Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Apr 1.
Published in final edited form as: Autism Res. 2011 Jan 19;4(2):132–142. doi: 10.1002/aur.179

Deficits in adults with Autism Spectrum Disorders when processing multiple objects in dynamic scenes

Kirsten O'Hearn 1, Laura Lakusta 2, Elizabeth Schroer 1, Nancy Minshew 1,3, Beatriz Luna 1
PMCID: PMC3074038  NIHMSID: NIHMS256379  PMID: 21254449

Scientific Abstract

People with autism spectrum disorder (ASD) process visual information in a manner that is distinct from typically developing individuals. They may be less sensitive to people's goals and, more generally, focus on visual details instead of the entire scene. To examine these differences, people with and without ASD were asked to detect changes in dynamic scenes with multiple elements. Participants viewed a brief video of a person or an inanimate object moving from one object to another; after a delay, they reported whether a second video was the same or different. Possible changes included the figure, the object the figure was moving from, or the object it was moving toward (the ‘goal’). We hypothesized that individuals with ASD would be less sensitive to changes in scenes with people, particularly elements that might be the person's goal. Alternately, people with ASD might attend to fewer elements regardless of whether the scene included a person. Our results indicate that, like controls, people with ASD noticed a change in the ‘goal’ object at the end of a person's movement more often than the object at the start. However, the group with ASD did not undergo the developmental improvement when detecting changes in start and end objects that was evident typically. This led to deficits in adults with ASD that were not specific to scenes with people or to ‘goals’. Improvements in visual processing that underlie mature representation of scenes may not occur in ASD, suggesting late developing brain processes are affected.

Keywords: ASD, change detection, development, developmental disorder, people perception, social cognition

As people interact with the environment, they visually process an array of distinct objects. To behave effectively, they must recognize these objects and understand their relationship to each other and the surroundings. The ability to attend to all the important (i.e., meaningful) aspects of a visual scene, and understand their significance, requires complex visual processing that may differ in neurodevelopmental disorders such as autism spectrum disorder (ASD). Individuals with ASD have a unique pattern of visual processing. They tend to focus on individual elements, possibly undermining their ability to process global configuration (Jemel et al., 2006; Dakin & Frith, 2005; Behrmann et al., 2006; Brosnan et al., 2004; Scherf et al., 2008). Several theories have attempted to characterize this tendency, including Weak Central Coherence (WCC) (Frith & Happe, 1994; Happe, 1999) and Enhanced Perceptual Functioning (EPF) (Mottron et al., 2006). Both theories highlight a local bias in visual processing, leading to superior performance on some tasks (i.e, embedded figures task, visual search; Shah & Frith, 1983; O'Riordan & Plaisted, 2001). WCC integrates the local bias with a deficit in processing global configuration in ASD, potentially contributing to deficits on important visual tasks such as face processing (Jemel et al., 2006; Behrmann et al., 2006; Deruelle et al., 2008). WCC predicts that there would be limitations when visually processing complex scenes and objects, and these differences are sometimes evident in ASD (Nakahachi et al., 2008; Jolliffe & Baron-Cohen, 2001).

Data from change detection paradigms also indicates that there are differences in how individuals with ASD “see” complex naturalistic scenes. Change detection paradigms can assess sensitivity to different elements within a scene; changes to more ‘central’ or ‘meaningful’ elements may be detected faster or more accurately than other changes (Rensink, 2002; Ro et al., 2001). A limitation of change detection paradigms is that it is difficult to identify at what level of processing a change is detected (e.g., percept, attended information, matching). However, Rensink and colleagues argue that focused attention may be the dominant factor, and is necessary (though not sufficient) to detect a change (Rensink, 2002; Simons & Rensink, 2005). These investigators also discuss the importance of attentional management – in other words, using attention in a way that is sensitive to the task at hand and utilizes high-level knowledge (including semantic knowledge and the overall visual layout of the scene). There is some evidence that such processes differ in ASD. Adults with ASD were impaired at detecting changes in objects that were semantically marginal in the scene, but not those that were central, compared to typical adults (Fletcher-Watson et al., 2006). Adults with ASD were also less likely to detect changes to objects that were ‘out of place’ in a scene (e.g., a briefcase changing to a toaster in an office; Loth et al., 2008). These findings suggest that the focus on individual element(s) in ASD may affect their ability to integrate elements together into a coherent representation of complex scene. In contrast, a study of adolescents with ASD indicated that they were more sensitive to changes than typically developing individuals, presumably due to their strong visual search abilities (Smith & Milne, 2009).

This evidence suggests that there are differences in how individuals with and without ASD view all complex scenes, including those without people. However, another view suggests that these differences would be more evident in scenes with people, reflecting impairments in interpreting people's behavior and intentions/goals, and possibly contributing to the social deficits that are a core feature of ASD (Sasson, 2006). A person's intention or goal is often evident from their direction of movement and gaze – people generally move toward and look at their goals. Individuals with ASD are sensitive to gaze direction (Fletcher-Watson et al., 2008b; Falck-Ytter, 2010; Freeth et al., 2010a; Rutherford & Krysko, 2008), but there are subtle differences in this ability in ASD (de Jong et al., 2008; Fletcher-Watson et al., 2008a; Freeth et al., 2010b; Wallace et al., 2006), along with distinct patterns of brain function (Dichter & Belger, 2007; Pelphrey et al., 2005). For instance, recent work indicates that, while adults with ASD looked at the people in scenes as much as controls, they were less likely to look at what that person was gazing at – the likely goal of that person's thoughts or actions (Fletcher-Watson et al., 2008a; Freeth et al., 2010b). These differences in interpreting gaze may contribute to the joint attention deficits in ASD (Leekam et al., 1997; Leekam et al., 2000; Mundy et al., 1986). Finally, individuals with ASD may be less likely to interpret some actions as intentional (Rutherford et al., 2006; Castelli et al., 2002; Campbell et al., 2006; for reviews, see Tuchman, 2003; Pelphrey et al., 2004). So, while people with ASD are generally more able to process mental state information than originally hypothesized (Perner & Leekam, 2008; Tager-Flusberg & Sullivan, 1994), a lack of sensitivity to gaze and other social information may lead to decreased attention to people's intentions and their goals.

Therefore, an important question is whether limitations in interpreting scenes in ASD reflect a lack of sensitivity to people and their behavior/goals or a greater focus on individual elements than on global configuration. This general visual pattern could also undermine comprehension of complex social scenes, for instance, by affecting the sensitivity to the relationships between elements. To examine this, the current study investigated whether individuals in ASD were less sensitive to changes in scenes with people, or whether the pattern of performance in ASD was similar in scenes with and without people. The former result would be consistent with theories positing that visual differences in ASD mainly reflect difficulties in interpreting human behavior; the latter would support theories positing that more general visual differences in ASD affect the ability to interpret visual stimuli. The task focused on specific aspects of visual processing thought to be affected in ASD, including the use of movement and gaze to highlight potentially important elements (i.e., goals), and attention to non-central (i.e., marginal) components of a scene. A change detection task with dynamic video clips was used. Dynamic videos were chosen because 1) previous work has shown striking inability to detect changes in video clips typically (Simons et al., 2000; Simons & Rensink, 2005), 2) differences in ASD in eye movements may be particularly evident in dynamic scenes (Speer et al., 2007), and 3) dynamic scenes are more naturalistic and allow movement cues to goals. Based on evidence that ASD affects sensitivity to goals (as suggested by eye gaze; Freeth et al., 2010b; Pierno et al., 2006; Fletcher-Watson et al., 2008a), we predicted that group differences would be more notable in the scenes that involved people than those without. In particular, we thought that the group with ASD would be less likely to detect changes in the object being gazed at. However, if the differences in ASD on this task reflect more general visual tendencies, group differences would be similar across conditions.

Methods

Participants

Participants included 37 people with ASD and 37 typically developing people matched individually to the participants with ASD on age (1.5 years for children, 3 years for adults) and IQ (15 points; see Table 1). Individuals with ASD were recruited through the University of Pittsburgh Autism Center of Excellence (ACE) Subject Core (HD# 055748), and were originally identified via announcements at parent meetings and autism group newsletters. Participants were diagnosed with ASD using the structured research diagnostic instruments, namely the Autism Diagnostic Interview (ADI; Lord et al., 1994) and Autism Diagnostic Observation Schedule-G (ADOS; Lord et al., 1989), with the DSM-IV scoring algorithm (American Psychiatric Association, 2000). Individuals with PDD-NOS were excluded, as were those with full-scale IQs < 80 on the Wechsler Abbreviated Scale of Intelligence (WASI; Wechsler, 1999). The diagnosis was also confirmed by expert assessment using the established clinical description of high functioning autism (Minshew, 1996). There was no known basis, for instance a genetic or infectious etiology, for the disorder. Typically developing participants were recruited through the ACE Subject Core and other ongoing projects in the laboratory. Participants were generally healthy, and had no history of head trauma, birth complications, seizures, or psychiatric disorder. Informed consent and assent was obtained from all participants or their legal guardians prior to the study, which was approved by the Institutional Review Board at the University of Pittsburgh.

Table 1. Demographics.

Variable Children ASD Control Adolescents ASD Control Adults ASD Control
M SD M SD M SD M SD M SD M SD
n 8 8 15 15 14 14
Age 11 1.07 11.13 0.835 15.2 1.7 15.07 1.83 22.14 4.15 22 4.51
Full scale IQ 100.25 6.69 105.63 9.74 105.87 13.15 106.07 7.44 105.86 12.05 105.43 10.63
Verbal IQ 104.25 13.25 104.5 9.26 104.87 12.28 105.33 8.47 105 10.14 103.5 10.35
Performance IQ 96 6.21 105.38 10.72 105.07 14.45 106.2 7.87 105.29 13.73 105.93 11.47
ADOS social 9.63 1.6 9.13 2.1 9 2.25
ADOS comm 4.75 1.58 4.6 0.99 5.07 1.21
ADOS total 14.38 2.88 13.73 2.43 14.07 3.17
Open in a new tab

Both children and adults with ASD were tested. While little is known about how visual processing of dynamic scenes develops, recent work suggests that change detection improves with age typically (Fletcher-Watson et al., 2009; Shore et al., 2006). Developmental changes may differ in ASD; Burack and colleagues found that, while performance was comparable across groups, detection became more rapid and accurate with developmental level (as measured by a matrices task) in the typical children but not in those with ASD (Burack et al., 2009). Thus, Age (children/adolescents, adults) was included as a between subjects factor in the initial omnibus ANOVA.

Stimuli

Stimuli were short video clips with a person or an object moving from one object (the Start object) to another (the End object; Figure 1). There were 3 conditions (Look forward, Look back, and Physical). The stimuli in the Look forward and Look back conditions were scenes with people while the Physical condition included only inanimate objects. The comparison of Look forward and Look back allowed us to examine whether participants used the direction of eye gaze to highlight ‘goal’ objects, and how this interacted with people's movement. The order of conditions was counterbalanced, and the counterbalanced conditions were matched across groups.

Figure 1.

Figure 1

Open in a new tab

a) Video stills from the three conditions and the elements that could change in “different” trials (Start object, Figure, Motion, End object, b) Trial design.

Look forward condition

Participants were shown 37 videotaped motion events where the Figure (a person) moved from the Start object to the End object (the “goal”). The Figure looked at the End object that they were moving towards, making it likely that the End object was construed as the goal of the movement. All components in the event remained on the screen throughout the event. The Start and End objects were real objects, such as a table or chair. Left-right location of the Start and End objects were counterbalanced over events, as was the direction that the actor moved. After the delay, participants saw a second video event that ‘matched’ the initial video in all or most ways. The matched event was either the same video as in the target event (No change; N = 9); or had a different Start object (N = 9), a different End object (N = 9), or a different Figure (N = 5). On some trials (N = 5), there was a change in motion (such as jumping or hopping); these filler trials were included to increase the difficulty of the task and to avoid participants adopting a strategy of attending only to the Figure, and Start and End objects. Changes in the Start and End objects were equated, in that the same set of changes (e.g., a TV to a cart) was used for both cases. This allowed us to examine the hypothesis that noticing changes in the End object, presumably the person's goal, would be more difficult for those with ASD.

Look Back Condition

These videos used the same background, objects and Figure as the Look forward condition except that, while the Figure was still moving from the Start object to the End object, they were looking back at the Start object instead of looking forward to the End object. This separated the direction of gaze from the direction of movement, providing conflicting cues to the person's goal. This manipulation enabled us to examine whether the two groups weighted movement and gaze cues similarly, and whether differences between groups reflected gaze direction per se, since the perception and interpretation of gaze are commonly reported deficits in ASD (Pelphrey et al., 2005).

Physical Condition

These 32 motion events included an inanimate object (the Figure) being rolled (N = 16) or blown (N = 16) from a Start object to an End object. Start and End objects were real objects, such as a cup or folder. The moving Figures were objects that could be easily blown (e.g., tissue, leaf) or rolled (e.g., pen, battery) by an external force (e.g., a hairdryer positioned off screen). Like the other conditions, each of the 32 events had a ‘matched’ event that was either the same video as the target event (No change; N = 8); or had a different Start object (N = 8), End object (N = 8), or moving Figure (N = 8). To confirm that the motion was not considered intentional, five adults were asked to describe the events and their descriptions were coded for mentalistic language (Heider & Simmel, 1944). Mentalistic language was not used, confirming that these events were not considered animate by typical adults. This condition allowed us to examine whether the pattern of performance in the other conditions was specific to “intentional” events with people, or generalized to events with an inanimate Figure that was not construed as intentional.

Procedure

The method utilizes a change blindness paradigm, expanding on evidence that people fail to notice all the information in a visual scene (Simons & Rensink, 2005; Simons & Levin, 1997). Individuals saw two videos and had to decide if the videos were the same or different. In each trial (Figure 1), participants viewed a video of a dynamic event for 2.5 s, a black screen for 10 s, and a second event for 2.5 s, after which they judged whether the two videos were the same or different. The participant responded verbally and the experimenter circled same or different for each response. We also asked participants what had changed and recorded these answers. However, careful inspection of these answers did not yield any informative data because many people did not answer these questions, or had a difficult time reporting exactly what changed (e.g., item on the left instead of the actual object), consistent with previous work showing that changes are often detected before they can be accurately identified (Rensink, 2005). Thus, we were unable to analyze the actual answers.

We wanted to ensure that participants utilized visual processes and did not rely on encoding the information linguistically. For example, participants might verbally represent all the objects in a scene by repeating “the man went from the desk to the ladder”. If participants did this, even if they didn't visually notice a change in the second video, they could go through each item (verbally encoded from the first video, i.e., “man, desk, ladder”) to examine whether it was in the second video, thus detecting the change without using a visual representation. To ensure that participants did not use this sort of verbal representation, participants were required to verbally shadow a sequence of numbers and words during the video clips. Such verbal output – repeating words out loud – makes it difficult to remember the items by repeating them verbally (see also Hermer-Vazquez et al., 2008). We also asked trivia questions during the ten-second delay. Pilot data suggested that this manipulation, along with the delay, was necessary to prevent typical adults from performing at ceiling; this also ensured there was no verbal encoding during the delay.

Spatial Span

Since the change detection paradigm has a significant working memory component, participants also did the spatial span to assess working memory. We administered the spatial span from the Cambridge Neuropsychological Test Automated Battery (CANTAB) via a desktop touch-screen computer (Mitsubishi Precise Point 8705). In this task, an array of white squares in a random pattern is presented. Some of the squares briefly change color and then return to white; participants are instructed to remember the location and the sequence of the squares that change color. Immediately after the final square returns to white, participants touch the squares that changed color in the exact sequence that they remember it having occurred. The number of squares to be remembered progressively increases from two to nine. Participants must perform one of the three trials correctly to progress to the next level of difficulty, which requires remembering one more square. Previous studies indicate that people with ASD perform more poorly than controls on this task (Williams et al., 2005; Luna et al., 2007; Luna et al., 2002). Thus the inclusion of this measure as a covariate helps to control for general differences in working memory between groups in the analyses. This task was generally administered after the experiment.

Analyses

Since two of the trial type/condition combinations (i.e., Figure change, Look forward; No change, Physical) exhibited substantial negative skew in the distribution of the proportion of correct responses, the proportion correct data was transformed using a natural log on the reflected data (Kleinbaum, Kupper & Muller, 1988; also thanks to Dr. Richard Lowry's very informative website at http://faculty.vassar.edu/lowry). This transformed ‘proportion of correct responses’ was more normally distributed and was used as the dependent variable.1 The primary analyses used repeated measures analysis of variance (ANOVAs). Since there was a trend for Age (Children, Adults) to interact with Group in this initial ANOVA, the sample was split into separate groups for further analyses using repeated measures ANOVAS. After these ANOVAs, planned comparisons compared group differences in each trial and condition separately, to examine our hypotheses that scenes with people - and, more specifically, the objects of people's gaze and actions – might be represented differently in individuals with ASD. Finally, ANOVAs were used to examine developmental change separately in the typically developing group and the group with ASD.

While signal detection analysis could not be performed on this data (See footnote 1), that performance was similar on the no change trials suggests that all groups mistakenly identified a change when there wasn't one at similar levels. That this occurred relatively often is interesting. A limitation of this current study is that we cannot be sure whether participants were correct about what changed when there was a change, or what they thought changed when they reported a change in the no change condition. Further work will hopefully address these issues.

Results

Preliminary analysis

Spatial span was significantly lower in individuals with ASD than in typically developing individuals (t[72]= 3.28, p=.002) consistent with previous studies (Steele et al., 2007; Williams et al., 2006). Thus, in the ANOVAs described below, spatial span scores were used as a covariate to minimize the influence of working memory on the results.2 Although well-matched on IQ, we also co-varied for full-scale IQ in the primary analyses.

Primary analyses

Omnibus ANOVA

To examine the effects of group on performance in the change detection task, a repeated measures ANOVA was used. Group (2 levels; Control, ASD) and Age (2 levels; Children/adolescents, Adults) were between subject factors. Condition (3 levels; Look forward, Look back and Inanimate) and, nested within Condition, Trial type (4 levels; No change, Source change, Goal change, Figure change) were within subject factors. Both spatial span and full-scale IQ were included as covariates. The Greenhouse-Geisser corrected F values are reported, to correct for some violations of sphericity. This initial, omnibus ANOVA indicated a main effect of Condition (F(1.99, 135.42)=4.18, p =.02), Group (F(1, 68)=4.84, p=.03), and a trend for Age (F(1, 68)=3.61, p=.06). Importantly, these effects were mitigated by a borderline interaction between Group and Age (F1, 68)=3.17, p=.08), suggesting that the divergence in performance between groups differed in children/adolescents and adults. No other factors were significant. Therefore, to better understand group differences, the two age groups were analyzed separately using repeated measures ANOVAs with the between group factor of Group, within group factors of Condition and, nested, Trial Type, with spatial span and IQ as covariates. When age in years was used as a covariate, there was a trend for a main effect of Group (F(1, 69)=3.59, p = 06) but none of the interactions reached significance. This probably reflects, in part, that Age does not affect performance on this task in a linear, continuous manner, especially since this sample contains a substantial number of adults. A larger sample is needed to identify the true developmental trajectory.

ANOVA – Children/adolescents 9 -18 years old (N=23/group)

There were no significant main effects or interactions between Group, Condition, and Trial Type, including no main effect of Group (F[1,42]= .08, p=.78). Exploratory analyses indicated that this pattern remained the same when the children (9 to 12; N=8) and the adolescents (13 to 17; N=15) were analyzed separately (Group effect, Children F(1,12)=.001, p=.97; Adolescents F(1,26)=1.0, p=.33), though caution is stressed due to the small sample size of children. Since there was such a large age range, we also performed an ANOVA with age as a covariate. This analysis also failed to identify a main effect of Group (F(1,41)=.73, p=.40) or any interactions with Group, though there was a main effect of Age (F(1, 41)=5.07, p=.03) on performance.

ANOVA - Adults (N=14/group)

A main effect of group indicated that adults with ASD performed more poorly than typically developing adults (Group: F[1,24]=9.05, p=.006). There was a borderline main effect of Condition (F[2,43.52]=3.31, p=.05), reflecting that performance was slightly better in Condition 2 (Look Back) than in the 1 (Look Forward) or 3 (Physical) but post-hoc comparisons did not reveal significance differences between conditions (all p's > .18). There was also an interaction between Group and Trial Type (F[2.36,56.53]=3.19, p=.04), indicating that group differences were more evident on some Trial Types than on others. The same pattern was evident when Age was used as a covariate (main effect of Group: F(1,23)=9.06, p=.006; Group x Trial interaction F(2.35, 53.98)=3.14, p<.05). No other main effects or interactions reached significance. We tested our hypothesis -- that individuals with ASD would be less sensitive to changes in potential goals -- using planned comparisons, specifically t-tests to examine group differences on each trial type/condition combination separately. These revealed that, regardless of condition, the group with ASD tended to perform more poorly than controls on the Start object change trials (Look forward condition, t[19.9]= 1.98, p=.06; Look back condition, t[21.40]= 2.37, p=.03; Physical condition, t[24.92]= 3.12, p=.005) and End object changes trials (Look forward condition, t[21.19]=2.30, p=.03; Look back condition, t[14.86]= 4.84, p<.001; Physical condition, t[21.64]= 2.09, p<.05). There was also a trend for a significant difference on the figure change trials in the Look back condition (t[24.86]= 1.99, p=.06).

ANOVA – Development

To examine the developmental profile in the groups (TD, ASD) separately, we examined performance in each group using an ANOVA with Condition and Trial Type as within subject factors and Age (children/adolescents, adults) as between subject factor. In typically developing individuals, this analysis revealed a main effect of Age (F(1,35)=6.76, p=.014), and borderline interaction between Age and Trial Type (F(1.98, 69.55)=2.91, p=.06). Age did not interact with any other variables. Post-hoc comparisons using independent t-tests (equal variances not assumed) examined this interaction further. Improvements in typical development between childhood and adulthood tended to occur on the Start and End object change trials (though not in the End object trials in the Physical condition). Performance improved with age on the Start and End object trials in the Look forward condition (t(34.9)=-2.67, p=.01; t(34.91)=-2.89, p=.007) and the Look back condition (t(34.9)=-2.07, p=.05; t(25.52)=-3.24, p=.003), as well as the Start object trials in the Physical condition (t(33.17)=-2.54, p=.02). In the group with ASD, Age did not have a main effect, nor did it interact with Condition or Trial (all p's > .47). The pattern was the same when spatial span and full-scale IQ were used as covariates in each ANOVA. Thus, typically developing individuals improved with age, but individuals with ASD did not, on 5 of the 6 trial types that showed a significant difference between the ASD and TD groups in adulthood.

Object of gaze

To explore how gaze and movement cues affected change detection in individuals with ASD, and whether they weighted the gaze and movement cues similarly to typically developing people, we analyzed a composite score that compared change detection performance for Start and End objects. If movement is a stronger cue to what to look at than gaze, people should detect changes in End objects better than Start objects for both event types (Look Forward and Look Back); if gaze is a stronger cue, than this ‘goal bias’ pattern would be evident in the Look forward but not the Look back condition. To examine this, we computed a difference score on the accuracy (End object – Start object). To examine whether there were differences across Group or Age on this variable, we first did an ANOVA on this difference score with Condition as a within subject factor, Group and Age as a between subjects factor, and spatial span/full-scale IQ as covariates. While this analysis revealed a main effect of Condition (F(1.97, 133.96)=7.01, p=.001), there were no main effects or interactions with Group or Age (all p's > .28). To examine the Condition effect further, one sample t-tests were used to identify whether the difference score was significantly higher than zero. Changes in End objects were detected better than Start objects in both the Look forward and Look back conditions (control group, Look forward t[36]= 4.09, p<.001, Look back t[36] =3.77, p = .001; ASD group, Look forward t[36]=2.74, p=.01], Look back t[36] =2.23, p = .03]), but not in the Physical condition, with an inanimate object as the Figure (control group t[36] = .58, p = .57; ASD group t[36]=.24, p=.81). Thus, overall, both groups were more likely to detect changes in the End object than the Start object when the Figure was a person, presumably because the intentional movement toward the End object was a strong indicator that this object was important. That this occurred in the Look back condition, when Figures were looking at the Start object, indicates that movement was a stronger cue than gaze as to which object was important in these videos. That this did not occur in the Physical condition suggests that neither group viewed the End object of the inanimate Figure's movement as particularly important, and that the better accuracy for the End object was not simply a recency effect.

Improvement across a condition3

One possibility is that participants may have learned what objects were likely to change, ‘priming’ them to attend to these objects. To examine this, we divided our data into first third and last third of each condition; this analysis is exploratory, because each of these ‘thirds’ contained between 2 and 5 trials of each trial type. We analyzed the data with Condition, Trial Type and Timing (first 1/3, last 1/3) as within subject factors, and Group (ASD, controls) and Age (children/adolescents, adults) as between subject factors. Timing had a significant main effect (F(1,70)= 93.00, p< .001), but it did not interact with Group, Age, or Group x Age (all p's > .42), indicating that this effect was similar across groups. Timing did interact with both Condition (F(1.86, 126.27)=50.19, p< .001), and Trial type (F(2.35, 165.11)=70.74, p< .001); these affects are mitigated by a significant 3 way interaction between Timing, Condition and Trial type (F(1.96, 291.05)=28.95, p< .001). No other interactions were significant.

We then examined this 3 way interaction using paired t-tests comparing performance on the first and last third of the trials in each condition/trial type combination. This was done across the entire group, since the ANOVA indicated that timing affected different groups and ages similarly. We found that individuals tended to perform similarly across the first third and the last third of the no change trials (Look forward t(73)=1.34, p=.18; Physical t(73)=-.33, p=.75) with the exception of the difficult Look back condition (t(73)=-7.42, p<.001). Participants grew better at detecting a Figure change when the Figure was a person (Look forward; t(73)=-9.27, p<.001; Look back; t(73)=-17.97, p<.001) but not when the Figure was an inanimate object (t(73)=-.43, p=.67). Participants grew better at detecting an End object change – what the figure was moving toward -- when the person was looking at the End object (t(73)=-4.14, p<.001) but grew significantly worse when the person was looking at the Start object or the Figure was inanimate (Look back; t(73)= 7.51, p<.001: Physical; t(73)= 2.87, p=.005). They also better at detecting Start object changes when the person was looking at the Start object (t(73)= -6.55, p<.001), grew worse when the Figure was inanimate (t(73)= 3.24, p=.002), and did not change when the Figure was looking forward (t(73)=-1.82, p=-.07). This analysis indicates that – as a condition went on – participants became more sensitive to people (as the Figure) and the object of their gaze, sometimes to the detriment of the other potential changes. Interestingly, this increasing sensitivity to people and the objects of their gaze (perhaps as participants noted the direction of gaze across a condition, as suggested by the reviewer) was similar across group and age.

Summary of the results

The results indicate that adults with ASD had difficulty encoding the Start and End objects in all conditions, regardless of whether the condition included people (Look forward, Look back) or not (Physical). Adults with ASD are less sensitive to changes in these peripheral objects; the similarity across condition indicates that this difference does not reflect decreased sensitivity to human ‘goals’ or intentions. This group difference was not evident between the children/adolescents with and without ASD. This suggests there are typical developmental improvements when representing peripheral but still important objects (the Start and End objects) through adolescence, but that these improvements do not occur in ASD. This possibility supported by our developmental analyses in each group separately which showed significant development on these types of change trials in controls but not in ASD. The exception was End object changes in the Physical condition: adults with ASD performed more poorly than control adults but there was not significant typical development on this trial type. There was also a trend for adults with ASD to perform more poorly than typical adults at detecting Figure changes in the Look back condition. This insensitivity to a change in a moving person may reflect that this condition had conflicting cues (looking one way, moving another) that added complexity to the task, potentially making it more difficult for adults with ASD to represent all the potentially important objects. In addition to the group differences, these results also revealed some similarities between the groups with and without ASD. Both groups were more likely to attend to the End object than to the Start object in the people conditions (Look forward and Look back conditions), regardless of whether the person was looking at the Start or End object, but not in the Physical condition. This indicates that both groups were more likely to use the ‘moving toward’ cue, compared to the ‘looking at’ cue, to allocate their attention.

Conclusions and Discussion

These findings indicate that adults with ASD have more difficulty than typical adults at detecting changes in all the potentially relevant objects in scene, in particular peripheral objects that the figure is moving to and from. The three conditions allowed us to examine whether the differences in ASD were specific to scenes with people or, even more precisely, the objects of people's gaze. In contrast to our hypothesis, adults with ASD performed more poorly than typical adults when detecting changes in the object at the beginning and the end of a movement, regardless of whether the moving object was a person or an inanimate object. Therefore, differences in adults with ASD in this paradigm are more likely to reflect their general visual profile (Dakin & Frith, 2005; Mottron et al., 2006; Plaisted et al., 1999) than difficulty in interpreting people's goals or actions. This impairment is consistent with recent evidence indicating that detection of changes in marginal objects was impaired in adults with ASD, while detection of the central changes in a scene was similar to typical adults (Fletcher-Watson et al., 2006). These authors suggest that individuals with ASD have difficulty attending to multiple objects, particularly when there is an engaging central figure, as in the current study. Decreased sensitivity to peripheral objects may impact the ability to interpret mental states, as understanding a person's goals often requires attending to peripheral and/or multiple objects (e.g., the object being gazed at; relations between objects).

No differences were found between children/adolescents with ASD and typically developing children, suggesting that the difficulties observed for adults with ASD may be attributed to ‘atypical’ development. Specifically, in typical development, the ability to detect changes in the important but peripheral objects that a moving figure goes from and to generally tended to improve with age. There was no developmental improvement in detecting any type of change in the group with ASD, leading to impairments in detecting changes in the peripheral objects by adulthood. In addition, adults with ASD tended to perform more poorly than typical adults when a person who was looking at one object but moving toward another (Look back condition). This may reflect the incongruent cues to intentionality and what is important in the scene. In other words, the group differences when the person changed in this condition may reflect that attention needs to be spread across objects as participants try to interpret the scene, and people with ASD have difficulty with this. This result is consistent with prior evidence indicating that people with autism may have particular difficulty when gaze direction is inconsistent with their expectations (Pelphrey et al., 2005) and with more complex stimuli (Minshew et al., 1999).

All groups paid better attention to changes in the ‘goal’ object – the object the person was moving toward – than they did to changes in the ‘start object’ – the object the person was moving from – but only when the moving figure was a person. This suggests that people with ASD processed scenes with people relatively typically, attending well to the central person and the object that they were gazing at and/or moving toward, consistent with recent reports on the representation of people in individuals with ASD (New et al., 2009; Fletcher-Watson et al., 2008b; Fletcher-Watson et al., 2008a). However, there are limitations to the present study. Though dynamic, our scenes were relatively simple (compared to, for instance, New et al., 2009, or the real world). It is possible that differences in ASD specific to viewing people would be evident with more complex stimuli, for instance, scenes with multiple people (Speer et al., 2007). In addition, there are limitations to the change detection paradigm itself. In particular, in this present experiment, we could not ensure that the changes detected were indeed the correct changes. Many people detected changes on ‘no change’ trials, and it is difficult to know what this reflects. Clearly, more work needs to be done to examine whether this impairment in adults with ASD, and the pattern of development in the two groups, generalizes to other visual stimuli. In addition, little is known about the typical development of the neural substrates underlying visual processing of dynamic scenes, so it is difficult to guess what developmental changes might not be occurring in ASD. While speculative, we suggest that this type of visual processing might require the coordination of a large network of brain regions, a process that appears to have a long developmental trajectory (Fair et al., 2009; Luna et al., 2001) and might be affected by ASD (Just et al., 2007; Minshew & Williams, 2007).

In summary, group differences were not specific to scenes with people, or objects of their gaze. While social deficits in ASD might affect the interpretation of social scenes throughout development, the currents findings do not appear to be associated with the social impairments in ASD per se. Instead, these findings seem to reflect less sensitivity in ASD to changes in peripheral but still important objects in a scene (and presumably less attention to; Rensink, 2002), a difference that may reflect the distinct visual pattern with a focus on individual elements described in ASD (Happe, 1999; Brosnan et al., 2004; Dakin & Frith, 2005; Mottron et al., 2006). This general visual difference could impact the interpretation of complex social scenes. Another scenario is that interest in social scenes supports late developments in attentional management (Rensink, 2002), leading to improvements during adolescence typically but not in ASD. That differences between groups became evident only with age is consistent with recent evidence indicating that late maturing visual processes are atypical in ASD, from both behavioral (Luna et al., 2007; Rump et al., 2009; Scherf et al., 2008, 2009; O'Hearn et al., 2010) and neuroimaging (Lee et al., 2007; Raznahan et al., 2009; Scherf et al., 2010) methods. That these differences were only evident in adulthood suggests that interventions during adolescence may be helpful for visual differences that people with ASD find problematic. Finally, these findings indicate that age differences across samples may contribute to the variability in findings on visual processing. Further study using larger sample sizes and possibly longitudinal data will be helpful in examining the development of the ability to represent complex, real-world scenes in both typical and atypical populations.

Figure 2.

Figure 2

Open in a new tab

Proportion correct in children and adolescents.

Figure 3.

Figure 3

Open in a new tab

Proportion correct in adults.

Acknowledgments

This work was done at the University of Pittsburgh and supported by Autism Speaks Grant 04593 to BL, NIH HD 055748 to NM, NIMH 5 R01 MH067924 to BL, NIMH KO1 081191 to KOH. We are very grateful to the participants and their families, the staff at the CeFar (Now ACE) in Pittsburgh, and the Autism Speaks organization, as well as Catherine Wright for help on this manuscript.

Abbreviations

ADI

Autism Diagnostic Interview

ADOS

Autism Diagnostic Observation Schedule

WASI

Wechsler Abbreviated Scale of Intelligence

Footnotes

Autism Speaks 04593, NIH HD 055748, NIMH R01 MH067924, NIMH KO1 081191

1

Signal detection analysis, taking into account the number of hits and false alarms, was not performed on these data because the no change trials (‘false alarms’) were the same for all types of change trials.

2

Two subjects with ASD did not take the Spatial Span task due to experiment malfunction; thus, these individuals are not included in the ANOVA. However, the pattern was similar when all participants were included, and spatial span was not a covariate.

3

We thank an anonymous reviewer for suggesting this analysis.

References

  1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) 2000. [Google Scholar]
  2. Behrmann M, Avidan G, Leonard GL, Kimchi R, Luna B, Humphreys K, Minshew N. Configural processing in autism and its relationship to face processing. Neuropsychologia. 2006;44:110–129. doi: 10.1016/j.neuropsychologia.2005.04.002. [DOI] [PubMed] [Google Scholar]
  3. Brosnan MJ, Scott FJ, Fox S, Pye J. Gestalt processing in autism: failure to process perceptual relationships and the implications for contextual understanding. J Child Psychol Psychiatry. 2004;45:459–469. doi: 10.1111/j.1469-7610.2004.00237.x. [DOI] [PubMed] [Google Scholar]
  4. Burack J, Joseph S, Russo N, Shore DI, Porporino M, Enns JT. Change detection in naturalistic pictures among children with autism. Journal of Autism & Developmental Disorders. 2009;39:471–479. doi: 10.1007/s10803-008-0647-6. [DOI] [PubMed] [Google Scholar]
  5. Campbell R, Lawrence K, Mandy W, Mitra C, Jeyakuma L, Skuse D. Meanings in motion and faces: developmental associations between the processing of intention from geometrical animations and gaze detection accuracy. Developmental Psychopathology. 2006;18:99–118. doi: 10.1017/S0954579406060068. [DOI] [PubMed] [Google Scholar]
  6. Castelli F, Frith C, Happe F, Frith U. Autism, Asperger syndrome and brain mechanisms for the attribution of mental states to animated shapes. Brain. 2002;125:1839–1849. doi: 10.1093/brain/awf189. [DOI] [PubMed] [Google Scholar]
  7. Dakin S, Frith U. Vagaries of visual perception in autism. Neuron. 2005;48:497–507. doi: 10.1016/j.neuron.2005.10.018. [DOI] [PubMed] [Google Scholar]
  8. de Jong MC, van Engeland H, Kemner C. Attentional effects of gaze shifts are influenced by emotion and spatial frequency, but not in autism. J Am Acad Child Adolesc Psychiatry. 2008;47:443–454. doi: 10.1097/CHI.0b013e31816429a6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Deruelle C, Rondan C, Salle-Collemiche X, Bastard-Rosset D, Da Fonseca D. Attention to low- and high-spatial frequencies in categorizing facial identities, emotions and gender in children with autism. Brain and Cognition. 2008;66:115–123. doi: 10.1016/j.bandc.2007.06.001. [DOI] [PubMed] [Google Scholar]
  10. Dichter GS, Belger A. Social stimuli interfere with cognitive control in autism. Neuroimage. 2007;35:1219–1230. doi: 10.1016/j.neuroimage.2006.12.038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fair DA, Cohen AL, Power JD, Dosenbach NU, Church JA, Miezin FM, Schlaggar B, Peterson S. Functional brain networks develop from a “local to distributed” organization. PLoS Comput Biol. 2009;5:e1000381. doi: 10.1371/journal.pcbi.1000381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Falck-Ytter T. Young children with autism spectrum disorder use predictive eye movements in action observation. Biol Lett. 2010;6:375–378. doi: 10.1098/rsbl.2009.0897. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fletcher-Watson S, Collis JM, Findlay JM, Leekam SR. The development of change blindness: children's attentional priorities whilst viewing naturalistic scenes. Developmental Science. 2009;12:438–445. doi: 10.1111/j.1467-7687.2008.00784.x. [DOI] [PubMed] [Google Scholar]
  14. Fletcher-Watson S, Leekam SR, Benson V, Frank MC, Findlay JM. Eye-movements reveal attention to social information in autism spectrum disorder. Neuropsychologia. 2008a;47:248–257. doi: 10.1016/j.neuropsychologia.2008.07.016. [DOI] [PubMed] [Google Scholar]
  15. Fletcher-Watson S, Leekam SR, Findlay JM, Stanton EC. Brief Report: Young Adults with Autism Spectrum Disorder Show Normal Attention to Eye-Gaze Information-Evidence from a New Change Blindness Paradigm. J Autism Dev Disord. 2008b;38:1785–1790. doi: 10.1007/s10803-008-0548-8. [DOI] [PubMed] [Google Scholar]
  16. Fletcher-Watson S, Leekam SR, Turner MA, Moxon L. Do people with autistic spectrum disorder show normal selection for attention? Evidence from change blindness. British Journal of Psychology. 2006;97:537–554. doi: 10.1348/000712606X114057. [DOI] [PubMed] [Google Scholar]
  17. Freeth M, Ropar D, Chapman P, Mitchell P. The eye gaze direction of an observed person can bias perception, memory, and attention in adolescents with and without autism spectrum disorder. Journal of Experimental Child Psychology. 2010;105:20–37. doi: 10.1016/j.jecp.2009.10.001. [DOI] [PubMed] [Google Scholar]
  18. Freeth M, Chapman P, Ropar D, Mitchell P. Do gaze cues in complex scenes capture and direct the attention of high functioning adolescents with ASD? Evidence from eye-tracking. J Autism Dev Disord. 2010;40:534–547. doi: 10.1007/s10803-009-0893-2. [DOI] [PubMed] [Google Scholar]
  19. Frith U, Happe F. Autism: beyond “theory of mind”. Cognition. 1994;50:115–132. doi: 10.1016/0010-0277(94)90024-8. [DOI] [PubMed] [Google Scholar]
  20. Happe F. Autism: cognitive deficit or cognitive style? Trends in Cognitive Sciences. 1999;3:216–222. doi: 10.1016/s1364-6613(99)01318-2. [DOI] [PubMed] [Google Scholar]
  21. Heider F, Simmel M. An experimental study of apparent behavior. American Journal of Psychology. 1944;57:243–259. [Google Scholar]
  22. Hermer-Vazquez L, Spelke E, Katsnelson AS. Sources of flexibility in human cognition: dual-task studies of space and language. Cognit Psychol. 2008;39:3–36. doi: 10.1006/cogp.1998.0713. [DOI] [PubMed] [Google Scholar]
  23. Jemel B, Mottron L, Dawson M. Impaired face processing in autism: fact or artifact? J Autism Dev Disord. 2006;36:91–106. doi: 10.1007/s10803-005-0050-5. [DOI] [PubMed] [Google Scholar]
  24. Jolliffe T, Baron-Cohen S. A test of central coherence theory: can adults with high-functioning autism or Asperger syndrome integrate fragments of an object? Cogn Neuropsychiatry. 2001;6:193–216. doi: 10.1080/13546800042000124. [DOI] [PubMed] [Google Scholar]
  25. Just MA, Cherkassky VL, Keller TA, Kana RK, Minshew NJ. Functional and anatomical cortical underconnectivity in autism: evidence from an FMRI study of an executive function task and corpus callosum morphometry. Cerebral Cortex. 2007;17:951–961. doi: 10.1093/cercor/bhl006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kleinbaum DG, Kupper LL, Muller KE. Applied Regression Analysis and other Multivariable Methods. Duxbury Press; Belmont, CA: 1988. [Google Scholar]
  27. Lee JE, Bigler ED, Alexander AL, Lazar M, DuBray MB, Chung MK, Johnson M, Morgan J, Miller JN, McMahon WM, Lu J, Jeong EK, Lainhart JE. Diffusion tensor imaging of white matter in the superior temporal gyrus and temporal stem in autism. Neurosci Lett. 2007;424:127–132. doi: 10.1016/j.neulet.2007.07.042. [DOI] [PubMed] [Google Scholar]
  28. Leekam SR, Baron-Cohen S, Perrett D, Milders M, Brown S. Eye-direction detection : A dissociation between geometric and joint attention skills in autism. British Journal of Developmental Psychology. 1997;15:77–95. [Google Scholar]
  29. Leekam SR, Lopez B, Moore C. Attention and joint attention in preschool children with autism. Dev Psychol. 2000;36:261–273. doi: 10.1037//0012-1649.36.2.261. [DOI] [PubMed] [Google Scholar]
  30. Lord C, Rutter M, Couteur AL. Autism diagnostic interview-revised: a revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. Journal of Autism and Developmental Disorders. 1994;24:659–685. doi: 10.1007/BF02172145. [DOI] [PubMed] [Google Scholar]
  31. Lord C, Rutter M, Goode S. Autism diagnostic observation schedule: a standardized observation of communicative and social behavior. Journal of Autism and Developmental Disorders. 1989;19:185–212. doi: 10.1007/BF02211841. [DOI] [PubMed] [Google Scholar]
  32. Loth E, Gómez CJ, Happé F. Detecting changes in naturalistic scenes: contextual inconsistency does not influence spontaneous attention in high-functioning people with autism spectrum disorder. Autism Res. 2008;1:179–188. doi: 10.1002/aur.19. [DOI] [PubMed] [Google Scholar]
  33. Luna B, Doll SK, Hegedus SJ, Minshew NJ, Sweeney JA. Maturation of executive function in autism. Biological psychiatry. 2007;61:474–481. doi: 10.1016/j.biopsych.2006.02.030. [DOI] [PubMed] [Google Scholar]
  34. Luna B, Minshew NJ, Garver KE, Lazar NA, Thulborn KR, Eddy WF, et al. Neocortical system abnormalities in autism: an fMRI study of spatial working memory. Neurology. 2002;59:834–40. doi: 10.1212/wnl.59.6.834. [DOI] [PubMed] [Google Scholar]
  35. Luna B, Thulborn KR, Munoz DP, Merriam EP, Garver KE, Minshew NJ, Keshavan MS, Genovese CR, Eddy WF, Sweeney JA. Maturation of widely distributed brain function subserves cognitive development. NeuroImage. 2001;13:786–793. doi: 10.1006/nimg.2000.0743. [DOI] [PubMed] [Google Scholar]
  36. Minshew NJ. Autism. In: Berg BO, editor. Principles of Child Neurology. New York: McGraw-Hill; 1996. pp. 1713–1729. [Google Scholar]
  37. Minshew NJ, Luna B, Sweeney JA. Oculomotor evidence for neocortical systems but not cerebellar dysfunction in autism. Neurology. 1999;52:917–922. doi: 10.1212/wnl.52.5.917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Minshew NJ, Williams DL. The new neurobiology of autism: cortex, connectivity, and neuronal organization. Archives of Neurology. 2007;64:945–950. doi: 10.1001/archneur.64.7.945. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Mottron L, Dawson M, Soulieres I, Hubert B, Burack J. Enhanced perceptual functioning in autism: an update, and eight principles of autistic perception. J Autism Dev Disord. 2006;36:27–43. doi: 10.1007/s10803-005-0040-7. [DOI] [PubMed] [Google Scholar]
  40. Mundy P, Sigman M, Ungerer J, Sherman T. Defining the social deficits of autism: the contribution of non-verbal communication measures. J Child Psychol Psychiatry. 1986;27:657–669. doi: 10.1111/j.1469-7610.1986.tb00190.x. [DOI] [PubMed] [Google Scholar]
  41. Nakahachi T, Yamashita K, Iwase M, Ishiqami W, Tanaka C, Toyonaga K, Maeda S, Hirotsune H, Tei Y, Yokoi K, Okajima S, Shimizu A, Takeda M. Disturbed holistic processing in autism spectrum disorders verified by two cognitive tasks requiring perception of complex visual stimuli. Psychiatry Res. 2008;159:330–338. doi: 10.1016/j.psychres.2005.08.028. [DOI] [PubMed] [Google Scholar]
  42. New JJ, Schultz RT, Wolf J, Niehaus JL, Klin A, German TC, Scholl BJ. The scope of social attention deficits in autism: Prioritized orienting to people and animals in static natural scenes. Neuropsychologia. 2010;48:51–9. doi: 10.1016/j.neuropsychologia.2009.08.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. O'Hearn K, Schroer E, Minshew N, Luna B. Lack of developmental improvement on a face memory task during adolescence in autism. Neuropsychologia. 2010;48:3955–60. doi: 10.1016/j.neuropsychologia.2010.08.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. O'Riordan M, Plaisted K. Enhanced discrimination in autism. Q J Exp Psychol A. 2001;54:961–79. doi: 10.1080/713756000. [DOI] [PubMed] [Google Scholar]
  45. Pelphrey K, Adolphs R, Morris JP. Neuroanatomical substrates of social cognition dysfunction in autism. Ment Retard Dev Disabil Res Rev. 2004;10:259–271. doi: 10.1002/mrdd.20040. [DOI] [PubMed] [Google Scholar]
  46. Pelphrey KA, Morris JA, McCarthy G. Neural basis of eye gaze processing deficits in autism. Brain. 2005;128:1038–1048. doi: 10.1093/brain/awh404. [DOI] [PubMed] [Google Scholar]
  47. Perner J, Leekam S. The curious incident of the photo that was accused of being false: issues of domain specificity in development, autism, and brain imaging. Q J Exp Psychol (Colchester) 2008;61:76–89. doi: 10.1080/17470210701508756. [DOI] [PubMed] [Google Scholar]
  48. Pierno AC, Mari M, Glover S, Georgiou I, Castiello U. Failure to read motor intentions from gaze in children with autism. Neuropsychologia. 2006;44:1483–1488. doi: 10.1016/j.neuropsychologia.2005.11.013. [DOI] [PubMed] [Google Scholar]
  49. Raznahan A, Toro R, Daly E, Robertson D, Murphy C, Deeley Q, Bolton PF, Paus T, Murphy DG. Cortical Anatomy in Autism Spectrum Disorder: An In Vivo MRI Study on the Effect of Age. Cereb Cortex. 2009 doi: 10.1093/cercor/bhp198. [DOI] [PubMed] [Google Scholar]
  50. Rensink RA. Change detection. Annu Rev Psychol. 2002;53:245–277. doi: 10.1146/annurev.psych.53.100901.135125. [DOI] [PubMed] [Google Scholar]
  51. Ro T, Russell C, Lavie N. Changing faces: a detection advantage in the flicker paradigm. Psychol Sci. 2001;12:94–99. doi: 10.1111/1467-9280.00317. [DOI] [PubMed] [Google Scholar]
  52. Rump KM, Giovannelli JL, Minshew NJ, Strauss MS. The development of emotion recognition in individuals with autism. Child Development. 2009;80:1434–1447. doi: 10.1111/j.1467-8624.2009.01343.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Rutherford MD, Krysko KM. Eye direction, not movement direction, predicts attention shifts in those with autism spectrum disorders. Journal of Autism and Developmental Disorders. 2008;38:1958–1965. doi: 10.1007/s10803-008-0592-4. [DOI] [PubMed] [Google Scholar]
  54. Rutherford MD, Pennington BF, Rogers SJ. The perception of animacy in young children with autism. J Autism Dev Disord. 2006;36:983–992. doi: 10.1007/s10803-006-0136-8. [DOI] [PubMed] [Google Scholar]
  55. Sasson NJ. The development of face processing in autism. J Autism Dev Disord. 2006;36:381–394. doi: 10.1007/s10803-006-0076-3. [DOI] [PubMed] [Google Scholar]
  56. Scherf K, Luna B, Minshew N, Behrmann M. Atypical development of face-related activation in Autism. Frontiers in Neuroscience. 2010 doi: 10.3389/fnhum.2010.00026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Scherf KS, Behrmann M, Kimchi R, Luna B. Emergence of global shape processing continues through adolescence. Child Dev. 2009;80:162–177. doi: 10.1111/j.1467-8624.2008.01252.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Scherf KS, Luna B, Kimchi R, Minshew N, Behrmann M. Missing the big picture: impaired development of global shape processing in autism. Autism Res. 2008;1:114–129. doi: 10.1002/aur.17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Shah A, Frith U. An islet of ability in autistic children: a research note. J Child Psychol Psychiatry. 1983;24:613–620. doi: 10.1111/j.1469-7610.1983.tb00137.x. [DOI] [PubMed] [Google Scholar]
  60. Shore D, Burack JA, Miller D, Joseph S, Enns JT. The development of change detection. Developmental Science. 2006;9:490–497. doi: 10.1111/j.1467-7687.2006.00516.x. [DOI] [PubMed] [Google Scholar]
  61. Simons DJ, Franconeri SL, Reimer RL. Change blindness in the absence of a visual disruption. Perception. 2000;29:1143–1154. doi: 10.1068/p3104. [DOI] [PubMed] [Google Scholar]
  62. Simons DJ, Rensink RA. Change blindness: past, present, and future. Trends Cogn Sci. 2005;9:16–20. doi: 10.1016/j.tics.2004.11.006. [DOI] [PubMed] [Google Scholar]
  63. Simons DJ, Levin DT. Change blindness. Trends in Cognitive Sciences. 1997;1:261–267. doi: 10.1016/S1364-6613(97)01080-2. [DOI] [PubMed] [Google Scholar]
  64. Smith H, Milne E. Reduced change blindness suggests enhanced attention to detail in individuals with autism. The Journal of Child Psychology and Psychiatry. 2009;50:300–306. doi: 10.1111/j.1469-7610.2008.01957.x. [DOI] [PubMed] [Google Scholar]
  65. Speer LL, Cook AE, McMahon WM, Clark E. Face processing in children with autism: effects of stimulus contents and type. Autism. 2007;11:265–277. doi: 10.1177/1362361307076925. [DOI] [PubMed] [Google Scholar]
  66. Steele S, Minshew N, Luna B, Sweeney J. Spatial working memory deficits in autism. Journal of Autism and Developmental Disorders. 2007;37:605–612. doi: 10.1007/s10803-006-0202-2. [DOI] [PubMed] [Google Scholar]
  67. Tager-Flusberg H, Sullivan K. A second look at second-order belief attribution in autism. Journal of Autism & Developmental Disorders. 1994;24:577–586. doi: 10.1007/BF02172139. [DOI] [PubMed] [Google Scholar]
  68. Tuchman R. Autism. Neuro Clin. 2003;21:915–932. doi: 10.1016/s0733-8619(03)00011-2. [DOI] [PubMed] [Google Scholar]
  69. Wallace S, Coleman M, Pascalis O, Bailey A. A study of impaired judgment of eye-gaze direction and related face-processing deficits in autism spectrum disorders. Perception. 2006;35:1651–1664. doi: 10.1068/p5442. [DOI] [PubMed] [Google Scholar]
  70. Wechsler D. Wechsler Abbreviated Scale of Intelligence. San Antonio, TX: The Psychological Corporation; 1999. [Google Scholar]
  71. Williams DL, Goldstein G, Minshew N. Impaired memory for faces and social scenes in autism: clinical implications of memory dysfunction. Arch Clin Neuropsychol. 2005;20:1–15. doi: 10.1016/j.acn.2002.08.001. [DOI] [PubMed] [Google Scholar]
  72. Zalla T, Labruyere N, Georgieff N. Goal-directed action representation in autism. J Autism Dev Disord. 2006;36:527–540. doi: 10.1007/s10803-006-0092-3. [DOI] [PubMed] [Google Scholar]

RESOURCES