Cognitive Predictors of Social Interaction in Daily Contexts Among Children With Autism Spectrum Disorder

Abstract

Importance: Theory of mind (ToM) and executive function (EF) are often used to explain social interaction deficits in children with autism spectrum disorder (ASD), but no behavioral studies have specifically examined their relationship.

Objective: To investigate the cognitive correlates of social interaction in daily contexts by ToM as well as cool and hot EFs among children with ASD.

Design: Cross-sectional study.

Setting: Teaching hospitals, clinics, and developmental centers.

Participants: One hundred thirty-two children with ASD and their caregivers.

Outcomes and Measures: Measurements included the Vineland Adaptive Behavior Scales, Theory of Mind Task Battery, Dimensional Change Card Sort, and Children’s Gambling Task, respectively, for children’s daily social interaction, ToM, cool EF, and hot EF. Pearson’s correlation analyses and three hierarchical regression models were conducted to identify the significant predictors of daily social interaction while controlling for verbal comprehension, measured using the Verbal Comprehension Index (VCI) of two Wechsler scales.

Results: ToM and cool EF were significant cognitive predictors of social interaction in daily contexts in children with ASD whose verbal comprehension was average or above average.

Conclusions and Relevance: Our results suggest that ToM and cool EF are predictors of social interaction in daily contexts when considering children’s verbal comprehension. Hot EF was not a significant predictor, contrary to our hypothesis. This behavioral study fills a research gap by enhancing the understanding of important cognitive correlates of social interaction in daily contexts for children with ASD to improve evaluation and intervention planning with this population.

What This Article Adds: This study identified two cognitive predictors, ToM and cool EF, of social interaction in daily contexts for children with ASD. In addition to verbal comprehension, occupational therapy practitioners should assess ToM and cool EF to inform more comprehensive evaluation and intervention planning to improve social interaction in daily contexts for children with ASD.

This study found that theory of mind and cool executive function are the cognitive predictors of social interaction in daily contexts among children with ASD whose verbal comprehension is average or above.

Social interaction deficits are a core feature of autism spectrum disorder (ASD; American Psychiatric Association, 2013). In children with ASD, these deficits present as a wide range of maladaptive social behaviors, such as a lack of awareness of the reactions of others, low use of joint attention in dyadic and triadic interactions, inappropriate social behaviors, social isolation, poor use of nonverbal communication, difficulties in sharing experiences with others, and the inability to form relationships (Charman & Baird, 2002; Christensen et al., 2018; Mendelson et al., 2016; Wong et al., 2015). The social interaction deficits of children with ASD lead to difficulties participating in daily social life. Increased social interactions with other children, however, have been shown to improve both social play behaviors and language skills (Butler, 2016). The benefits of social participation are well documented and include increased prosocial skills (Eisenberg et al., 2015), increased academic achievement (Nakamoto & Schwartz, 2010), reduced school dropout risk, and reduced risk of later adjustment problems (Malecki & Elliot, 2002; Parker & Asher, 1987).

Complementing the behavioral research, research investigating the neural correlates of social interaction (Sakaiya et al., 2013) has found that theory of mind (ToM) and executive function (EF) may play critical roles in social interaction deficit and may also influence social interaction in daily contexts among children with ASD. This area is worthy of further investigations.

ToM is a social cognitive ability (Adolphs, 2001); the term has been used interchangeably with terms such as “perspective taking” and “social cognition.” The ventromedial prefrontal cortex (VMPFC) and precuneus seem to be important neural processing centers for ToM (Sakaiya et al., 2013). With ToM, a person can understand and then predict the behaviors and actions of others. Therefore, ToM is a prerequisite skill for developing reciprocal communication and social interaction (Schneider et al., 2020; Wellman, 1990). ToM helps people understand and distinguish emotional states and complex social information to participate in social life. Previous studies have determined that people with ASD have impaired ToM (Baron-Cohen, 2000), which may bring about impairment in the social cognitive skills needed for daily social interaction (Sally & Hill, 2006).

EF is a set of cognitive and behavioral skills that work together in the execution of purposeful, goal-oriented tasks (Lezak et al., 2004), including social activities. EF encompasses the ability to suppress responses, keep and manipulate information in mind, change strategies, and plan ahead. Currently, EF includes two types: cool EF and hot EF (Krain et al., 2006; Tsermentseli & Poland, 2016). Cool EF, the more cognitive aspect of EF, is related to the dorsolateral prefrontal cortex (Frith et al., 1991; Krain et al., 2006) and is involved in processing relatively abstract, decontextualized, nonemotional, and analytical conditions. Hot EF, the affective aspect of EF, is related to the orbitofrontal cortex (a part of the VMPFC) and other medial regions (Kerr & Zelazo, 2004; Krain et al., 2006) and is involved in dealing with conditions that induce emotion and motivation.

Both cool and hot EFs are needed for effective social interaction. Cool EF is involved in the continuous updating, evaluation, and selection of appropriate responses to verbal and nonverbal information in social behaviors (Landa & Goldberg, 2005). Hot EF might play an important role in social contexts because reappraisal, a function of hot EF, is required for rapidly changing social contexts (Happaney et al., 2004; Rolls, 2004). However, little research has directly examined the relation between cool EF and social interaction in daily contexts, and no studies have investigated the relation between hot EF and daily social interaction, especially among children with ASD.

On the basis of a previous literature review on social interaction, it appears that ToM and EF are often used to explain social interaction deficits among children with ASD (Roselló et al., 2017); however, these two theories may not be able to fully explain the social interaction deficits of these children. Moreover, although the regions controlling ToM and EF have been found to be the neural correlates of social interaction, no behavioral studies have examined ToM and EF simultaneously in daily social interaction in children with ASD. Therefore, we aimed to determine whether ToM, cool EF, and hot EF are cognitive correlates of social interaction in daily contexts for children with ASD. The research hypothesis of this study was that ToM, cool EF, and hot EF are predictors of social interaction in real-life contexts among children with ASD.

Method

Participants

A total of 132 children with ASD (M age = 97.3 mo, SD = 24.4) were recruited from January 2015 to September 2018 at teaching hospitals, clinics, and developmental centers in Taiwan. Data included in this study were collected as part of a larger study involving social interaction among children with ASD. Some data from this study have been published (Lai et al., 2020; Yu et al., 2021). The following inclusion criteria were used: (1) formal diagnosis of autistic disorder or Asperger’s disorder based on criteria in the Diagnostic and Statistical Manual of Mental Disorders (DSM; 4th ed., text rev.; American Psychiatric Association [APA], 2000) or formal diagnosis of ASD based on DSM–5 (APA, 2013) criteria by a trained psychiatrist or pediatrician, (2) ages 4 to 12 yr, (3) ability to follow directions, (4) no organic brain dysfunction (e.g., seizures), and (5) no uncorrectable impairments of vision or hearing. This study was approved by the institutional review boards of two teaching hospitals, National Cheng Kung University Hospital and Kaohsiung Municipal Kai-Syuan Psychiatric Hospital. Written informed consent was obtained from each child’s caregiver and from children age 7 or older. The sample size was calculated by G*Power 3.1.9.7 software (Faul et al., 2007) for an effect size (f ²) of 0.15, α value of .05, and power of 0.8 in a regression model. A minimum of 109 children were required for the regression model analysis in this study.

Measures

Vineland Adaptive Behavior Scales

Children’s social interaction in daily life was measured with the Socialization domain of the Vineland Adaptive Behavior Scales (VABS; Sparrow et al., 1984). The VABS is a semistructured interview with a primary caregiver, and it is used to assess the personal and social sufficiency of children ages 3 to 12 yr. The Socialization domain has three subdomains: Interpersonal Relationships, Play/Leisure Time, and Coping Skills. These three socialization subdomains pertain to those skills needed to get along with others and to regulate emotions and behaviors as well as the skills involved in leisure activities such as play. Each item is scored from 0 to 2 according to frequency. The VABS has been translated into Chinese, and the Chinese version has good split-half reliability, test–retest reliability, interrater reliability, and construct validity (Wu et al., 2004).

Theory of Mind Task Battery

The Theory of Mind Task Battery (ToMTB; Hutchins et al., 2014) was used to measure children’s ToM in this study. The ToMTB consists of 15 test questions within 9 tasks designed to measure a range of ToM tasks. Tasks are presented as short vignettes with illustrations and accompanying text, in a storybook format, arranged in ascending difficulty. Each item includes memory control questions and test questions scored 1 (pass) or 0 (fail). Children are awarded points on the test questions only if they pass the corresponding control questions. Children can answer the questions verbally or point to a picture to show their answer; thus, the ToMTB is appropriate for both verbal and nonverbal children with ASD. The ToMTB has good internal consistency, test–retest reliability, and criterion-related validity (Hutchins et al., 2012, 2014).

Dimensional Change Card Sort

Children’s cool EF was assessed with the computerized version of the Dimensional Change Card Sort (DCCS; Dichter et al., 2010; Zelazo, 2006). In this version of the DCCS, the child is presented with seven pairs of target cards, of which there are two kinds: “blue rabbit” and “red boat.” They must follow the experimenter’s sorting rule to sort the cards by either color or shape. First, the child is provided a rule for sorting the cards according to one dimension (e.g., color). After sorting several pairs according to that dimension, the child is directed to sort cards according to the other one (e.g., shape; Zelazo, 2006). After practicing for several trials, they perform seven blocks of test trials with no feedback. The criterion for each block is either color or shape. In Blocks 1, 3, and 6, one criterion is used; the other is used in Blocks 2, 5, and 7. The order is counterbalanced across children. Block 4 is a mixed block, wherein the color and shape sorting rules are mixed (pseudo randomly intermixed color and shape trials). Each normal block includes 10 trials, and the mixed block has 20 trials (i.e., 7 switch trials and 13 nonswitch trials). At the beginning of each trial, model pictures (i.e., a blue rabbit and a red boat) are presented to indicate the keys to press, and a cue picture indicates the matching criterion for the trial (color or shape). In this study, accuracy and reaction time were recorded with E-Prime software (Version 2.0; Sharpsburg, PA). The shifting cost was computed as the difference in accuracy and reaction time between the last two correct trials in the first block and the first two trials in the second block.

Children’s Gambling Task

The Children’s Gambling Task (CGT; Kerr & Zelazo, 2004) was used for assessing hot EF. This task is a simplified version of a similar task for assessing adults, the Iowa Gambling Task (IGT; Kerr & Zelazo, 2004). The CGT is used for assessing affective decision making in hot EF among children (Kerr & Zelazo, 2004). It resembles the IGT in that it is a psychological task that simulates decision making in real life. The materials for the CGT are two decks of 50 cards, one advantageous and one disadvantageous. Choosing from the former consistently earns the child a net gain in rewards (usually candies), whereas choosing from the latter consistently causes a net loss. The advantageous cards each provide a gain of one reward and either zero or one loss. The disadvantageous cards have more variation, providing a gain of two rewards along with losses of zero, four, five, or six rewards. The gains and losses of each card in each deck are proportional to those in the IGT, and the cards are always presented in the same order. Relatively advantageous affective decisions are indicated by positive CGT scores; disadvantageous decisions are indicated by negative CGT scores.

Wechsler Intelligence Scale for Children–Fourth Edition and Wechsler Preschool and Primary Scale of Intelligence–Fourth Edition

To measure the children’s verbal comprehension, we used the Verbal Comprehension Index (VCI) of the Chinese version of the Wechsler Intelligence Scale for Children–Fourth Edition (WISC–IV; Chen & Chen, 2007) for children ages 6 yr, 0 mo to 16 yr, 11 mo; for those ages 2 yr, 6 mo to 7 yr, 7 mo, we used the VCI of the Wechsler Preschool and Primary Scale of Intelligence–Fourth Edition (WPPSI–IV; Chen & Chen, 2013). The VCI of the WPPSI–IV is composed of two main subscales: Similarities and Vocabulary. The VCI of the WISC–IV contains the subtests of Vocabulary, Similarities, and Comprehension. The standard mean score is 100, and the standard deviation is 15. The Chinese versions of the WISC–IV and WPPSI–IV are reported to have acceptable to excellent psychometric properties (Chen & Chen, 2007 ; Chen & Chen, 2013).

Procedures

Because prolonged assessments could result in fatigue, the children visited the clinic twice for data collection. Each visit was sufficiently long to complete the assessments (about 1 hr). On the first visit, the WISC–IV or WPPSI–IV was administered, and those who scored <70 on the VCI of either assessment were excluded. Those who met the inclusion and exclusion criteria were then assessed with the ToMTB, and a basic information sheet was filled out by their caretakers. On the second visit, the DCCS and the CGT were conducted, and the socialization part of the VABS was completed by the caretakers.

Statistical Analysis

The demographic and experimental data of the children were processed with descriptive analysis. Pearson’s correlation analyses (two tailed) were conducted for the associations among ToM, cool EF, hot EF, social interaction, and verbal comprehension. Then, for identification of the significant predictors of daily social interaction three hierarchical regression models were conducted, controlling for verbal comprehension. The four VABS variables were the dependent variables. We added the independent variables to the regression models in two steps. All of the variables that were controlled for, including the VCIs, were first entered as the baseline. The variables of the ToMTB total score, DCCS, and CGT were then added to the baseline model as the augmented model. The coefficient of determination (R ²) and R ² change (ΔR ²), the F test of overall significance and model comparison, and the regression coefficients were calculated to determine the significance of ToM, cool EF, and hot EF. The significance level was set at α < .05. All the collected data were analyzed with IBM SPSS Statistics (Version 17.0) for Windows.

Results

The descriptive results (Table 1) indicated that these children had average verbal comprehension and that their adaptive level in socialization was low to moderate. On average, their ToM development was at a basic stage. In the DCCS, children performed less well in the mixed block than in the single block, as indicated by lower accuracy and longer reaction time. The mean score of their CGT indicated that these children chose more cards from the advantageous deck than from the disadvantageous deck.

Table 1.

Participant Characteristics (N = 132)

Characteristic	M (SD)	Range
Chronological age, mo	97.3 (24.4)	57–143
Gender (boys), n (%)	115 (87.1)
WISC–IV and WPPSI–IV VCI scores	106.4 (18.5)	60–154
ToMTB scores	9.9 (2.8)	3–15
DCCS scores
Single block accuracy, %	88.2 (11.9)	35–100
Reaction time, ms	1,589.6 (751.3)	640.3–6,406.7
Mixed block accuracy, %	82.3 (13.3)	45–100
Reaction time, ms	1,803.4 (791.7)	699.0–5,328.3
Shifting cost, ms	−1,254.0 (2,278.7)	−11,506.0–2,727.0
CGT scores	9.0 (16.7)	−40–50
VABS–Socialization domain scores
Interpersonal Relationships subdomain	19.3 (6.5)	4–34
Play/Leisure Time subdomain	21.5 (7.1)	4–36
Coping Skills subdomain	13.9 (7.8)	0–32
Total domain score	54.3 (19.6)	2–95

Note. CGT = Children’s Gambling Task; DCCS = Dimensional Change Card Sort; ToMTB = Theory of Mind Task Battery; VABS = Vineland Adaptive Behavior Scales; VCI = Verbal Comprehension Index; WISC–IV = Wechsler Intelligence Scale for Children–Fourth Edition; WPPSI–IV = Wechsler Preschool and Primary Scale of Intelligence–Fourth Edition.

The results of the correlational analysis are shown in Table 2. Except for the CGT score, all the other cognitive variables were found to be significantly (p < .01) associated with the Socialization domain of the VABS and its three subdomains. Table 3 shows the results of the hierarchical regression models. The augmented models of the VABS–Interpersonal Relationships were significant, F(2, 129) = 12.85, p < .001. In addition to the VCI, the variable of the ToMTB score significantly predicted children’s interpersonal relationships and additionally explained 3.9% of the variance. For the augmented models of the VABS– Play/Leisure Time, in addition to the VCI, the ToMTB score, DCCS–mixed block accuracy, and DCCS–shifting cost were all significant predictors, F(4, 122) = 15.02, p < .001, additionally explaining 21.1% of the variance in children’s participation in their play and leisure time. For the augmented models of the VABS–Coping Skills, DCCS–mixed block accuracy and DCCS–shifting cost were significant in explaining children’s daily performance of coping skills in socializing while controlling for the VCI, F(3, 123) = 14.81, p < .001, which explained an additional 11.6% of the variance. The augmented models of the VABS–Socialization total were significant, F(3, 123) = 15.25, p < .001. In addition to the VCI, the variables of the DCCS–mixed block accuracy and DCCS–shifting cost were significant in predicting children’s total social interaction, additionally explaining 10.8% of the variance.

Table 2.

Correlations on Measurements of Interest Among Children With Autism Spectrum Disorder (N = 132)

Variable	1	2	3	4	5	6	7	8
1. WISC–IV and WPPSI–IV VCI	—
2. ToMTB	.526**	—
3. DCCS–Single block accuracy	.342**	.394**	—
4. DCCS–Mixed block accuracy	.289**	.275**	.701**	—
5. DCCS–Single block reaction time	−.320**	−.417**	−.499**	−.413**	—
6. DCCS–Mixed block reaction time	−.258**	−.297**	−.235**	−.141	.788**	—
7. DCCS–Shifting cost	.146	.196*	.152	.050	−.409**	−.490**	—
8. CGT	.095	.246**	.224*	.144	−.099	−.073	.119	—
9. VABS–Interpersonal Relationships	.356**	.356**	.284**	.271**	−.220*	−.199*	.230**	.156
10. VABS–Play/Leisure Time	.366**	.375**	.389**	.398**	−.288**	−.258**	.352**	.118
11. VABS–Coping Skills	.402**	.320**	.350**	.387**	−.294**	−.253**	.239**	.098
12. VABS–Socialization total	.419**	.347**	.371**	.362**	−.278**	−.255**	.265**	.124

Note. Correlations between the VABS subdomains are not presented. CGT = Children’s Gambling Task; DCCS = Dimensional Change Card Sort; ToMTB = Theory of Mind Task Battery; VABS = Vineland Adaptive Behavior Scales; VCI = Verbal Comprehension Index; WISC–IV = Wechsler Intelligence Scale for Children–Fourth Edition; WPPSI–IV = Wechsler Preschool and Primary Scale of Intelligence–Fourth Edition.

p < .05.

p < .01.

Table 3.

Hierarchical Regression Models of Daily Social Interaction for Children With Autism Spectrum Disorder (N = 132)

Model and Predictor	β	SE (B)	R ² or ΔR ²	F ^a	p
Models for VABS–Interpersonal Relationships
Baseline
Constant	5.898	3.125	.127	18.860	<.001
VCI	0.126**	0.029
Augmented
Constant	5.160	3.080	.166	12.853	<.001
VCI	0.082*	0.033
ToMTB	0.541*	0.219
Model comparison			.039	6.104	.015
Models for VABS–Play/Leisure Time
Baseline
Constant	7.627	3.421	.119	16.848	<.001
VCI	0.130**	0.032
Augmented
Constant	0.406	3.953	.330	15.015	<.001
VCI	0.046	0.033
ToMTB	0.440*	0.217
DCCS–Mixed block accuracy	15.649**	4.103
DCCS–Shifting cost	0.001**	0.000
Model comparison			.211	12.812	<.001
Models for VABS–Coping Skills
Baseline
Constant	−3.674	3.764	.150	21.975	<.001
VCI	0.164**	0.035
Augmented
Constant	−12.227**	4.617	.265	14.807	<.001
VCI	0.116**	0.034
DCCS–Mixed block accuracy	17.532**	4.738
DCCS–Shifting cost	0.001*	0.000
Model comparison			.116	9.695	<.001
Models for VABS–Socialization Total
Baseline
Constant	8.365	9.396	.163	24.346	<.001
VCI	0.430**	0.087
Augmented
Constant	−9.588	11.572	.271	15.250	<.001
VCI	0.316**	0.086
DCCS–Mixed block accuracy	39.275**	11.874
DCCS–Shifting cost	0.002**	0.001
Model comparison			.108	9.120	<.001

Note. DCCS = Dimensional Change Card Sort; ToMTB = Theory of Mind Task Battery; VABS = Vineland Adaptive Behavior Scales; VCI = Verbal Comprehension Index.

For the baseline model, the degrees of freedom are (1, 130); for the augmented model, they are (2, 129).

p < .05.

p < .01.

Discussion

In this study, we identified the cognitive predictors of social interaction in daily contexts among children with ASD while controlling for verbal comprehension ability. Our results suggest that both ToM and cool EF are significant cognitive predictors of daily social interaction when considering children’s verbal comprehension as an important correlate. However, hot EF seems not to be a significant predictor. This study a gap in previous research and provides a basis for targeting interventions to improve social interaction in daily contexts for children with ASD.

In this study, we used the VABS to measure four aspects of daily social interaction—(1) interpersonal relationships, (2) play and leisure participation, (3) coping skills, and (4) overall socialization—to examine their respective predictors. First, as expected, ToM and verbal comprehension played important roles in interpersonal relationships for children with ASD. The Interpersonal Relationships subdomain of the VABS focuses on emotion expression, mimicking, and friendship abilities (Sparrow et al., 1984). ToM enables a person to choose interpersonal responses that are more likely to lead to successful interactions in social encounters and, over time, to achieve a higher level of social attainment and build interpersonal relationships in daily social contexts. ToM is a powerful cognitive tool for determining whether a person is willing to cooperate and reciprocate or intends to manipulate others’ responses.

In addition, the results of our study show that for children with ASD, interpersonal relationships rely on good verbal comprehension. ToM is about mapping social contexts onto others’ behaviors in terms of their underlying mental states, and verbal comprehension is about mapping sentences to meaning. Cool EF was also correlated with interpersonal relationships among children with ASD. Hot EF, however, is required in rapidly changing contexts to reappraise the importance of the interpersonal relationships. Surprisingly, hot EF was not significantly correlated with interpersonal relationships in this study. This finding could be explained by the fact that the Interpersonal Relationships subdomain of the VABS measures only the frequency of social behaviors (Sparrow et al., 1984), not their quality, and thus is unable to detect children’s motivation-related social interaction in daily life.

Second, in addition to verbal comprehension, both ToM and cool EF are considered to be important for play and leisure participation among children with ASD. Social participation in play and leisure is the collective representation of ToM and EF. In play and leisure contexts, children use their ToM to infer and predict the thoughts and behaviors of their playmates in order to build and sustain the activities, especially pretend play. Despite being an important ability, ToM cannot fully explain the play and leisure performance in this study because play and leisure are an interactive and complex process that requires multiple abilities to accomplish, such as cool EF (Rosenblum et al., 2017) and verbal comprehension (Chen et al., 2019). In measuring cool EF in this study, we assessed not only shifting ability, including accuracy, reaction time, and shifting cost, but also sustained attention and impulse control.

Children with higher accuracy in shifting attention, briefer shifting time, longer sustained attention, and better impulse control may obtain higher scores on the Play/Leisure Time subdomain. Thus, children who perform better on a cool EF task have a better ability to inhibit their impulses and sustain and shift their attention correctly to the focus of their play. This finding is consistent with a previous report of cool EF being needed for better effective play and leisure participation, and possibly being involved in the continuous updating, evaluation, and selection of appropriate responses to verbal and nonverbal information in play and leisure activities (Landa & Goldberg, 2005). Moreover, small to moderate correlations between ToM and cool EF were also found, indicating that cool EF is specifically involved in attributing mental states (i.e., ToM). This finding suggests that cool EF functions in coordinating perspectives in play and leisure activities, which is consistent with results from a previous study (Fizke et al., 2014).

Third, in addition to verbal comprehension, only cool EF was an important predictor of children’s daily social interaction. Although ToM was correlated with coping skills, when compared with cool EF and verbal comprehension, ToM was not an important predictor. In the VABS, the Coping Skills subdomain assesses the extent to which children demonstrate compliance with social norms, such as “follows school rules,” “follows community rules,” and “keeps secrets or confidences” (Sparrow et al., 1984); these behaviors are more closely related to children’s cool EF (i.e., attention and inhibition) than to ToM.

As for the overall Socialization domain, in addition to verbal comprehension, only cool EF was correlated with predicting children’s daily social interaction. The reason could be that the social items in the VABS focus more on general social interaction (e.g., returning objects to others, proper table manners; Sparrow et al., 1984) and less on ToM-related social interaction, especially social performance involving advanced ToM. Therefore, the correlates for overall socialization are similar to those for coping skills.

Two limitations may concern readers. First, the findings cannot be generalized to children with ASD with poor verbal comprehension because the children included in this study had average verbal comprehension. Second, the VABS does not contain items addressing emotion- or motivation-based social behaviors to measure children’s social interaction involving hot EF. Therefore, future studies should recruit children with ASD with poor verbal comprehension and use social interaction assessments that also measure hot EF–related social behaviors to further explore the predictors of social interaction in daily contexts among children with ASD.

Implications for Occupational Therapy Practice

In this study, we identified cognitive predictors of social interaction in daily contexts among children with ASD. This knowledge can guide evaluation and intervention by occupational therapy practitioners seeking to improve the daily performance of social interaction among children with ASD. Among the predictors, children’s verbal comprehension has commonly been identified and considered in previous clinical work and research. However, this study found that ToM and cool EF, which have been considered less often, are also important considerations. The following implications for occupational therapy practice should also be considered:

In addition to verbal comprehension, ToM and cool EF should be assessed for more comprehensive evaluation and intervention planning to improve social interaction in daily contexts for children with ASD.

Contrary to our hypothesis, hot EF was not a significant predictor of daily social interaction; future studies on the relation of hot EF to social interaction in daily contexts are still warranted.

This behavioral study provides evidence of important cognitive correlates of social interaction in daily contexts, providing a foundation for intervention protocols to improve social interaction in children with ASD.

Conclusion

The results of this study show that ToM and cool EF are cognitive predictors of social interaction in daily contexts in children with ASD whose verbal comprehension is average or above average. By conducting this behavioral study, we have filled a gap in the research by enhancing the understanding of important cognitive correlates of daily social interaction in children with ASD to improve evaluation and intervention planning for this population.

Footnotes

Acknowledgments

We thank all the caregivers and children for participating in this study. This research was supported by Ministry of Science and Technology Grants 103-2410-H-006-060, 104-2410-H-006-065, 105-2410-H-006-047-MY2, and 107-2314-B-006-023-MY3 awarded to Kuan-Lin Chen.

References

Adolphs

(2001). The neurobiology of social cognition. Current Opinion in Neurobiology , 11 , 231–239. https://doi.org/10.1016/S0959-4388(00)00202-6

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.).

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.).

Baron-Cohen

(2000). Theory of mind and autism: A fifteen year review. In Baron-Cohen

Tager-Flusberg

, & Cohen

D. J.

(Eds.), Understanding other minds: Perspectives from developmental cognitive neuroscience (2nd ed., pp. 3–20). Oxford University Press.

Butler

C. W.

(2016). Talk and social interaction in the playground. Routledge. https://doi.org/10.4324/9781315241876

Charman

, & Baird

(2002). Practitioner review: Diagnosis of autism spectrum disorder in 2- and 3-year-old children. Journal of Child Psychology and Psychiatry, and Allied Disciplines , 43 , 289–305. https://doi.org/10.1111/1469-7610.00022

Chen

H. Y.

, & Chen

Y. H.

(2013). Wechsler Preschool and Primary Scale of Intelligence–Fourth Edition: Chinese version manual. Chinese Behavioral Science Corporation.

Chen

K.-L.

Chen

C.-T.

Lin

C.-H.

Huang

C.-Y.

, & Lee

Y.-C.

(2019). Prediction of playfulness by pretend play, severity of autism behaviors, and verbal comprehension in children with autism spectrum disorder. Neuropsychiatric Disease and Treatment , 15 , 3177–3186. https://doi.org/10.2147/NDT.S223681

Chen

Y. H.

, & Chen

H. Y.

(2007). Wechsler Intelligence Scale for Children–Fourth Edition: Chinese version manual. Chinese Behavioral Science Corporation.

10.

Christensen

D. L.

Braun

K. V. N.

Baio

Bilder

Charles

Constantino

J. N.

, . . . Yeargin-Allsopp

(2018). Prevalence and characteristics of autism spectrum disorder among children aged 8 years—Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2012. Morbidity and Mortality Weekly Report: Surveillance Summaries , 65 , 1–23. https://doi.org/10.15585/mmwr.ss6513a1

11.

Dichter

G. S.

Radonovich

K. J.

Turner-Brown

L. M.

Lam

K. S.

Holtzclaw

T. N.

, & Bodfish

J. W.

(2010). Performance of children with autism spectrum disorders on the Dimension-Change Card Sort Task. Journal of Autism and Developmental Disorders , 40 , 448–456. https://doi.org/10.1007/s10803-009-0886-1

12.

Eisenberg

Spinrad

T. L.

, & Knafo‐Noam

(2015). Prosocial development. In Lerner

R. M.

(Ed.), Handbook of child psychology and developmental science (7th ed., pp. 1–47). Wiley. https://doi.org/10.1002/9781118963418.childpsy315

13.

Faul

Erdfelder

Lang

A.-G.

, & Buchner

(2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods , 39 , 175–191. https://doi.org/10.3758/BF03193146

14.

Fizke

Barthel

Peters

, & Rakoczy

(2014). Executive function plays a role in coordinating different perspectives, particularly when one’s own perspective is involved. Cognition , 130 , 315–334. https://doi.org/10.1016/j.cognition.2013.11.017

15.

Frith

C. D.

Friston

Liddle

P. F.

, & Frackowiak

R. S.

(1991). Willed action and the prefrontal cortex in man: A study with PET. Proceedings: Biological Sciences , 244 , 241–246. https://doi.org/10.1098/rspb.1991.0077

16.

Happaney

Zelazo

P. D.

, & Stuss

D. T.

(2004). Development of orbitofrontal function: Current themes and future directions. Brain and Cognition , 55 , 1–10. https://doi.org/10.1016/j.bandc.2004.01.001

17.

Hutchins

T. L.

Prelock

P. A.

, & Bonazinga

(2012). Psychometric evaluation of the Theory of Mind Inventory (ToMI): A study of typically developing children and children with autism spectrum disorder. Journal of Autism and Developmental Disorders , 42 , 327–341. https://doi.org/10.1007/s10803-011-1244-7

18.

Hutchins

T. L.

Prelock

P. A.

, & Bouyea

L. B.

(2014). Technical manual for the Theory of Mind Inventory–2 (13th ed.). University of Vermont.

19.

Kerr

, & Zelazo

P. D.

(2004). Development of “hot” executive function: The Children’s Gambling Task. Brain and Cognition , 55 , 148–157. https://doi.org/10.1016/S0278-2626(03)00275-6

20.

Krain

A. L.

Wilson

A. M.

Arbuckle

Castellanos

F. X.

, & Milham

M. P.

(2006). Distinct neural mechanisms of risk and ambiguity: A meta-analysis of decision-making. NeuroImage , 32 , 477–484. https://doi.org/10.1016/j.neuroimage.2006.02.047

21.

Lai

S.-S.

Tsai

C.-H.

C.-C.

Chen

C.-T.

H.-J.

, & Chen

K.-L.

(2020). Identifying the cognitive correlates of reciprocity in children with autism spectrum disorder. Journal of Autism and Developmental Disorders , 50 , 2007–2018. https://doi.org/10.1007/s10803-019-03957-7

22.

Landa

R. J.

, & Goldberg

M. C.

(2005). Language, social, and executive functions in high functioning autism: A continuum of performance. Journal of Autism and Developmental Disorders , 35 , 557–573. https://doi.org/10.1007/s10803-005-0001-1

23.

Lezak

M. D.

Howieson

D. B.

Loring

D. W.

, & Fischer

J. S.

(2004). Neuropsychological assessment (4th ed.). Oxford University Press.

24.

Malecki

C. K.

, & Elliot

S. N.

(2002). Children’s social behaviors as predictors of academic achievement: A longitudinal analysis. School Psychology Quarterly , 17 , 1–23. https://doi.org/10.1521/scpq.17.1.1.19902

25.

Mendelson

J. L.

Gates

J. A.

, & Lerner

M. D.

(2016). Friendship in school-age boys with autism spectrum disorders: A meta-analytic summary and developmental, process-based model. Psychological Bulletin , 142 , 601–622. https://doi.org/10.1037/bul0000041

26.

Nakamoto

, & Schwartz

(2010). Is peer victimization associated with academic achievement? A meta-analytic review. Social Development , 19 , 221–242. https://doi.org/10.1111/j.1467-9507. 2009.00539.x

27.

Parker

J. G.

, & Asher

S. R.

(1987). Peer relations and later personal adjustment: Are low-accepted children at risk? Psychological Bulletin , 102 , 357–389. https://doi.org/10.1037/0033-2909.102.3.357

28.

Rolls

E. T.

(2004). The functions of the orbitofrontal cortex. Brain and Cognition , 55 , 11–29. https://doi.org/10.1016/S0278-2626(03)00277-X

29.

Roselló

Berenguer

Navío

Baixauli

, & Miranda

(2017). Executive functioning, social cognition, pragmatics, and social interaction in attention deficit hyperactivity disorder and autism spectrum disorder. Current Developmental Disorders Reports , 4 , 72–77. https://doi.org/10.1007/s40474-017-0114-1

30.

Rosenblum

Waissman

, & Diamond

G. W.

(2017). Identifying play characteristics of pre-school children with developmental coordination disorder via parental questionnaires. Human Movement Science , 53 , 5–15. https://doi.org/10.1016/j.humov.2016.11.003

31.

Sakaiya

Shiraito

Kato

Ide

Okada

Takano

, & Kansaku

(2013). Neural correlate of human reciprocity in social interactions. Frontiers in Neuroscience , 7 , 239. https://doi.org/10.3389/fnins.2013.00239

32.

Sally

, & Hill

(2006). The development of interpersonal strategy: Autism, theory-of-mind, cooperation and fairness. Journal of Economic Psychology , 27 , 73–97. https://doi.org/10.1016/j.joep.2005.06.015

33.

Schneider

Myin

, & Myin-Germeys

(2020). Is theory of mind a prerequisite for social interactions? A study in psychotic disorder. Psychological Medicine , 50 , 754–760. https://doi.org/10.1017/S0033291719000540

34.

Sparrow

S. S.

Balla

D. A.

Cicchetti

D. V.

Harrison

P. L.

, & Doll

E. A.

(1984). Vineland Adaptive Behavior Scales. American Guidance Service.

35.

Tsermentseli

, & Poland

(2016). Cool versus hot executive function: A new approach to executive function. Encephalos , 53(1), 11–14. http://www.encephalos.gr/pdf/53-1-02e.pdf

36.

Wellman

H. M.

(1990). The child’s theory of mind. MIT Press.

37.

Wong

Odom

S. L.

Hume

K. A.

Cox

A. W.

Fettig

Kucharczyk

, . . . Schultz

T. R.

(2015). Evidence-based practices for children, youth, and young adults with autism spectrum disorder: A comprehensive review. Journal of Autism and Developmental Disorders , 45 , 1951–1966. https://doi.org/10.1007/s10803-014-2351-z

38.

W. T.

Chang

C. F.

T. H.

, & Chiu

S. C.

(2004). Vineland Adaptive Behavior Scales: Chinese version manual. Chinese Behavioral Science Corporation.

39.

Y.-T.

H.-J.

Tsai

C.-H.

Lin

C.-H.

Lai

S.-S.

, & Chen

K.-L.

(2021). Cool executive function and verbal comprehension mediate the relation of hot executive function and theory of mind in children with autism spectrum disorder. Autism Research , 14 , 921–931. https://doi.org/10.1002/aur.2412

40.

Zelazo

P. D.

(2006). The Dimensional Change Card Sort (DCCS): A method of assessing executive function in children. Nature Protocols , 1 , 297–301. https://doi.org/10.1038/nprot.2006.46