial (location) and two non-spatial working memory tasks on a touch screen. The visuo-spatial task consisted of a sequence of two to nine stimuli (shapes and faces with a neutral expression; size 100 × 128 pixels) that appeared sequentially at random, non-symmetrical locations for 2 s, with no inter-stimulus interval. Empty blue boxes then appeared in the same location as the preceding stimuli and participants were asked to touch the boxes in the same order as the stimuli had appeared. In both non-spatial tasks, participants were presented with one image (either complex shapes or faces) at a time, centrally on the computer screen. The stimuli were presented in a random order at the bottom of the computer screen and participants were asked to touch the stimuli in the order that they had appeared in. If successful in three same-length sequences, one more stimulus was introduced up to a maximum of nine stimuli/ levels. Stimulus size and delivery rate were identical in all three tasks and working memory accuracy and span (defined as the maximum number of levels completed; i.e., two to nine) were recorded for each task.
Analysis Demographic information and intellectual functioning (WASI) were examined using independent samples t-tests and chi-square analysis. A (2 × 4) mixed between-within subjects analysis of variance (ANOVA) was conducted to analyse mean scores from the PST. The between-factor had two levels for participant group (22q11DS vs. TD control) and the within-factor had four levels for story type (social-script vs. mechanical vs. capture vs. false-belief). An index of general sequencing ability was created by averaging the scores for social-script, mechanical
and capture stories. A one-way analysis of covariance (ANCOVA) was performed to assess possible effects of general sequencing ability on false-belief scores. EAT scores were analysed using a one-way ANOVA.1 A one-way ANCOVA was performed to assess the effect of emotion recognition accuracy on EAT scores. A one-way ANOVA was also used to assess scores on the planning task and working memory task. Multiple regression analysis was used to assess (1) the ability of general sequencing ability, emotion attribution ability, planning, working memory2 and full-scale IQ to predict performance on false-belief across groups; and (2) the ability of general sequencing ability, planning, false-belief, working memory, and full-scale IQ to predict performance on emotion attribution. The threshold set for significance was P < 0.05.
Results Demographic data The 22q11DS and TD control groups did not differ significantly on gender distribution or age (P’s > 0.05). The 22q11DS group reported significantly lower IQ scores than the TD group t(49) = −7.99, P < 0.001; IQ differences were greater than two SDs between the TD control and 22q11DS groups and were therefore considered a group-defining characteristic. Table 1 describes means, SDs and patterns of significance.
Strengths and Difficulties Questionnaire When assessing SDQ scores, a significant main effect was found for group, F(1, 47) = 28.59, P < 0.001, partial eta squared = 0.38; and for subscale type F(4, 44) = 34.63, P = 0.001, partial eta squared = 0.76. Further, a significant group by subscale interaction effect was found, F(4, 44) = 11.16, P < 0.001, partial eta squared = 0.5. Compared with the TD control group, the 22q11DS group scored particularly high for emo1 Response times for the EAT were not assessed because of faulty timing data for a large proportion of the participants. 2 The working memory sub-tests were highly correlated with one another; therefore, where working memory was entered as a covariate in analysis, we used only the face accuracy sub-scale of this task.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 851 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
22q11DS
Gender (% male) Age (years) Intelligence quotient (WASI)** SDQ Emotional symptoms** Conduct problems Hyperactivity** Peer problems** Prosocial behaviour†
TD control
Mean (SD)
Min–Max
Mean (SD)
Min–Max
41.7 16.75 (3.14) 75.88 (14.93)
N/A 12–21 56–115
48.1 16.26 (3.65) 108.48 (14.21)
N/A 8–22 83–141
5.33 (2.96) 2.04 (1.88) 4.63 (2.06) 4.54 (2.45) 6.79 (2.78)
1–10 0–6 0–8 1–10 0–10
1.64 (2.27) 1.36 (1.63) 2.36 (2.08) 1.40 (1.94) 7.76 (2.07)
0–8 0–5 0–7 0–7 2–10
Table 1 Means and SDs of demographic data for 22q11DS and control groups
* P < 0.05, ** P < 0.001. † Reversed scored. N/A, not applicable; SD, standard deviation; SDQ, Strengths and Difficulties Questionnaire; TD, typically developing; WASI, Wechsler Abbreviated Scale of Intelligence.
tional symptoms, hyperactivity/inattention, and peer relationship problems (Ps < 0.001), but scored comparably on conduct problems and prosocial behaviour (Ps > 0.1). See Table 1 for means and SDs.
Picture sequencing task Position scores When assessing mean position scores, there was a significant main effect of group, F(1,49) = 23.51, P < 0.001, partial eta squared = 0.32; and story type, F(3, 47) = 37.88, P < 0.001, partial eta squared = 0.71; however, there was no significant group by story type interaction, F(3, 47) = 0.71, P = 0.55, partial eta squared = 0.04. Participants with 22q11DS made significantly more positioning errors than TD controls across all story types (Ps ≤ 0.005). All participants made the most positioning errors for capture, followed by false-belief and social-script, and made the least errors for mechanical. Furthermore, across all participants, positioning scores differed significantly between almost all story types (Ps ≤ 0.001 for social-script/false-belief, capture/ false-belief, social-script/capture, and capture/ mechanical; and P ≤ 0.01 for false-belief/mechanical) with the exception of scores between social-script and mechanical stories (P = 0.13). See Table 2 for means, SDs and patterns of statistical significance.
An ANCOVA was undertaken to assess the effect of general sequencing ability on the false-belief scores of 22q11DS and TD control participants. This revealed that general sequencing ability was a significant predictor of false-belief scores, F(1, 48) = 21.94, P < 0.001, and after correcting for general sequencing ability, the difference in falsebelief scores between 22q11DS participants and TD controls was no longer significant (F < 1).
Response times To assess mean RTs for correctly sequenced trials, a one-way ANOVA was performed due to missing data for 12 participants.3 Participants with 22q11DS were significantly slower than TD controls at sequencing the false-belief F(1, 45) = 5.91, P < 0.05, social-script F(1, 48) = 10.85, P < 0.01, and mechanical F(1, 47) = 8.17, P < 0.01 stories; however, there was no difference between the two groups in RTs to
3 Twelve participants (11× 22q11DS and 1× TD control) had RTs excluded for at least one story type because of having no correctly sequenced trials for that story type. The control participant had their capture RTs excluded, and for the 22q11DS participants, RTs were excluded for 1× social-script, 2× mechanical, 5× false-belief and 10× capture stories.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 852 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
Table 2 Group means and standard deviations for the picture sequencing task and emotion attribution task
22q11DS Mean (SD) Picture sequencing task Position scores False-belief** Social-script*** Mechanical*** Capture*** Response times (seconds) False-belief* Social-script** Mechanical** Capture Emotion attribution task Emotion identification accuracy** Overall emotion attribution accuracy
TD control Min–Max
Mean (SD)
Min–Max
16.67 (6.63) 19.79 (4.43) 18.54 (5.49) 10.83 (3.89)
1–24 6–24 3–24 0–17
20.96 (3.32) 23.04 (1.83) 22.52 (2.68) 16.26 (5.24)
12–24 18–24 14–24 8–24
26.31 (8.04) 18.56 (4.69) 23.16 (10.14) 29.89 (10.84)
15.95–45.71 13.22–29.99 11.29–49.35 11.75–54.63
19.89 (9.56) 14.10 (4.84) 16.44 (6.16) 29.33 (12.51)
8.95–61.60 7.16–28.45 6.75–39.75 13.65–61
23.54 (2.06) 2.54 (1.69)
18–27 0–5
25.77 (3.05) 3.96 (1.09)
16–29 1–5
* P < 0.05, ** P < 0.01, *** P ≤ 0.001. SD, standard deviation; TD, typically developing.
sequence the capture stories (P = 0.89; see Table 2 for means, SDs and range of RT scores). Across all participants, RTs were significantly different among all four story types (Ps < 0.001 for social-script/ mechanical, social-script/false-belief, social-script/capture and false-belief/mechanical, and P ≤ 0.05 for capture/ false-belief and capture/mechanical). All participants were quickest to correctly sequence social-script stories followed by mechanical stories, then falsebelief stories and were slowest to correctly sequence capture stories.
Emotion attribution task Emotion identification accuracy Participants with 22q11DS made significantly more errors in identifying the facial affect of the cartoon characters compared with TD control participants, F(1, 48) = 8.99, P < 0.01. See Table 2 for means and SDs.4
4 Emotion identification accuracy score is missing for one TD control participant.
Emotion attribution accuracy Participants with 22q11DS were significantly less accurate on emotion attribution than the TD control participants, F(1,49) = 12.98, P = 0.001. See Table 2 for means and SDs. Emotion identification accuracy was included in an ANCOVA analysis to assess whether emotion identification accuracy had an effect on the ability to correctly attribute emotion. The results showed that emotion identification accuracy was a significant predictor of emotion attribution scores, F(1, 47) = 5.78, P < 0.05, and after correcting for emotion identification ability, the difference in emotion attribution scores between 22q11DS participants and TD controls lowered, but remained statistically significant, F(1, 47) = 6.51, P = 0.01. General sequencing ability was then included as a covariate to examine its effect on emotion attribution accuracy scores between the two groups. The results showed that general sequencing ability was a significant predictor of emotion attribution scores, F(1, 48) = 5.87, P < 0.05. After adjusting for general sequencing ability, the difference in emotion attribution scores between 22q11DS participants and TD controls were no longer significant, F(1, 48) = 2.63, P = 0.11.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 853 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
Table 3 Group means and standard deviations for the planning task and working memory task
22q11DS
Planning task Pre-planning time for correct trials (ms)** Move time for correct trials (ms)* Number of moves above minimum*** Working memory task Spatial accuracy*** Object accuracy*** Facial accuracy*** Spatial span*** Object span*** Facial span***
TD control
Mean (SD)
Min–Max
Mean (SD)
Min–Max
7539.35 (2639.76) 34518.87 (45969.07) 33.95 (14.89)
4718–13109.67 3116.5–200352.25 15–74
12186.22 (7752.6) 13483.69 (1478.89) 14.12 (8.28)
3834.27–35248.13 1518.93–37986.15 1–36
13.88 (3.84) 11.44 (3.42) 11.24 (3.03) 6.04 (1.4) 5.16 (1.25) 11.24 (3.03)
1–21 4–17 5–16 1–8 3–7 5–16
9.27 (4.2) 6.64 (3.43) 5.91 (2.65) 4.41 (1.47) 3.5 (1.26) 5.91 (2.65)
1–16 3–16 1–9 2–7 2–7 1–9
* P < 0.05, ** P ≤ 0.01, *** P < 0.001. SD, standard deviation; TD, typically developing.
Working memory variable
1
2
3
4
5
1 2 3 4 5 6
– 0.983*** 0.634*** 0.663*** 0.541*** 0.564***
– 0.646*** 0.674*** 0.561*** 0.586***
– 0.982*** 0.621*** 0.659***
– 0.602*** 0.635***
– 0.978***
Spatial accuracy Spatial span Object accuracy Object span Facial accuracy Facial span
Table 4 Pearson’s r correlations between all variables of the working memory task, n = 47
*** P < 0.001.
Executive functioning Planning task A one-way ANOVA revealed significant group differences on all three levels of the planning task (see Table 3 for means and SDs). Compared with TD controls, 22q11DS participants spent significantly less time planning their moves F(1, 46) = 7.18, P ≤ 0.01, but took a longer time F(1, 46) = 5.27, P < 0.5 and made a greater number of moves F(1, 46) = 33.87, P < 0.001 in completing the task.
trols on all types of the working memory task (spatial = F(1, 45) = 15.42, P < 0.001; object = F(1, 45) = 23.05, P < 0.001; facial = F(1, 45) = 40.62, P < 0.001). Consequently, 22q11DS participants had significantly shorter working memory span than TD controls across working memory tasks (spatial = F(1, 45) = 15.18, P < 0.001; object = F(1, 45) = 20.48, P < 0.001; facial = F(1, 45) = 35.68, P < 0.001). See Table 3 for means and SDs. A Pearson’s r analysis revealed that all accuracy and span scores were positively correlated with one another (see Table 4).
Working memory A one-way ANOVA revealed that 22q11DS participants were significantly less accurate in remembering the location of stimuli than TD con-
Multiple regression analysis Preliminary analyses were conducted to ensure no violation of the assumptions of normality, linearity,
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 854 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
multicollinearity and homoscedasticity. Multiple regression was used, firstly to assess the ability of general sequencing, emotion attribution, planning, working memory and full-scale IQ to predict performance on false-belief across groups. The total variance explained by the model as a whole was 56.5%, F(6, 45) = 8.45, P < 0.001. The variable with the strongest significant unique contribution to explaining performance of false-belief was general sequencing ability beta = 0.44, P < 0.01 while emotion attribution ability was slightly lower beta = 0.28, P = 0.06. Secondly, the ability of general sequencing ability, planning, false-belief, working memory and fullscale IQ to predict performance on emotion attribution ability were investigated. The total variance explained by the model as a whole was 52.1%, F(5, 45) = 8.69, P < 0.001. The variable with the strongest significant unique contribution to explaining differences in emotion attribution was false-belief beta = −33, P < 0.05. Thirdly, across groups, the variables mentioned earlier were included in a model to investigate the predictive ability of peer problems. The variable with the strongest significant unique contribution to explaining differences in peer relationships was intellectual functioning beta = −0.72, P < 0.005, while working memory was the second largest beta = 0.35, P = 0.09. Follow-up within group multiple regression analyses revealed that the best significant predictor of false-belief in the 22q11DS group was emotion attribution beta = 0.39, P < 0.05. The total variance explained by the model was 65.6%, F(4, 23) = 9.06, P < 0.001. Meanwhile, the best predictor of emotion attribution in the 22q11DS group was false-belief beta = 0.42, P = 0.12. The total variance explained by the model was 65.6%, F(5, 20) = 3.67, P < 0.05. Subsequent analyses revealed that the best predictors of peer relationships in the 22q11DS group were working memory (indexed using the face accuracy subscale; beta = 0.55, P < 0.02, intellectual functioning beta = −0.94, P < 0.005, and emotion attribution beta = 0.69, P < 0.05. The total variance explained by the model was 79.9%, F(7, 20) = 8.24, P < 0.05. The multiple regression analysis for the TD group did not reach statistical significance (P > 0.2).
Discussion 22q11DS is associated with a range of cognitive and social cognition difficulties (Henry et al. 2002; Niklasson et al. 2002). The present study’s findings are consistent with previous research and support our first hypothesis by showing impairment of social cognition, as measured by false-belief and EATs, and EF, as measured by planning and working memory tasks, by individuals with 22q11DS compared with TD control participants. However, the measures employed in the current study were more sensitive and allowed for overall performance on the social cognitive tasks to be broken down into components. Our findings indicate that the socialcognitive impairments observed in 22q11DS are not specific to false-belief or indeed emotion attribution dysfunction. In the current study, we confirmed the findings of previous research that people with 22q11DS display impaired performance on a range of social cognition (Niklasson et al. 2002; Campbell et al. 2011; Ho et al. 2012; Jalbrzikowski et al. 2012) and EF (Swillen et al. 1999; Woodin et al. 2001; Henry et al. 2002; Sobin et al. 2005; Lewandowski et al. 2007; Campbell et al. 2010b) tasks. More specifically, the data revealed that the young people with 22q11DS performed poorer on each of the four story types in the PST indicating a general underlying dysfunction, including problems with understanding cause and effect, the ability to logically reason about social scenarios (without needing to understand false-beliefs), and also to inhibit responses to salient, but misleading cues. Although the 22q11DS group consistently performed at lower levels compared with TD controls, the errors and RTs across story types were similar in both groups. These findings suggest that the 22q11DS group displayed a generalised deficit coupled with slower responses. Thus, a methodological strength of the present study is the sensitivity of the tasks selected to examine ToM and EF. The suitability of EF tasks to groups with mild intellectual deficits is a necessary consideration for 22q11DS researchers. Langdon et al.’s (2006) PST is well suited to ID groups because it is designed in such a way that participants, to some degree, act as their own control. Specifically, the capture stories are incorporated to account for differential
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 855 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
sensitivity to increasing task difficulty, while in combination with social-script (logical sequencing) and mechanical (cause/effect) these components of the task act as an index of EF (general sequencing ability). The inclusion of these components permitted the examination of patterns and hierarchies of performance on these tasks showing that our 22q11DS group responded in the same way to our HC group, only slower and with less accuracy. Similarly, on the EAT, it was found that although the TD group were better at completing the task, the performance of the young people with 22q11DS was partially contingent on their ability to recognise emotions. Further analysis revealed that the strongest predictor of performance on the social cognition task examining ToM measures of false-belief (as indexed by the PST) was general sequencing ability, an approximate measure of EF, while the ability to understand false-belief, in turn, predicted the accuracy of emotion attribution. These findings confirmed our second hypothesis that deficits in social cognition in 22q11DS would be accounted for by impaired EF. Finally, contrary to our predictions, we found only partial support for our third hypothesis that both ToM ability and EF would predict social competence, with none of the measures of social cognition predicting social competence (as indexed by peer relationship problems in the SDQ). Instead, intellectual ability and working memory were the strongest unique predictors of peer relationships in 22q11DS. Performance by participants with 22q11DS on the social cognition tasks reported here indicates evidence of a general underlying dysfunction. Difficulty grasping cause and effect, interpreting social scenarios and inhibiting responses to salient, but misleading cues demand further exploration. Disturbance of lower-order sensory and cognitive functions may be associated with higher-order social cognition processing deficits reported in 22q11DS. For instance, Simon (2008) suggests that a ‘hypergranularity’ of attention processes in 22q11DS may best explain performance impairments on measures of attention, which they identify as associated with higher-order processes such as magnitude and numerical processing (Simon 2008). Similarly, studies in schizophrenia groups
describe relationships between sensory and cognitive impairment, which show associations between basic sensory processes such as pitch perception and emotion recognition (Leitman et al. 2010; Gold et al. 2012), as well as spatial frequency with object recognition (Calderone et al. 2013). Although the present findings did not focus on basic sensory processing, we did find support for the notion of generalised impairments, whereby impaired executive control may best explain impairment on tests of social cognition, rather than specific difficulties with social stimuli per se. Future research will benefit from the continued exploration of the relationship between basic cognitive and sensory processes and higher-order cognitions. Our finding that working memory was a significant predictor of social functioning problems (second to intellectual functioning) was somewhat unexpected. Although this finding is supported by earlier studies that report working memory as one of the strongest predictors of social skills in 22q11DS, but not sibling controls (Kiley-Brabeck & Sobin 2006), in the current study, we expected that ToM measures would be more likely to explain differences in peer relationship problems (our measure of social functioning). From a cognitive perspective, social functioning includes a complex network of processes that are reliant on working memory features such as the timely retrieval and integration of information under conditions that include competing responses and that are likely to be vulnerable to distraction. The limited capacity of people with 22q11DS to hold and manipulate information in working memory would likely impair their ability to react and respond to complex social cues in an efficient manner.
Clinical implications The findings from the present study suggest that conclusions of generalised impairment in 22q11DS are likely far more nuanced. The patterns of similarity between 22q11DS and TD control participants on the SDQ indicate that generalised impairment does not extend to some aspects of social functioning with similar rates of conduct problems and prosocial behaviour reported by 22q11DS participants as their age-matched peers. Coupled with
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 856 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
evidence of similar although poorer patterns of accuracy and slower reaction times on measures of emotion attribution, EF and ToM, these findings also raise important questions regarding whether EF and social cognition should be considered as single or separate constructs for the purpose of designing effective interventions in this population. As a neurodevelopmental disorder, 22q11DS is associated with a wide range of atypical cognitive functions and it seems unlikely that models of cognition, developed on the basis of TD people or indeed people with very specific cognitive difficulties such as autism, will offer an adequate explanation for the social problems among young people with 22q11DS. Indeed, while social perception, social knowledge, attribution bias and emotion processing are associated with ToM, they do depend on other cognitive mechanisms such as EF, all of which start to develop in very early childhood. Hence, in order to get a full picture of the social difficulties among children with 22q11DS, it will also be important to investigate cognitive and perceptual precursors to the higher-order social cognition deficits reported in 22q11DS.
Limitations As noted in previous studies conducted by our group (Campbell et al. 2010a; McCabe et al. 2013), significant differences in IQ between the 22q11DS and TD control groups were reported. 22q11DS is associated with significant intellectual impairment and is considered a group-defining characteristic. Thus, it was not controlled for statistically in our analysis. It is also necessary to consider the potential influences of medication on the measures used in this study. Several of the medications with potential sedative side effects (e.g., risperedone, valproate, amitriptyline) may have impaired performance. In addition, the possibility of different subtypes is worth considering when doing research into 22q11DS as this may affect our ability to reliably identify behavioural characteristics. This is particularly interesting as Sinderberry et al. (2013) identified that one 22q11DS subtype was primarily characterised by EF deficits. In the future, larger sample sizes and more sophisticated statistical analysis should be
carried out to identify specific behavioural phenotypes within the syndrome.
Conclusions The present study would seem to suggest support for a domain-general rather than domain-specific interpretation of social cognition deficits in 22q11DS because, while we have reported ToM deficits, when we accounted for the contribution of EF deficits, these were no longer significant. However, when this relates to measures of functioning (social competence indexed using a measure of peer problems), other domains (intellectual ability, working memory) best explain dysfunction.
Acknowledgements We would like to thank Dr Kathryn Leadbeater for her assistance in recruitment and testing for the current study and A/Prof Robyn Langdon for her advice during the preparation of this paper. Funding for the current study was received from the Hunter Medical Research Institute (HMRI) in the form of a Port-Waratah Coal Services post-doctoral fellowship, and from the National Health and Medical Research Council (NMMRC) in the form of an Australian Training Fellowship (455624). Infrastructure support was also obtained from the Schizophrenia Research Institute. The funding bodies had no input in the conduct of the study or in the analysis and reporting of the data. We thank our participants, with and without 22q11.2 deletion syndrome and their families, for their generosity of time and information, and support for our research.
References Abell F., Happe F. & Frith U. (2000) Do triangles play tricks? Attribution of mental states to animated shapes in normal and abnormal development. Cognitive Development 15, 1–16. Baron-Cohen S., Leslie A. M. & Frith U. (1985) Does the autistic child have a ‘theory of mind. Cognition 21, 37–46. Botto L., May K., Fernhoff P., Correa A., Coleman K., Rasmussen S. et al. (2003) A population-based study of
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 857 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
the 22q11.2 deletion: phenotype, incidence, and contribution to major birth defects in the population. Pediatrics 112, 101–7.
rated health of English men and women with mild and moderate intellectual disabilities: cross sectional survey. European Journal of Public Health 18, 31–7.
Calderone D. J., Hoptman M. J., Martinez A., Nair-Collins S., Mauro C. J., Bar M. et al. (2013) Contributions of low and high spatial frequency processing to impaired object recognition circuitry in schizophrenia. Cerebral Cortex 23, 1849–58.
Feinstein C., Eliez S., Blasey C. & Reiss A. (2002) Psychiatric disorders and behavioral problems in children with velocardiofacial syndrome: usefulness as phenotypic indicators of schizophrenia risk. Biological Psychiatry 51, 312–18.
Campbell L., McCabe K., Leadbeater K., Schall U., Loughland C. M. & Rich D. (2010a) Visual scanning of faces in 22q11.2 deletion syndrome: attention to the mouth or the eyes? Psychiatry Research 177, 211– 15.
First M. B., Spitzer R. L., Gibbon M. & Williams J. B. W. (2002) Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Research Version, Patient Edition With Psychotic Screen (SCID-I/P W/ PSY SCREEN). Biometrics Research, New York.
Campbell L. E., Stevens A., Daly E., Toal F., Azuma R., Karmiloff-Smith A. et al. (2009) A comparative study of cognition and brain anatomy between two neurodevelopmental disorders: 22q11.2 deletion syndrome and Williams syndrome. Neuropsychologia 47, 1034–44.
Fuerst K. B., Dool C. B. & Rourke B. P. (1995) Velocardiofacial syndrome. In: Syndrome of Nonverbal Learning Disabilities; Neurodevelopmental Manifestations (ed. B. Rourke), pp. 119–37. Guilford Press, New York.
Campbell L. E., Azuma R., Ambery F., Stevens A., Smith A., Morris R. G. et al. (2010b) Executive functions and memory abilities in children with 22q11.2 deletion syndrome. The Australian and New Zealand Journal of Psychiatry 44, 364–71.
Glaser B., Lothe A., Chabloz M., Dukes D., Pasca C., Redoute J. et al. (2012) Candidate socioemotional remediation program for individuals with intellectual disability. American Journal on Intellectual and Developmental Disabilities 117, 368–83.
Campbell L. E., Stevens A. F., McCabe K., Cruickshank L., Morris R. G., Murphy D. G. et al. (2011) Is theory of mind related to social dysfunction and emotional problems in 22q11.2 deletion syndrome (velo-cardiofacial syndrome). Journal of Neurodevelopmental Disorders 3, 152–61.
Gold R., Butler P., Revheim N., Leitman D. I., Hansen J. A., Gur R. C. et al. (2012) Auditory emotion recognition impairments in schizophrenia: relationship to acoustic features and cognition. American Journal of Psychiatry 169, 424–32.
Castelli F., Happe F., Frith U. & Frith C. (2000) Movement and mind: a functional imaging study of perception and interpretation of complex intentional movement patterns. Neuroimage 12, 314–25. Chow E. W., Watson M., Young D. A. & Bassett A. S. (2006) Neurocognitive profile in 22q11 deletion syndrome and schizophrenia. Schizophrenia Research 87, 270–8.
Golding-Kushner K., Weller G. & Shprintzen R. (1985) Velo-cardio-facial syndrome: language and psychological profiles. Journal of Craniofacial Genetics and Developmental Biology 5, 259–66. Goodman R., Ford T., Simmons H., Gatward R. & Meltzer H. (2000) Using the Strengths and Difficulties Questionnaire (SDQ) to screen for child psychiatric disorders in a community sample. British Journal of Psychiatry 177, 534–9.
Driscoll D. A., Spinner N. B., Budarf M. L., McDonald-McGinn D. M., Zackai E. H., Goldberg R. B. et al. (1992) Deletions and microdeletions of 22q11 in velo-cardio-facial syndrome. American Journal of Medical Genetics 44, 261–8.
Green M. F. & Leitman D. I. (2008) Social cognition in schizophrenia. Schizophrenia Bulletin 34, 670–2.
Dykens E. M. (2000) Annotation: psychopathology in children with intellectual disability. Journal of Child Psychology and Psychiatry 41, 407–17.
Henry J. C., Van Amelsvoort T., Morris R. G., Owen M. J., Murphy D. G. & Murphy K. C. (2002) An investigation of the neuropsychological profile in adults with velo-cardio-facial syndrome (VCFS). Neuropsychologia 40, 471–8.
Emerson E. (2003) Prevalence of psychiatric disorders in children and adolescents with and without intellectual disability. Journal of Intellectual Disability Research 47, 51–8.
Happé F. (1994) Autism, An Introduction to Psychological Theory. UCL Press, Ltd., London.
Emerson E. & Hatton C. (2007) Mental health of children and adolescents with intellectual disabilities in Britain. The British Journal of Psychiatry 191, 493–9.
Ho J. S., Radoeva P. D., Jalbrzikowski M., Chow C., Hopkins J., Tran W. C. et al. (2012) Deficits in mental state attributions in individuals with 22q11.2 deletion syndrome (velo-cardio-facial syndrome). Autism Research 5, 407–18.
Emerson E. & Hatton C. (2008) Socioeconomic disadvantage, social participation and networks and the self-
Jalbrzikowski M., Carter C., Senturk D., Chow C., Hopkins J. M., Green M. F. et al. (2012) Social
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 858 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
cognition in 22q11.2 microdeletion syndrome: relevance to psychosis? Schizophrenia Research 142, 99–107. Kaufman J., Birmaher B., Brent D., Rao U., Flynn C., Moreci P. et al. (1997) Schedule for affective disorders and schizophrenia for school-age children-present and lifetime version (K-SADS-PL): initial reliability and validity data. Journal of the American Academy of Child and Adolescent Psychiatry 36, 980–8. Kiley-Brabeck K. & Sobin C. (2006) Social skills and executive function deficits in children with the 22q11 deletion syndrome. Applied Neuropsychology 13, 258–68. Langdon R., Coltheart M. & Ward P. (2006) Empathetic perspective-taking is impaired in schizophrenia: evidence from a study of emotion attribution and theory of mind. Cognitive Neuropsychiatry 11, 133–55. Leitman D. I., Laukka P., Juslin P. N., Saccente E., Butler P. & Javitt D. C. (2010) Getting the cue: sensory contributions to auditory emotion recognition impairments in schizophrenia. Schizophrenia Bulletin 36, 545– 56. Lewandowski K., Shashi V., Berry P. & Kwapil T. (2007) Schizophrenic-like neurocognitive deficits in children and adolescents with 22q11 deletion syndrome. American Journal of Medical Genetics. Part B, Neuropsychiatric Genetics 144, 27–36. Looman W. S., Thurmes A. K. & O’Conner-Von S. K. (2010) Quality of life among children with velocardiofacial syndrome. Cleft Palate Craniofacial Journal 47, 273–83. McCabe K., Rich D., Loughland C. M., Schall U. & Campbell L. E. (2011) Visual scanpath abnormalities in 22q11.2 deletion syndrome: is this a face specific deficit? Psychiatry Research 189, 292–8. McCabe K. L., Melville J. L., Rich D., Strutt P. A., Cooper G., Loughland C. M. et al. (2013) Divergent patterns of social cognition performance in autism and 22q11.2 deletion syndrome (22q11DS). Journal of Autism and Developmental Disorders 43, 1926–34. Miller S. J. & Marcovitch S. (2012) How theory of mind and executive function co-develop. Review of Philosophical Psychology 3, 597–625. Miyake A., Friedman N., Emerson M., Witzki A., Howerter A. & Wagner T. (2000) The unity and diversity of executive functions and their contributions to complex ‘frontal lobe’ tasks: a latent variable analysis. Cognitive Psychology 41, 49–100. Murphy K., Jones L. & Owen M. (1999) High rates of schizophrenia in adults with velo-cardio-facial syndrome. Archives of General Psychiatry 56, 940–5. Niklasson L., Rasmussen P., Oskarsdottir S. & Gillberg C. (2002) Chromosome 22q11 deletion syndrome (CATCH 22): neuropsychiatric and neuropsychological
aspects. Developmental Medicine and Child Neurology 44, 44–50. Nyden A., Hjelmquist E. & Gillberg C. (2000) Autism spectrum and attention-deficit disorders in girls. Some neuropsychological aspects. European Child & Adolescent Psychiatry 9, 180–5. Roizen N., Antshel K., Fremont W., Abdulsabur N., Higgins A. M., Shprintzen R. J. et al. (2007) 22q11.2DS deletion syndrome: developmental milestones in infants and toddlers. Journal of Developmental & Behavioral Pediatrics 28, 119–24. Scambler P. J., Kelly D., Lindsay E., Williamson R., Goldberg R., Shprintzen R. et al. (1992) Velo-cardiofacial syndrome associated with chromosome 22 deletions encompassing the DiGeorge locus. Lancet 339, 1138–9. Shashi V., Veerapandiyan A., Schoch K., Kwapil T. R., Keshavan M. S., Ip E. et al. (2012) Social skills and associated psychopathology in children with chromosome 22q11.2 deletion syndrome: implications for interventions. Journal of Intellectual Disability Research 56, 865–78. Shprintzen R. J. (2008) Velo-cardio-facial syndrome: 30 years of study. Developmental Disabilities Research Reviews 14, 3–10. Simon T. J. (2008) A new account of the neurocognitive foundations of impairments in space, time and number processing in children with chromosome 22q11.2 deletion syndrome. Developmental Disabilities Research Reviews 14, 52–8. Sinderberry B., Brown S., Hammond P., Stevens A. F., Schall U., Murphy D. G. et al. (2013) Subtypes in 22q11.2 deletion syndrome associated with behaviour and neurofacial morphology. Research in Developmental Disabilities 34, 116–25. Sobin C., Kiley-Brabeck K., Daniels S., Khuri J., Taylor L., Blundell M. et al. (2005) Neuropsychological characteristics of children with the 22q11 deletion syndrome: a descriptive analysis. Child Neuropsychology: A Journal on Normal and Abnormal Development in Childhood and Adolescence 11, 39–53. Swillen A., Devriendt K., Legius E., Eyskens B., Dumoulin M., Gewillig M. et al. (1997) Intelligence and psychosocial adjustment in velocardiofacial syndrome: a study of 37 children and adolescents with VCFS. Journal of Medical Genetics 34, 453–8. Swillen A., Vandeputte L., Cracco J., Maes B., Ghesquiere P., Devriendt K. et al. (1999) Neuropsychological, learning and psychosocial profile of primary school aged children with the velocardio-facial syndrome (22q11 deletion): evidence for a nonverbal learning disability? Child Neuropsychology 5, 230–41. Swillen A., Devriendt K., Ghesquiere P. & Fryns J. (2001) Children with a 22q11 deletion versus children with a
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 9 septemBer 2015
Journal of Intellectual Disability Research 859 L.E. Campbell et al. • Social cognition dysfunction in 22q11.2 deletion syndrome
speech-language impairment and learning disability: behavior during primary school age. Genetic Counseling 12, 309–17. Wechsler D. (1999) Wechsler Abbreviated Scale of Intelligence. The Psychological Corporation, San Antonio, TX.
Woodin M., Wang P., Aleman D., McDonald-McGinn D., Zackai E. & Moss E. (2001) Neuropsychological profile of children and adolescents with the 22q11.2 microdeletion. Genetics in Medicine 3, 34–9.
Accepted 30 November 2014.
© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
