Table 3 Overview of the reviewed studies which used eye-tracking to measure social cognition in intellectual disability

Expression discrimination. Gaze towards salient facial features across different emotional facial expressions

Campbell et al. (2010) [78]

10

22q11.2 deletion syndrome (17)

17.2 ± 3.2

FSIQ 72.8 ± 13.2

Neurotypical (17) *Gender, CA

35 images of human facial expressions (neutral, happy, sad, surprise, angry, fear, and disgust). Participants were asked to explicitly identify the expression viewed.

Across expressions, people with 22q11.2 deletion syndrome looked less at the eyes, more at the mouth, and had fewer fixations/shorter scan paths.

Crawford et al. (2015a) [65]

11

Fragile X syndrome

(13)

19.7 ± 9.07

VABSr 357.9 ± 95.6

Autism (15)

*VABSr

38 images of human facial expressions (happy, disgust) paired against neutral faces.

People with fragile X syndrome looked less at the eye region of neutral faces than autistic people. Both groups showed a spontaneous preference for emotional expressions.

Crawford et al. (2015b) [66]

11

Cornelia de Lange syndrome (15)

18.4 ± 9.8

VABSc 59.9 ± 25.0

* CA, gender, autism traits, VABSc

38 images of human facial expressions (happy, disgust) paired against neutral faces.

People with Cornelia de Lange and Rubinstein-Taybi syndrome looked similarly at the eye and mouth region and showed spontaneous preference for emotional expressions.

Rubinstein-Taybi syndrome (17)

17.3 ± 10.1

VABSc 58.5 ± 15.1

Dalton et al. (2008) [79]

8

Fragile X syndrome

(9)

20.7 ± 2.8

FSIQ 66.1 ± 23.8

Autism (14)

Images of human emotional (happy, fear, disgust) and neutral facial expressions. Participants were asked to explicitly identify the expression viewed.

People with fragile X syndrome looked similarly at facial features as autistic people, but less at the eye region than neurotypical people.

Neurotypical (15)

Debladis et al. (2019) [80]

12

Prader-Willi syndrome (39)

28.0 ± 8.0

FSIQ 57.0 ± 10.0

Neurotypical (20)

35 images of human emotional (happy, sad, fear) and neutral facial expressions. Participants were asked to explicitly identify the expression viewed.

People with PWS (mUPD) looked less at the eye region and had more difficulty with spontaneous discrimination of emotional expressions than those with PWS (deletion) and neurotypical controls.

Djukic et al. (2014) [81]

11

Rett syndrome (37)

10.0 ± 7.7

VABSc 45.9 ± 12.5

Neurotypical (34) *Gender, CA

12 images of human emotional (happy, sad, fear) and neutral facial expressions. Participants were asked to explicitly identify the expression viewed.

People with Rett syndrome had difficulty spontaneously discriminating emotional expressions and spent less time fixating on salient facial features, unlike the neurotypical group.

Farzin et al. (2009) [82]

7

Fragile X syndrome (16)

17.0 ± 6.8

FSIQ 58.4 ± 9.8

Neurotypical (16) *Gender, CA

60 images of human facial expressions (calm, happy, fear) and 60 scrambled versions of the faces. Participants were asked to explicitly identify the expression viewed.

People with fragile X syndrome showed increased pupil reactivity to facial expressions and looked less at the eye region across faces, compared to neurotypical comparison groups.

Farzin et al. (2011) [83]

10

Fragile X syndrome (15)

18.8 ± 10.7

FSIQ 57.5 ± 14.5

Neurotypical (20) *CA

Replication of Farzin et al (2009).

High test-retest reliability of gaze and pupillometry measures.

Franchini et al. (2016) [84]

11

22q11.2 deletion syndrome (35)

18.2 ± 5.9

FSIQ 57.0 ± 10.0

Neurotypical (20)

*Gender, CA

48 videos of dynamic avatar facial expressions (anger, fear, happy, sad). Participants were asked to explicitly identify the expression viewed.

People with 22q11DS were slower to recognise emotional expressions than neurotypical controls. They also spent less time looking at the nose during happy and fearful faces.

Gomez et al. (2020) [85]

11

Williams syndrome (22)

12.4 ± 3.8

NA

Neurotypical (21)

*Gender, CA

48 pairs of images of avatars with ‘trustworthy’ and ‘untrustworthy’ expressions. Spontaneous preference was measured alongside explicit appraisal of trustworthiness.

People with Williams syndrome did not exhibit a preference for trustworthy faces, unlike neurotypical people.

Hanley et al. (2013) [70]

11

Williams syndrome (15)

21.9 ± 9.3

FSIQ 72.8 ± 13.2

Neurotypical (15) *CA

Image and video depictions of eight mental states (e.g. relieved, surprised, worried) expressed by an actor. Participants were asked to explicitly identify the actor’s mental state.

In contrast to both comparison groups. people with Williams syndrome looked less at salient facial features whilst making judgements about mental states.

Neurotypical (14) *BPVS

Kirk et al. (2013) [86]

10

Williams syndrome (13)

23.6 ± 6.9

BPVSr 132.0 ± 18.9

Neurotypical (13) *CA

20 images of human emotional (angry, happy, sad, fear) and neutral facial expressions. Participants were asked to explicitly identify the expression viewed.

People with Williams syndrome fixated on the eye region of faces similarly to comparison groups. High levels of behavioural anxiety associated with reduced gaze towards the eye regions of threatening facial expressions in Williams syndrome.

Neurotypical (13) *BPVS

McCabe et al. (2011) [87]

10

22q11.2 deletion syndrome (18)

17.4 ± 3.1

FSIQ 73.8 ± 13.6

Neurotypical (17)

*Gender, CA

28 images of human emotional (happy, sad, surprise, disgust, afraid, angry) and neutral faces. Discrimination of 35 non-face stimuli (weather scenes) also measured. Participants were asked to explicitly identify the expression/weather viewed.

People with 22q11.2 deletion syndrome demonstrate fewer and longer fixations across facial expression and non-face (weather scenes) stimuli compared to a neurotypical comparison group.

McCabe et al. (2013) [88]

7

22q11.2 deletion syndrome (20)

16.8 ± 3.7

FSIQ 72.1 ± 13.0

Autism (14) *CA

28 images of human emotional (happy, sad, surprise, disgust, afraid, angry) and neutral faces. Discrimination of 35 non-face stimuli (weather scenes) also measured. Participants were asked to explicitly identify the expression/weather viewed.

For faces, the 22q11.2 deletion syndrome and autism groups demonstrated lower emotion recognition accuracy and fewer fixations compared to the neurotypical group. People with 22q11.2 deletion syndrome looked more at weather scenes than autistic people yet had more difficulty with explicit appraisal than autistic and neurotypical groups.

Neurotypical (20) *CA

Porter et al. (2010) [89]

9

Williams syndrome (16)

25.1 ± 11.7

FSIQ 61.0 ± 15.0

Neurotypical (16) *Gender, FSIQ

24 images of human emotional (happy, sad, surprise, disgust, afraid, angry) and neutral faces. Participants were asked to explicitly identify the expression viewed.

People with Williams syndrome did not look at the eye region faster than neurotypical people. But, once attended people with Williams syndrome spent more time looking at the eye region, People with William syndrome’s scan paths were similar across the different facial expressions but showed most difficulty with explicit recognition of anger.

Shaw & Porter (2013) [90]

11

Fragile X syndrome

(16)

24.8 ± 12.9

FSIQ 64.0 ± 13.7

Neurotypical (16) *CA, gender

24 images of human emotional (happy, sad, surprise, disgust, afraid, angry) and neutral faces. Participants were asked to explicitly identify the expression viewed.

Whilst people with FXS displayed reduced fixations on the eyes and scanned facial expressions significantly differently compared to the CA-matched neurotypical comparison group, scan paths were similar to the MA-matched neurotypical comparison group.

Neurotypical (16) *MA, gender

Facial recognition. Gaze towards novel faces when presented alongside a familiar face.

Glaser et al. (2010) [98]

12

22q11.2 deletion syndrome

(26)

12.4 ± 1.9

FSIQ 74.2 ± 11.8

Neurotypical (22) *CA

Paired images of human faces with neutral expressions, with either configural differences (30 trials) or featural differences (30 trials). Participants were asked to identify if the faces were the ‘same’ or ‘different’ to each other.

Few differences in explicit recognition accuracy between people with 22q11.2 deletion syndrome and neurotypical comparison group. People with 22q11.2 deletion syndrome spent more time looking at the mouth region than the eye region than those with non-syndromic ID.

Non-syndromic ID

(17)

11.5 ± 2.7

FSIQ 68.4 ± 10.2

Guillory et al. (2021) [99]

9

Phelan-McDermid syndrome (8)

9.2 ± 3.4

DQ 35.8 ± 19.8

Neurotypical (26) *CA

Identical images of human faces (happy) presented in pairs (familiarisation), which switched with an image of a novel face (pre-switch). The two images then swap sides (post-switch). Nine trials in total. Repeated with non-social images.

Rate of looking back-and-forth between images was lowest in the non-syndromic ID (+ autism) group. Across indices, people with Phelan-McDermid syndrome (+ autism) looked more similarly to people with Phelan-McDermid syndrome-alone, rather than the non-syndromic ID (+ autism) group.

Phelan-McDermid syndrome (+ autism) (14)

9.9 ± 4.2

DQ 17.3 ± 12.2

Non-syndromic ID (+ autism) (7)

7.8 ± 2.7

DQ 45.5 ± 21.1

Rose et al. (2013) [100]

10

Rett syndrome (27)

10.6 ± 6.8

VABSc 44.2 ± 10.1

Neurotypical (30) *Gender, CA

Identical images of human faces (happy) presented in pairs (familiarisation), which switched with an image of a novel face (pre-switch). The two images then swap sides (post-switch). Nine trials in total. Repeated with non-social images.

Recognition of novel faces was poorer in Rett syndrome compared to neurotypical comparison group. People with Rett syndrome’s gaze was characterised by fewer and longer fixations. They all tended to ignore the mouth/nose region.

Yi et al. (2014) [101]

11

Non-syndromic ID (22)

23.6 ± 3.1

CRTr 23.8 ± 8.6

Neurotypical (28) *CA

74 images of neutral human faces: 37 own race and 37 other race. Presented individually, followed by same/different face. Participants were asked to identify if the faces were the ‘same’ or ‘different’ to each other.

Non-syndromic ID (+ autism) group looked more at the nose region. Non-syndromic ID groups scanned the whole face less than neurotypical comparison group.

Non-syndromic ID (+ autism) (19)

20.8 ± 3.3

CRTr 23.7 ± 9.4

Yi et al. (2015) [73]

11

Non-syndromic ID (26)

23.0 ± 3.1

CRTr 22.0 ± 8.8

Neurotypical (30) *CA

36 images of neutral human faces. Presented individually, followed by same/different face. Participants were asked to identify if the faces were the ‘same’ or ‘different’ to each other.

Non-syndromic ID groups displayed better recognition of own race faces. All groups displayed more looking at the eye region of other-race faces relative to own race faces.

Non-syndromic ID (+ autism) (24)

20.7 ± 3.9

CRTr 22.4 ± 9.0

Zaharia et al. (2018) [102]

12

22q11.2 deletion syndrome (time 1)

(75)

12.8 ± 3.6

FSIQ 72.0 ± 11.3

Neurotypical (time 1) (84) *CA

Paired images of human faces with neutral expressions, with either configural differences (30 trials) or featural differences (30 trials). Participants were asked to identify if the faces were the ‘same’ or ‘different’ to each other.

People with 22q11.2 deletion syndrome look more at the mouth region and demonstrate restricted scan paths, with a reduced number of transitions between faces and longer fixations compared to the neurotypical comparison group. Similar scan paths evident across time 1 and 2 in 22q11.2 deletion syndrome.

22q11.2 deletion syndrome (time 2) (36)

16.5 ± 3.1

FSIQ 69.1 ± 11.6

Neurotypical (time 2) (30)*CA

Social preference. Proportion of gaze towards paired social versus non-social (e.g. geometric shapes) stimuli.

Cooke et al. (2019) [64]

8

Phelan-McDermid syndrome (15)

8.9 ± 0.8

NA

Autism (24)

8 trials in which a human face was displayed among an array of four non-social (e.g. car, bird) images.

The majority of children with Phelan-McDermid syndrome and autistic children showed reduced: 1) first looks to the face and 2) less overall looking at face, compared to neurotypical children.

Neurotypical (27)

Crawford et al. (2016) [67]

11

Fragile X syndrome (15)

18.2 ± 5.6

VABSc 51.3 ± 17.4

Autistic and neurotypical children included in separate study to establish baseline.

Paired videos of actor (social) or object (non-social) moving towards/away from the camera. 28 trials in total.

Autistic people looked less at social versus non-social videos only when stimuli were moving towards them. Individuals in the three genetic syndrome groups showed similar looking-time but differences in fixation latency for social stimuli moving towards them.

Cornelia de Lange syndrome (14)

20.9 ± 11.9

VABSc 47.9 ± 16.0

Rubinstein-Taybi syndrome (19)

24.2 ± 8.6

VABSc 47.8 ± 14.6

Hirai et al. (2016a) [92]

10

Williams syndrome (21)

16.2 ± 7.1

RCPMr 18.3 ± 5.0

Neurotypical (21)

*RCPMr

120 trials in which non-social images (e.g. clock, apple) presented in array alongside: 1) a human face, 2) a butterfly, 3) a face and butterfly, 4) neither face nor butterfly. Participants asked to explicitly identify when butterfly was present.

Unlike neurotypical comparison groups, people with Williams syndrome looked more at the face during later search stages. Slower explicit recognition of the butterfly correlated with increased fixation on face in Williams syndrome.

Neurotypical (21)

*CA

Hirai et al. (2016b) [93]

9

Williams syndrome (17)

16.7 ± 6.5

RCPMr 18.4 ± 5.1

Neurotypical (17)

*RCPMr

120 trials in which non-social images (e.g. clock, apple) presented in array alongside: 1) an inverted human face, 2) a butterfly, 3) an inverted face and butterfly, 4) neither face nor butterfly. Participants asked to explicitly identify when butterfly was present.

When the butterfly and inverted face were present in the same search array, explicit recognition of the butterfly was similar in Williams syndrome relative to neurotypical comparison groups. Neither group looked preferentially towards inverted faces.

Neurotypical (17)

*CA

Hong et al. (2017) [93]

9

Angelman syndrome (8)

12.4 ± 10.7

VABSc 44.8 ± 20.7

Neurotypical (N not reported)

*Gender, CA

Three trials in which a video clip of children interacting (social) was paired with geometric shapes (non-social).

Relative to neurotypical comparison groups, Angelman syndrome looked less at social stimuli, and showed increased pupil dilation for non-social stimuli. No difference between people with Angelman syndrome and autistic children.

Autism (N not reported)

*Gender, CA

Hong et al. (2019) [71]

10

Fragile X syndrome (17)

16.6 ± 6.1

VABSc 61.2 ± 12.0

Neurotypical (17)

*Gender, CA

Three trials where a video clip of children interacting (social) was paired with geometric shapes (non-social). Also, twelve trials in which an image of face was paired with scrambled (/geometric) face.

Whilst autistic comparison groups showed significantly less social preference, people with fragile X syndrome displayed social preference similar to the neurotypical comparison group.

Autism (17)

*Gender, CA

Ponson et al. (2018) [94]

7

Phelan-McDermid syndrome (18)

12.7 ± 9.2

DQ 25.4 (10-55)

Neurotypical (N not reported)

Ten trials, in which an image of a neutral face (social) was paired with an object (non-social).

Unlike the neurotypical comparison group, people with Phelan-McDermid syndrome showed increased pupil dilatation when viewing the non-social versus social stimuli.

Autism (N not reported)

Riby & Hancock (2009a) [95]

9

Williams syndrome (14)

8.75 - 28.0

RCPMr 14

Neurotypical (14)

*RCPMr

Images include 1) scrambled pictures containing faces (20 trials) and 2) pictures of scenes with embedded faces (9 trials) or no faces (9 trials).

People with Williams syndrome showed prolonged face gaze across tasks relative to neurotypical comparison group, whereas autistic comparison group showed reduced face gaze.

Autism (14)

*RCPMr

Schwartzman et al. (2015) [96]

7

Rett syndrome (14)

12.4 ± 7.3

NA

Neurotypical 17)

Three images: 1) two children, 2) human face (happy) and 3) child and clock.

Proportion of fixations on social stimuli was higher in Rett syndrome relative to comparison groups.

Autism (11)

Vivanti et al. (2017a) [97]

7

Williams syndrome (21)

4.3 ± 1.4

VABSc 69.9 ± 9.8

Neurotypical (20) *CA

Five images of scene including social and non-social target. Three videos of paired social and non-social stimuli moving in unison.

Williams syndrome and neurotypical comparison group looked more at social stimuli compared to autistic group.

DQ 56.1 ± 16.5

Autism (36) *CA, DQ

Social scene scanning. Gaze allocation (e.g. people, objects, background) within a social scene.

Crawford et al. (2017) [68]

8

Fragile X syndrome (11)

26.3 ± 9.1

BPVSr 87.0 ± 27.2

Neurotypical (11)

*BPVSr

Twenty images of scenes where human actors engaged in natural activities.

Looking patterns were similar in fragile X syndrome and neurotypical comparison group. However, in fragile X syndrome gaze upon the face was associated with heightened anxiety and fewer social communication difficulties.

Debladis et al. (2019) [80]

12

Prader-Willi syndrome (39)

28.0 ± 8.0

FSIQ 57.0 ± 10.0

Neurotypical (20)

Three video clips of actors interacting whilst a third is standing in the background.

People with Prader-Willi syndrome looked similarly at social scenes as neurotypical people, but their gaze patterns became more atypical as social content increased.

Guy et al. (2020) [103]

11

Fragile X syndrome

(6)

14.3 ± 2.0

BPVSr 126.3 ± 23.5

Neurotypical (15)

*CA

Social scenes including a social (32 trials) and non-social (32 trials) distractor. Scenes included a unique target. Participants were asked to explicitly report when they found the target.

People with fragile X syndrome increased first looks to the target in later trials. However, overall attention to nonsocial distractor was higher in fragile X syndrome relative to comparison groups.

Neurotypical (16)

*BPVSr

RCPMr 22.3 ± 4.5

Neurotypical (16)

*RCPMr

Liang & Wilkinson (2017) [104]

9

Down syndrome (10)

19.0 ± 6.0

PPVTr 51

Autism (10) *CA, PPVTr

Sixteen pairs of images of social scenes, with two (8 trials) or three (8 trials) people. Each pair depicted an interaction where sharing was present and absent.

Gaze patterns were similar across groups. People with Down syndrome were quicker than comparison groups to view image where sharing was present.

Neurotypical (10) *CA

Riby & Hancock (2008) [105]

10

Williams syndrome (16)

17.5 ± 6.3

RCPMr 13.0 ± 4.0

Neurotypical (10) *CA

Twenty images of scenes where human actors engaged in natural activities.

Whilst autistic people spend less time viewing people/faces than is typical, people with Williams syndrome demonstrate exaggerated fixations towards the eyes.

Neurotypical (16) *RCPMr

Autism (20)

Riby & Hancock (2009b) [106]

9

Williams syndrome (16)

17.5 ± 6.3

RCPMr 16.0 ± 5.0

Neurotypical (10) *CA

Twenty images from actors interacting. The position and number of characters varied. Video clips including human actors (3 trials) and cartoons (3 trials).

Autistic people attended to faces less than was typical, whereas individuals with Williams syndrome attended to the face for longer than is typical. Atypical gaze behaviours in Williams syndrome were restricted to human actors.

Neurotypical (16) *RCPMr

Autism (20)

Wilkinson & Light (2014) [107]

7

Down syndrome (5)

17.8 ± 2.1

PPVTr 27

Neurotypical (5)

Eight images of an actor in a scene.

Fixation towards actor was similar across all groups.

Non-syndromic ID (3)

15.5 ± 0.7

PPVTr 36

Autism (5)

Williams et al. (2013) [108]

11

Williams syndrome (14)

22.2 ± 8.7

FSIQ 56.0 ± 13.1

Neurotypical (14) *Gender, CA

Eighteen images of social scenes including an actor who was presented centrally or peripherally.

People with fragile X syndrome avoided the actor presented centrally, at least initially. People with Williams syndrome took longer than is typical to disengage from actor.

Fragile X syndrome (14)

23.0 ± 10.5

FSIQ 64 ± 14.7

Neurotypical (14) *Gender, FSIQ

Gaze-following. Gaze towards a target object following an actor’s gaze cue.

Benjamin et al. (2014) [63]

8

Fragile X syndrome (11)

7.7 ± 1.9

PPVTr 70.71 ± 34.5

Autism (17)

*PPVTr

Video of an actor presenting one of two novel objects. The actor gazes towards the target object and then points towards the target object. Two novel objects presented within each of the four trials.

The fragile X group showed significantly increased face gaze rather than gaze-following towards the target object, whereas the comparison groups showed similar amounts of relative gaze towards the face and objects. The act of pointing increased gaze toward the target objects compared to when there was no pointing in all groups.

Leiter NVIQ 58.6 ± 13.5

Neurotypical (18)

*PPVTr

Riby et al. (2013) [109]

10

Williams syndrome (14)

13.5 ± 5.8

RCPMr 15 ± 5

Neurotypical (14) *RCPMr

Fourteen different images of an actor in a complex scene (e.g. office). The actor’s gaze was directed to a target item in the scene. Target object was presented alongside a plausible and implausible target. Participants were asked to explicitly identify the target object.

People with Williams syndrome looked more at the face and eye region compared to comparison groups. When cued to follow gaze, people with Williams syndrome looked towards the target, unlike autistic people who had greater difficulty identifying the target object.

Autism (24) *RCPMr

Vivanti et al. (2017a) [69]

11

Williams syndrome (21)

4.3 ± 1.4

VABSc 69.9 ± 9.8

Neurotypical (20) *CA

Six videos of an actor gazing towards one of two objects. The actor looks up to establish direct gaze and then turns head to gaze at the target object.

People with Williams syndrome looked less at the target object compared to neurotypical comparison group.

Autism (35) *CA, DQ

DQ 56.1 ± 16.5

Attention to the eye region. Gaze specifically towards or away from the eye region of neutral faces.

Kleberg et al. (2022) [111]

8

Williams syndrome (37)

23.4 ± 12.2

NA

Neurotypical infants (37)

Image of human face displaying an angry, happy, or neutral facial expression. 60 trials in total, 30 each condition. Fixation cross, so point of gaze towards eyes, then mouth (eye cued condition and mouth cued condition).

People with Williams syndrome were less likely, and slower, to orient to the eyes compared with neurotypical comparison groups of all ages except infants.

Neurotypical children (44)

Neurotypical adolescents (36)

Neurotypical adults (50)

Klusek et al. (2019) [72]

11

Fragile X syndrome (24)

19.3 ± 2.7

FSIQ 39.4 ± 5.8

Neurotypical (23) *CA

Video of avatar displaying gaze which is directed (16 trials) or averted (16 trials).

People with fragile X syndrome had shorter first fixation on the eyes relative to the neurotypical comparison group. Gaze direction (directed/averted) did not change orienting to the eye region in either group.

Face scanning. Gaze towards salient facial features whilst speaking.

D’Souza et al. (2015) [115]

8

Down syndrome (infant) (22)

1.4 ± 0.2

MSELma 8.5 ± 2.5

Neurotypical (25)

*MSELma

Paired video clips of human actor talking. The mouth movement was either congruent or incongruent to auditory stimuli heard.

Toddlers with fragile X and Williams syndrome who had a relatively large receptive vocabulary made more fixations to the eyes (rather than the mouth) of the incongruent face. In Down syndrome, fixations to the actor’s overall face predicted vocabulary size.

Down syndrome (toddler) (21)

2.4 ± 0.6

MSELma 15.9 ± 4.5

Williams syndrome (infant) (12)

1.3 ± 0.2

MSELma 8.7 ± 1.9

Williams syndrome (toddler) (25)

2.5 ± 0.7

MSELma 16.1 ± 4.5

Fragile X syndrome (toddler) (14)

2.9 ± 0.7

MSELma 15.3 ± 4.4

Hall et al. (2015) [69]

10

Fragile X syndrome (51)

20.2 ± 3.8

VABSc 58.5 ± 23.5

CA, VABSc, SCQ

An experimenter sat opposite the participant engaging with the participant in conversation about familiar topics (e.g. friends, family, hobbies). Participant was prompted to maintain eye contact with experimenter.

Participants with fragile X syndrome spent significantly less time looking at the face, including eyes, nose, and mouth individually, and had shorter episodes (and longer inter-episodes) of social gaze than comparison group.

Non-syndromic ID (19)

19.4 ± 2.9

VABSc 57.7 ± 16.8

Overimitation. Gaze allocation during video demonstration of an action by an actor.

Vivanti et al. (2016) [112]

7

Williams syndrome (21)

4.4 ± 1.4

VABSc 69.9 ± 10.1

Autism (36)

*VABSc, DQ

Eight videos of actor performing an action with one of eight objects presented on a table. Two conditions: 1) playful (4 trials) and 2) neutral (4 trials).

Unlike autistic children, those with Williams syndrome increased their gaze towards actor who was demonstrating the action in a socially engaging manner.

DQ 56.44 ± 16.9

Vivanti et al. (2017b) [113]

7

Williams syndrome (18)

4.4 ± 1.4

VABSc 69.9 ± 9.8

Autism (36)

*CA, DQ

Three videos in which an actor demonstrated a causally relevant and irrelevant action.

Children with Williams syndrome and neurotypical children were more likely to increase their attention to the actor’s face during demonstration of causally irrelevant actions, compared to autistic children.

DQ 56.1 ± 16.5

False-belief reasoning. Anticipatory gaze towards the location an actor saw an object last.

Van Herwegen et al. (2015) [114]

9

Williams syndrome (14)

7.5 ± 1.6

RCPMr 13.3 ± 2.4

Neurotypical (15) *CA

Autism (13) *CA

Change location false-belief reasoning task, where an object is moved whilst the actor leaves the room. Participants were asked to explicitly identify where the actor would look when they returned.

Unlike neurotypical who looked longer at the original location of the object upon the actors return, children with Williams syndrome and autistic children do not show such contrasts. Children with Williams syndrome spend longer looking at the actor.

BPVSr 49.3 ± 18.4