Skip to main content
NCBI home page
International Journal of Developmental Disabilities logoLink to International Journal of Developmental Disabilities
. 2022 May 24;70(2):268–277. doi: 10.1080/20473869.2022.2077078

Attentional bias toward pictures related to circumscribed interests in children with autism spectrum disorder

Lei Wang 1, Huizhong He 2,✉, Jianxin Feng 3, Tingzhao Wang 4
PMCID: PMC10930113  PMID: 38481455

Abstract

Background

Circumscribed interests (CIs) are regarded as one of the common symptoms for individuals with autism spectrum disorder (ASD). Although some studies have found attentional bias toward CI-related stimuli for individuals with ASD, few studies have directly explored the reasons for these findings.

Method

Children with ASD (n = 15) and age-matched typically developing (TD) children (n = 15) completed a Stroop-like task and a modified dot-probe task, and their reaction times were measured. The stimuli of these tasks included neutral objects and objects related to CI for individuals with ASD.

Results

Children with ASD had longer reaction times to name colors of CI-related stimuli than neutral stimuli in the Stroop-like task. Similarly, in the modified dot-probe task, if the CI-related stimulus was presented simultaneously with the neutral stimulus, children with ASD responded faster to the target that appeared in the previous location of the CI-related stimulus (i.e. congruent condition) than to the neutral stimulus (i.e. incongruent condition). Importantly, the reaction times of children with ASD in the incongruent condition were significantly slower than in the baseline condition, where both neutral stimuli were displayed simultaneously with stimulus-onset asynchrony (SOA) of 500 ms. In contrast, there was no significant difference in reaction times to CI-related stimuli and neutral stimuli in both tasks for TD children.

Conclusions

These findings suggest children with ASD have attentional bias toward CI-related stimuli, because of difficulty disengaging attention from CI-related stimuli with SOA of 500 ms.

Keywords: Autism spectrum disorder, attentional bias, Stroop-like task, modified dot-probe task, circumscribed interests, disengaging attention

Introduction

Autism spectrum disorder (ASD) is primarily characterized by social interaction and communication deficits with restricted and repetitive behaviors (American Psychiatric Association 2013). Circumscribed interests (CI) are an important subtype of restricted and repetitive behaviors, defined as intense and focused interest in a narrow range of topics (Lepper et al. 2017). CI are considered one of the common features for individuals with ASD (Morrison et al. 2018), with 75–95% of individuals with ASD experiencing CI (Grove et al. 2018, Zheng et al. 2019), and play an important role in early identification and revealing the cause of ASD. For example, previous studies have found that CI may be one of the efficient and effective biomarkers of ASD at an early stage (Zivan et al. 2021). Notably, if CI-related stimuli excessively and initially attract the attention of children with ASD, it may occupy other types of stimulus input channels important to the individuals, such as social stimuli (e.g. faces), which, in turn, may exacerbate their restricted and repetitive behaviors, and social interaction and language difficulties (Foss-Feig et al. 2016). The universality and importance of CI in ASD have aroused researchers’ interest in attentional bias toward CI-related objects among individuals with ASD.

An attentional bias is present when individuals give different attentional allocations to certain types of stimuli than to neutral stimuli, which is often a form of hyperattention to threatening material (Cisler and Koster 2010). Support for the attentional bias toward CI-related objects in ASD comes from the data collected by subjective clinical observations and questionnaire surveys, noting a preference for CI-related objects (e.g. trains) in children with ASD (Baron-Cohen and Wheelwright 1999, Harrop et al. 2017, Richler et al. 2010). South et al. (2005), using the Repetitive Behaviors Interview (RBI) and Yale Special Interests Interview (YSII), found that children and adolescents with ASD were biased toward CI-related stimuli and summarized the common CI-related stimuli among individuals with ASD. With the development of eye-tracking technology that objectively and directly detects attention behavior, researchers have explored the passive viewing paradigm to prove attentional bias toward CI-related objects for individuals with ASD. In this paradigm, some studies have found that children and adolescents with ASD had earlier and longer fixation time on CI-related objects than on other types of stimuli (Manyakov et al. 2018, Sasson et al. 2008, Sun et al. 2021). Compared with typically developing (TD) children, preschool children with ASD have demonstrated relatively more exploration and perseverative attention to CI-related stimuli (Cai et al. 2020, Sasson and Touchstone 2014, Sasson et al. 2011, Troyb et al. 2016). The materials of these studies included social images, neutral images, and CI-related images, and were presented to individuals with ASD simultaneously. The results show that visual avoidance of faces in children with ASD was most likely to occur when viewing social stimuli (Ma et al. 2021). Hence, it is difficult to directly verify whether individuals with ASD have attentional biases toward CI-related objects in passive viewing tasks including social objects, as the difference could be caused by the need for viewers to shift visual attention to CI-related objects to avoid steadily looking at social stimuli (Safra et al. 2018). Therefore, the following questions arise: would children with ASD still exhibit attentional bias toward CI-related stimuli if the social stimuli were eliminated and the CI-related stimuli were presented simultaneously with neutral stimuli? If so, what is the cause of attentional bias toward CI-related stimuli in individuals with ASD?

The reasons for attentional bias are facilitated attention, difficulty disengaging attention, or attentional avoidance. Facilitated attention refers to the relative ease or speed with which attention is drawn to a preferred stimulus. However, difficulty disengaging attention is the extent to which a preferred stimulus captures the attention and impairs the switching of attention from one stimulus to another. Attentional avoidance is a relatively recent empirical phenomenon in which attention is preferentially allocated toward locations opposite the first location (Cisler and Koster 2010). Zhao et al. (2016) found that children with ASD responded significantly faster to the target presented at the previous location of disgusted facial expressions compared to neutral facial expressions in the spatial-cueing task, indicating that children facilitated their attention oriented to the disgusted faces. May et al. (2015) used the modified dot-probe task, and discovered that individuals with ASD responded more slowly to the probe on the neutral expression in angry-neutral trials than in neutral-neutral trials, reflecting their difficulty in disengaging from the angry expression. However, few studies have investigated the reasons for attentional bias toward CI-related stimuli among individuals with ASD. To the best of our knowledge, only one study found that children with ASD had difficulty in disengaging from CI-related objects by using a gap-overlap visual attention paradigm, suggesting that difficulty disengaging attention results in attentional bias toward CI-related objects (Mo et al. 2019).

The emotional Stroop task and the modified dot-probe task are often used to study attentional bias. The emotional Stroop task, derived from the original Stroop task, has an 80-year history of development and is the earliest and most common paradigm regarding attentional bias (Adams and Jarrold 2012). The Stroop task is about the interference effect between the meaning of a written word and the color of its font, which essentially uses two dimensions to stimulate mutual conflict. The emotional stimuli in the emotional Stroop task replace the words used in the Stroop task with emotional facial expressions or words, and the participant is asked to name the color of the emotional expressions or words as quickly as possible. The participants who are relatively slow to recognize the color of an emotional stimulus relative to the neutral stimulus are identified as positive for the emotional Stroop effect, reflecting the participants’ attentional bias toward the emotional stimuli (Yasumura et al. 2014).

The emotional Stroop task does not distinguish the causes of attentional bias, whereas the modified dot-probe task can reveal specific aspects of attentional bias by adding pairs of neutral stimuli (Puls and Rothermund 2018). The modified dot-probe task first presents pairs of pictures, such as emotional-neutral or neutral-neutral facial expressions, on the right and the left sides, respectively, of a central fixation point. Then, the pair of pictures disappear and a probe appears at the location of one of the previously viewed pictures. The participants are asked to indicate the location (i.e. left or right) of the probe as quickly as possible. The extent of attentional bias is quantified by calculating the difference between the reaction times (RTs) to the probe that replaced the emotional stimulus (i.e. congruent condition) and that replaced the neutral stimulus (i.e. incongruent condition). The participants are identified as having an attentional bias toward the emotional expression or word when the RTs in the congruent condition are significantly faster than in the incongruent condition. Specially, the modified dot-probe task distinguishes aspects of attentional bias using pairs of neutral-neutral stimuli. If the participant’s RTs in the congruent trial are faster than the average RTs for pairs of neutral-neutral stimuli, it demonstrates that the participant’s facilitated attention is oriented to emotional stimuli. However, if the participant’s RTs in the incongruent condition are slower, it reflects their difficulty in disengaging attention from emotional stimuli. Given their critical advantages, using a Stroop-like task and a modified dot-probe task is particularly vital for studying attentional bias in individuals with ASD.

This study aimed to reveal whether there was an attentional bias toward CI-related objects compared to neutral objects and to focus on the aspects of attentional bias in children with ASD by using two tasks. The first task used a Stroop-like task, in which CI-related pictures (instead of the emotional facial expressions or words used in the emotional Stroop task) were used to determine whether there was an attentional bias toward these pictures. The second task used a modified dot-probe task to replicate the findings of the first task and to further reveal the reasons for attentional bias toward CI-related pictures. Experimental materials were selected for the CI-related objects and neutral (unrelated to CI) objects to ensure homogeneity. Based on the results of previous studies, we developed the following hypotheses.

Hypothesis 1: Children with ASD, but not TD children, have longer RTs to CI-related pictures than to neutral color-naming pictures in the Stroop-like task.

Hypothesis 2: Children with ASD, but not TD children, have significantly faster RTs in the congruent trials than in the incongruent trials in the modified dot-probe task.

Hypothesis 3: Only children with ASD have a significantly slower RTs in incongruent trials than the average RTs for pairs of neutral-neutral pictures in the modified dot-probe task, which would demonstrate difficulty disengaging attention resulting in attentional bias toward CI-related stimuli in children with ASD.

Methods

Participants

Seventeen children with ASD and 15 TD children participated in this study, however, two children with ASD who did not successfully complete one or the other tasks were excluded from the study. Thus, the final sample comprised 15 children diagnosed with ASD and 15 TD children. Sample size was determined by availability, and was larger than that of other studies (Pierce and Courchesne, 2001, Sasson et al. 2011, Traynor et al. 2019). Furthermore, based on the results of a prior study, we calculated the sample size required in order to observe a significant interaction between group membership (ASD vs. TD) and condition type (CI-related picture vs. neutral picture). We conducted this analysis with G*Power (Faul et al. 2007), using an alpha of 0.05, power of 0.80, correlation among repeated measures of 0.5, and the effect size found in the previous experiment (ηp2 = 0.19) (Sasson et al. 2011). A total sample of at least 12 individuals was required. Nevertheless, we used 30 participants, which yielded 99% power in both tasks.

The children with ASD were recruited from a special education and primary school in Xi’an, China. ASD diagnoses were made by certified psychiatrists using the DSM-V and Chinese Classification and Diagnoses Criteria of Mental Disorders-III, which were confirmed based on observations and parent interviews after administering standard autism diagnostic assessments (e.g. the Autism Diagnostic Interview-Revised and the Childhood Autism Rating Scale). Diagnosis documentation of all 15 children was provided by the hospital before the experiment. The intelligence quotient of participants was not an exclusion criterion because CI manifest in individuals with ASD with different intelligence levels (Lam et al. 2008). The interviews with the children’s parents or caregivers confirmed the CI characteristics of the children.

TD participants were recruited from a primary school, and the teachers and parents reported no concern about any potential developmental or psychiatric disorder. All participants had normal or corrected-to-normal vision, no history of seizure disorder, no acute medical or genetic condition, and had not previously participated in the same or a similar experiment. The 15 children with ASD (13 males, two females; mean age (SD) = 142.13 (14.14) months, range = 123–167 months) and 15 TD children (14 males, one female; mean age (SD) = 133.87 (8.68) months, range = 125–153 months) did not significantly differ in chronological age (t (28) = 1.93, p > 0.05) or gender (χ2 = 0.37, p > 0.05). All participants’ parents or legal guardians provided written informed consent for their children to participate in the study. All experimental protocols were approved by the Ethics Committee of Shaanxi Normal University.

Materials

The CI-related pictures were based on types of restricted interests common to ASD, and were obtained from the combination of the RBI and YSII, such as vehicles, airplanes, blocks, computer-related items, sports items, road signs, household electronics, and trains (South et al. 2005). The neutral pictures included eight categories of everyday objects not often related to CI, such as clothing, furniture, food, school supplies, bathroom supplies, bags, hats, and plants (Sasson et al. 2008). These CI-related pictures had been found to preferentially capture the visual attention of children and adults with ASD (Cai et al. 2020, Kohls et al. 2018, Parsons et al. 2017, Sun et al. 2021), and had been rated as relatively pleasing by children with ASD (Sasson et al. 2012), whereas neutral pictures were not. In addition, the CI-related pictures and neutral pictures were confirmed by interviews with parents or caregivers with ASD, including the content and frequency of CI-related stimuli in their daily life, to ensure that the experimental material was effective.

The CI-related and neutral pictures were downloaded from the Internet and were similar to each other in size and complexity. All pictures were grayscale, 216 × 188 pixels in size, and the luminance was controlled using Photoshop. A colored border of 10 pixels (i.e. red or blue) was added to each picture in the Stroop-like task.

Procedure

Each participant was seated in a comfortable chair, 65 cm from the computer display, and was asked to complete a Stroop-like task and a modified dot-probe task written in E-prime 2.0.

In the Stroop-like task, the participants first completed 16 practice trials to help them understand the procedure of the practice trials. All the participants in the experimental phase demonstrated 80% or better accuracy in the practice trials. The experiment comprised 58 CI-related pictures and 58 neutral pictures with respectively red and blue borders, for 116 trials. The experimental trials were identical to those of the practice phase. First, the participants read the instructions; then, they pressed the “Q” key on the keyboard to start the task. In the task, a gray-colored central fixation cross appeared for 1,000 ms. Then, a CI-related picture or a neutral picture with a horizontal viewing angle of 4.4° and a vertical viewing angle of 5.0° was randomly displayed in the center of the display for 2,000 ms. The participants were asked to ignore the content of the picture and press the key that corresponded to the color of the picture’s border as quickly and accurately as possible. When the border was red, they were to press the “F” key, and when the border was blue, they were to press the “J” key. The picture on the display did not disappear until the participant pressed a key or until 2,000 ms had passed. Immediately after the picture disappeared, a white noise mask was presented for 2,000 ms. Lastly, an empty display was shown for 2,000 ms, after which the next trial began immediately. A time gap occurred after 48 experimental trials, and the participants pressed the “Q” key to continue the task. The entire task was completed in about 13 min.

In the modified dot-probe task, all participants who successfully finished the 12-trial practice series and had an accuracy rate higher than 80%, completed the experimental phase. The stimuli used in the practice series were neutral stimuli that were not used in the experimental phase. The experiment involved 108 CI-neutral pictures pairs and 108 neutral-neutral pictures pairs, which were each presented twice. The location of the target probe was balanced in each picture pair. Each trial started with a 1,000 ms exposure to a central fixation cross, followed by picture pairs (CI-neutral or neutral-neutral) on the left and right sides of the fixation cross for 500 ms. Each picture had a 4.4° horizontal visual angle, and a 5.6° vertical visual angle. The center of each picture was about 6 cm from the central fixation cross. Then, the pairs of pictures disappeared and the target probe was randomly presented where one of the pictures was presented. The participants were instructed to press either the “F” key (to indicate the left side) or the “J” key (to indicate the right side) to identify the target’s location. The last step was an empty display shown for 2,000 ms, and the next trial started immediately thereafter. A time gap was presented after 50 trials had been completed, and the participant continued the task by pressing the “Q” key. The total task lasted for about 20 min.

Results

The dependent variable was the RTs to the target stimulus in the correct trials, and the cut-off value of statistical significance for all statistical tests was set at p < 0.05. Incorrect trials (ASD: 5.34%, TD: 4.83%) and trials with RTs values three standard deviations above or below the mean RTs by the group were excluded (ASD: 3.22%, TD: 1.55%) in the Stroop-like task, whereas incorrect trials (ASD: 5.19%, TD: 1.79%) and trials with RTs three standard deviations above or below the mean RTs by the group were excluded (ASD: 3.83%, TD: 1.3%) in the modified dot-probe task. Mean and standard deviations of participants in task 1 and 2 are presented in Table 1.

Table 1.

Mean RTs ± SDs for the two groups (ms).

  Task 1 Task 2
  CI-related picture Neutral picture Congruent Incongruent Neutral-neutral
ASD 918.28 ± 183.91 852.79 ± 190.63 645.40 ± 160.95 706.20 ± 162.73 650.68 ± 186.50
TD 687.04 ± 105.49 681.03 ± 104.79 483.13 ± 69.39 478.96 ± 76.72 485.70 ± 75.30
Open in a new tab

Note: RTs, Reaction times; ASD, Autism spectrum disorder; TD, Typically developing children

In the Stroop-like task, RTs were analyzed using a repeated-measures analysis of variance (ANOVA) with Group membership (ASD vs. TD) as the between-subjects factor and Condition type (CI-related picture vs. neutral picture) as the within-subjects factor by SPSS 23.0 and JASP. The ANOVA results (Figure 1) revealed that the main effect of Group membership was statistically significant, F (1, 28) = 13.69, p < 0.01, ηp2 = 0.33, showing that participants with ASD had longer RTs compared to TD children. The main effect of Condition type was significant, F (1, 28) = 11.66, p < 0.01, ηp2 = 0.29, indicating that the RTs were slower for the CI-related condition than the neutral condition. Importantly, the Group membership × Condition type interaction effect was statistically significant, F (1, 28) = 8.07, p < 0.01, ηp2 = 0.22. BF10 (group membership × condition type) of 4.93, indicated moderate evidence for the interaction term according to the classification scheme for the interpretation of Bayes factors BF10 by Wagenmakers et al. (2018). Follow-up simple effects analyses found that children with ASD had slower RTs to CI-related than to neutral pictures, F (1, 28) = 19.57, p < 0.001, ηp2 = 0.41. BF10 of 83.06, indicated very strong evidence in favor of H1 (the alternative hypothesis) and meant that the probability of the present data under H1 was 83.06 times higher than that under H0 (the null hypothesis). However, TD children had similar RTs for CI-related and neutral pictures, F (1, 28) = 0.17, p > 0.05, ηp2 = 0.01. BF10 of 0.28, indicated moderate evidence in favor of H0 and meant that the observed data were 0.28 times more likely under H1 than that under H0. The above results show that children with ASD exhibited an attentional bias towards CI-related stimuli, as they took more time to name the color of the CI-related pictures compared to neutral pictures; TD children did not exhibit this attentional bias because there was no difference in RTs between the CI-related pictures and neutral pictures.

Figure 1.

Figure 1.

Open in a new tab

Mean RTs (ms) and SDs for the two groups in the Stroop-like task.

In the modified dot-probe task, the RTs were analyzed using a repeated-measures ANOVA with Group membership (ASD vs. TD) as the between-subjects factor and Congruency (congruent vs. incongruent) as the within-subject factor. The ANOVA results (Figure 2) indicated that there was a significant main effect of Group membership, F (1, 28) = 18.34, p < 0.001, ηp2 = 0.40, showing that the RTs for participants with ASD were longer than for TD participants. There was also a significant main effect of Congruency, F (1,28) = 23.02, p < 0.001, ηp2 = 0.45, indicating that RTs in the incongruent condition were higher than in the congruent condition. Notably, a significant interaction effect was found between Group membership and Congruency, F (1,28) = 30.30, p < 0.001, ηp2 = 0.52. BF10 (group membership × congruency) = 912.19, indicated extreme evidence in favor of the interaction term. Follow-up simple effects analyses found that the RTs were significantly faster when the target was cued by a CI-related picture than by a neutral picture in children with ASD, F (1,28) = 53.07, p < 0.001, ηp2 = 0.66. BF10 = 1126.44, indicated extreme evidence in favor of H1 and meant that the observed data were 1126.44 times more likely under H1 than that under H0. However, this effect was absent in the TD group, F (1,28) = 0.25, p > 0.05, ηp2 = 0.01. BF10 = 0.31, indicated moderate evidence in favor of H0 and meant that the observed data were 0.31 times more likely under H1 than that under H0. The above results reveal an attentional bias toward CI-related stimuli only in children with ASD, as they had slower RTs in the incongruent condition than in the congruent condition.

Figure 2.

Figure 2.

Open in a new tab

RTs (ms) to the target probe for the two groups in the modified dot-probe task.

To detect the reason for attentional bias toward CI-related pictures in children with ASD, the mean RTs for pairs of neutral-neutral condition were compared to RTs in the congruent and in the incongruent condition, respectively. The t-test results found no significant difference between the mean RTs for pairs of neutral-neutral pictures and RTs in the congruent condition, t (14) = 0.39, p > 0.05, Cohen’s d = 0.10. BF10 of 0.28, indicated moderate evidence in favor of H0 and meant that the observed data were 0.28 times more likely under H1 than that under H0. However, there was a significant difference between the mean RTs for pairs of neutral-neutral trials and RTs in the incongruent condition, t (14) = 3.61, p < 0.01, Cohen’s d = 0.93. BF10 of 15.94, indicated strong evidence in favor of H1 and meant that the observed data were 15.94 times more likely under H1 than that under H0. This result showed that difficulty with attentional disengagement rather than hypervigilance to the CI-related stimuli caused attentional bias, with the 500 ms SOA, for children with ASD.

Discussion

CI are one of the key features of ASD. However, compared with social attention, few studies have focused on the attentional characteristics of CI-related stimuli to date. The findings of these studies discovered that children with ASD show atypical attentional features to CI-related stimuli. For example, one study showed that once individuals with ASD were obsessed with CI, they would ignore others and not allow them to say other things; otherwise, they experienced serious emotional and behavioral problems (Mercier et al. 2000). Unusually high-intensity interests require much time and energy from children with ASD and reduce their engagement and enthusiasm for other activities that may have a positive effect on learning and development of adaptive behaviors (Pierce and Courchesne 2001). Some researchers even argue that social interaction and communication deficits may be caused by CI (Zivan et al. 2021). Further, CI seem to be specific to ASD, while other restricted and repetitive behaviors are frequently observed in other neurodevelopmental disorders (Harrop et al. 2018). Therefore, researchers have sought to understand the mechanisms that underlie CI.

This study investigated whether children with ASD had attentional bias toward CI-related pictures compared to neutral pictures in the absence of social stimuli (Tasks 1 and 2) and the reasons for the attentional bias (Task 2). In the Stroop-like task, children with ASD had slower RTs to name color of CI-related pictures than neutral pictures, indicating that the participants with ASD had an attentional bias toward CI-related pictures; this result supported Hypothesis 1. Compared to the neutral pictures, CI-related pictures occupied relatively more attentional resources and reduced attentional allocation to the target task, which resulted in responding to CI-related pictures more slowly than to neutral pictures for children with ASD. Moreover, these results were verified in Task 2, that indicated faster RTs to target probe replacement of CI-related stimuli than of neutral stimuli in the CI-neutral trials. These findings suggested attentional bias toward CI-related stimuli for children with ASD, supported Hypothesis 2, and were consistent with previous studies’ findings (Harrop et al. 2018, Parsons et al. 2017, Sasson et al. 2008, 2011, 2012). However, the current results contradicted those of Parish-Morris et al. (2013), who concluded that six- to 17-year-old individuals with ASD showed no attentional bias toward CI-related stimuli, as they found no significant difference in total fixation duration between CI-related pictures and other types of stimuli. Differences in experimental materials might be an important reason for these diametrical outcomes. Parish-Morris et al. (2013) used ecologically interesting videos as their experimental materials, while this study used static pictures. Previous studies found that individuals with ASD often have rapid visual-motor integration deficits (Gepner and Mestre 2002). Therefore, dynamic materials might not attract a large number of attentional resources, and instead might limit the normal attentional processing of stimuli. Besides, one possible explanation for the attentional bias to CI-related stimuli in this study may be positive affective experience for individuals with ASD. Viewing images related to CI may activate neural reward circuitry (Dichter et al. 2012), and elevate affective responses (Wang et al. 2021), which reduce anxiety symptoms (Sterling et al. 2013) and improve subjective wellbeing and life satisfaction (Grove et al. 2018). Thus, children with ASD tend to pay more attention to CI-related stimuli.

In addition, in the modified-probe task, the RTs to the target probe in the incongruent condition were significantly slower than the average RTs for pairs of neutral-neutral stimuli, which reflected difficulty disengaging from CI-related pictures for children with ASD. CI-related pictures were more likely to capture attentional resources in the children with ASD than in the TD group. When identifying the target probe in the incongruent trials, participants with ASD had difficulty inhibiting their processing of CI-related stimuli. Thus, it took them more time to disengage from the CI-related stimuli and turn their attention toward the target probe on the other side of the display, creating longer RTs regarding the target probe presented at the incongruent position, which provided evidence of delayed disengagement from CI-related pictures. This result supported Hypothesis 3, which expected that the RTs in incongruent trials would be significantly slower than the RTs for pairs of neutral-neutral pictures. Difficulty disengaging attention from CI-related pictures might be influenced by attentional characteristics of children with ASD and the observed objects’ attributes. Regarding attentional characteristics, the deficits of inhibitory control and flexibility caused individuals with ASD to be unable to interrupt ongoing activities in a reasonable way to appropriately and timely regulate their attention (Borkowska 2016). Thus, their visual attention was “obsessed” with the current stimulus. Regarding the objects’ properties, the reward and arousal of the stimuli caused delayed disengagement. On the one hand, CI-related stimuli gave reward feedback (Kohls et al. 2018, Parsons et al. 2017), and stimuli with reward feedback created difficulties with individuals’ ability to disengage their attention (Frewen et al. 2008). On the other hand, Benning et al. (2016) found that individuals with ASD were more aroused by CI-related pictures, and that low-arousal objects were less likely than high-arousal objects to generate slow disengagement (Vogt et al. 2008).

Finally, there were no significant differences between RTs to probes in the congruent condition and the mean RTs for pairs of neutral-neutral stimuli in Task 2. Nonetheless, these results should be considered cautiously and it is difficult to conclude whether attentional bias to CI-related stimuli was not caused by facilitated attention or attentional avoidance in children with ASD, because stimulus onset asynchrony (SOA) selected was 500 ms in the current study. SOA is an important factor that affects the processing of attentional bias. According to the bi-processing theory of attention, attentional bias includes automatic processing and controlled processing stages, which correspond to bottom-up and top-down processes of attention regulation, respectively, with SOA as an important factor distinguishing these two stages (Fan et al. 2020). Generally speaking, automatic processing phase mostly occurs when the SOA is equal to or less than 500 ms, while controlled processing mostly occurs at more than 1000 ms. Further, with regards to the reasons for attentional bias, facilitated attention usually occurs at the supra-threshold presentation or at less than 100 ms, and difficulty disengaging attention tends to occur when SOA exceeds 200 ms (Koster et al. 2006). Attentional avoidance occurs when SOA exceeds 3,000 ms (Gamble and Rapee 2010). Thus, more specifically, the current data provide support for the attentional bias toward CI-related stimuli in automatic processing phase for children with ASD. These salient physical properties and biologically relevant information of CI-related stimuli drive children with ASD to process them automatically under unconscious and involuntary conditions. Importantly, the current results concerning difficulty disengaging attention oriented to CI-related stimuli were limited to an SOA of 500 ms. The attention engagement is not static, but is a dynamic process, and the reasons for attentional bias are more likely to switch for different time periods, which change dynamically with different SOA. For example, vigilance-maintenance pattern means early orienting and continuous maintenance bias, while avoidance-maintenance pattern means less likely to direct their initial orientation to stimuli, but once they noticed it, they showed maintenance bias. So, various SOA should be used in future studies to further reveal the dynamic characteristics of attentional bias toward CI-related stimuli in spatial processes for individuals with ASD.

Implications for practice

The results of this study support the existence of attentional bias toward CI-related stimuli among children with ASD. This information can be used by therapists in regularly practice and in developing interventions for children with ASD. They can take advantage of attentional bias and integrate CI-related stimuli into interventions through antecedent-based and consequence-based approaches, which have significantly positive effects in the field of teaching, social interaction, and emotion, to help transfer attention from CI-related stimuli to social stimuli (Grove et al. 2018). The antecedent-based approach considers CI as the center of intervention and organizes activities around CI-related stimuli, integrating targeted behaviors into activities. Further, it aims to promote the occurrence of adaptive behaviors in the activities, by triggering intrinsic motivation and interest in such activities among children with ASD. For example, Koegel et al. (2012) conducted games around the CI-related stimuli in children with ASD, and found that children with ASD had significantly increased their time with peers after the intervention. The consequence-based approach treats CI-related stimuli as reinforcements after children with ASD have completed the targeted behavior. Existing research has found that compared to ordinary stimuli, individuals with ASD had a higher accuracy rate of completing tasks and less frequent occurrence of challenging behaviors when the reinforcement was a CI-related stimulus (Charlop-Christy and Haymes 1998). It is noteworthy that educators should control the engagement time for the CI-related stimuli in children with ASD, to avoid “obsessing” because of the difficulty in disengaging attention and the emergence of negative emotions after the mandatory removal of the CI-related stimuli. At the same time, educators need to realize that it is more likely that the high-intensity of CI is caused by difficulty disengaging attention for children with ASD. Therefore, during the intervention process, educators need to gradually shorten the duration of these children's investment in CI-related stimuli and hence, reduce their dependence on it.

However, how to effectively select CI-related stimuli is a significant concern for both educators and researchers, as every individual with ASD has different CI. Nonetheless, even though there is heterogeneity in the CI of individuals with ASD, it needs to be determined that they still have a common aspect, such as CI-related stimuli are often non-social objects. We can filter out the CI-related stimuli in individuals with ASD with the help of scales with good reliability and validity. In addition, we can use eight common categories of CI-related stimuli proposed by South et al. (2005), which are based on RBI and YSII. A growing body of empirical studies has confirmed that these eight common types of CI-related stimuli are suitable for different age groups of American and Chinese children with ASD. The current study adopted the types of CI-related stimuli summarized by South et al. (2005), and the results of this study reconfirms the validity of the CI-related stimuli described by South et al. (2005), that is, children with ASD have attentional bias toward them. Therefore, these eight categories of stimuli can be good tools, for educators and researchers in identifying CI-related to stimuli in children with ASD.

Limitations and future directions

Although this study was a preliminary study that discovered the reason underlying CI in children with ASD, it had several limitations. One limitation of the study was that it did not distinguish SOA in the modified-dot probe task, which just selected for 500 ms. This restricted the generalization of the research conclusions to some extent, and hardly revealed the dynamic characteristics of attention bias toward CI-related stimuli for children with ASD. Additionally, this study did not consider that sex differences may have also influenced the results. Recent research has paid greater attention to how different sex may show ASD characteristics differently (Calderoni 2022). For example, Grove et al. (2018) found that females with ASD were significantly more likely than males with ASD to be interested in dolls and animals, while males with ASD were more likely than females with ASD to have an interest in vehicles and computers. The intensity of CI in males with ASD was significantly higher than that in females with ASD (Harrop et al. 2018). Therefore, gender differences should be explored in future studies to understand how these differences affect the attentional bias toward CI-related stimuli in children with ASD.

Finally, this study argued that the personal attention characteristics and objects’ attributes caused difficulty in disengaging attention from CI-related stimuli for children with ASD. However, the basis of this argument needs to be confirmed to understand whether there is indeed a relationship between attentional characteristics and objects’ properties and whether individuals with ASD have poorer inhibitory control and flexibility when processing high-reward stimuli, such as CI-related stimuli. If they do, which of the stimuli have the stronger and primary position of impacting attentional bias? These questions should also be explored in future research.

In sum, this study used a progressive approach with a Stroop-like task and a modified dot-probe task to investigate attentional bias toward CI-related stimuli in children with ASD. Compared to the control group, the children with ASD had an attentional bias toward CI-related pictures, which was caused by their slower disengagement from the CI-related pictures under the SOA of 500 ms condition.

Acknowledgements

We are grateful to all participating children and their families who participated in this study. Special thanks to the special education school for the subjects provided for this study.

Funding Statement

This research was supported by the National Social Science Fund of China [grant number 21&ZD293] and the Shanghai Major Innovation Project of Educational Science (Grant NO. 2019-01-07-00-05-E00007) awarded to Huizhong He.

Disclosure statement

No potential conflict of interest was reported by the authors.

References

  1. Adams, N. C. and Jarrold, C.. 2012. Inhibition in autism: Children with autism have difficulty inhibiting irrelevant distractors but not prepotent responses. Journal of Autism and Developmental Disorders, 42, 1052–1063. [DOI] [PubMed] [Google Scholar]
  2. American Psychiatric Association . 2013. Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing. [Google Scholar]
  3. Baron-Cohen, S. and Wheelwright, S.. 1999. ‘Obsessions’ in children with autism or Asperger syndrome. Content analysis in terms of core domains of cognition. The British Journal of Psychiatry, 175, 484–490. [DOI] [PubMed] [Google Scholar]
  4. Benning, S. D., Kovac, M., Campbell, A., Miller, S., Hanna, E. K., Damiano, C. R., Sabatino-DiCriscio, A., Turner-Brown, L., Sasson, N. J., Aaron, R. V., Kinard, J. and Dichter, G. S.. 2016. Late positive potential ERP responses to social and nonsocial stimuli in youth with autism spectrum disorder. Journal of Autism and Developmental Disorders, 46, 3068–3077. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Borkowska, A. R. 2016. The dynamics of attentional and inhibitory functions in the presence of distracting stimuli in children with attention-deficit/hyperactivity disorder, high-functioning autism and oppositional defiant disorder. Psychiatria i Psychologia Kliniczna, 16, 68–80. [Google Scholar]
  6. Cai, X. E., Shou, X. J. and Wang, X. X.. 2020. Eye-tracking study on circumscribed interest in preschool-aged children with autism spectrum disorder. Chinese Journal of Behavioral Medicine and Brain Science, 29, 840–846. [Google Scholar]
  7. Calderoni, S. 2022. Sex/gender differences in children with autism spectrum disorder: A brief overview on epidemiology, symptom profile, and neuroanatomy. Journal of Neuroscience Research. [DOI] [PubMed] [Google Scholar]
  8. Charlop-Christy, M. H. and Haymes, L. K.. 1998. Using objects of obsession as token reinforcers for children with autism. Journal of Autism and Developmental Disorders, 28, 189–198. [DOI] [PubMed] [Google Scholar]
  9. Cisler, J. M. and Koster, E. H.. 2010. Mechanisms of attentional biases towards threat in anxiety disorders: An integrative review. Clinical Psychology Review, 30, 203–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dichter, G. S., Richey, J. A., Rittenberg, A. M., Sabatino, A. and Bodfish, J. W.. 2012. Reward circuitry function in autism during face anticipation and outcomes. Journal of Autism and Developmental Disorders, 42, 147–160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fan, X. Z., Duan, Y. W., Yi, L. X. and He, H. Z.. 2020. Attentional bias toward threatening emotional faces in individuals with autism spectrum disorder: A meta-analysis on reaction time tasks. Research in Autism Spectrum Disorders, 78, 101646. [Google Scholar]
  12. Faul, F., Erdfelder, E., Lang, A.-G. and Buchner, A.. 2007. G∗Power 3: A flexible statistical power analysis program for the social, behavioral, biomedical sciences. Behavior Research Methods, 39, 175–191. [DOI] [PubMed] [Google Scholar]
  13. Foss-Feig, J. H., McGugin, R. W., Gauthier, I., Mash, L. E., Ventola, P. and Cascio, C. J.. 2016. A functional neuroimaging study of fusiform response to restricted interests in children and adolescents with autism spectrum disorder. Journal of Neurodevelopmental Disorders, 8, 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Frewen, P. A., Dozois, D. J. A., Joanisse, M. F. and Neufeld, R. W. J.. 2008. Selective attention to threat versus reward: Meta-analysis and neural-network modeling of the dot-probe task. Clinical Psychology Review, 28, 307–337. [DOI] [PubMed] [Google Scholar]
  15. Gamble, A. L. and Rapee, R. M.. 2010. The time-course of attention to emotional faces in social phobia. Journal of Behavior Therapy and Experimental Psychiatry, 41, 39–44. [DOI] [PubMed] [Google Scholar]
  16. Gepner, B. and Mestre, D.. 2002. Rapid visual-motion integration deficit in autism. Trends in Cognitive Sciences, 6, 455–455. [DOI] [PubMed] [Google Scholar]
  17. Grove, R., Hoekstra, R. A., Wierda, M. and Begeer, S.. 2018. Special interests and subjective wellbeing in autistic adults. Autism Research, 31, 12–18. [DOI] [PubMed] [Google Scholar]
  18. Harrop, C., Green, J. and Hudry, K.. 2017. Play complexity and toy engagement in preschoolers with autism spectrum disorder: Do girls and boys differ? Autism, 21, 37–50. [DOI] [PubMed] [Google Scholar]
  19. Harrop, C., Jones, D., Zheng, S., Nowell, S., Boyd, B. A. and Sasson, N.. 2018. Circumscribed interests and attention in autism: The role of biological sex. Journal of Autism and Developmental Disorders, 48, 3449–3459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Koegel, L. K., Vernon, T., Koegel, R. L., Koegel, B. L. and Paullin, A. W.. 2012. Improving social engagement and initiations between children with autism spectrum disorder and their peers in inclusive settings. Journal of Positive Behavior Interventions, 14, 220–230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kohls, G., Antezana, L., Mosner, M. G., Schultz, R. T. and Yerys, B. E.. 2018. Altered reward system reactivity for personalized circumscribed interests in autism. Molecular Autism, 9, 9–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Koster, E. H. W., Crombez, G., Verschuere, B. and Houwer, J. D.. 2006. Attention to threat in anxiety-prone individuals: Mechanisms underlying attentional bias. Cognitive Therapy and Research, 30, 635–643. [Google Scholar]
  23. Lam, K. S. L., Bodfish, J. W. and Piven, J.. 2008. Evidence for three subtypes of repetitive behavior in autism that differ in familiality and association with other symptoms. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 49, 1193–1200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lepper, T. L., Devine, B. and Petursdottir, A. I.. 2017. Application of a lag contingency to reduce perseveration on circumscribed interests. Developmental Neurorehabilitation, 20, 313–316. [DOI] [PubMed] [Google Scholar]
  25. Manyakov, N. V., Bangerter, A., Chatterjee, M., Mason, L., Ness, S., Lewin, D., Skalkin, A., Boice, M., Goodwin, M. S., Dawson, G., Hendren, R., Leventhal, B., Shic, F. and Pandina, G.. 2018. Visual exploration in autism spectrum disorder: Exploring age differences and dynamic features using recurrence quantification analysis. Autism Research, 11, 1554–1566. [DOI] [PubMed] [Google Scholar]
  26. Ma, X., Gu, H. and Zhao, J.. 2021. Atypical gaze patterns to facial feature areas in autism spectrum disorders reveal age and culture effects: A meta-analysis of eye-tracking studies. Autism Research, 14, 2625–2639. [DOI] [PubMed] [Google Scholar]
  27. May, T., Cornish, K. and Rinehart, N. J.. 2015. Mechanisms of anxiety related attentional biases in children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 45, 3339–3350. [DOI] [PubMed] [Google Scholar]
  28. Mercier, C., Mottron, L. and Belleville, S.. 2000. A psychosocial study on restricted interests in high-functioning persons with pervasive developmental disorders. Autism, 4, 406–425. [Google Scholar]
  29. Mo, S., Liang, L., Bardikoff, N. and Sabbagh, M. A.. 2019. Shifting visual attention to social and non-social stimuli in autism spectrum disorders. Research in Autism Spectrum Disorders, 65, 56–64. [Google Scholar]
  30. Morrison, K. E., Chambers, L. K., Faso, D. J. and Sasson, N. J.. 2018. The content and function of interests in the broad autism phenotype. Research in Autism Spectrum Disorders, 49, 25–33. [Google Scholar]
  31. Parish-Morris, J., Chevallier, C., Tonge, N., Letzen, J., Pandey, J. and Schultz, R. T.. 2013. Visual attention to dynamic faces and objects is linked to face processing skills: A combined study of children with autism and controls. Frontiers in Psychology, 4, 185–192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Parsons, O. E., Bayliss, A. P. and Remington, A.. 2017. A few of my favorite things: Circumscribed interests in autism are not accompanied by increased attentional salience on a personalized selective attention task. Molecular Autism, 8, 20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Pierce, K. and Courchesne, E.. 2001. Evidence for a cerebellar role in reduced exploration and stereotyped behavior in autism. Biological Psychiatry, 49, 655–664. [DOI] [PubMed] [Google Scholar]
  34. Puls, S. and Rothermund, K.. 2018. Attending to emotional expressions: No evidence for automatic capture in the dot-probe task. Cognition & Emotion, 32, 450–454. [DOI] [PubMed] [Google Scholar]
  35. Richler, J., Huerta, M., Bishop, S. L. and Lord, C.. 2010. Developmental trajectories of restricted and repetitive behaviors and interests in children with autism spectrum disorders. Development and Psychopathology, 22, 55–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Safra, L., Ioannou, C., Amsellem, F., Delorme, R. and Chevallier, C.. 2018. Distinct effects of social motivation on face evaluations in adolescents with and without autism. Scientific Rports, 8, 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sasson, N. J., Dichter, G. S. and Bodfish, J. W.. 2012. Affective responses by adults with autism are reduced to social images but elevated to images related to circumscribed interests. PLoS One, 7, e42457–457. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sasson, N. J., Elison, J. T., Turner-Brown, L. M., Dichter, G. S. and Bodfish, J. W.. 2011. Brief report: Circumscribed attention in young children with autism. Journal of Autism and Developmental Disorders, 41, 242–247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sasson, N. J. and Touchstone, E. W.. 2014. Visual attention to competing social and object images by preschool children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 44, 584–592. [DOI] [PubMed] [Google Scholar]
  40. Sasson, N. J., Turner-Brown, L. M., Holtzclaw, T. N., Lam, K. S. L. and Bodfish, J. W.. 2008. Children with autism demonstrate circumscribed attention during passive viewing of complex social and nonsocial picture arrays. Autism Research, 1, 31–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. South, M., Ozonoff, S. and Mcmahon, W. M.. 2005. Repetitive behavior profiles in Asperger syndrome and high-functioning autism. Journal of Autism and Developmental Disorders, 35, 145–158. [DOI] [PubMed] [Google Scholar]
  42. Sterling, L., Munson, J., Estes, A., Murias, M., Webb, S. J., King, B. and Dawson, G.. 2013. Fear-potentiated startle response is unrelated to social or emotional functioning in adolescents with autism spectrum disorders. Autism Research, 6, 320–331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Sun, B. B., Wang, X. Y. and Wei, Z.. 2021. Research on restrictive interest of high-risk children with autism spectrum disorder based on eye tracking technology. Chinese Journal of Child Health Care, 29, 932–936. [Google Scholar]
  44. Traynor, J. M., Gough, A., Duku, E., Shore, D. I. and Hall, G.. 2019. Eye tracking effort expenditure and autonomic arousal to social and circumscribed interest stimuli in autism spectrum disorder. Journal of Autism and Developmental Disorders, 49, 1988–2002. [DOI] [PubMed] [Google Scholar]
  45. Troyb, E., Knoch, K., Herlihy, L., Stevens, M. C., Chen, C.-M., Barton, M., Treadwell, K. and Fein, D.. 2016. Restricted and repetitive behaviors as predictors of outcome in autism spectrum disorders. Journal of Autism and Developmental Disorders, 46, 1282–1296. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Vogt, J., De Houwer, J., Koster, E. H. W., Van Damme, S. and Crombez, G.. 2008. Allocation of spatial attention to emotional stimuli depends upon arousal and not valence. Emotion, 8, 880–885. [DOI] [PubMed] [Google Scholar]
  47. Wagenmakers, E.-J., Love, J., Marsman, M., Jamil, T., Ly, A., Verhagen, J., Selker, R., Gronau, Q. F., Dropmann, D., Boutin, B., Meerhoff, F., Knight, P., Raj, A., van Kesteren, E.-J., van Doorn, J., Šmíra, M., Epskamp, S., Etz, A., Matzke, D., de Jong, T., van den Bergh, D., Sarafoglou, A., Steingroever, H., Derks, K., Rouder, J. N. and Morey, R. D.. 2018. Bayesian inference for psychology. Part II: Example applications with. Psychonomic Bulletin & Review, 25, 58–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Wang, Q., Hoi, S. P., Song, C., Li, T., Lam, C. M., Wang, Y. and Yi, L.. 2021. Circumscribed interest modulates attention to eyes in boys with and without autism spectrum disorder. Frontiers in Psychiatry, 12, e627365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Yasumura, A., Kokubo, N., Yamamoto, H., Yasumura, Y., Nakagawa, E., Kaga, M., Hiraki, K. and Inagaki, M.. 2014. Neurobehavioral and hemodynamic evaluation of Stroop and reverse Stroop interference in children with attention-deficit/hyperactivity disorder. Brain & Development, 36, 97–106. [DOI] [PubMed] [Google Scholar]
  50. Zhao, X., Zhang, P., Fu, L. and Maes, J. H.. 2016. Attentional biases to faces expressing disgust in children with autism spectrum disorders: An exploratory study. Scientific Reports, 6, 19381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Zheng, L., Grove, R. and Eapen, V.. 2019. Spectrum or subtypes? A latent profile analysis of restricted and repetitive behaviours in autism. Research in Autism Spectrum Disorders, 57, 46–54. [Google Scholar]
  52. Zivan, M., Morag, I., Yarmolovsky, J. and Geva, R.. 2021. Hyper-reactivity to salience limits social interaction among infants born pre-term and infant siblings of children with ASD. Frontiers in Psychiatry, 12, 14. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from International Journal of Developmental Disabilities are provided here courtesy of The British Society of Developmental Disabilities

RESOURCES