- Email: [email protected]
Performance on verbal and low-verbal false belief tasks: Evidence from children with Williams syndrome
Journal of Communication Disorders 46 (2013) 440–448
Contents lists available at ScienceDirect
Journal of Communication Disorders
Research paper
Performance on verbal and low-verbal false belief tasks: Evidence from children with Williams syndrome Jo Van Herwegen a,b,*, Dagmara Dimitriou c, Gabriella Rundblad b a
Department of Psychology, Kingston University, Penrhyn Road, Kingston-Upon-Thames KT1 2EE, UK Department of Education and Professional Studies, King’s College London, Waterloo Bridge Wing, Waterloo Road, SE1 9NH, UK c Department of Psychology and Human Development, Institute of Education, University of London, 20 Bedford Way, London WC1H 0AL, UK b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 27 September 2012 Received in revised form 1 October 2013 Accepted 28 October 2013 Available online 6 November 2013
Previous studies that have investigated the relationship between performance on theory of mind (ToM) tasks and verbal abilities in individuals with Williams syndrome (WS) have reported contradictory findings with some showing that language abilities aid performance on ToM tasks while others have found that participants with WS fail these tasks because of their verbal demands. The current study investigated this relationship again comparing performance on a classical change-location task to two newly developed low-verbal tasks, one change-location task and one unexpected content task. Thirty children with WS (aged 5–17;01 years) and 30 typically developing (TD) children (aged between 2;10 years and 9;09 years), who were matched for vocabulary comprehension scores were included in the study. Although performance in the WS group was significantly poorer compared to the TD group on all three tasks, performance was not predicted by their receptive vocabulary or grammatical ability scores. In addition, ToM abilities in both groups depended on the cognitive demands of the task at hand. This finding shows that performance on ToM tasks in WS is not necessarily hindered by their delayed language abilities but rather by the task administered. This could potentially affect the diagnosis of developmental disorders, such as Autism Spectrum Disorders, and comparison of ToM abilities across developmental disorders. Learning outcomes: Readers of this article should be able to (1) describe the current state of theory of mind research in Williams syndrome, (2) identify which cognitive abilities might explain performance on theory of mind tasks in both typically developing children and in children with Williams syndrome, and (3) interpret the importance of task demands when assessing children’s theory of mind abilities. ß 2013 Elsevier Inc. All rights reserved.
Keywords: Theory of mind Williams syndrome Verbal abilities
1. Introduction Theory of mind (ToM) is the ability to infer mental states, such as beliefs, desires, intentions and emotions, in order to predict and explain actions (Baron-Cohen, Leslie, & Frith, 1985). Typically developing (TD) children younger than 4 years old generally consider only what they have seen or know, while older children understand that people can hold different beliefs from their own beliefs (Flynn, 2006; Wellman, Cross, & Watson, 2001). The ability to reason about mental states is important
* Corresponding author at: Department of Psychology, Kingston University, Penrhyn Road, Kingston-Upon-Thames KT1 2EE, UK. Tel.: +44 020 8547 2803; fax: +44 020 8417 2292. E-mail addresses: [email protected] (J. Van Herwegen), [email protected] (D. Dimitriou), [email protected] (G. Rundblad). 0021-9924/$ – see front matter ß 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcomdis.2013.10.002
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
441
for social development as well as for language and communication. For example, in conversations we mentalize about the information already known to the listener and which information still needs to be communicated. Furthermore, ToM abilities allow us to monitor other people’s beliefs, reason about other people’s motives and manage reputations (Frith, 1989). Thus, impaired ToM abilities have been proposed as a possible explanation for some of the social communication difficulties observed in developmental disorders such as autism (Baron-Cohen, 2000). Although ToM abilities have been well researched in the past few decades, it is still unclear what abilities are necessary to develop ToM (Carlson, Moses, & Breton, 2002; Milligan, Astington, & Dack, 2007). For example, some have argued that domain specific abilities, such as certain vocabulary and grammatical abilities are necessary for the development of ToM (for a review see Milligan et al., 2007), whereas others have argued that general intelligence abilities and problem solving skills are better predictors for ToM abilities (Carlson et al., 2002; Carlson, Mandell, & Williams, 2004). In contrast to TD children, ToM abilities and those cognitive abilities that have been argued to be related to ToM understanding, often develop at different rates within developmental disorders and thus, developmental disorders allow investigation of how different abilities are related within development. Williams syndrome (WS) is a rare neurodevelopmental disorder which is caused by a 1.55 Mb deletion on the long arm of chromosome 7, affecting approximately 28 genes (Schubert, 2009). Individuals with WS show an uneven cognitive profile with lower performance on non-verbal abilities, such as drawing, visuospatial, planning and number processing, compared to verbal ones (Bellugi, Lichtenberger, Jones, Lai, & St-George, 2000; Brock, 2007; Van Herwegen, Rundblad, Davelaar, & Annaz, 2011). However, the onset of language is delayed and thus, a verbal advantage is present in adults but not in very young children (Jarrold, Baddeley, & Hewes, 1998). Furthermore, language development is atypical in that TD children generally start pointing before they start talking, while young children with WS started to point only after they had acquired words (Laing et al., 2002). Socially, individuals with WS can be recognised by their extreme friendliness and their overt social behaviour (Doyle, Bellugi, Korenberg, & Graham, 2004; Jones et al., 2001). Although individuals with WS have an excessive desire towards social contact, they have little contact with peers (Howlin, Davies, & Udwin, 1998) and they have difficulty in making and sustaining friendships (Einfeld, Tonge, & Florio, 1997). In addition, individuals with WS show difficulties in pragmatics (Annaz et al., 2009), especially with conversational skills such as turn taking, conversational coherence and appreciation of conversational context (Laws & Bishop, 2004). It has been suggested that these difficulties with social relationships and pragmatic abilities are caused by deficits in social understanding and cognition (Sullivan & Tager-Flusberg, 1999). However, evidence has been mixed with some claiming that social understanding is at a level expected for general intellectual abilities in WS (Karmiloff-Smith, Klima, Bellugi, Grant, & Baron-Cohen, 1995), while others have found that ToM abilities are impaired in this population (Sullivan & Tager-Flusberg, 1999; Tager-Flusberg & Sullivan, 2000; Tager-Flusberg, Sullivan, & Boshart, 1997). Several reasons have been proposed to explain these discrepancies. Tager-Flusberg and Sullivan (2000) suggested that ToM comprises of two different components: a social perceptual component that develops early in development and uses facial and bodily expressions to make judgements about people’s mental states, and a social cognitive component that builds on the social perceptual component and involves more complex inferences about mental states. It has been suggested that the social perceptual component (e.g., the eyes task in Tager-Flusberg, Boshart, & Baron-Cohen, 1998) is at the expected developmental level in WS, but not the social cognitive component. Thus, studies that used tasks that rely on the social perceptual component would not report any delays or impairments for ToM, while those that employed a more cognitive demanding task would find impaired ToM abilities in WS. However, Tager-Flusberg and Sullivan (2000) acknowledge that it is difficult to find tasks that completely separate these two components. Other discrepancies between studies might be caused by differences in task demands and comparison groups. Although recent studies have investigated these issues (Porter, Coltheart, & Langdon, 2008), further exploration of task demands and the use of a developmental approach is required in order to obtain a complete understanding of ToM abilities in WS. These issues will be discussed below. Most previous studies investigating ToM abilities in developmental disorders, including WS, have used the Sally–Anne false belief task (Baron-Cohen et al., 1985). This task involves children to understand a narrative that is acted out using props and dolls, after which they answer some questions about this narrative. Thus, failing to understand the narrative could explain why children fail this task. Indeed, several studies have identified that ToM abilities depend on language abilities (de Villiers & de Villiers, 2000; de Villiers & Pyers, 2002; Lohmann & Tomasello, 2003). For example, studies in autism have shown that those children who passed false belief tasks had higher verbal mental abilities (Happe´, 1995). Thus, the poor performance in WS on verbal ToM tasks might be caused by difficulties in comprehension of narratives and language rather than ToM abilities per se, since language development is generally delayed, as well as atypical in WS (Brock, 2007). In order to explore the fact that the language abilities in WS might hinder performance on ToM tasks, low-verbal ToM tasks1 need to be administered. Using a low-verbal ToM task, Tager-Flusberg et al. (1998) compared 13 adults with WS to age matched TD adults and 13 adults with Prader–Willi Syndrome (PWS) on a task in which participants had to select the mental state that matched the expression of the eyes in 25 photographs. The results showed that both the TD and the WS group performed significantly above chance (at least 17 out of 25 correct), while participants with PWS did not, and thus ToM abilities are not impaired in
1 The term low-verbal task is used rather than non-verbal task as most tasks labelled as non-verbal in previous studies still included verbal instructions, testing, and control questions.
442
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
WS (Tager-Flusberg et al., 1998). However, it has been debated whether the eye-task is more akin to an emotional perception task rather than a ToM task (Porter et al., 2008). Porter et al. (2008) carried out a low-verbal ToM task in 31 individuals with WS between the ages of 5;04 and 43;08 years old. In this study participants were asked to arrange four picture cards in the correct order as to make a story that made sense. The WS group performed significantly worse on the false belief story, but not on stories including intension or pretence, compared to typical mental age matched controls. Pretence stories include stories in which one object is replaced with something else (e.g., a banana is used as a telephone in pretend play) and do not require a child to understand the beliefs or intentions of the person. Thus, pretence stories did not necessary involve ToM understanding, while the false belief stories did. This suggests that ToM abilities in WS are below the level that is expected for their overall cognitive abilities (Porter et al., 2008). Yet, these previous studies included low-verbal tasks that were very dissimilar from typically administered verbal theory of mind tasks (i.e. understanding a narration story presented by the experimenter in contrast to creating a story yourself that makes sense to you) and thus, any differences observed might have been differences in task demands (e.g., difficulty of language or narrative used, memory demands, familiarity with social events and schemas) rather than evidence that verbal abilities relate to ToM abilities. For example, research in children with autism, SLI and TD control children has shown that the type and presentation mode of verbal false belief task results in different performance outcomes in the disorder groups, but not the TD group (Van Buijsen, Hendriks, Ketelaars, & Verhoeven, 2011). This can be explained by the fact that cognitive profiles in clinical groups, including WS, are often uneven and delayed (Van Herwegen et al., 2011) and thus performance on ToM tasks depends on the children’s cognitive and linguistic abilities, as well as on the task demands. The fact that TD children’s cognitive and linguistic abilities are age appropriate means that they had sufficient verbal and non-verbal abilities to cope and complete the tasks. A second reason why previous studies have reported mixed results about ToM abilities in WS might be caused by the age of the participants and the controls included. For example, Santos and Deruelle (2009) investigated performance of 19 WS participants aged 7–26 years old on both a verbal as well as a matching low-verbal picture ToM task. The results showed that individuals with WS performed better on the verbal compared to the low-verbal ToM task. This has lead to the argument that those with WS use their verbal abilities to ‘bootstrap’ performance on verbal ToM tasks (Porter et al., 2008). However, previous studies (Santos & Deruelle, 2009; Porter et al., 2008) included participants of a large age range with the control participants being matched for full-scale IQs. Yet, it has been shown that language abilities in WS develop at a faster rate than non-verbal abilities (Jarrold et al., 1998) and thus the language abilities of participants with WS would have been underestimated when matching on full IQ scores, while their non-verbal abilities would have been over-estimated. This could explain why the WS participants made more errors in the visual condition compared to the verbal one while no such differences in modality were observed in the MA matched controls (Santos & Deruelle, 2009). The current study therefore aimed to explore the relationship between verbal abilities, such as receptive vocabulary and grammar comprehension abilities and ToM tasks in children with WS. In contrast to previous studies, this study compared performance on the classic verbal Sally–Anne task to two newly developed low-verbal ToM tasks: Video task and Balloon task. While the Video task was directly comparable to the verbal Sally–Anne task (i.e., both are change-location tasks), the Balloon task was not (i.e., an unexpected content task). This allowed direct investigation of whether poor performance on Sally–Anne reported in previous studies was caused by the verbal demands of the task and whether results of previous studies could be explained by other factors, such as the type of task administered or the involvement of existing knowledge about certain social situations and scripts. In addition, the current study matched the participants with WS to control participants based upon performance on a receptive vocabulary task rather than full IQ. It was predicted that if language abilities predict performance on ToM tasks, those with higher scores on a receptive vocabulary comprehension or grammar comprehension task should pass the verbal ToM tasks in contrast to those with lower language abilities, but there should be no advantage for participants with higher language abilities on the low-verbal ToM tasks. 2. Method 2.1. Participants Thirty participants with WS (12 males, 18 females) who lived within Greater London and were aged between 5 and 17;01 years old were recruited via the Williams Syndrome Foundation UK. All participants had been positively diagnosed for the genetic deletion on the long arm of chromosome 7 using Flourescene In Situ Hybridisation (FISH). Verbal ability was measured using receptive vocabulary scores from the British Picture Vocabulary Scale (BPVS II: Dunn, Dunn, Whetton, & Burley, 1997) and the Pattern Construction (PC) subtest from the British Ability Scales was administered to measure visuospatial abilities (Elliot, Smith, & McCulloch, 1996). An additional language measure was obtained for a sub-sample of the WS group, namely receptive grammatical ability scores from Test for Reception of Grammar (TROG-2; Bishop, 2003). This task consists of 20 blocks which each containing 4 trials related to one grammatical construct. A block is passed when the participant fails no more than one trial per block. The number of blocks passed was recorded. For the control group, TD children were recruited from schools in Greater London and parental meeting groups in local libraries. Thirty TD children (14 males, 16 females) were selected and pairwise matched on the BPVS raw scores (within 4 points) to the WS group (t(58) = 0.145, p = 0.885). This ensured that both groups had similar receptive vocabulary comprehension abilities and allowed investigation whether a similar verbal advantage could be observed in both groups. This meant that the TD children were significantly younger (aged between 2;10 and 9;9 years) than the WS group
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
443
Table 1 Participant details: chronological age, British Picture Vocabulary Scales, Test For Reception of Grammar and pattern construction raw scores. Task
CAc in months BPVSd raw score TROG-2e blocks passed PCf raw score a b c d e f
TDa
WSb
N
Mean (SD)
Range
N
Mean (SD)
Range
30 30 N/A 24
73.03 (24.20) 63.67 (20.32) N/A 18.50 (10.45)
34–117 24–112 N/A 1–38
30 30 18 25
119.60 64.43 5.78 8.60
60–205 25–112 2–12 1–24
(39.52) (20.72) (3.19) (6.73)
Typical developing group. Williams syndrome. Chronological age. British Picture Vocabulary Scale. Test for reception of grammar. Pattern construction.
(t(48.059) = 5.504, p < 0.001). Yet, the PC raw scores for TD participants were significantly higher in comparison to the WS group; t(47) = 3.958, p < 0.001, which means that the TD group had better visuo-spatial abilities than the WS group. None of the TD children had been diagnosed with specific learning or behavioural difficulties according to a parental questionnaire. In addition, the TD children had a similar family background and SES compared to the WS group, as measured by the highest level of education of the mother and the number of single parents in each group. An overview of the ages, receptive vocabulary scores (BPVS) and visuo-spatial ability scores (PC raw scores) for each group can be found in Table 1. All parents were sent an invitation to take part in the study including the information letter about the study and the consent form and all parents and children consented to take part prior to the study. 2.2. Materials and methods 2.2.1. Sally–Anne The Sally–Anne task was carried out using props and dolls in accordance with the scenario illustrated in Frith (1989, 83). In this scenario, the child was first introduced to two dolls, Sally and Anne. When the child could tell the experimenter which doll was Anne and which one was Sally, the experimenter told the following narrative: ‘‘Sally has a marble. She puts the marble into her basket. Sally goes out for a walk. While Sally is away, Anne takes the marble out of the basket, and puts it into the box. Then, Anne goes home for her dinner. Now Sally is back from her walk. She wants to play with her marble.’’ Next, the participant was asked a first-order belief question (‘‘Where will Sally look for the marble?’’) followed by two control questions: a reality (‘‘Where is the marble really?’’) and a memory (‘‘Where was the marble in the beginning?’’). Participants received a score of zero or one for the belief question. However, only the data from participants who answered both the memory and control question correctly were included in the analyses below. This ensured that participants did not fail the belief questions in the tasks due to attention or memory problems. 2.2.2. Video task This task was inspired from a previous study by Newton and de Villiers (2007). In this task the participant watched a false belief video similar to the scenario in the Sally–Anne task, with the contrast that this task did not include a narrative. In order to avoid gender preferences, there were two conditions of the video, one including two male protagonists and a game console and one including two female protagonists and an apple. Participants were randomly allocated to each condition, so that half of the participants watched the male condition and the other half watched the female condition. In the video two protagonists (either two girls or two boys) are in a room, where there is a table with a basket and a box on (see Fig. 1). Protagonist A has an object (either an apple or a game console). Protagonist A puts the object in the basket. Protagonist A yawns, stretches, and leaves the room. While protagonist A is away, protagonist B goes to the basket and moves the object into the box. Then protagonist B leaves the room through a different door. Protagonist A returns and the participant was asked the exact same three questions as in the classic Sally–Anne task, namely ‘‘Where will this girl/boy look for the apple/game?’’ (Belief question), a reality question (‘‘Where is the apple/game now?’’), and a memory question (‘‘Where did the girl/boy put the apple/game?’’). Participants answered these questions by pointing to a picture out of three options (a picture of protagonist A looking into the basket, protagonist A looking into the box, and protagonist A looking under the table) while the experimenter recorded the participant’s response. These pictures were presented from left to right on the screen in randomised order. Similar to the Sally–Anne task participants received a score of zero or one for the belief question, but only those scores for participants who had the control questions correct were included in the analyses. 2.2.3. Balloon task The Balloon task was adapted from Woolfe, Want, and Siegal (2002). To start, each participant was given two practice trials to establish whether he/she was familiar with the concept of a thought balloon. Each participant was given a card on which two pictures of a protagonist were depicted next to one another. In one of the pictures, the protagonist was holding an object, while in the second picture the protagonist was thinking about the object which was illustrated using a
444
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
Fig. 1. A photo shot of the non-verbal Video task.
thought-balloon. The participant was asked, in alternating order, to select the picture where the protagonist was thinking about the object and where the protagonist actually had the object in his hand. After the first trial, regardless of whether the questions were answered correctly, each participant was explicitly told that a thought-balloon shows what a person is thinking (Wellman, Hollander, & Schult, 1996). Then, a second card was shown and the participant was asked the same questions. Participants who failed to answer the questions for the second card, and thus failed to understand that a thoughtballoon depicts someone’s thoughts, were excluded from the study. Next, the participant was administered one false-belief trial. To avoid gender preferences, two conditions were created: one including a boy looking at a birdhouse that contained a mobile phone and one including a girl approaching a letterbox that contained an apple. Half of the participants were administered the girl approaching the letterbox and the other half completed the condition in which the boy looked at the birdhouse. In each condition, the participant was shown a picture in which a flap was hiding an object (e.g. the participant sees a picture of a girl next to a letter box, but a flap over the letter box hides what is actually in the letter box) (see Fig. 2). The participant was instructed to cover the eyes of the protagonist in the picture and to have a look what was behind the flap. Then, the participant was shown a separate picture depicting only the face of the protagonist with a blank thought balloon. Below the thought balloon there were three pictures (a control picture, a picture of what was really behind the flap, and a picture depicting what would be most probably there and thus what the protagonist believed was under there). The control picture depicted a possible, but unrelated answer (e.g. a flower). The child was asked a ToM belief question (‘‘What does X think is behind the flap?’’) and a control question (i.e. ‘‘What was actually behind the flap?’’). The participant answered by pointing to one of the three pictures and this choice was recorded by the experimenter. Participants received a score of one or zero for the belief question. However, scores of participants who answered the control question incorrectly were not included in the data analyses reported below.
Fig. 2. Example of a low-verbal Balloon trial.
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
445
Table 2 Number of participants per task and group that passed or failed the false belief question. Group
Pass (%)
Fail (%)
Total N of participantsa
Sally–Anne task
b
TD WSc
20 (67) 6 (21)
10 (33) 22 (79)
30 28
Video task
TD WS
17 (74) 5 (29)
6 (26) 12 (71)
23 17
Balloon task
TD WS
17 (68) 9 (35)
8 (32) 17 (65)
25 26
Task
a b c
Participants were excluded because they failed the control and/or memory question. For none of the cases these were the youngest or less able children. Typical developing group. Williams syndrome.
3. Results 3.1. Performance on control and memory questions As shown in Table 2, in both groups more children had to be excluded for the Video task than the Sally–Anne task (TD: p < 0.05; WS: p < 0.01, Fisher’s exact test), even though the control and memory questions in both tasks were identical. Fisher exact tests showed that more TD participants had to be excluded for the Balloon task than the Sally–Anne task (p < 0.01), but there were no differences with the Video task (p > 0.05). In contrast, fewer WS participants had to be excluded for the Balloon task in contrast to the Video task, but there were no differences with the Sally–Anne task (p > 0.05). It is important to note that those TD who failed the control questions in the Balloon task also failed these questions in the Video task (N = 5) while in the WS group this was not the case (N = 1). 3.2. Performance on belief question Comparison of overall performance between the two groups for the Sally–Anne task showed that more participants in the WS group (79%) failed the false belief question in contrast to the TD group (33%) (the Fisher’s exact test: p = 0.001) (see Table 2). As shown in Table 2, more TD participants passed the Video task compared to the WS group (p = 0.010, Fisher’s exact test). Subsequently, it was investigated whether reducing the verbal demands in the Video task had resulted in more or less successful performance on the belief question. Although three WS participants (25%) and one TD participant (17%) who failed the Sally–Anne task passed the belief question in the Video task, Fisher exact test did not show any significant improvements for either group (all p > 0.05). Thus, performance on the Video task did not differ from the Sally–Anne task, even though overall language and memory demands were reduced in the Video task. For the Balloon task, more WS participants failed the belief question compared to the TD group (p = 0.037, Fisher’s exact test). Comparison of performance on the Balloon task to the Sally–Anne task showed no significant differences in either group (p > 0.05, Fisher’s exact test). However, for the TD group more participants failed the belief question in the Balloon compared to the Video task. This shows that the Balloon task was significantly harder than the Video task (Fisher exact, p = 0.026), while there were no such differences for the WS group (p > 0.05, Fisher’s exact). 3.3. Performance on belief question and relationship to age, receptive vocabulary, receptive grammatical comprehension, and visuo-spatial abilities It was investigated whether chronological age was a good predictor for passing each of the three tasks. In the TD group, those who passed the ToM tasks were older than those who failed; Sally–Anne: t(28) = 2.953, p = 0.006; Video task: t(21) = 2.560, p = 0.018, Balloon task: t(24) = 3.044, p = 0.006. There were no differences for chronological age for the WS group for any of the three ToM tasks (all p’s > 0.05). An overview is provided in Table 3. Next, it was investigated whether receptive vocabulary abilities were a good predictor for passing each of the three tasks. There was a significant difference for BPVS raw scores in the TD group between those participants who passed the belief question and those who failed the belief question for each of the three tasks (Sally–Anne t(28) = 2.953, p = 0.006; Video task t(21) = 2.096, p = 0.048; Balloon task t(18) = 2.799, p = 0.012). However, this was not the case for the WS group (all p’s > 0.05). In addition, there were no significant differences between those in the WS group who passed and those who failed the Sally–Anne task; t(14) = 0.021, p = 0.837, the Video task; t(8) = 1.691, p = 0.129, or the Balloon task; t(16) = 0.862, p = 0.401 for receptive grammatical ability scores as measured by the amount of blocks passed on the TROG-2. Finally, the relationship between passing or failing the belief questions on the three ToM tasks and visuo-spatial abilities, as measured by PC, was examined. In the TD group there was a significant effect between those who passed and failed only for the Sally–Anne task; t(22) = 2.380, p = 0.026. There were no significant differences for the Video or the Balloon task (all p’s > 0.05). Again, there were no significant differences for PC scores between those who passed or failed the belief tasks for either of the ToM tasks in the WS group (all p’s > 0.05).
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
446
Table 3 Chronological age, BPVS raw scores, TROG blocks passed, and PC raw scores (means and standard deviation) per group for those who passed and those who failed the belief question in (a) Sally–Anne task, (b) Video task, and (c) Balloon task. Task
TDa Passed
WSb Failed
Mean (SD)
Passed
Failed
Mean (SD)
CAc in months
Sally–Anne task Video task Balloon task
81.80 (20.91) 85.18 (20.73) 86.24 (16.54)
55.50 (21.25) 60.50 (18.87) 62.22 (23.49)
134.17 (34.97) 114.80 (50.87) 140.13 (31.84)
118.59 (41.23) 112.25 (31.82) 120.19 (42.89)
BPVSd raw scores
Sally–Anne task Video task Balloon task
70.55 (18.14) 72.65 (18.82) 74.00 (12.15)
49.90 (17.88) 53.33 (21.17) 51.67 (24.06)
74.14 (19.34) 78.20 (29.71) 69.69 (25.18)
61.48 (20.60) 61.42 (16.60) 64.56 (20.38)
TROGe blocks passed
Sally–Anne task Video task Balloon task
N/A N/A N/A
N/A N/A N/A
6.14 (3.81) 9.00 (5.20) 7.00 (3.74)
5.78 (3.15) 5.00 (2.82) 5.43 (3.08)
PCf raw scores
Sally–Anne task Video task Balloon task
21.47 (9.55) 21.07 (11.19) 20.80 (9.87)
11.29 (9.46) 16.20 (8.758) 15.14 (11.57)
10.17 (8.70) 15.00 (8.98) 8.00 (8.21)
8.50 (6.12) 7.11 (5.65) 9.58 (6.75)
a b c d e f
Typical developing group. Williams syndrome. Chronological age. British Picture Vocabulary Scale. Test for reception of grammar. Pattern construction.
4. Discussion The current study aimed to clarify whether receptive language abilities, such as vocabulary and grammar comprehension abilities, or the task demands would influence performance on ToM tasks, since some previous studies found evidence that WS rely on their language abilities to pass verbal ToM tasks (Porter et al., 2008; Santos & Deruelle, 2009), while other studies suggested that participants with WS specifically fail verbal ToM tasks due to their language demands (Sullivan & TagerFlusberg, 1999; Tager-Flusberg & Sullivan, 2000; Tager-Flusberg et al., 1997). Yet, in contrast to previous studies, the current study aimed to compare performance on the classic Sally–Anne task to a low-verbal change-location task that was similar (Video task) and one unexpected-content task that was dissimilar in design (Balloon task) in order to investigate differences in task demands. In addition, the current study matched participants based on their receptive vocabulary abilities rather than full scale IQ since participants with WS show uneven cognitive profiles and matching on full-IQ might result in underestimation of the language abilities in WS. In line with previous studies, the results showed that participants with WS performed worse on the Sally–Anne task compared to control participants who had similar receptive vocabulary scores but were chronologically younger than the WS participants (Tager-Flusberg & Sullivan, 2000; Tager-Flusberg et al., 1997). However, performance in the WS group did not worsen or improve when they were assessed on a low-verbal task (Video task) that was similar in design to the Sally– Anne task. In addition, while participants in the TD group who passed the belief questions in both tasks had higher vocabulary comprehension scores than those who failed, there was no such difference in the WS group. For a sub-sample of the WS-group, receptive grammatical comprehension scores measured by TROG-2 were available. Comparisons between those who passed and those who failed the three ToM tasks in this study showed that there were no differences for TROG-2 scores. This shows that performance on ToM tasks is not driven by poor or proficient language abilities in the WS group in contrast to what some previous studies have suggested. Comparison between the three tasks showed that a significant number of TD and WS participants had to be excluded from the data analyses in the Video task compared to the Sally–Anne task as they failed to answer the control and memory questions correctly. The fact that these questions were identical to the Sally–Anne task shows that explicit verbal narration aided children’s context understanding or engagement with the task in general which allowed more children to pass the control and memory questions. Santos and Deruelle (2009) reported that participants with WS only showed a deficit on visual ToM trials but not on trials that included language. However, their study did not include any control questions to see whether the participant was engaged with the task at hand. As the current study shows, low-verbal visual tasks include less contextual information and are thus more difficult. This might increase the risk of participants losing interest in the task at hand and this should be taken into account in future visual tasks. Finally, the current study also compared performance on the Sally–Anne task to a low-verbal task that was dissimilar in structure (i.e., unexpected content versus change-location task) and task demands (verbal versus low-verbal). Again, participants in either group did not perform any better or worse on the belief question in the Balloon task compared to the Sally–Anne task and there were no differences in the WS group for receptive vocabulary or grammar abilities between those who passed and those who failed the belief question in the Balloon task. Yet, the TD group found the Balloon task much
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
447
harder than the Video task, in that not only did more children fail to understand the concept of thought balloons, there were also more children who failed the belief question in the Balloon task in contrast to the Video task. So while the WS group seemed to struggle with the Video task, more TD children had to be excluded or failed the belief question in the Balloon task. This shows that task demands of the ToM tasks employed need to be taken into account when assessing children’s ToM abilities. In all three tasks the questions were posed in verbal format and children answered by pointing or providing a verbal answer (whichever they preferred). In addition, the Sally–Anne and Video task included the exact same questions. Therefore, how children were assessed in the three tasks was similar, yet the tasks differed in the kind of information children had to infer from the task itself and how the task was presented, i.e. whether further narratives were provided or not. This finding is in line with results from Van Buijsen et al. (2011) which have shown that ToM performance in children with autism and SLI is also dependent upon the task employed. One possible reason why the TD children performed worse on the Balloon task compared to the Video task is that the Balloon task relies upon knowledge of scripts (i.e., the knowledge that a letter box would typically contain a letter and a birdhouse would not generally contain a phone) and these scripts would develop or become more set as children grow older (Nelson, 1986). In contrast, the fact that the WS participants were older than the TD group might therefore explain why they did not find the Balloon task harder. This explanation is supported by the fact that chronological age, but not visuo-spatial abilities, was a good predictor for performance on the Balloon task in the TD group. 5. Conclusions In sum, the current study did not find any evidence that performance on ToM tasks is driven by language abilities, as measured by a receptive vocabulary and grammar abilities, in participants with WS or that participants with WS do better on low-verbal compared to verbal ToM tasks. However, the current study did find evidence that performance depends upon the overall cognitive demands a task poses. For example, tasks that included explicit narratives of what was happening proved easier than those tasks in which children had to infer what was happening. In addition, our evidence suggests that knowledge of certain schemas was an important factor for performance on low-verbal ToM tasks. This finding has implications for clinicians and researchers who wish to establish ToM abilities in developmental disorders or who would like to make comparisons between different groups, in that the outcomes will depend upon the kind of task administered. In addition, reducing the verbal narrative of the ToM task reduced the amount of information available about what is happening in the situation and this seems to affect the level of difficulty or engagement with the task in children. However, the current study only included controls that were matched for receptive vocabulary abilities using BPVS scores and scores for grammatical abilities were only available for a sub-sample of our WS group. Although, both provided some indication of language abilities in children, future studies should include control participants matched for overall language abilities, including story comprehension abilities, as these might be better predictors for ToM performance. In addition, future studies should also explore why participants perform worse on visual or low-verbal ToM tasks. For example, the use of eye-tracking methodology would allow investigating engagement to the task as well as differences in looking behaviour. Acknowledgements We thank the Williams Syndrome Foundation UK and all the children who took part in this study. This research was supported by an ESRC open-competition grant (PTA-030-2006-00235), a grant from the University of London Central Research Fund (REGCRF06-07), a Departmental Scholarship from King’s College London, a grant from the Williams Syndrome Foundation, UK and a grant from the Nuffield Foundation. We also thank reviewers for their comments on an earlier draft of this manuscript. Conflict of interest statement The authors received financial supportfor this project from ESRC open-competition grant (PTA-030-2006-00235), a grant from the University of London Central Research Fund (REGCRF06-07), a Departmental Scholarship from King’s College London, a grant from the Williams Syndrome Foundation, UK and a grant from the Nuffield Foundation. There are no nonfinancial relationships to disclose. Appendix A. Continuing education CEU Questions 1. Which of the following statements about language development in Williams syndrome is correct: a. Language abilities and non-verbal abilities are equally impaired in Williams syndrome b. Language development in Williams syndrome is delayed and atypical. c. Language development in Williams syndrome is the same as in typically developing controls d. Language abilities are better in comparison to typically developing controls of a similar age 2. Sally–Anne is an example of a low-verbal Theory of Mind task: TRUE/FALSE 3. The current study found that children with Williams syndrome:
448
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
a. Performed better than the controls on all three tasks b. Performed worse than the controls but only on the verbal task c. Performed worse than the controls but only on the low-verbal tasks d. Performed worse than the controls on all three tasks 4. Receptive grammar abilities predict performance on verbal theory of mind tasks in Williams syndrome: TRUE/FALSE 5. Which two statements are correct for the control group? a. The video task is harder because it does not contain language b. The Sally–Anne task is harder because it contains language c. Both low-verbal tasks are harder because they do not contain language d. The balloon task is harder because the control children are too young and do not have enough world knowledge
References Annaz, D., Van Herwegen, J., Thomas, M., Fishman, R., Karmiloff-Smith, A., & Rundblad, G. (2009). Comprehension of metaphor and metonymy in children with Williams syndrome. International Journal of Language and Communication Disorders, 44(6), 962–978. Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a ‘theory of mind’? Cognition, 21, 37–46. Baron-Cohen, S. (2000). Theory of mind and autism: A fifteen year review. In S. Baron-Cohen, H. Tager-Flusberg, & D. J. Cohen (Eds.), Understanding other minds: Perspectives from developmental cognitive neuroscience (pp. 3–20). Oxford: Oxford University Press. Bellugi, U., Lichtenberger, L., Jones, W., Lai, Z., & St-George, M. (2000). The neurocognitive profile of WS: A complex pattern of strengths and weaknesses. Journal of Cognitive Neuroscience, 12(Suppl.), 7–29. Bishop, D. M. V. (2003). Test for reception of grammar, version-2 (TROG-2). London: Psychological Corporation. Brock, J. (2007). Language abilities in Williams syndrome: A critical review. Development and Psychopathology, 19, 97–127. Carlson, S., Mandell, D., & Williams, L. (2004). Executive function and theory of mind: Stability and prediction from ages 2 to 3. Developmental Psychology, 40, 1105– 1122. Carlson, S., Moses, L., & Breton, C. (2002). How specific is the relation between executive function and theory of mind? Contributions of inhibitory control and working memory. Infant and Child Development, 11, 73–92. de Villiers, J. G., & de Villiers, P. A. (2000). Linguistic determinism and the understanding of false beliefs. In P. Mitchell & K. Riggs (Eds.), Children’s reasoning and the mind (pp. 189–226). Hove, UK: Psychology Press. de Villiers, J., & Pyers, J. (2002). Complements to cognition: A longitudinal study of the relationship between complex syntax and false-belief-understanding. Cognitive Development, 17, 1037–1060. Doyle, T. F., Bellugi, U., Korenberg, J. R., & Graham, J. (2004). ‘‘Everybody in the world is my friend’’ Hypersociability in young children with Williams syndrome. American Journal of Medical Genetics, 124, 263–273. Dunn, L., Dunn, L., Whetton, C., & Burley, J. (1997). British Picture Vocabulary Scale II. Windsor: NFER-Nelson Publishing Company Limited. Einfeld, S. L., Tonge, B. J., & Florio, T. (1997). Behavioural and emotional disturbance in individuals with Williams Syndrome. American Journal on Mental Retardation, 102(1), 45–53. Elliot, C. D., Smith, P., & McCulloch, K. (1996). British Ability Scales Second Edition (BAS II). Windsor: NFER-Nelson Publishing Company Limited. Flynn, E. (2006). A microgenetic investigation of stability and continuity in theory of mind development. British Journal of Developmental Psychology, 24, 631–654. Frith, U. (1989). Autism and theory of mind. In C. Gillberg (Ed.), Diagnosis and treatment of autism (pp. 33–52). New York: Plenum Press. Happe´, F. (1995). The role of age and verbal ability in the theory of mind task performance of subjects with autism. Child Development, 66(3), 843–855. Howlin, P., Davies, M., & Udwin, O. (1998). Syndrome specific characteristics in Williams syndrome: To what extent do early behavioural patterns persist into adult life? Journal of Applied Research in Intellectual Disabilities, 11(3), 207–226. Jarrold, C., Baddeley, A. D., & Hewes, A. K. (1998). Verbal and non-verbal abilities in the Williams syndrome phenotype: Evidence for diverging developmental trajectories. Journal of Child Psychology and Psychiatry, 39, 511–524. Jones, W., Bellugi, U., Lai, Z., Chiles, M., Reilly, J., Lincoln, A., et al. (2001). Hypersociability: The social and affective phenotype of Williams syndrome. In U. Bellugi & M. George (Eds.), Journey from cognition to brain to gene: perspectives from Williams syndrome (pp. 43–72). Massachusetts: MIT Press. Karmiloff-Smith, A., Klima, E. S., Bellugi, U., Grant, J., & Baron-Cohen, S. (1995). Is there a social module? Language, face processing, and theory of mind in subjects with Williams syndrome. Journal of Cognitive Neuroscience, 7(2), 196–208. Laing, E., Butterworth, G., Ansari, D., Gso˝ dl, M., Longhi, E., Panagiotaki, G., et al. (2002). Atypical development of language and social communication in toddlers with Williams syndrome. Developmental Science, 5(2), 233–246. Laws, G., & Bishop, D. V. M. (2004). Verbal deficits in specific language impairment and Down syndrome: A comparative review. International Journal of Language and Communication Disorders, 39, 423–451. Lohmann, H., & Tomasello, M. (2003). The role of language in the development of false belief understanding: A training study. Child Development, 74, 1130–1144. Milligan, K., Astington, J. W., & Dack, L. A. (2007). Language and theory of mind: Meta-analysis of the relation between language ability and false-belief understanding. Child Development, 78(2), 622–646. Nelson, K. (1986). Event knowledge. Structure and function in development. Hillsdale, NJ: Lawrence Erlbaum Associates. Newton, A. M., & de Villiers, J. G. (2007). Thinking while talking: Adults fail nonverbal false-belief reasoning. Psychological Science, 18(7), 574–579. Porter, M. A., Coltheart, M., & Langdon, R. (2008). Theory of mind in Williams syndrome assessed using a nonverbal task. Journal of Autism and Developmental Disorders, 38, 806–814. Santos, A., & Deruelle, C. (2009). Verbal peaks and visual valleys in theory of mind ability in Williams syndrome. Journal of Autism and Developmental Disorders, 39(4), 651–659. Schubert, C. (2009). The genomic basis of the Williams–Beuren syndrome. Cellular and Molecular Life Science, 66, 1178–1197. Sullivan, K., & Tager-Flusberg, H. (1999). Second-order belief attribution in Williams syndrome: Intact or impaired? American Journal on Mental Retardation, 104(6), 523–532. Tager-Flusberg, H., Boshart, J., & Baron-Cohen, S. (1998). Reading the windows to the soul: Evidence of domain-specific sparing in Williams syndrome. Journal of Cognitive Neuroscience, 10, 631–639. Tager-Flusberg, H., & Sullivan, K. (2000). A componential view of theory of mind: Evidence from Williams syndrome. Cognition, 76, 59–89. Tager-Flusberg, H., Sullivan, K., & Boshart, J. (1997). Executive functions and performance on false belief tasks. Developmental Neuropsychology, 13(4), 487–493. Van Buijsen, M., Hendriks, A., Ketelaars, M., & Verhoeven, L. (2011). Assessment of theory of mind in children with communication disorders: Role of presentation mode. Research in Developmental Disabilities, 32, 1038–1045. Van Herwegen, J., Rundblad, G., Davelaar, E. J., & Annaz, D. (2011). Variability and standardised test profiles in typically developing children and children with Williams syndrome. British Journal of Developmental Psychology, 29(4), 883–894. Wellman, H., Cross, D., & Watson, J. (2001). Meta-analysis of theory of mind development: The truth about false-belief. Child Development, 72(3), 655–684. Wellman, H., Hollander, M., & Schult, C. (1996). Young children’s understanding of thought bubbles and of thoughts. Child Development, 67(3), 768–788. Woolfe, T., Want, S. C., & Siegal, M. (2002). Signposts to development: Theory of mind in deaf children. Child Development, 73(3), 768–778.
Contents lists available at ScienceDirect
Journal of Communication Disorders
Research paper
Performance on verbal and low-verbal false belief tasks: Evidence from children with Williams syndrome Jo Van Herwegen a,b,*, Dagmara Dimitriou c, Gabriella Rundblad b a
Department of Psychology, Kingston University, Penrhyn Road, Kingston-Upon-Thames KT1 2EE, UK Department of Education and Professional Studies, King’s College London, Waterloo Bridge Wing, Waterloo Road, SE1 9NH, UK c Department of Psychology and Human Development, Institute of Education, University of London, 20 Bedford Way, London WC1H 0AL, UK b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 27 September 2012 Received in revised form 1 October 2013 Accepted 28 October 2013 Available online 6 November 2013
Previous studies that have investigated the relationship between performance on theory of mind (ToM) tasks and verbal abilities in individuals with Williams syndrome (WS) have reported contradictory findings with some showing that language abilities aid performance on ToM tasks while others have found that participants with WS fail these tasks because of their verbal demands. The current study investigated this relationship again comparing performance on a classical change-location task to two newly developed low-verbal tasks, one change-location task and one unexpected content task. Thirty children with WS (aged 5–17;01 years) and 30 typically developing (TD) children (aged between 2;10 years and 9;09 years), who were matched for vocabulary comprehension scores were included in the study. Although performance in the WS group was significantly poorer compared to the TD group on all three tasks, performance was not predicted by their receptive vocabulary or grammatical ability scores. In addition, ToM abilities in both groups depended on the cognitive demands of the task at hand. This finding shows that performance on ToM tasks in WS is not necessarily hindered by their delayed language abilities but rather by the task administered. This could potentially affect the diagnosis of developmental disorders, such as Autism Spectrum Disorders, and comparison of ToM abilities across developmental disorders. Learning outcomes: Readers of this article should be able to (1) describe the current state of theory of mind research in Williams syndrome, (2) identify which cognitive abilities might explain performance on theory of mind tasks in both typically developing children and in children with Williams syndrome, and (3) interpret the importance of task demands when assessing children’s theory of mind abilities. ß 2013 Elsevier Inc. All rights reserved.
Keywords: Theory of mind Williams syndrome Verbal abilities
1. Introduction Theory of mind (ToM) is the ability to infer mental states, such as beliefs, desires, intentions and emotions, in order to predict and explain actions (Baron-Cohen, Leslie, & Frith, 1985). Typically developing (TD) children younger than 4 years old generally consider only what they have seen or know, while older children understand that people can hold different beliefs from their own beliefs (Flynn, 2006; Wellman, Cross, & Watson, 2001). The ability to reason about mental states is important
* Corresponding author at: Department of Psychology, Kingston University, Penrhyn Road, Kingston-Upon-Thames KT1 2EE, UK. Tel.: +44 020 8547 2803; fax: +44 020 8417 2292. E-mail addresses: [email protected] (J. Van Herwegen), [email protected] (D. Dimitriou), [email protected] (G. Rundblad). 0021-9924/$ – see front matter ß 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcomdis.2013.10.002
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
441
for social development as well as for language and communication. For example, in conversations we mentalize about the information already known to the listener and which information still needs to be communicated. Furthermore, ToM abilities allow us to monitor other people’s beliefs, reason about other people’s motives and manage reputations (Frith, 1989). Thus, impaired ToM abilities have been proposed as a possible explanation for some of the social communication difficulties observed in developmental disorders such as autism (Baron-Cohen, 2000). Although ToM abilities have been well researched in the past few decades, it is still unclear what abilities are necessary to develop ToM (Carlson, Moses, & Breton, 2002; Milligan, Astington, & Dack, 2007). For example, some have argued that domain specific abilities, such as certain vocabulary and grammatical abilities are necessary for the development of ToM (for a review see Milligan et al., 2007), whereas others have argued that general intelligence abilities and problem solving skills are better predictors for ToM abilities (Carlson et al., 2002; Carlson, Mandell, & Williams, 2004). In contrast to TD children, ToM abilities and those cognitive abilities that have been argued to be related to ToM understanding, often develop at different rates within developmental disorders and thus, developmental disorders allow investigation of how different abilities are related within development. Williams syndrome (WS) is a rare neurodevelopmental disorder which is caused by a 1.55 Mb deletion on the long arm of chromosome 7, affecting approximately 28 genes (Schubert, 2009). Individuals with WS show an uneven cognitive profile with lower performance on non-verbal abilities, such as drawing, visuospatial, planning and number processing, compared to verbal ones (Bellugi, Lichtenberger, Jones, Lai, & St-George, 2000; Brock, 2007; Van Herwegen, Rundblad, Davelaar, & Annaz, 2011). However, the onset of language is delayed and thus, a verbal advantage is present in adults but not in very young children (Jarrold, Baddeley, & Hewes, 1998). Furthermore, language development is atypical in that TD children generally start pointing before they start talking, while young children with WS started to point only after they had acquired words (Laing et al., 2002). Socially, individuals with WS can be recognised by their extreme friendliness and their overt social behaviour (Doyle, Bellugi, Korenberg, & Graham, 2004; Jones et al., 2001). Although individuals with WS have an excessive desire towards social contact, they have little contact with peers (Howlin, Davies, & Udwin, 1998) and they have difficulty in making and sustaining friendships (Einfeld, Tonge, & Florio, 1997). In addition, individuals with WS show difficulties in pragmatics (Annaz et al., 2009), especially with conversational skills such as turn taking, conversational coherence and appreciation of conversational context (Laws & Bishop, 2004). It has been suggested that these difficulties with social relationships and pragmatic abilities are caused by deficits in social understanding and cognition (Sullivan & Tager-Flusberg, 1999). However, evidence has been mixed with some claiming that social understanding is at a level expected for general intellectual abilities in WS (Karmiloff-Smith, Klima, Bellugi, Grant, & Baron-Cohen, 1995), while others have found that ToM abilities are impaired in this population (Sullivan & Tager-Flusberg, 1999; Tager-Flusberg & Sullivan, 2000; Tager-Flusberg, Sullivan, & Boshart, 1997). Several reasons have been proposed to explain these discrepancies. Tager-Flusberg and Sullivan (2000) suggested that ToM comprises of two different components: a social perceptual component that develops early in development and uses facial and bodily expressions to make judgements about people’s mental states, and a social cognitive component that builds on the social perceptual component and involves more complex inferences about mental states. It has been suggested that the social perceptual component (e.g., the eyes task in Tager-Flusberg, Boshart, & Baron-Cohen, 1998) is at the expected developmental level in WS, but not the social cognitive component. Thus, studies that used tasks that rely on the social perceptual component would not report any delays or impairments for ToM, while those that employed a more cognitive demanding task would find impaired ToM abilities in WS. However, Tager-Flusberg and Sullivan (2000) acknowledge that it is difficult to find tasks that completely separate these two components. Other discrepancies between studies might be caused by differences in task demands and comparison groups. Although recent studies have investigated these issues (Porter, Coltheart, & Langdon, 2008), further exploration of task demands and the use of a developmental approach is required in order to obtain a complete understanding of ToM abilities in WS. These issues will be discussed below. Most previous studies investigating ToM abilities in developmental disorders, including WS, have used the Sally–Anne false belief task (Baron-Cohen et al., 1985). This task involves children to understand a narrative that is acted out using props and dolls, after which they answer some questions about this narrative. Thus, failing to understand the narrative could explain why children fail this task. Indeed, several studies have identified that ToM abilities depend on language abilities (de Villiers & de Villiers, 2000; de Villiers & Pyers, 2002; Lohmann & Tomasello, 2003). For example, studies in autism have shown that those children who passed false belief tasks had higher verbal mental abilities (Happe´, 1995). Thus, the poor performance in WS on verbal ToM tasks might be caused by difficulties in comprehension of narratives and language rather than ToM abilities per se, since language development is generally delayed, as well as atypical in WS (Brock, 2007). In order to explore the fact that the language abilities in WS might hinder performance on ToM tasks, low-verbal ToM tasks1 need to be administered. Using a low-verbal ToM task, Tager-Flusberg et al. (1998) compared 13 adults with WS to age matched TD adults and 13 adults with Prader–Willi Syndrome (PWS) on a task in which participants had to select the mental state that matched the expression of the eyes in 25 photographs. The results showed that both the TD and the WS group performed significantly above chance (at least 17 out of 25 correct), while participants with PWS did not, and thus ToM abilities are not impaired in
1 The term low-verbal task is used rather than non-verbal task as most tasks labelled as non-verbal in previous studies still included verbal instructions, testing, and control questions.
442
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
WS (Tager-Flusberg et al., 1998). However, it has been debated whether the eye-task is more akin to an emotional perception task rather than a ToM task (Porter et al., 2008). Porter et al. (2008) carried out a low-verbal ToM task in 31 individuals with WS between the ages of 5;04 and 43;08 years old. In this study participants were asked to arrange four picture cards in the correct order as to make a story that made sense. The WS group performed significantly worse on the false belief story, but not on stories including intension or pretence, compared to typical mental age matched controls. Pretence stories include stories in which one object is replaced with something else (e.g., a banana is used as a telephone in pretend play) and do not require a child to understand the beliefs or intentions of the person. Thus, pretence stories did not necessary involve ToM understanding, while the false belief stories did. This suggests that ToM abilities in WS are below the level that is expected for their overall cognitive abilities (Porter et al., 2008). Yet, these previous studies included low-verbal tasks that were very dissimilar from typically administered verbal theory of mind tasks (i.e. understanding a narration story presented by the experimenter in contrast to creating a story yourself that makes sense to you) and thus, any differences observed might have been differences in task demands (e.g., difficulty of language or narrative used, memory demands, familiarity with social events and schemas) rather than evidence that verbal abilities relate to ToM abilities. For example, research in children with autism, SLI and TD control children has shown that the type and presentation mode of verbal false belief task results in different performance outcomes in the disorder groups, but not the TD group (Van Buijsen, Hendriks, Ketelaars, & Verhoeven, 2011). This can be explained by the fact that cognitive profiles in clinical groups, including WS, are often uneven and delayed (Van Herwegen et al., 2011) and thus performance on ToM tasks depends on the children’s cognitive and linguistic abilities, as well as on the task demands. The fact that TD children’s cognitive and linguistic abilities are age appropriate means that they had sufficient verbal and non-verbal abilities to cope and complete the tasks. A second reason why previous studies have reported mixed results about ToM abilities in WS might be caused by the age of the participants and the controls included. For example, Santos and Deruelle (2009) investigated performance of 19 WS participants aged 7–26 years old on both a verbal as well as a matching low-verbal picture ToM task. The results showed that individuals with WS performed better on the verbal compared to the low-verbal ToM task. This has lead to the argument that those with WS use their verbal abilities to ‘bootstrap’ performance on verbal ToM tasks (Porter et al., 2008). However, previous studies (Santos & Deruelle, 2009; Porter et al., 2008) included participants of a large age range with the control participants being matched for full-scale IQs. Yet, it has been shown that language abilities in WS develop at a faster rate than non-verbal abilities (Jarrold et al., 1998) and thus the language abilities of participants with WS would have been underestimated when matching on full IQ scores, while their non-verbal abilities would have been over-estimated. This could explain why the WS participants made more errors in the visual condition compared to the verbal one while no such differences in modality were observed in the MA matched controls (Santos & Deruelle, 2009). The current study therefore aimed to explore the relationship between verbal abilities, such as receptive vocabulary and grammar comprehension abilities and ToM tasks in children with WS. In contrast to previous studies, this study compared performance on the classic verbal Sally–Anne task to two newly developed low-verbal ToM tasks: Video task and Balloon task. While the Video task was directly comparable to the verbal Sally–Anne task (i.e., both are change-location tasks), the Balloon task was not (i.e., an unexpected content task). This allowed direct investigation of whether poor performance on Sally–Anne reported in previous studies was caused by the verbal demands of the task and whether results of previous studies could be explained by other factors, such as the type of task administered or the involvement of existing knowledge about certain social situations and scripts. In addition, the current study matched the participants with WS to control participants based upon performance on a receptive vocabulary task rather than full IQ. It was predicted that if language abilities predict performance on ToM tasks, those with higher scores on a receptive vocabulary comprehension or grammar comprehension task should pass the verbal ToM tasks in contrast to those with lower language abilities, but there should be no advantage for participants with higher language abilities on the low-verbal ToM tasks. 2. Method 2.1. Participants Thirty participants with WS (12 males, 18 females) who lived within Greater London and were aged between 5 and 17;01 years old were recruited via the Williams Syndrome Foundation UK. All participants had been positively diagnosed for the genetic deletion on the long arm of chromosome 7 using Flourescene In Situ Hybridisation (FISH). Verbal ability was measured using receptive vocabulary scores from the British Picture Vocabulary Scale (BPVS II: Dunn, Dunn, Whetton, & Burley, 1997) and the Pattern Construction (PC) subtest from the British Ability Scales was administered to measure visuospatial abilities (Elliot, Smith, & McCulloch, 1996). An additional language measure was obtained for a sub-sample of the WS group, namely receptive grammatical ability scores from Test for Reception of Grammar (TROG-2; Bishop, 2003). This task consists of 20 blocks which each containing 4 trials related to one grammatical construct. A block is passed when the participant fails no more than one trial per block. The number of blocks passed was recorded. For the control group, TD children were recruited from schools in Greater London and parental meeting groups in local libraries. Thirty TD children (14 males, 16 females) were selected and pairwise matched on the BPVS raw scores (within 4 points) to the WS group (t(58) = 0.145, p = 0.885). This ensured that both groups had similar receptive vocabulary comprehension abilities and allowed investigation whether a similar verbal advantage could be observed in both groups. This meant that the TD children were significantly younger (aged between 2;10 and 9;9 years) than the WS group
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
443
Table 1 Participant details: chronological age, British Picture Vocabulary Scales, Test For Reception of Grammar and pattern construction raw scores. Task
CAc in months BPVSd raw score TROG-2e blocks passed PCf raw score a b c d e f
TDa
WSb
N
Mean (SD)
Range
N
Mean (SD)
Range
30 30 N/A 24
73.03 (24.20) 63.67 (20.32) N/A 18.50 (10.45)
34–117 24–112 N/A 1–38
30 30 18 25
119.60 64.43 5.78 8.60
60–205 25–112 2–12 1–24
(39.52) (20.72) (3.19) (6.73)
Typical developing group. Williams syndrome. Chronological age. British Picture Vocabulary Scale. Test for reception of grammar. Pattern construction.
(t(48.059) = 5.504, p < 0.001). Yet, the PC raw scores for TD participants were significantly higher in comparison to the WS group; t(47) = 3.958, p < 0.001, which means that the TD group had better visuo-spatial abilities than the WS group. None of the TD children had been diagnosed with specific learning or behavioural difficulties according to a parental questionnaire. In addition, the TD children had a similar family background and SES compared to the WS group, as measured by the highest level of education of the mother and the number of single parents in each group. An overview of the ages, receptive vocabulary scores (BPVS) and visuo-spatial ability scores (PC raw scores) for each group can be found in Table 1. All parents were sent an invitation to take part in the study including the information letter about the study and the consent form and all parents and children consented to take part prior to the study. 2.2. Materials and methods 2.2.1. Sally–Anne The Sally–Anne task was carried out using props and dolls in accordance with the scenario illustrated in Frith (1989, 83). In this scenario, the child was first introduced to two dolls, Sally and Anne. When the child could tell the experimenter which doll was Anne and which one was Sally, the experimenter told the following narrative: ‘‘Sally has a marble. She puts the marble into her basket. Sally goes out for a walk. While Sally is away, Anne takes the marble out of the basket, and puts it into the box. Then, Anne goes home for her dinner. Now Sally is back from her walk. She wants to play with her marble.’’ Next, the participant was asked a first-order belief question (‘‘Where will Sally look for the marble?’’) followed by two control questions: a reality (‘‘Where is the marble really?’’) and a memory (‘‘Where was the marble in the beginning?’’). Participants received a score of zero or one for the belief question. However, only the data from participants who answered both the memory and control question correctly were included in the analyses below. This ensured that participants did not fail the belief questions in the tasks due to attention or memory problems. 2.2.2. Video task This task was inspired from a previous study by Newton and de Villiers (2007). In this task the participant watched a false belief video similar to the scenario in the Sally–Anne task, with the contrast that this task did not include a narrative. In order to avoid gender preferences, there were two conditions of the video, one including two male protagonists and a game console and one including two female protagonists and an apple. Participants were randomly allocated to each condition, so that half of the participants watched the male condition and the other half watched the female condition. In the video two protagonists (either two girls or two boys) are in a room, where there is a table with a basket and a box on (see Fig. 1). Protagonist A has an object (either an apple or a game console). Protagonist A puts the object in the basket. Protagonist A yawns, stretches, and leaves the room. While protagonist A is away, protagonist B goes to the basket and moves the object into the box. Then protagonist B leaves the room through a different door. Protagonist A returns and the participant was asked the exact same three questions as in the classic Sally–Anne task, namely ‘‘Where will this girl/boy look for the apple/game?’’ (Belief question), a reality question (‘‘Where is the apple/game now?’’), and a memory question (‘‘Where did the girl/boy put the apple/game?’’). Participants answered these questions by pointing to a picture out of three options (a picture of protagonist A looking into the basket, protagonist A looking into the box, and protagonist A looking under the table) while the experimenter recorded the participant’s response. These pictures were presented from left to right on the screen in randomised order. Similar to the Sally–Anne task participants received a score of zero or one for the belief question, but only those scores for participants who had the control questions correct were included in the analyses. 2.2.3. Balloon task The Balloon task was adapted from Woolfe, Want, and Siegal (2002). To start, each participant was given two practice trials to establish whether he/she was familiar with the concept of a thought balloon. Each participant was given a card on which two pictures of a protagonist were depicted next to one another. In one of the pictures, the protagonist was holding an object, while in the second picture the protagonist was thinking about the object which was illustrated using a
444
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
Fig. 1. A photo shot of the non-verbal Video task.
thought-balloon. The participant was asked, in alternating order, to select the picture where the protagonist was thinking about the object and where the protagonist actually had the object in his hand. After the first trial, regardless of whether the questions were answered correctly, each participant was explicitly told that a thought-balloon shows what a person is thinking (Wellman, Hollander, & Schult, 1996). Then, a second card was shown and the participant was asked the same questions. Participants who failed to answer the questions for the second card, and thus failed to understand that a thoughtballoon depicts someone’s thoughts, were excluded from the study. Next, the participant was administered one false-belief trial. To avoid gender preferences, two conditions were created: one including a boy looking at a birdhouse that contained a mobile phone and one including a girl approaching a letterbox that contained an apple. Half of the participants were administered the girl approaching the letterbox and the other half completed the condition in which the boy looked at the birdhouse. In each condition, the participant was shown a picture in which a flap was hiding an object (e.g. the participant sees a picture of a girl next to a letter box, but a flap over the letter box hides what is actually in the letter box) (see Fig. 2). The participant was instructed to cover the eyes of the protagonist in the picture and to have a look what was behind the flap. Then, the participant was shown a separate picture depicting only the face of the protagonist with a blank thought balloon. Below the thought balloon there were three pictures (a control picture, a picture of what was really behind the flap, and a picture depicting what would be most probably there and thus what the protagonist believed was under there). The control picture depicted a possible, but unrelated answer (e.g. a flower). The child was asked a ToM belief question (‘‘What does X think is behind the flap?’’) and a control question (i.e. ‘‘What was actually behind the flap?’’). The participant answered by pointing to one of the three pictures and this choice was recorded by the experimenter. Participants received a score of one or zero for the belief question. However, scores of participants who answered the control question incorrectly were not included in the data analyses reported below.
Fig. 2. Example of a low-verbal Balloon trial.
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
445
Table 2 Number of participants per task and group that passed or failed the false belief question. Group
Pass (%)
Fail (%)
Total N of participantsa
Sally–Anne task
b
TD WSc
20 (67) 6 (21)
10 (33) 22 (79)
30 28
Video task
TD WS
17 (74) 5 (29)
6 (26) 12 (71)
23 17
Balloon task
TD WS
17 (68) 9 (35)
8 (32) 17 (65)
25 26
Task
a b c
Participants were excluded because they failed the control and/or memory question. For none of the cases these were the youngest or less able children. Typical developing group. Williams syndrome.
3. Results 3.1. Performance on control and memory questions As shown in Table 2, in both groups more children had to be excluded for the Video task than the Sally–Anne task (TD: p < 0.05; WS: p < 0.01, Fisher’s exact test), even though the control and memory questions in both tasks were identical. Fisher exact tests showed that more TD participants had to be excluded for the Balloon task than the Sally–Anne task (p < 0.01), but there were no differences with the Video task (p > 0.05). In contrast, fewer WS participants had to be excluded for the Balloon task in contrast to the Video task, but there were no differences with the Sally–Anne task (p > 0.05). It is important to note that those TD who failed the control questions in the Balloon task also failed these questions in the Video task (N = 5) while in the WS group this was not the case (N = 1). 3.2. Performance on belief question Comparison of overall performance between the two groups for the Sally–Anne task showed that more participants in the WS group (79%) failed the false belief question in contrast to the TD group (33%) (the Fisher’s exact test: p = 0.001) (see Table 2). As shown in Table 2, more TD participants passed the Video task compared to the WS group (p = 0.010, Fisher’s exact test). Subsequently, it was investigated whether reducing the verbal demands in the Video task had resulted in more or less successful performance on the belief question. Although three WS participants (25%) and one TD participant (17%) who failed the Sally–Anne task passed the belief question in the Video task, Fisher exact test did not show any significant improvements for either group (all p > 0.05). Thus, performance on the Video task did not differ from the Sally–Anne task, even though overall language and memory demands were reduced in the Video task. For the Balloon task, more WS participants failed the belief question compared to the TD group (p = 0.037, Fisher’s exact test). Comparison of performance on the Balloon task to the Sally–Anne task showed no significant differences in either group (p > 0.05, Fisher’s exact test). However, for the TD group more participants failed the belief question in the Balloon compared to the Video task. This shows that the Balloon task was significantly harder than the Video task (Fisher exact, p = 0.026), while there were no such differences for the WS group (p > 0.05, Fisher’s exact). 3.3. Performance on belief question and relationship to age, receptive vocabulary, receptive grammatical comprehension, and visuo-spatial abilities It was investigated whether chronological age was a good predictor for passing each of the three tasks. In the TD group, those who passed the ToM tasks were older than those who failed; Sally–Anne: t(28) = 2.953, p = 0.006; Video task: t(21) = 2.560, p = 0.018, Balloon task: t(24) = 3.044, p = 0.006. There were no differences for chronological age for the WS group for any of the three ToM tasks (all p’s > 0.05). An overview is provided in Table 3. Next, it was investigated whether receptive vocabulary abilities were a good predictor for passing each of the three tasks. There was a significant difference for BPVS raw scores in the TD group between those participants who passed the belief question and those who failed the belief question for each of the three tasks (Sally–Anne t(28) = 2.953, p = 0.006; Video task t(21) = 2.096, p = 0.048; Balloon task t(18) = 2.799, p = 0.012). However, this was not the case for the WS group (all p’s > 0.05). In addition, there were no significant differences between those in the WS group who passed and those who failed the Sally–Anne task; t(14) = 0.021, p = 0.837, the Video task; t(8) = 1.691, p = 0.129, or the Balloon task; t(16) = 0.862, p = 0.401 for receptive grammatical ability scores as measured by the amount of blocks passed on the TROG-2. Finally, the relationship between passing or failing the belief questions on the three ToM tasks and visuo-spatial abilities, as measured by PC, was examined. In the TD group there was a significant effect between those who passed and failed only for the Sally–Anne task; t(22) = 2.380, p = 0.026. There were no significant differences for the Video or the Balloon task (all p’s > 0.05). Again, there were no significant differences for PC scores between those who passed or failed the belief tasks for either of the ToM tasks in the WS group (all p’s > 0.05).
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
446
Table 3 Chronological age, BPVS raw scores, TROG blocks passed, and PC raw scores (means and standard deviation) per group for those who passed and those who failed the belief question in (a) Sally–Anne task, (b) Video task, and (c) Balloon task. Task
TDa Passed
WSb Failed
Mean (SD)
Passed
Failed
Mean (SD)
CAc in months
Sally–Anne task Video task Balloon task
81.80 (20.91) 85.18 (20.73) 86.24 (16.54)
55.50 (21.25) 60.50 (18.87) 62.22 (23.49)
134.17 (34.97) 114.80 (50.87) 140.13 (31.84)
118.59 (41.23) 112.25 (31.82) 120.19 (42.89)
BPVSd raw scores
Sally–Anne task Video task Balloon task
70.55 (18.14) 72.65 (18.82) 74.00 (12.15)
49.90 (17.88) 53.33 (21.17) 51.67 (24.06)
74.14 (19.34) 78.20 (29.71) 69.69 (25.18)
61.48 (20.60) 61.42 (16.60) 64.56 (20.38)
TROGe blocks passed
Sally–Anne task Video task Balloon task
N/A N/A N/A
N/A N/A N/A
6.14 (3.81) 9.00 (5.20) 7.00 (3.74)
5.78 (3.15) 5.00 (2.82) 5.43 (3.08)
PCf raw scores
Sally–Anne task Video task Balloon task
21.47 (9.55) 21.07 (11.19) 20.80 (9.87)
11.29 (9.46) 16.20 (8.758) 15.14 (11.57)
10.17 (8.70) 15.00 (8.98) 8.00 (8.21)
8.50 (6.12) 7.11 (5.65) 9.58 (6.75)
a b c d e f
Typical developing group. Williams syndrome. Chronological age. British Picture Vocabulary Scale. Test for reception of grammar. Pattern construction.
4. Discussion The current study aimed to clarify whether receptive language abilities, such as vocabulary and grammar comprehension abilities, or the task demands would influence performance on ToM tasks, since some previous studies found evidence that WS rely on their language abilities to pass verbal ToM tasks (Porter et al., 2008; Santos & Deruelle, 2009), while other studies suggested that participants with WS specifically fail verbal ToM tasks due to their language demands (Sullivan & TagerFlusberg, 1999; Tager-Flusberg & Sullivan, 2000; Tager-Flusberg et al., 1997). Yet, in contrast to previous studies, the current study aimed to compare performance on the classic Sally–Anne task to a low-verbal change-location task that was similar (Video task) and one unexpected-content task that was dissimilar in design (Balloon task) in order to investigate differences in task demands. In addition, the current study matched participants based on their receptive vocabulary abilities rather than full scale IQ since participants with WS show uneven cognitive profiles and matching on full-IQ might result in underestimation of the language abilities in WS. In line with previous studies, the results showed that participants with WS performed worse on the Sally–Anne task compared to control participants who had similar receptive vocabulary scores but were chronologically younger than the WS participants (Tager-Flusberg & Sullivan, 2000; Tager-Flusberg et al., 1997). However, performance in the WS group did not worsen or improve when they were assessed on a low-verbal task (Video task) that was similar in design to the Sally– Anne task. In addition, while participants in the TD group who passed the belief questions in both tasks had higher vocabulary comprehension scores than those who failed, there was no such difference in the WS group. For a sub-sample of the WS-group, receptive grammatical comprehension scores measured by TROG-2 were available. Comparisons between those who passed and those who failed the three ToM tasks in this study showed that there were no differences for TROG-2 scores. This shows that performance on ToM tasks is not driven by poor or proficient language abilities in the WS group in contrast to what some previous studies have suggested. Comparison between the three tasks showed that a significant number of TD and WS participants had to be excluded from the data analyses in the Video task compared to the Sally–Anne task as they failed to answer the control and memory questions correctly. The fact that these questions were identical to the Sally–Anne task shows that explicit verbal narration aided children’s context understanding or engagement with the task in general which allowed more children to pass the control and memory questions. Santos and Deruelle (2009) reported that participants with WS only showed a deficit on visual ToM trials but not on trials that included language. However, their study did not include any control questions to see whether the participant was engaged with the task at hand. As the current study shows, low-verbal visual tasks include less contextual information and are thus more difficult. This might increase the risk of participants losing interest in the task at hand and this should be taken into account in future visual tasks. Finally, the current study also compared performance on the Sally–Anne task to a low-verbal task that was dissimilar in structure (i.e., unexpected content versus change-location task) and task demands (verbal versus low-verbal). Again, participants in either group did not perform any better or worse on the belief question in the Balloon task compared to the Sally–Anne task and there were no differences in the WS group for receptive vocabulary or grammar abilities between those who passed and those who failed the belief question in the Balloon task. Yet, the TD group found the Balloon task much
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
447
harder than the Video task, in that not only did more children fail to understand the concept of thought balloons, there were also more children who failed the belief question in the Balloon task in contrast to the Video task. So while the WS group seemed to struggle with the Video task, more TD children had to be excluded or failed the belief question in the Balloon task. This shows that task demands of the ToM tasks employed need to be taken into account when assessing children’s ToM abilities. In all three tasks the questions were posed in verbal format and children answered by pointing or providing a verbal answer (whichever they preferred). In addition, the Sally–Anne and Video task included the exact same questions. Therefore, how children were assessed in the three tasks was similar, yet the tasks differed in the kind of information children had to infer from the task itself and how the task was presented, i.e. whether further narratives were provided or not. This finding is in line with results from Van Buijsen et al. (2011) which have shown that ToM performance in children with autism and SLI is also dependent upon the task employed. One possible reason why the TD children performed worse on the Balloon task compared to the Video task is that the Balloon task relies upon knowledge of scripts (i.e., the knowledge that a letter box would typically contain a letter and a birdhouse would not generally contain a phone) and these scripts would develop or become more set as children grow older (Nelson, 1986). In contrast, the fact that the WS participants were older than the TD group might therefore explain why they did not find the Balloon task harder. This explanation is supported by the fact that chronological age, but not visuo-spatial abilities, was a good predictor for performance on the Balloon task in the TD group. 5. Conclusions In sum, the current study did not find any evidence that performance on ToM tasks is driven by language abilities, as measured by a receptive vocabulary and grammar abilities, in participants with WS or that participants with WS do better on low-verbal compared to verbal ToM tasks. However, the current study did find evidence that performance depends upon the overall cognitive demands a task poses. For example, tasks that included explicit narratives of what was happening proved easier than those tasks in which children had to infer what was happening. In addition, our evidence suggests that knowledge of certain schemas was an important factor for performance on low-verbal ToM tasks. This finding has implications for clinicians and researchers who wish to establish ToM abilities in developmental disorders or who would like to make comparisons between different groups, in that the outcomes will depend upon the kind of task administered. In addition, reducing the verbal narrative of the ToM task reduced the amount of information available about what is happening in the situation and this seems to affect the level of difficulty or engagement with the task in children. However, the current study only included controls that were matched for receptive vocabulary abilities using BPVS scores and scores for grammatical abilities were only available for a sub-sample of our WS group. Although, both provided some indication of language abilities in children, future studies should include control participants matched for overall language abilities, including story comprehension abilities, as these might be better predictors for ToM performance. In addition, future studies should also explore why participants perform worse on visual or low-verbal ToM tasks. For example, the use of eye-tracking methodology would allow investigating engagement to the task as well as differences in looking behaviour. Acknowledgements We thank the Williams Syndrome Foundation UK and all the children who took part in this study. This research was supported by an ESRC open-competition grant (PTA-030-2006-00235), a grant from the University of London Central Research Fund (REGCRF06-07), a Departmental Scholarship from King’s College London, a grant from the Williams Syndrome Foundation, UK and a grant from the Nuffield Foundation. We also thank reviewers for their comments on an earlier draft of this manuscript. Conflict of interest statement The authors received financial supportfor this project from ESRC open-competition grant (PTA-030-2006-00235), a grant from the University of London Central Research Fund (REGCRF06-07), a Departmental Scholarship from King’s College London, a grant from the Williams Syndrome Foundation, UK and a grant from the Nuffield Foundation. There are no nonfinancial relationships to disclose. Appendix A. Continuing education CEU Questions 1. Which of the following statements about language development in Williams syndrome is correct: a. Language abilities and non-verbal abilities are equally impaired in Williams syndrome b. Language development in Williams syndrome is delayed and atypical. c. Language development in Williams syndrome is the same as in typically developing controls d. Language abilities are better in comparison to typically developing controls of a similar age 2. Sally–Anne is an example of a low-verbal Theory of Mind task: TRUE/FALSE 3. The current study found that children with Williams syndrome:
448
J. Van Herwegen et al. / Journal of Communication Disorders 46 (2013) 440–448
a. Performed better than the controls on all three tasks b. Performed worse than the controls but only on the verbal task c. Performed worse than the controls but only on the low-verbal tasks d. Performed worse than the controls on all three tasks 4. Receptive grammar abilities predict performance on verbal theory of mind tasks in Williams syndrome: TRUE/FALSE 5. Which two statements are correct for the control group? a. The video task is harder because it does not contain language b. The Sally–Anne task is harder because it contains language c. Both low-verbal tasks are harder because they do not contain language d. The balloon task is harder because the control children are too young and do not have enough world knowledge
References Annaz, D., Van Herwegen, J., Thomas, M., Fishman, R., Karmiloff-Smith, A., & Rundblad, G. (2009). Comprehension of metaphor and metonymy in children with Williams syndrome. International Journal of Language and Communication Disorders, 44(6), 962–978. Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a ‘theory of mind’? Cognition, 21, 37–46. Baron-Cohen, S. (2000). Theory of mind and autism: A fifteen year review. In S. Baron-Cohen, H. Tager-Flusberg, & D. J. Cohen (Eds.), Understanding other minds: Perspectives from developmental cognitive neuroscience (pp. 3–20). Oxford: Oxford University Press. Bellugi, U., Lichtenberger, L., Jones, W., Lai, Z., & St-George, M. (2000). The neurocognitive profile of WS: A complex pattern of strengths and weaknesses. Journal of Cognitive Neuroscience, 12(Suppl.), 7–29. Bishop, D. M. V. (2003). Test for reception of grammar, version-2 (TROG-2). London: Psychological Corporation. Brock, J. (2007). Language abilities in Williams syndrome: A critical review. Development and Psychopathology, 19, 97–127. Carlson, S., Mandell, D., & Williams, L. (2004). Executive function and theory of mind: Stability and prediction from ages 2 to 3. Developmental Psychology, 40, 1105– 1122. Carlson, S., Moses, L., & Breton, C. (2002). How specific is the relation between executive function and theory of mind? Contributions of inhibitory control and working memory. Infant and Child Development, 11, 73–92. de Villiers, J. G., & de Villiers, P. A. (2000). Linguistic determinism and the understanding of false beliefs. In P. Mitchell & K. Riggs (Eds.), Children’s reasoning and the mind (pp. 189–226). Hove, UK: Psychology Press. de Villiers, J., & Pyers, J. (2002). Complements to cognition: A longitudinal study of the relationship between complex syntax and false-belief-understanding. Cognitive Development, 17, 1037–1060. Doyle, T. F., Bellugi, U., Korenberg, J. R., & Graham, J. (2004). ‘‘Everybody in the world is my friend’’ Hypersociability in young children with Williams syndrome. American Journal of Medical Genetics, 124, 263–273. Dunn, L., Dunn, L., Whetton, C., & Burley, J. (1997). British Picture Vocabulary Scale II. Windsor: NFER-Nelson Publishing Company Limited. Einfeld, S. L., Tonge, B. J., & Florio, T. (1997). Behavioural and emotional disturbance in individuals with Williams Syndrome. American Journal on Mental Retardation, 102(1), 45–53. Elliot, C. D., Smith, P., & McCulloch, K. (1996). British Ability Scales Second Edition (BAS II). Windsor: NFER-Nelson Publishing Company Limited. Flynn, E. (2006). A microgenetic investigation of stability and continuity in theory of mind development. British Journal of Developmental Psychology, 24, 631–654. Frith, U. (1989). Autism and theory of mind. In C. Gillberg (Ed.), Diagnosis and treatment of autism (pp. 33–52). New York: Plenum Press. Happe´, F. (1995). The role of age and verbal ability in the theory of mind task performance of subjects with autism. Child Development, 66(3), 843–855. Howlin, P., Davies, M., & Udwin, O. (1998). Syndrome specific characteristics in Williams syndrome: To what extent do early behavioural patterns persist into adult life? Journal of Applied Research in Intellectual Disabilities, 11(3), 207–226. Jarrold, C., Baddeley, A. D., & Hewes, A. K. (1998). Verbal and non-verbal abilities in the Williams syndrome phenotype: Evidence for diverging developmental trajectories. Journal of Child Psychology and Psychiatry, 39, 511–524. Jones, W., Bellugi, U., Lai, Z., Chiles, M., Reilly, J., Lincoln, A., et al. (2001). Hypersociability: The social and affective phenotype of Williams syndrome. In U. Bellugi & M. George (Eds.), Journey from cognition to brain to gene: perspectives from Williams syndrome (pp. 43–72). Massachusetts: MIT Press. Karmiloff-Smith, A., Klima, E. S., Bellugi, U., Grant, J., & Baron-Cohen, S. (1995). Is there a social module? Language, face processing, and theory of mind in subjects with Williams syndrome. Journal of Cognitive Neuroscience, 7(2), 196–208. Laing, E., Butterworth, G., Ansari, D., Gso˝ dl, M., Longhi, E., Panagiotaki, G., et al. (2002). Atypical development of language and social communication in toddlers with Williams syndrome. Developmental Science, 5(2), 233–246. Laws, G., & Bishop, D. V. M. (2004). Verbal deficits in specific language impairment and Down syndrome: A comparative review. International Journal of Language and Communication Disorders, 39, 423–451. Lohmann, H., & Tomasello, M. (2003). The role of language in the development of false belief understanding: A training study. Child Development, 74, 1130–1144. Milligan, K., Astington, J. W., & Dack, L. A. (2007). Language and theory of mind: Meta-analysis of the relation between language ability and false-belief understanding. Child Development, 78(2), 622–646. Nelson, K. (1986). Event knowledge. Structure and function in development. Hillsdale, NJ: Lawrence Erlbaum Associates. Newton, A. M., & de Villiers, J. G. (2007). Thinking while talking: Adults fail nonverbal false-belief reasoning. Psychological Science, 18(7), 574–579. Porter, M. A., Coltheart, M., & Langdon, R. (2008). Theory of mind in Williams syndrome assessed using a nonverbal task. Journal of Autism and Developmental Disorders, 38, 806–814. Santos, A., & Deruelle, C. (2009). Verbal peaks and visual valleys in theory of mind ability in Williams syndrome. Journal of Autism and Developmental Disorders, 39(4), 651–659. Schubert, C. (2009). The genomic basis of the Williams–Beuren syndrome. Cellular and Molecular Life Science, 66, 1178–1197. Sullivan, K., & Tager-Flusberg, H. (1999). Second-order belief attribution in Williams syndrome: Intact or impaired? American Journal on Mental Retardation, 104(6), 523–532. Tager-Flusberg, H., Boshart, J., & Baron-Cohen, S. (1998). Reading the windows to the soul: Evidence of domain-specific sparing in Williams syndrome. Journal of Cognitive Neuroscience, 10, 631–639. Tager-Flusberg, H., & Sullivan, K. (2000). A componential view of theory of mind: Evidence from Williams syndrome. Cognition, 76, 59–89. Tager-Flusberg, H., Sullivan, K., & Boshart, J. (1997). Executive functions and performance on false belief tasks. Developmental Neuropsychology, 13(4), 487–493. Van Buijsen, M., Hendriks, A., Ketelaars, M., & Verhoeven, L. (2011). Assessment of theory of mind in children with communication disorders: Role of presentation mode. Research in Developmental Disabilities, 32, 1038–1045. Van Herwegen, J., Rundblad, G., Davelaar, E. J., & Annaz, D. (2011). Variability and standardised test profiles in typically developing children and children with Williams syndrome. British Journal of Developmental Psychology, 29(4), 883–894. Wellman, H., Cross, D., & Watson, J. (2001). Meta-analysis of theory of mind development: The truth about false-belief. Child Development, 72(3), 655–684. Wellman, H., Hollander, M., & Schult, C. (1996). Young children’s understanding of thought bubbles and of thoughts. Child Development, 67(3), 768–788. Woolfe, T., Want, S. C., & Siegal, M. (2002). Signposts to development: Theory of mind in deaf children. Child Development, 73(3), 768–778.