RE S EARCH ART I C L E Bilingualism and second-order theory of mind development in autistic children over time: Longitudinal relations with language, executive functions, and intelligence Eleni Peristeri1 | Margreet Vogelzang2 | Ianthi Maria Tsimpli3 | Stephanie Durrleman4 1Department of Theoretical and Applied Linguistics, School of English, Aristotle University of Thessaloniki, Thessaloniki, Greece 2School of Psychology, Newcastle University, Newcastle upon Tyne, UK 3Department of Theoretical and Applied Linguistics, Faculty of Modern & Medieval Languages & Linguistics, University of Cambridge, Cambridge, UK 4Department of Science and Medicine, University of Fribourg, Fribourg, Switzerland Correspondence Eleni Peristeri, Department of Theoretical and Applied Linguistics, School of English, Aristotle University of Thessaloniki, Thessaloniki, Greece. Email: eperiste@enl.auth.gr Funding information Swiss National Science Foundation, Grant/Award Number: PR00P1_193104/1 Abstract Theory of Mind has long been studied as a core weakness in autism spectrum disorder due to its relationship with social reciprocity, while bilingualism has been shown to compensate for autistic individuals’ mentalizing weaknesses. However, our knowledge of the Theory of Mind developmental trajectories of bilingual and monolingual autistic children, as well as of the factors related to Theory of Mind development in autism spectrum disorder is still limited. The current study has examined first- and second-order Theory of Mind skills in 21 monolingual and 21 bilingual autistic children longitudinally across three time points, specifically at ages 6, 9, and 12, and also investigated associations between Theory of Mind trajectories and trajectories of the children’s lan- guage, intelligence and executive function skills. The results reveal that bilin- gual autistic children outperformed their monolingual peers in second-order Theory of Mind at ages 9 and 12, and that intelligence and, especially, expres- sive vocabulary skills played a pivotal role in advancing bilingual autistic chil- dren’s second-order Theory of Mind development. On the other hand, monolingual autistic children only managed to capitalize on their language and intelligence resources at age 12. The findings highlight the importance of investigating bilingualism effects on autistic children’s advanced cognitive abilities longitudinally. Lay Summary The current study offers evidence on the developmental trajectories of both first- and second-order Theory of Mind skills in 21 bilingual and 21 monolingual autis- tic children that were initially tested at age 6, and were then followed up at ages 9 and 12. The same children were also administered language, intelligence, and executive function tests across the three time points. The results show that bilin- gualism improved autistic children’s second-order Theory of Mind skills after age 9, while this boosting effect was driven by the children’s expressive vocabulary and intelligence skills. The findings emphasize the beneficial effects of bilingual- ism on autistic children’s advanced Theory of Mind, and also highlight the impor- tance of language and intelligence underlying this effect. KEYWORDS autism spectrum disorder, bilingualism, executive functions, expressive vocabulary, intelligence, longitudinal design, second-order theory of mind Received: 6 April 2024 Accepted: 1 August 2024 DOI: 10.1002/aur.3214 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2024 The Author(s). Autism Research published by International Society for Autism Research and Wiley Periodicals LLC. Autism Research. 2024;1–12. wileyonlinelibrary.com/journal/aur 1 https://orcid.org/0000-0002-1151-677X mailto:eperiste@enl.auth.gr http://creativecommons.org/licenses/by/4.0/ http://wileyonlinelibrary.com/journal/aur INTRODUCTION Autistic children often display difficulties in Theory of Mind (ToM) (Baron-Cohen et al., 1997; Gernsbacher & Yergeau, 2019; Tager-Flusberg, 2007), which is the understanding of mental states such as emotions, desires and beliefs, and the ability to predict ensuing behaviors. A functional ToM is fundamental for successful social interactions, explaining why difficulties in this realm con- tribute to various social challenges attested in autistic individuals (Bauminger et al., 2010; Durrleman, Peri- steri, & Tsimpli, 2022; Durrleman, Tsimpli, & Peristeri, 2022; Peristeri et al., 2021b, 2023; Rosello et al., 2020). ToM difficulties of autistic children can be alleviated by better vocabulary (Milligan et al., 2007), executive func- tions (EF) (Joseph & Tager–Flusberg, 2004), IQ, (Happé, 1994a) and the linguistic experience of bilingual- ism (Peristeri, Baldimtsi, et al., 2021). Research on bilin- gualism effects on ToM in autistic individuals so far has focused on cross-sectional comparisons of first-order ToM between autistic children and typically-developing (TD) peers (e.g., Jones et al., 2018; Mazza et al., 2017; Peristeri et al., 2024; Peterson et al., 2016). First-order ToM is the ability to grasp that someone can have a belief about the state of the world, while second-order ToM is the ability to grasp that someone can have a belief about another person’s belief about the state of the world. The current longitudinal study aims to investigate for the first time the trajectory of second-order ToM development in monolingual and bilingual autistic chil- dren, and also explore linguistic and cognitive factors that influence ToM development. ToM development passes through different phases (Marinis et al., 2023; Peterson & Wellman, 2019; Wellman, 2017; Wellman et al., 2001). At about age 3–4, TD children start to become aware that others may not share their intentions and beliefs, while between ages 4 and 6, they start to understand that someone can hold a belief that is different from reality, and that their acts may be guided by false beliefs. Similarly, Piaget (1947) notices that the first evidence of perspective-taking in children is antici- pated around the 6th year of age (i.e., at the onset of the concrete operational stage), and that around 9–10 years (i.e., just before the onset of the formal operational stage) the child is fully aware of possessing a viewpoint that may be distinct from that of other individuals, and is able to coordinate multiple points of view. First-order ToM skills have been usually measured through the use of mental state terms in narrative production (e.g., Losh & Capps, 2003; Peristeri et al., 2020; Siller et al., 2014) and/or inferential questions in narrative comprehension (e.g., Happé, 1994b; Nuske & Bavin, 2015), which have both been particularly challenging for autistic children. First-order ToM has also been tested via short, simple stories that are illustrated by means of puppets and often end with a binary choice, such as the classical Sally-Ann (change of location) task by Baron-Cohen et al. (1985). While children can be successful at first-order ToM tasks around the ages of 4–6 years, they only succeed on second-order ToM tasks after age 7 (Astington et al., 2002; Baron-Cohen, 2001; Duval et al., 2011). Such tasks involve longer stories with richer vocabulary, multi- ple protagonists, and complex questions about the char- acters’ belief about another character’s belief. As an illustration, consider the Faux Pas test (Baron-Cohen et al., 2001; Happé, 1994b). In this test, children learn that a class took part in a story competition that Emma really wanted to win. Whilst Emma was away from school, the results of the competition were announced: Alice was the winner. The next day, when Alice saw Emma, she said “I’m sorry about your story” to which Emma asks: “What do you mean?”. Alice then replies: “Oh nothing.” The child is then asked a series of ques- tions to test their memory and their understanding of second-order belief reasoning: Who won the story compe- tition? Did someone say something they should not have said? Did Alice realize that Emma had not heard the results of the competition? Various factors have been found to improve chil- dren’s performance on ToM tasks. For instance, those with a larger vocabulary as measured by general vocabu- lary or specific mental state vocabulary (e.g., think, believe) are more likely to succeed in ToM tasks, mainly because reasoning through these tasks involves an inter- nal monologue that enables the individual to represent others’ perspectives. Richer vocabulary also allows TD (e.g., Farrar et al., 2017; Tompkins et al., 2019) and autistic children (e.g., Buac & Kaushanskaya, 2020) to reap more from surrounding verbal interactions, which in turn enables them to understand that people have minds which may differ from one another’s and their own. As such, both a rich general vocabulary and a specific reper- toire of mental state terms can increase one’s perfor- mance on ToM tasks. Another factor that has been found to enhance per- formance in ToM tasks is EF (e.g., Jones et al., 2018; Wade et al., 2018). Reasoning though a ToM task requires paying attention to a narrative, and thus holding in working memory its events and the protagonists’ beliefs regarding one and the same situation, inhibiting one in order to reply from the perspective of another. For instance, in the Sally-Ann task, a child needs inhibitory control to suppress her/his response to the actual location of the object. Standard false belief tasks can thus pose a challenge to young children whose EF are not yet fully developed. Working memory is also implicated in false belief attribution tasks, since children need to keep track of changing locations or contents in the false belief stories in order to be able to respond correctly (Wellman & Bartsch, 1994). Yet, another factor that has been found to influence performance in ToM tasks is general intelligence, whether measured nonverbally via performance IQ (PIQ), or verbally, that is, though verbal IQ (VIQ). IQ as 2 PERISTERI ET AL. a proxy for global cognitive skills has been shown to be a boost in various tasks including ToM in both autistic (Buitelaar et al., 1999; Happé, 1994a; Peristeri et al., 2022; Yirmiya et al., 1998) and non-autistic popu- lations (e.g., Baker & Tenenbaum, 2014; Charlton et al., 2009). This could stem from the fact that individ- uals with higher IQ tend to have stronger general cogni- tive resources, including attention and memory, as well as increased social knowledge (Jones & Day, 1997; Schweizer & Moosbrugger, 2004). This arguably allows for a more efficient integration of the information included in ToM tasks. The contribution of intelligence to ToM understanding in autistic individuals is particu- larly important, as intellectual ability represents a pri- mary aspect of heterogeneity within the autistic phenotype and is a strong predictor of outcomes in autis- tic individuals (Gillespie-Lynch et al., 2012; Munson et al., 2008; Peristeri & Andreou, 2024). As such, longitu- dinal changes in the IQ performances of autistic children may predict changes in their ToM development. Growing up bilingual has also been associated with enhanced ToM performance in children (Goetz, 2003; Nguyen & Astington, 2014; Yu et al., 2021). Bilinguals need to constantly select the appropriate language for their interlocutor or context; the ability to understand that other individuals have different linguistic knowledge may lead to a better understanding that others have other perspectives their theirs. Put differently, considering others’ perspectives–operationalized in terms of language switching–may in turn boost bilingual children’s ToM. Nowadays, there is increasing evidence in favor of enhanced ToM performance in bilingual TD children as compared to their monolingual peers (see Schroeder, 2018 for a meta-analysis). In recent work, ToM advantages have been also reported for bilingual autistic children as compared to their monolingual peers, however, relevant evidence has been drawn from first- order ToM tasks and cross-sectional studies only (Andreou et al., 2020; Durrleman, Tsimpli, & Peristeri, 2022; Durrleman, Peristeri, & Tsimpli, 2022; Peristeri et al., 2021a). ToM benefits for bilingual autistic children have been mainly attributed to their stronger EF skills as compared to monolinguals, since switching between languages presupposes the inhibition of one language in favor of another (Bialystok, 2011; Poulin- Dubois et al., 2011).Though previous research with bilin- gual autistic children has revealed relations between bilingualism and EF skills (Gonzalez-Barrero & Nadig, 2019; Montgomery et al., 2022; Peristeri et al., 2020; Romero & Uddin, 2021), and between EF and ToM (specifically, first-order false belief attribution) (Andreou et al., 2020; Peristeri et al., 2021a), the question whether bilingualism has a beneficial effect on autistic children’s second-order ToM skills is still unanswered. Another gap in our knowledge of bilingualism effects on ToM in autistic children regards the way the relations between ToM, language, IQ and EF unfold over time, since most work in this field has been cross-sectional. It thus remains to be determined if benefits of bilingualism on the social cognition of autistic children, and the fac- tors underlying this relation, might arise at specific time points over children’s development. The aim of the current study is to investigate the developmental phases of second-order ToM in autistic bilinguals and monolinguals across three time points, spe- cifically at ages 6, 9, and 12 years of age. The rationale for the three time points stems from the fact that according to previous research (Astington et al., 2002; Baron-Cohen, 2001; Duval et al., 2011; Piaget, 1947), coordination of multiple viewpoints as a proxy for second-order ToM is not possible at age 6 (i.e., the cur- rent study’s first time point), but begins to fully develop around 9–10 years and beyond, which coincides with time points 2 and 3. The current study also aims to eluci- date the relations between ToM and expressive vocabu- lary, IQ, and EF that have been shown to influence ToM development in children. We seek then to address the fol- lowing two research questions: 1. Do bilingual autistic children outperform monolin- gual peers on second-order ToM at certain phases and, if so, when? 2. How do expressive vocabulary, IQ, and working memory influence ToM development in monolingual and bilingual autistic children? METHOD Participants The current longitudinal study included 21 bilingual autistic children (henceforth, ASDbi; 16 males), and 21 age-, socioeconomic status (SES)-, and VIQ-matched monolingual autistic children (henceforth, ASDmono; 15 males). The children were recruited from the geo- graphical region of central Greece, and were referred by Centers for Differential Diagnosis, Assessment, Counsel- ing, and Evaluation (KEDASY) that constitute the offi- cial state centers responsible for the diagnosis and assessment of autism in Greece. All children received a formal clinical diagnosis of autism at preschool age at KEDASY on the basis of the DSM-V, and ICD-10 cri- teria (American Psychiatric Association, 2013; World Health Organization, 1993), a record review conducted by teams with diverse expertise (psychiatrist, clinical psy- chologist, specialized educator, social worker, speech lan- guage pathologist), as well as the Autism Diagnostic Interview-Revised (ADI-R; Lord et al., 1994). SES was measured in terms of the number of years that the chil- dren’s mothers had spent in education. The ASDmono group included Greek-speaking children, while the ASDbi group was made up of Albanian-Greek (�dominan children that were born to mixed marriages, PERISTERI ET AL. 3 and attended Greek mainstream schools. Both cohorts were followed for three time points, specifically, at ages 6, 9, and 12. Alongside the critical measures of second- order ToM, a series of background measures assessing expressive vocabulary, VIQ, and PIQ (Wechsler Intelli- gence Scale for Children (WISC-III); Wechsler, 1991; adapted to and standardized in Greek by Georgas et al., 1997), working memory, and first-order ToM were administered to the children at all three time points. Only children who scored a minimum PIQ score of 70 at time point 1 were included in the sample, as children with lower PIQ scores may have had cognitive difficulties which could have prevented participation in the tasks proposed (Aylward, 2002; Rommelse et al., 2015). All study procedures were conducted according to the Hel- sinki declaration and were approved by the Aristotle University of Thessaloniki Institutional review board (IRB) (IRB protocol number: 39928). Procedure At all three time points, the 42 ASDmono and ASDbi children completed the same tasks, including an expres- sive vocabulary test (Vogindroukas et al., 2009; adapta- tion from Renfrew, 1997), the WISC-III (Wechsler, 1991; adapted to and standardized in Greek by Georgas et al., 1997), a 2-back task, two first-order ToM tests, that is, the Sally-Ann (Baron-Cohen et al., 1985) and the Smarties test (Perner et al., 1989), and two second-order ToM tests (Astington et al., 2002). The tests were admin- istered in two sessions that took place on different days at the children’s home or school. Session 1 included the expressive vocabulary, the WISC-III, and the 2-back task, and session 2 included the first- and second-order ToM tests. All the tasks were administered in Greek. Materials Expressive vocabulary test (Vogindroukas et al., 2009; adaptation from Renfrew, 1997). Children’s expressive vocabulary in Modern Greek was assessed with an expressive vocabulary test, which has been standardized for 3- to 10-year-old Greek-speaking monolingual chil- dren. It includes 50 black-and-white pictures of common objects that each child was asked to name individually. Each correct answer earned 1 point, with a maximum score of 50. The test was terminated in case the partici- pant failed to respond correctly to five consecutive trials. Intelligence Children’s VIQ and PIQ (standardized) scores were mea- sured through the Greek version of the WISC-III (Wechsler, 1991; adapted to and standardized in Greek by Georgas et al., 1997). Executive function task 2-back test Children’s working memory skills were measured using a variation of the classic N-back task (Smith & Jonides, 1999). In this version, children viewed a sequence of digits on a computer screen and were instructed to press a certain button (specifically, the ‘J’ key on the keyboard), when the number on the screen was the same as the one that appeared two trials (2-back) before, and not press any button if the item was different. Though the n-back task is regarded as a standard “execu- tive” WM measure, it requires a cascade of cognitive pro- cesses, including inhibition and updating, that have been implicated in ToM development (e.g., Jones et al., 2018; Wade et al., 2018; Wellman & Bartsch, 1994). Children completed 20 practice trials before completing the actual task. Each digit was presented for 500 ms with an inter- stimulus interval (ISI) of 2500 ms. Across 60 trials, 20 were the “correct hit” trials. Errors included children’s target misses and nontarget false hits (see the Appendix A of the supporting information). Accuracy scores (i.e., composite score (%) of the number of corrects hits minus false hits) indexed children’s working memory skills. Theory of mind tasks First-order Theory of Mind tests Children’s first-order ToM skills were measured through the Sally-Ann task and the Smarties test, which were administered in Greek. The Sally-Ann test (adapted from “Sally-Ann Test”– Baron-Cohen et al., 1985) was acted out by the first author with two dolls, called Maria and Irene, a basket, a box and a ball, while commenting: “Maria is putting her ball in the basket and briefly leaves the room, during which time Irene enters and moves the ball to the box. Irene then leaves the room, and Maria comes back.” Then the child was asked the following questions: a. “Where did Maria put the ball?”; “Where is the ball now?” (reality control questions) b. “Where does Maria believe that the ball is?” (first- order false belief question); “Why?” (justification question) c. “When Maria returns to the room, where will she look for her ball first?” (test question); “Why?” (justifica- tion question) 4 PERISTERI ET AL. In the Smarties task (adapted from Perner et al., 1989), children were shown a Smarties package that contained a pencil, rather than the expected candy, and then the child was asked what her/his best friend would think is in the package without having looked inside the package. The following questions were asked: a. “What do you think that this package has inside?” (reality control question). Let’s see what is inside. Look! There is a pencil inside the package! b. “What did you think that the package contained before I opened it?” OR “Your best friend now enters the room. What would s/he say about the content of the package before we open it?” (first-order false belief question); “Why?” (justification question). Correctly responding to both tests’ reality control questions was an inclusionary criterion. A child was awarded a score of 2 if s/he made no errors in the reality control questions, while correct responses in the first- order false belief and test questions were scored with 1 point each. The maximum accuracy score was 4 for each test. Second-order Theory of Mind test The test consists of two stories (adapted from Astington et al., 2002) which were read aloud by the examiner. In each story, children were asked to predict what one char- acter would think another character would do or say, when the first character held a false belief about the other character’s belief. For example, in the first story a girl moved an object while a boy was out of the room and she was not aware that the boy saw her putting the object into the new location as he was coming back into the room. Children were asked three control questions to ensure that they understood the story, and a first-order question to ensure that they understood the girl’s false belief about the boy’s knowledge state. They were then asked to predict where the girl thought the boy would look for the object. The two stories were administered to all children. In each story, there were two control ques- tions, one first-order ToM question, one second-order ToM question, and finally a justification question. Each question was counted as 1 point each. The maximum accuracy score for both stories was 10 points. Analysis plan All data analyses were performed in the statistical com- puting software R (version 3.6.2; R Core Team, 2019). The ASDmono group was coded as the reference level using deviation coding (�0.5/0. for all analyses. First, independent/unpaired t-tests were used to compare the two groups’ SES (only at time point 1), VIQ and PIQ scores, and first-order ToM (at all three time points) to investigate whether these background measures were matched between the two groups (see Table 1). The two groups were matched on SES as well as on VIQ and age at all three time points, and at the first time point (i.e., age 6) they were also matched on PIQ. Note that the groups did differ on PIQ later on, namely at ages 9 and 12; these differences were taken into account in the ana- lyses. The two groups were also matched on scores (%) from the two first-order ToM tests, namely the Sally-Ann (Baron-Cohen et al., 1985) and the Smarties test (Perner et al., 1989), although monolingual children showed numerically better performance. In the second step of the analysis, we investigated whether the two groups performed similarly in second- order ToM, and which factors influenced their perfor- mance over time. First, we compared performance on expressive vocabulary and EF between the groups at each time point with independent t-tests. Then, we ran linear growth curve analyses to investigate the relations between expressive vocabulary, EF, and PIQ, on the one hand, and ToM development, on the other. These mixed- effects models included participant as a random inter- cept, as well as the slope for time point over participant. First- and second-order ToM were the dependent vari- ables. We started with a full model including group and time point as the independent factors, and the covariates of expressive vocabulary, EF, and PIQ, as well as the interactions between group, time point, and the three covariates. The models were then simplified by testing whether inclusion of the covariates was warranted based on the models’ AIC score. Finally, independent t-tests for second-order ToM performance between the groups at each time point and Pearson correlations between the covariates and second-order ToM were run to confirm the interactions found in the growth curve model. Relative effect sizes are reported throughout, calcu- lated using Cohen’s d (“effsize” package in R for the t- test comparisons; Torchiano, 2019; “EMAtools” package in R for the growth curve analyses; Kleiman, 2021), with the interpretation of jdj < 0.2 as “negligible”, jdj < 0.5 as “small”, jdj < 0.8 as “medium”, otherwise “large.” RESULTS The ASDmono group consistently outperformed the ASDbi group on the expressive vocabulary measure, with large effect sizes. In contrast, the ASDbi group consis- tently outperformed the ASDmono group on the EF measure, with large effect sizes. Results of independent t- tests between the groups are provided in Table 2. Performance on first- and second-order ToM are shown in Figure 1. The full results of the growth curve models can be found in the Appendix B of the Supporting Infoarmation. For first-order ToM, children’s perfor- mance improved significantly with age (estimate = 13.49; PERISTERI ET AL. 5 t = 3.32; p = 0.001; d = �0.5 (medium)). There was also a significant effect of group (estimate = �43.60 t = �2.16 p = 0.03; d = 0.83 (large)), which was due to the fact that, collapsing over time points, the ASDmono outperformed the ASDbi group. However, the t-tests across the separate time points did not reach significance. Though the best model included the effect of EF, this effect did not reach significance. TABLE 1 Background measures across the ASDmono and ASDbi groups (means, with SDs in brackets). Time point 1 Time point 2 Time point 3 ASDmono ASDbi ASDmono ASDbi ASDmono ASDbi Age (years; months) 6;3 (0.3) 6;4 (0.3) 9;2 (0.2) 9;3 (0.2) 12;4 (0.3) 12;3 (0.3) t = �1.00 df = 40; p = 0.32 t = �2.66 df = 40; p = 0.79 t = 0.38; df = 40; p = 0.70 First-order ToM (%) 46.4 (24.1) 36.9 (23.9) 61.9 (22.2) 54.8 (20.7) 79.8 (19.2) 76.2 (22.7) t = 1.29; df = 40; p = 0.21; d = �0.4 t = 1.08; df = 40; p = 0.28; d = �0.3 t = 0.55; df = 39; p = 0.58; d = �0.1 PIQ 90.7 (10.2) 97.6 (14.4) 88.6 (15.6) 103.9 (16.1) 83.4 (16.7) 109.6 (10.2) t = �1.78 df = 36; p = 0.08; d = 0.55 t = �3.13 df = 40; p = 0.003; d = 0.96 t = �6.11 df = 33; p < 0.001; d = 1.89 VIQ 90.0 (15.3) 92.2 (18.7) 91.0 (15.7) 92.4 (17.2) 89.8 (11.5) 89.4 (13.3) t = �0.42 df = 39; p = 0.68; d = 0.13 t = �0.28 df = 40; p = 0.78; d = 0.09 t = 0.09; df = 39; p = 0.93; d = �0.0 SES (years of education) 9.3 (2.9) 9.8 (2.4) t = �0.57 df = 39; p = 0.57; d = 0.18 Note: The Table reports results from independent t-tests between the groups and Cohen’s d measures of effect size. Abbreviations: ASDbi, bilingual autistic children; ASDmono, monolingual autistic children; PIQ, Performance IQ; SES, socioeconomic status; VIQ, Verbal IQ. TABLE 2 Groups’ performances on expressive vocabulary and the 2-back task (means, with SDs in brackets). Time point 1 Time point 2 Time point 3 ASDmono ASDbi ASDmono ASDbi ASDmono ASDbi Expressive vocabulary (max. 50) 29.8 (3.3) 24.2 (3.6) 37.5 (4.7) 29.9 (4.1) 42.2 (4.6) 34.8 (4.4) t = 5.23; df = 40; p < 0.001; d = �1.6 t = 5.55; df = 39; p < 0.001; d = �1.7 t = 5.40; df = 40; p < 0.001; d = �1.6 2-back 16.3 (3.6) 25.6 (11.2) 16.6 (9.1) 26.9 (12.6) 19.2 (10.6) 30.9 (10.2) t = �3.60 df = 24; p = 0.001; d = 1.11 t = �3.03 df = 36; p = 0.005; d = 0.93 t = �18.14 df = 40; p < 0.001; d = 1.12 Abbreviations: ASDbi, bilingual autistic children; ASDmono, monolingual autistic children. F I GURE 1 Groups’ performance on the first- and second-order ToM over time points. Error bars indicate the standard error. ASDbi, bilingual autistic children; ASDmono, monolingual autistic children. 6 PERISTERI ET AL. Regarding second-order ToM, the growth curve model without any covariates showed a significant posi- tive effect of time point (estimate = 18.45; t = 10.05; p < 0.001; d = 1.82 (large)). There was no simple effect of group, but there were significant interactions between group and expressive vocabulary (estimate = 8.36; t = 3.68; p < 0.001; d = 0.84 (large)), as well as between group, time point, and expressive vocabulary (estimate = �2.82 t = �3.20 p = 0.001; d = �0.6 (medium)), which were due to the fact that correlations between expressive vocabulary score and ToM perfor- mance were stronger for the ASDbi as compared to the ASDmono group (see Figure 2). The correlational ana- lyses run at each time point separately also indicate that the ASDbi children showed a stronger correlation between expressive vocabulary scores and second-order ToM at time point 2, whereas at time points 1 and 3 the two groups showed similarly strong correlations (see Table 3). Regarding PIQ effects on second-order ToM, there was a significant effect of PIQ (estimate = �1.62 t = �2.60 p = 0.01; d = �0.5 (medium)) as well as an interaction between PIQ and time point (estimate = 0.73; t = 2.96; p = 0.004; d = 0.59 (medium)), which indicates a positive effect of PIQ on second-order ToM scores at later time points (see Figure 3). Though the three-way interaction between group, time point, and PIQ fell short of significance (estimate = 0.66; t = 1.33; p = 0.19; d = 0.27 (small)), the separate correlation analyses exam- ining the relation between second-order ToM perfor- mance and PIQ at each time point showed a significant correlation between PIQ and second-order ToM scores for the ASDbi children only (see Table 3). For second-order ToM performance, the best model included the effect of EF but this effect did not reach sig- nificance. Yet, the correlational analyses (Table 3) showed patterns similar to those found for expressive vocabulary and PIQ, namely, a correlation for ASDbi children at time point 2, and correlations for both ASDbi and ASDmono children at time point 3. Although the simple effect of group as well as the interaction of group with time point did not reach signifi- cance in the growth curve model for second-order ToM, the interactions with language, EFs, and intelligence may have concealed some differences between groups. For this reason, we finally present simple, separate t-tests between groups for second-order ToM at each time point. ASD- mono and ASDbi children performed similarly in second-order ToM at time point 1 (age 6; with scores of 4.8 (SD = 12.8; ASDmono) and 6.0 (SD = 10.9; ASDbi); t = �0.32 df = 39; p = 0.75; d = 0.10 (negligible)) (see Figure 1). However, at time points 2 and 3, the bilingual children outperformed their monolingual peers (time point 2: scores of 9.5 (SD = 16.7; ASDmono) and 35.7 F I GURE 2 Relations between second- order ToM performance and expressive vocabulary scores for the two groups at the three time points. ASDbi, bilingual autistic children; ASDmono, monolingual autistic children. TABLE 3 Correlation coefficients of Pearson correlations between second-order ToM performance for the ADmono and ASDbi groups at each of the three time points. Time point 1 Time point 2 Time point 3 ASDmono ASDbi ASDmono ASDbi ASDmono ASDbi Expressive vocabulary �0.0 n.s. �0.1 n.s. �0.0 n.s. 0.46* 0.53* 0.61** Executive function �0.1 n.s. 0.08 n.s. 0.09 n.s. 0.55** 0.55* 0.65** PIQ �0.0 n.s. �0.0 n.s. �0.0 n.s. 0.47* 0.89*** 0.67*** Abbreviations: ASDbi, bilingual autistic children; ASDmono, monolingual autistic children; n.s., not significant; PIQ, performance IQ. *p < 0.05. **p < 0.01. ***p < 0.001. PERISTERI ET AL. 7 (SD = 24.5; ASDbi); t = �4.05 df = 35; p < 0.001; d = 1.25 (large); time point 3: scores of 34.5 (SD = 21.6; ASDmono) and 50.0 (SD = 22.4; ASDbi); t = �2.28 df = 40; p = 0.03; d = 0.70 (medium)). We should note that besides t-tests, we ran an addi- tional MANOVA analysis to examine the differences between groups over time point on first-order ToM, second-order ToM, PIQ, expressive vocabulary, and EF within one model. The results of the MANOVA were in line with the t-test-based results presented above. More specifically, for first-order ToM, there was no difference between the groups. For second-order ToM, expressive vocabulary, PIQ, and EF, there were significant between- group differences, which were stable over time points. Only in PIQ the effect of group was modulated by an interaction with time point, indicating that the difference increased over time (i.e., it increased with age). These effects all survived Bonferroni corrections for 15 different comparisons, that is, decreasing the alpha to 0.003. In conclusion, ASDbi children scored lower on first- order ToM overall, but higher on second-order ToM at time points 2 and 3 compared to ASDmono children. For second-order ToM, increased performance was asso- ciated with expressive vocabulary scores and PIQ and EF. DISCUSSION Mentalizing abilities in bilingual autistic children have been typically studied in terms of their first-order ToM skills, that is, the ability to understand that people may have different beliefs about the world. Cross-sectional data have shown that bilingualism boosts autistic chil- dren’s ability to make inferences about the mental states of other people (Peristeri et al., 2021a). In the current study, we have capitalized on longitudinal data suited to test, first, if bilingualism has a positive effect on autistic children’s second-order ToM skills, and, second, if lan- guage ability, intelligence or/and EF influence second- order ToM development in bilingual (and monolingual) autistic children. Two groups of 21 bilingual and 21 monolingual autistic children were followed-up at three time points, namely, at 6, 9, and 12 years. Our results show that bilingual autistic children outperformed their VIQ- and SES-matched monolingual peers on second-order ToM at ages 9 and 12, and that both expressive vocabulary and PIQ seemed to play a causal role in advancing their second-order ToM development. We also observed that the monolingual autistic children managed to capitalize on their PIQ and expressive vocab- ulary skills to support their second-order ToM perfor- mance only at the latest testing time point, that is, at age 12. The results suggest that bilingualism boosts autistic children’s second-order ToM skills, and that this effect is related to the bilingual children’s earlier attunement to language (expressive vocabulary) and IQ (specifically, PIQ) resources as compared to their monolingual autistic peers. Our first research question focused on comparing the two autistic groups on their second-order ToM perfor- mances. At age 6, both monolingual and bilingual autis- tic children performed exactly the same on second-order ToM. However, at ages 9 and 12, bilinguals outper- formed their monolingual autistic peers in the second- order ToM task, whereas they were matched on their first-order ToM performances at both time points. Inter- estingly, Peristeri, Tsimpli, and Durrleman (2021); Peri- steri, Baldimtsi, et al. (2021) have recently found that bilingual autistic children outperformed their monolin- gual peers on a first-order ToM task, however, the ToM paradigm used was minimally verbal which may have strengthened bilingualism effects. In the current study, we used verbal first-order ToM tasks as a background measure mainly because the second-order ToM tasks were highly-verbal. The difference between monolingual and bilingual autistic children in the second-order ToM tasks at age 9 was quite large, highlighting clear differ- ences in the two groups’ developmental trajectories. This is a novel piece of evidence that shows that bilingualism F I GURE 3 Relations between second- order ToM performance and PIQ for the two groups at the three time points. ASDbi, bilingual autistic children; ASDmono, monolingual autistic children. 8 PERISTERI ET AL. effects at age 9 were pronounced in a way that offset autistic children’s second-order ToM difficulties. The study’s second research question was concerned with the extent to which children’s language, IQ and EF resources affected their second-order ToM performance over time. According to our findings, at age 9, there were strong links between expressive vocabulary and second- order ToM (and to a lesser extent with PIQ) for the bilin- gual but not the monolingual group, while both groups seemed to capitalize on their expressive vocabulary and PIQ resources at age 12. This suggests that bilingual autistic children managed to engage language and cogni- tive resources that benefited their second-order ToM per- formance earlier in time as compared to their monolingual peers. In contrast, there were no correla- tions between expressive vocabulary, PIQ and second- order ToM at age 6 for either group. As PIQ was actually significantly higher for the bilinguals as compared to the monolingual autistic group at both time points 2 and 3 (but not at time point 1), we may conclude that bilin- gual autistic children took advantage of their enhanced intellectual functioning resources to support advanced ToM reasoning. A similar pattern was observed for expressive vocabulary, namely that bilinguals capitalized on their lexical skills earlier than their monolingual peers, that is, at 9 while the monolinguals showed this trend at 12 years. However, in contrast to PIQ, the bilinguals scored lower than their monolingual peers on expressive vocabulary. This suggests that their second-order ToM performance was supported by a language skill, which was weaker compared to monolinguals. The finding that bilingual autistic children’s second- order ToM performance relied on expressive vocabulary, which has been found to be a vulnerable language domain across both autistic and non-autistic bilingual child populations (e.g., Hambly & Fombonne, 2014; Hoff et al., 2014), may imply a strategy to rely on lan- guage resources even when the latter is not an area of strength for bilingual children. We may further speculate that the bilingual autistic children in the present study have attuned their second-order ToM reasoning skills to their language resources once they have acquired the pre- requisite vocabulary at the age of 9. Finally, regarding EF, we found no compelling evidence of a correlation with second-order ToM, which is in line with the findings of previous research that focused on relations between ToM and EF in autistic and non-autistic children (Dahlgren et al., 2017; Peristeri et al., 2021a). However, PIQ can be understood to subsume EF to a certain degree as a broader, multidimensional cognitive measure. This may explain why PIQ was found to be a better pre- dictor of second-order ToM performance than perfor- mance in the 2-back task as a proxy for children’s EF. The overall findings reveal that bilingualism has boosted autistic children’s second-order ToM skills at ages 9 and 12. Regarding the mechanisms underlying the bilingualism effect, there were positive relationships between expressive vocabulary, PIQ and trajectories of second-order ToM skills, most noticeably at 9 and 12 years, for the bilingual autistic group only, while monolingual autistic children only managed to capitalize on their expressive vocabulary and PIQ resources later on, at the age of 12. While the bilingual autistic children showed higher EF across all three time points, this advantage did not seem to boost their second-order ToM performance over and above the influence of expressive vocabulary and PIQ. The findings underscore the impor- tance of bilingualism in yielding beneficial cascading effects on ToM development in autistic children, as well as the ways the bilingualism advantage is related to the autistic children’s language and intelligence skills. LIMITATIONS Although the findings of the current longitudinal study provide straightforward evidence for positive relations between bilingualism and second-order ToM skills in autistic children, as well as the ways ToM trajectories are (or are not) affected by the children’s language (expres- sive vocabulary) and cognitive (PIQ, EF) skills, our find- ings should be replicated by follow-up studies given the following limitations. For one, the sample size of the two groups comprising 21 children each was relatively small, and it is possible that the power of the growth curve models was too low to detect more effects or/and interac- tions between the critical variables of the study. Another limitation is the lack of data from control monolingual and bilingual TD groups, which could have provided the baseline characteristics of developmental changes in ToM, language, IQ and EF, and could further highlight differences between monolingual and bilingual autistic children. Furthermore, the study did not include expres- sive vocabulary measures in Albanian for the bilingual autistic group, since, due to the lack of validated lan- guage screening tests in Albanian, collecting data on this language was not feasible. Similarly, no tool of autism severity or adaptive functioning adapted to Greek is cur- rently available, so autistic children’s profiling did not include the specific dimensions. In addition, longitudinal studies examining second-order ToM skills in bilingual autistic children should be replicated with non-verbal ToM paradigms, thereby disentangling ToM reasoning and language processing demands in autism. Finally, it is possible that the groups’ ToM trajectories might have been modulated by other factors to those investigated in the current study; it is important to note that the stories that have been used to assess second-order ToM were highly-verbal, which might have burdened autistic chil- dren’s language and cognitive processing skills, so syntax or/and other EF components, including cognitive flexibil- ity and inhibition, should be included in future ToM studies in autism. Future research should expand meth- odological designs to include additional factors that may PERISTERI ET AL. 9 influence second-order ToM skills in bilingual autistic children. ACKNOWLEDGMENTS We would like to thank all the children and their parents for their willingness to participate in the study. FUNDING INFORMATION This work has benefitted from the support of the Swiss National Science Foundation (grant no. PR00P1_193104/1). CONFLICT OF INTEREST STATEMENT The authors declare no conflicts of interest. DATA AVAILABILITY STATEMENT The data that support the findings of this study are avail- able on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. ETHICS STATEMENT All the authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript. The study was approved by the Institutional Review Board (or Ethics Committee) of the Aristotle University of Thessaloniki (protocol-code 39928). The study was per- formed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki. ORCID Eleni Peristeri https://orcid.org/0000-0002-1151-677X REFERENCES American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (DSM-V). American Psychiatric Publishing. Andreou, M., Tsimpli, I. M., Durrleman, S., & Peristeri, E. (2020). The- ory of mind, executive functions, and syntax in bilingual children with autism spectrum disorder. Language, 5(4), 67. Astington, J. W., Pelletier, J., & Homer, B. (2002). Theory of mind and epistemological development: The relation between children’s second-order false-belief understanding and their ability to reason about evidence. New Ideas in Psychology, 20(2–3), 131–144. Aylward, G. P. (2002). Cognitive and neuropsychological outcomes: More than IQ scores. Mental Retardation and Developmental Dis- abilities Research Reviews, 8(4), 234–240. Baker, C. L., & Tenenbaum, J. B. (2014). Modeling human plan recog- nition using Bayesian theory of mind. Plan, activity, and intent recognition. Theory and Practice, 7(177–204), 86. Baron-Cohen, S. (2001). Theory of mind in normal development and autism. Prisme, 34(1), 74–183. Baron-Cohen, S., Jolliffe, T., Mortimore, C., & Robertson, M. (1997). Another advanced test of theory of mind: Evidence from very high functioning adults with autism or Asperger syndrome. Journal of Child Psychology and Psychiatry, 38(7), 813–822. Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a “theory of mind”? Cognition, 21(1), 37–46. Baron-Cohen, S., Wheelwright, S., Spong, A., Scahill, V., & Lawson, J. (2001). Studies of theory of mind: Are intuitive physics and intuitive psychology independent. Journal of Developmental and Learning Disorders, 5(1), 47–78. Bauminger, N., Solomon, M., & Rogers, S. J. (2010). Predicting friend- ship quality in autism spectrum disorders and typical development. Journal of Autism and Developmental Disorders, 40, 751–761. Bialystok, E. (2011). Reshaping the mind: The benefits of bilingualism. Canadian Journal of Experimental Psychology/Revue Canadienne de Psychologie expérimentale, 65(4), 229–235. Buac, M., & Kaushanskaya, M. (2020). Predictors of theory of mind performance in bilingual and monolingual children. International Journal of Bilingualism, 24(2), 339–359. Buitelaar, J. K., van der Wees, M., Swaab-Barneveld, H., & van der Gaag, R. J. (1999). Verbal memory and performance IQ predict theory of mind and emotion recognition ability in children with autistic spectrum disorders and in psychiatric control children. The Journal of Child Psychology and Psychiatry and Allied Disciplines, 40(6), 869–881. Charlton, R. A., Barrick, T. R., Markus, H. S., & Morris, R. G. (2009). Theory of mind associations with other cognitive functions and brain imaging in normal aging. Psychology and Aging, 24(2), 338–348. Dahlgren, S., Almén, H., & Dahlgren Sandberg, A. (2017). Theory of mind and executive functions in young bilingual children. The Journal of Genetic Psychology, 178(5), 303–307. Durrleman, S., Peristeri, E., & Tsimpli, I. M. (2022). The language- communication divide: Evidence from bilingual children with atypical development. Evolutionary Linguistic Theory, 4(1), 5–51. Durrleman, S., Tsimpli, I. M., & Peristeri, E. (2022). Domino effects of bilingualism in autism Spectrum disorders? Executive functions, complement clauses and theory of mind. In Y. Gong & F. Kpogo (Eds.), Proceedings of the 46th Boston University conference on lan- guage development (pp. 180–193). Cascadilla Press. Duval, C., Piolino, P., Bejanin, A., Eustache, F., & Desgranges, B. (2011). Age effects on different components of theory of mind. Consciousness and Cognition, 20(3), 627–642. Farrar, M. J., Benigno, J. P., Tompkins, V., & Gage, N. A. (2017). Are there different pathways to explicit false belief understanding? General language and complementation in typical and atypical children. Cognitive Development, 43, 49–66. Georgas, J., Paraskevopoulos, I. N., Besevegis, E., & Giannitsas, N. D. (1997). The Hellenic WISC-III. Psychometric Laboratory, Univer- sity of Athens. Gernsbacher, M. A., & Yergeau, M. (2019). Empirical failures of the claim that autistic people lack a theory of mind. Archives of Scien- tific Psychology, 7(1), 102–118. Gillespie-Lynch, K., Sepeta, L., Wang, Y., Marshall, S., Gomez, L., Sigman, M., & Hutman, T. (2012). Early childhood predictors of the social competence of adults with autism. Journal of Autism and Developmental Disorders, 42, 161–174. Goetz, P. J. (2003). The effects of bilingualism on theory of mind devel- opment. Bilingualism: Language and Cognition, 6(1), 1–15. Gonzalez-Barrero, A. M., & Nadig, A. S. (2019). Can bilingualism miti- gate set-shifting difficulties in children with autism spectrum disor- ders? Child Development, 90(4), 1043–1060. Hambly, C., & Fombonne, E. (2014). Factors influencing bilingual expressive vocabulary size in children with autism spectrum disor- ders. Research in Autism Spectrum Disorders, 8(9), 1079–1089. Happé, F. G. (1994a). Wechsler IQ profile and theory of mind in autism: A research note. Journal of Child Psychology and Psychia- try, 35(8), 1461–1471. Happé, F. G. (1994b). An advanced test of theory of mind: Understand- ing of story characters’ thoughts and feelings by able autistic, men- tally handicapped, and normal children and adults. Journal of Autism and Developmental Disorders, 24(2), 129–154. Hoff, E., Rumiche, R., Burridge, A., Ribot, K. M., & Welsh, S. N. (2014). Expressive vocabulary development in children from bilin- gual and monolingual homes: A longitudinal study from two to four years. Early Childhood Research Quarterly, 29(4), 433–444. 10 PERISTERI ET AL. https://orcid.org/0000-0002-1151-677X https://orcid.org/0000-0002-1151-677X Jones, C. R., Simonoff, E., Baird, G., Pickles, A., Marsden, A. J., Tregay, J., Happé, F., & Charman, T. (2018). The association between theory of mind, executive function, and the symptoms of autism spectrum disorder. Autism Research, 11(1), 95–109. Jones, K., & Day, J. D. (1997). Discrimination of two aspects of cognitive-social intelligence from academic intelligence. Journal of Educational Psychology, 89(3), 486–497. Joseph, R. M., & Tager–Flusberg, H. E. L. E. N. (2004). The relation- ship of theory of mind and executive functions to symptom type and severity in children with autism. Development and Psychopa- thology, 16(1), 137–155. Kleiman, E. (2021). EMAtools: Data management tools for real-time monitoring/ecological momentary assessment data. R package ver- sion 0.1.4. https://CRAN.R-project.org/package=EMAtools Lord, C., Rutter, M., & Le Couteur, A. (1994). Autism diagnostic interview-revised: A revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. Journal of Autism and Developmental Disorders, 24(5), 659–685. Losh, M., & Capps, L. (2003). Narrative ability in high-functioning children with autism or Asperger’s syndrome. Journal of Autism and Developmental Disorders, 33, 239–251. Marinis, T., Andreou, M., Bagioka, D. V., Baumeister, F., Bongartz, C., Czypionka, A., Colegos, A., Peristeri, E., Skrimpa, V., Durrleman, S., & Terzi, A. (2023). Development and validation of a task battery for verbal and non-verbal first-and second-order theory of mind. Frontiers in Language Sciences, 1, 1052095. Mazza, M., Mariano, M., Peretti, S., Masedu, F., Pino, M. C., & Valenti, M. (2017). The role of theory of mind on social informa- tion processing in children with autism spectrum disorders: A mediation analysis. Journal of Autism and Developmental Disor- ders, 47, 1369–1379. Milligan, K., Astington, J. W., & Dack, L. A. (2007). Language and theory of mind: Meta-analysis of the relation between language ability and false-belief understanding. Child Development, 78(2), 622–646. Montgomery, L., Chondrogianni, V., Fletcher-Watson, S., Rabagliati, H., Sorace, A., & Davis, R. (2022). Measuring the impact of bilingualism on executive functioning via inhibitory con- trol abilities in autistic children. Journal of Autism and Develop- mental Disorders, 52(8), 3560–3573. Munson, J., Faja, S., Meltzoff, A., Abbott, R., & Dawson, G. (2008). Neurocognitive predictors of social and communicative develop- mental trajectories in preschoolers with autism spectrum disorders. Journal of the International Neuropsychological Society, 14(6), 956–966. Nguyen, T. K., & Astington, J. W. (2014). Reassessing the bilingual advantage in theory of mind and its cognitive underpinnings. Bilin- gualism: Language and Cognition, 17(2), 396–409. Nuske, H. J., & Bavin, E. L. (2015). Narrative comprehension in 4– 7-year-old children with autism: Testing the weak central coher- ence account. International Journal of Language & Communication Disorders, 1–12. Peristeri, E., & Andreou, M. (2024). Intellectual development in young children with autism spectrum disorders: A longitudinal study. Autism Research, 17, 543–554. https://doi.org/10.1002/aur.3089 Peristeri, E., Andreou, M., Ketseridou, S. N., Machairas, I., Papadopoulou, V., Stravoravdi, A. S., Bamidis, P. D., & Frantzidis, C. A. (2023). Animacy processing in autism: Event- related potentials reflect social functioning skills. Brain Sciences, 13(12), 1656. Peristeri, E., Baldimtsi, E., Andreou, M., & Tsimpli, I. M. (2020). The impact of bilingualism on the narrative ability and the executive functions of children with autism spectrum disorders. Journal of Communication Disorders, 85, 105999. Peristeri, E., Baldimtsi, E., Tsimpli, I. M., & Durrleman, S. (2024). Bilingualism matters: A study of children with autism spectrum disorder and developmental language disorder. In K. Grohmann (Ed.), Multifaceted multilingualism (pp. 206–233). John Benjamins Publishing Company. Peristeri, E., Baldimtsi, E., Vogelzang, M., Tsimpli, I. M., & Durrleman, S. (2021). The cognitive benefits of bilingualism in autism spectrum disorder: Is theory of mind boosted and by which underlying factors? Autism Research, 14(8), 1695–1709. Peristeri, E., Silleresi, S., & Tsimpli, I. M. (2022). Bilingualism effects on cognition in autistic children are not all-or-nothing: The role of socioeconomic status in intellectual skills in bilingual autistic chil- dren. Autism, 26(8), 2084–2097. Peristeri, E., Tsimpli, I. M., & Durrleman, S. (2021). Positive effects of bilingualism on social cognition in autism Spectrum disorder: A study of social exclusion and theory of mind. In D. Dionne & L. A. Vidal Covas (Eds.), BUCLD 45: Proceedings of the 44th annual Boston University conference on language development (pp. 626–638). Cascadilla Press. Perner, J., Frith, U., Leslie, A. M., & Leekam, S. R. (1989). Exploration of the autistic child’s theory of mind: Knowledge, belief, and com- munication. Child Development, 60, 689–700. Peterson, C., Slaughter, V., Moore, C., & Wellman, H. M. (2016). Peer social skills and theory of mind in children with autism, deafness, or typical development. Developmental Psychology, 52(1), 46–57. Peterson, C., & Wellman, H. M. (2019). Longitudinal theory of mind (ToM) development from preschool to adolescence with and with- out ToM delay. Child Development, 90(6), 1917–1934. Piaget, J. (1947). La psychologie de l’ intelligence. Armand Colin. Poulin-Dubois, D., Blaye, A., Coutya, J., & Bialystok, E. (2011). The effects of bilingualism on toddlers’ executive functioning. Journal of Experimental Child Psychology, 108(3), 567–579. R Core Team. (2019). R: A language and environment for statistical com- puting. R Foundation for Statistical Computing. Renfrew, C. (1997). Word finding vocabulary test (the Renfrew language scales). Winslow Press. Romero, C., & Uddin, L. Q. (2021). Bilingualism, executive function, and the brain: Implications for autism. Neurobiology of Language, 2(4), 513–531. Rommelse, N., Langerak, I., van der Meer, J., de Bruijn, Y., Staal, W., Oerlemans, A., & Buitelaar, J. (2015). Intelligence may moderate the cognitive profile of patients with ASD. PLoS One, 10(10), e0138698. Rosello, B., Berenguer, C., Baixauli, I., García, R., & Miranda, A. (2020). Theory of mind profiles in children with autism spectrum disorder: Adaptive/social skills and pragmatic competence. Fron- tiers in Psychology, 11, 567401. Schroeder, S. R. (2018). Do bilinguals have an advantage in theory of mind? A meta-analysis. Frontiers in Communication, 3, 36. Schweizer, K., & Moosbrugger, H. (2004). Attention and working mem- ory as predictors of intelligence. Intelligence, 32(4), 329–347. Siller, M., Swanson, M. R., Serlin, G., & Teachworth, A. G. (2014). Internal state language in the storybook narratives of children with and without autism spectrum disorder: Investigating relations to theory of mind abilities. Research in Autism Spectrum Disorders, 8(5), 589–596. Smith, E. E., & Jonides, J. (1999). Storage and executive processes in the frontal lobes. Science, 283(5408), 1657–1661. Tager-Flusberg, H. (2007). Evaluating the theory-of-mind hypothesis of autism. Current Directions in Psychological Science, 16(6), 311–315. Tompkins, V., Farrar, M. J., & Montgomery, D. E. (2019). Speaking your mind: Language and narrative in young children’s theory of mind development. Advances in Child Development and Behavior, 56, 109–140. Torchiano, M. (2019). Effsize: Efficient Effect Size Computation, R pack- age version 0.8.1. https://CRAN.R-project.org/package=effsize Vogindroukas, I., Protopapas, A., & Sideridis, G. (2009). Experiment on the expressive vocabulary (Greek version of Renfrew word finding vocabulary test). Glafki. PERISTERI ET AL. 11 https://cran.r-project.org/package=EMAtools https://doi.org/10.1002/aur.3089 https://cran.r-project.org/package=effsize Wade, M., Prime, H., Jenkins, J. M., Yeates, K. O., Williams, T., & Lee, K. (2018). On the relation between theory of mind and execu- tive functioning: A developmental cognitive neuroscience perspec- tive. Psychonomic Bulletin & Review, 25, 2119–2140. Wechsler, D. (1991). Wechsler intelligence scale for children, third edi- tion, WISC-III. The Psychological Corporation. Wellman, H. M. (2017). The development of theory of mind: Historical reflections. Child Development Perspectives, 11(3), 207–214. Wellman, H. M., & Bartsch, K. (1994). Before belief: Children’s early psychological theory. In C. Lewis & P. Mitchell (Eds.), Children’s early understanding of mind: Origins and development (pp. 331– 354). Lawrence Erlbaum Associates, Inc. Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of theory-of-mind development: The truth about false belief. Child Development, 72(3), 655–684. World Health Organization (WHO). (1993). The ICD-10 classification of mental and behavioural disorders. World Health Organization. Yirmiya, N., Erel, O., Shaked, M., & Solomonica-Levi, D. (1998). Meta-analyses comparing theory of mind abilities of individuals with autism, individuals with mental retardation, and normally developing individuals. Psychological Bulletin, 124(3), 283–307. Yu, C. L., Kovelman, I., & Wellman, H. M. (2021). How bilingualism informs theory of mind development. Child Development Perspec- tives, 15(3), 154–159. SUPPORTING INFORMATION Additional supporting information can be found online in the Supporting Information section at the end of this article. How to cite this article: Peristeri, E., Vogelzang, M., Tsimpli, I. M., & Durrleman, S. (2024). Bilingualism and second-order theory of mind development in autistic children over time: Longitudinal relations with language, executive functions, and intelligence. Autism Research, 1–12. https://doi.org/10.1002/aur.3214 12 PERISTERI ET AL. https://doi.org/10.1002/aur.3214 Bilingualism and second-order theory of mind development in autistic children over time: Longitudinal relations with langua... INTRODUCTION METHOD Participants Procedure Materials Intelligence Executive function task 2-back test Theory of mind tasks First-order Theory of Mind tests Second-order Theory of Mind test Analysis plan RESULTS DISCUSSION LIMITATIONS ACKNOWLEDGMENTS FUNDING INFORMATION CONFLICT OF INTEREST STATEMENT DATA AVAILABILITY STATEMENT ETHICS STATEMENT REFERENCES