LP762-L2 week 3

Journal of Experimental Child Psychology 149 (2016) 146–158

Contents lists available at ScienceDirect

Journal of Experimental Child Psychology

journal homepage: www.elsevier .com/locate/ jecp

Deaf and hearing children’s development of theory of mind, peer popularity, and leadership during middle childhood

http://dx.doi.org/10.1016/j.jecp.2015.11.008 0022-0965/� 2015 Elsevier Inc. All rights reserved.

⇑ Corresponding author. Fax: +61 7 3365 4466. E-mail address: [email protected] (C.C. Peterson).

Candida C. Peterson a,⇑, Karin O’Reilly a, Henry M. Wellman b

a School of Psychology, University of Queensland, St Lucia, Queensland 4072, Australia bDepartment of Psychology, University of Michigan, Ann Arbor, MI 48109, USA

a r t i c l e i n f o

Article history: Available online 8 January 2016

Keywords: Theory of mind Popularity Peers Leadership Middle childhood Atypical development

a b s t r a c t

This study had two primary aims. First, we compared deaf and hearing children during middle and late childhood on (a) cognitive understanding of basic and advanced theory of mind (ToM) and (b) social dimensions of peer group relations, including popularity, iso- lation, leadership, and the disposition to interact positively with peers. Second, using correlational analyses, we examined ToM’s connections with these social variables to see whether and how ToM impacts children’s social lives. A total of 57 children (36 deaf children of hearing parents and 21 hearing children) 6 to 14 years of age completed a 6-step developmental ToM Scale, and their teachers reported on the social variables. Hearing children outper- formed deaf children on ToM and all teacher-rated variables. For deaf children, popularity correlated positively, and social isolation correlated negatively, with ToM even after controlling for age, gen- der, and language ability. For hearing children, the only ToM link was a weak correlation with leadership. Possible reasons for the differences between deaf and hearing groups are discussed, together with the likelihood of bidirectional causal links and impli- cations for deaf children’s social development in school.

� 2015 Elsevier Inc. All rights reserved.

C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158 147

Introduction

Theory of mind (ToM), or the understanding of others’ thoughts and feelings and their behavioral consequences, has been extensively studied in hearing typically developing (TD) preschool children over several decades, including research demonstrating the influence of ToM on preschool children’s social lives. But much less is known about connections between the ToM and social dispositions of school-age children and in particular children with deafness and other disabilities. Promising emerg- ing research on ToM development with TD youths for these post-preschool periods certainly exists (Devine & Hughes, 2013; Miller, 2009), but studies comparing typical developers and those with delay seem particularly promising and needed. One issue for such research is how to measure later ToM developments and later social proclivities for school-age children. Preschool ToM is prototypically assessed with false belief tests requiring children to infer the behavior, speech, or thoughts of protag- onists with false beliefs. Most TD children pass these by 5 years of age (Wellman, Cross, & Watson, 2001), but certain atypical groups, such as deaf children from hearing families (e.g., Peterson & Siegal, 1995; Peterson & Siegal, 1999) and children with autism (Baron-Cohen, 1995), are often sub- stantially delayed, routinely continuing to fail throughout middle childhood and the early teens (see Happé, 1995; Peterson, Slaughter, & Paynter, 2007, and Siegal & Peterson, 2008, for reviews).

Documentation of these false belief delays is abundant and clear, but important theoretical ques- tions remain as to their nature, basis, and later consequences. One view holds that, for deaf children of hearing parents (i.e., non-native signers), restricted social and conversational experience during early childhood in a hearing, non-signing family delays the development of preschool ToM concepts such as false belief. Entry into a community of fluently signing peers and teachers in primary school can sub- sequently assist the development of ToM understanding, although on a delayed timetable. The fact that natively signing deaf children with deaf parents (a 5% minority of children born deaf) develop ToM on the same early timetable as hearing children (see Peterson, 2009, for a review) is consistent with this theory.

However, another view is equally possible. For example, to explain ToM delays in children with autism, it is often argued (e.g., Leslie & Thaiss, 1992; Scholl & Leslie, 2001) that circumscribed congen- ital damage to a neurological ToM module limits the extent to which ToM development will ever be possible up to and/or beyond false belief. In this form, this explanation is unlikely to apply to deaf chil- dren of hearing parents even though they are often as delayed in ToM achievements as are high- functioning children with autism during late childhood. However, a plausible extension of such related but broader maturational views would be that normatively preschool achievements could exhibit a critical period for ToM (e.g., Morgan & Kegl, 2006) just as for language (e.g., Newport, 1991). In this view, if they are not attained early in life, they might be impossible to achieve later. This could apply to children with autism, but more obviously deaf children provide a special opportunity to address these issues and hypotheses. Doing so requires a means of measuring ToM growth as a con- tinuing developmental process extending beyond false belief mastery. A reliable and valid way in which to measure later ToM development in deaf children, and in TD children as well, would provide an opportunity to address other critical questions as well such as the relationships between later ToM developments and later social proclivities for school-age children, including those with deafness.

To begin to address these issues, we took advantage of a new 6-step ToM Scale (Peterson, Wellman, & Slaughter, 2012) that builds on, and extends developmentally beyond, the 5-step sequential pre- school ToM Scale devised by Wellman and Liu (2004). Peterson et al. (2012) found that the new scale contained concepts that continued to challenge hearing children at 7 to 10 years of age and to chal- lenge even those deaf children who had already successfully mastered false belief. One subsidiary focus of the current research, given the increased awareness of the importance of replication in psy- chological research, is further validation of this scale and its findings with a fresh sample of older deaf and TD children.

More substantively, however, we aimed to go beyond the laboratory assessment of children’s ToM understanding to examine some of the possible real-world correlates and consequences of ToM growth. When deaf and hearing children enter primary school, they join a new social world via full-day immersion in a classroom peer group. In this new and challenging social environment, peer

148 C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158

group dynamics becomes a major force in children’s lives (Hartup, 1970; Hartup, 1992). For deaf and hearing children alike, coping effectively with the fresh opportunities and demands of peer relation- ships can serve as a stimulus to social growth during middle childhood (Hartup, 1992; Rubin, Bukowski, & Parker, 2006) and of course could serve as a major challenge as well. At the very least, effective integration into the classroom peer group socially challenges children to gain peer accep- tance, adopt leader or follower roles, forge friendships, cooperate versus conflict with the group, com- municate effectively, and so on. Consequently, a central question for ToM research on hearing and deaf children alike is whether individual differences among children in their skills and modes of coping with the complex dynamics of peer groups interconnect with their levels of cognitive ToM under- standing. This is the question focal to the current research.

Studies of younger children, particularly hearing preschoolers, have suggested some connections between ToM understanding and selected aspects of their social relations with individual peers, including (a) skilled sharing of social pretend play (e.g., Astington & Jenkins, 1995; Lalonde & Chandler, 1995), (b) skilled deception via games such as hide-and-seek (e.g., Peskin & Ardino, 2003), (c) friendship versus friendlessness (e.g., Fink, Begeer, Peterson, Slaughter, & de Rosnay, 2015), and (d) effective bullying of other children (Sutton, Smith, & Swettenham, 1999). Yet, expected links are not always apparent. After reviewing a large body of such research, Banerjee, Watling, and Caputi (2011) noted, ‘‘The evidence for the ‘real-life’ manifestations of individual differences in typi- cally developing children’s social understanding is less than overwhelming” (p. 1888).

Furthermore, during later childhood when the dynamics of children’s peer relations become com- plex and gain greater salience in children’s lives (e.g., Rubin et al., 2006), evidence on connections between ToM and TD children’s social actions is scarce and also mixed. However, there are some important exceptions, and in particular ToM understanding has been conclusively shown to connect with at least one key aspect of peer group relations, namely a child’s popularity status within the classroom peer group. Popular children are well liked by most members of the peer group, in sharp contrast to those who are socially isolated from, or rejected by, most of their peers. In theory, howwell children understand their peers’ minds could help to determine group popularity versus isolation, and indeed numerous studies have examined this for TD children, both during preschool and subse- quently. Recently, Slaughter, Imuta, Peterson, and Henry (2015) conducted a meta-analysis of 20 such studies sampling more than 2000 TD children from 2 to 10 years of age. Results revealed a significant overall effect for higher peer group popularity to correlate with higher ToM test scores. A correspond- ing negative effect (lower ToM understanding correlated with greater rejection by peers) was also sta- tistically significant, although of weaker absolute magnitude. These effects applied equally at all ages within this broad range and were just as apparent when popularity was measured using teacher rat- ings as when it was measured with sociometric peer nomination procedures in which children them- selves identify others in the peer group whom they like and dislike (Rubin et al., 2006).

What about children with ToM delay? Whether the same patterns apply is unknown. Conse- quently, a key goal of the current study was to examine whether individual differences in ToM are linked with variations in levels of peer popularity among deaf children of hearing parents. These chil- dren are known to encounter difficulties and delays in both social popularity and cognitive ToM domains, but to our knowledge no previous study has explored the connection between the two. To help fill this gap, we assessed ToM understanding among deaf children of hearing parents (non- native signers) as a possible predictor of their social status as popular members of the peer group ver- sus as socially isolated from peers.

Although no known past study has looked at this, several studies have examined popularity by itself in deaf children. Their results have often shown reduced peer popularity and heightened lone- liness, together with social neglect and social isolation, among deaf preschool and school-age children relative to hearing peers (e.g., Nunes, Pretzlik, & Olsson, 2001). In theory, other possible factors besides ToM may contribute to this either in association with ToM or independently. Thus, we included a pre- liminary examination of several of these possible additional predictors, including deaf children’s (a) receptive and productive language abilities (in sign), (b) motivational dispositions to interact posi- tively with peers (e.g., willingness to communicate and cooperate), and (c) having versus not having a cochlear implant. We were not primarily interested in examining the effects of cochlear implanta- tion (CI) on children’s social lives, but given that our data were able to provide needed information

C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158 149

with regard to the ToM–popularity connection, and given the increased prevalence of, and research interest in, the technological innovation of CI for deaf children (e.g., Martin, Bat-Chava, Lalwani, & Waltzmann, 2011), we deemed the CI variable an important factor to consider in a preliminary fash- ion, at the very least as a basis for future investigation.

Leadership within the peer group is another social variable of clear everyday importance for chil- dren’s everyday social lives. Although overlapping to some degree with popularity, leadership is clearly both conceptually and empirically distinct (Hartup, 1970). Both popularity and leadership are reciprocal relations involving the entire peer group, in contrast to other peer involvements such as having a playmate or friend (Fink et al., 2015) that need only to involve a dyad. Leadership also requires active choice and commitment on the part of a group of followers rather than simply being the product of the individual’s own level of social skill. Furthermore, over and above mere popularity (i.e., being well liked by a group of peers), the leader must exercise complex social judgment and ini- tiative. Children who lead their groups are able to direct their peers’ attitudes and activities and to inspire a following while minimizing dissent, discord, alienation and non-compliance. Thus, leader- ship is clearly not simply a synonym for popularity, although how closely the two variables may cor- relate with one another is an intriguing empirical question for deaf and hearing children alike.

Although far less frequently investigated among children now than it was several decades ago (see Hartup, 1970, for a review), the leadership concept today continues to have clear practical relevance to children’s daily social lives in school. The early literature on the attributes of child leaders showed that they often possessed exceptional social insight and dispositions to be inclusive and communicative as well as other social and personality attributes such as extraversion and assertiveness. However, because no known previous study has examined peer leadership in relation to ToM even for hearing TD children, let alone for deaf children, our study also explored whether ToM and leadership correlate with one another in deaf and/or hearing groups.

Method

Participants

The sample of 57 Australian children comprised 36 prelingually severely or profoundly deaf chil- dren of hearing parents with a mean age of 9.76 years (range = 5.67–14.17; 24 boys) and 21 age- matched TD hearing children with a mean age of 8.87 years (range = 7.33–13.42; 9 boys), all recruited from government-funded primary schools and high schools.

The deaf children came from three different types of school situation. All 36 of them were main- streamed for at least some part of each school day in ordinary classrooms where all or most other pupils were hearing children. In these mainstream classrooms, the medium of instruction was stan- dard English (spoken and written) rather than any form of sign language, and with very rare excep- tions hearing fellow pupils were unable to sign. The 36 deaf children in our sample varied on a continuum of predominance of this mainstream oral-only schooling. For 5 (14%) of them (denoted School Type 1), this oral mainstreaming occupied all of their class time, and even outside of class they had contact with hearing–speaking peers only because their schools had no other deaf pupils. At the other extreme, a further 22 deaf children (61%) were in School Type 3. These children were main- streamed for only some of their classes each day. They also spent some daily class time in specialist bilingual units located on the grounds of their mainstream primary schools. These units used both Auslan (Australian Sign Language) and spoken (or written) English as modes of instruction. Further- more, because the units included other signing deaf pupils, all children in Type 3 schooling interacted daily with fluently signing deaf peers and teachers as well as with hearing teachers and children in their mainstream classes. The remaining 9 children (25%, School Type 2) had an intermediate situa- tion. They attended mainstream schools without deafness units. However, unlike Type 1, some other deaf pupils also attended the same school, although none in exactly the same school grade. Thus, even though most daily classroom time was with hearing classmates, Type 2 children had daily playground access to other deaf signers, and in addition most attended weekly Auslan classes together with other deaf children from their own or nearby schools. We were not focally interested in children’s school

150 C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158

arrangements. However, in preliminary analyses, we examined these in case they influenced the pri- mary relationship of interest—the connections between ToM and children’s peer-related social differences.

Of the total deaf sample, 18 (50%) had a cochlear implant. None had any deaf family members. All 36 of the deaf children (including those in School Type 1) were being taught Auslan either in classes with other deaf children or via private lessons (either in person or via the internet) with an adult Aus- lan signer. All children in our sample had been learning Auslan for more than 1 school year, and all used it with at least moderate skill (see language test data in Results). Most also attempted some oral communication on at least some occasions. Nonetheless, signing was the preferred communication modality for many; thus, given these varied backgrounds, we tested all of the deaf children bilingually so that the testing mode would remain uniform, and maximally comprehensible, across the whole sample. This ensured that all deaf children had full access to identical signed-plus-oral input from the experimenters (see ‘‘Procedure” section below) and also enabled the deaf children themselves to sign their responses, an option that all children in this sample used for at least some tasks and questions.

No deaf child in the sample had any disability apart from hearing loss. Hearing children were selected to roughly match deaf children in age. None had any disability. Parents of all children in both groups supplied written informed consent and had English as their sole or primary language.

Tasks and scoring

ToM Scale Peterson et al.’s (2012) 6-Step ToM Scale was presented and scored exactly as described in their

article. In brief, the six sequentially developing ToM concepts assessed by the scale were (1) Diverse Desires (DD: awareness that different people may want different things), (2) Diverse Beliefs (DB: awareness that different people may hold different potentially true beliefs or opinions), (3) Knowledge Access (KA: awareness that not seeing leads to ignorance), (4) False Belief (FB: a standard misleading container task), (5) Hidden Emotion (HE: awareness that people’s facial expressions might not match their true feelings), and (6) Sarcasm Understanding (SARC: awareness that the literal meanings of peo- ple’s words might not match their intended meanings). The order of difficulty of these items is reliably 1 through 6 (with DD being easier than DB, etc.). Thus, HE and SARC, as tested on this scale, are reliably more difficult than false belief (Peterson et al., 2012; Wellman, Fang, & Peterson, 2011). Each task included not only a focal test question but also a comprehension control question, both of which needed to be answered correctly to pass any given task. Further details of tasks and scoring can be found in Peterson et al. (2012). Children received a ToM Scale total score of 0 to 6 to reflect the number of tasks they passed.

Peer popularity Consistent with many studies of peer popularity in hearing children (see the meta-analysis by

Slaughter et al., 2015), we used teacher-report measures of a child’s status as (a) a popular member of the peer group and (b) socially isolated from peers. The specific items we used came from the Mea- dow–Kendall Inventory (MKI; Meadow, 1983). For popularity, the teacher marked a 4-point scale— very true (scored 4), a bit true (scored 3), a bit untrue (scored 2), and very untrue (scored 1)—in response to this item: ‘‘This child relates well to peers and is accepted by them.” For social isolation, the item was, ‘‘This child is isolated from the peer group and has few or no friends,” with response choices again ranging from very true (scored 4) to very untrue (scored 1).

Leadership To examine the child’s leadership status in the peer group, we also used an item from the MKI.

Teachers rated the child on this item: ‘‘Other children look to this child as a leader” (again from 4 = very true to 1 = very untrue).

C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158 151

Positive social dispositions We included a further three-item teacher-rated measure designed to examine the child’s motiva-

tional dispositions for engaging in cooperative peer interaction. It was likewise adapted from the MKI by rewording three of the items as follows: (a) ‘‘Conciliatory: This child performs well in peer groups by contributing to group cohesion rather than conflict”; (b) ‘‘Tolerant/inclusive: This child accepts dif- ferences in other people, does not tease or exclude on the basis of race, appearance, handicap, etc.”; (c) ‘‘Eager communicator: This child tries to communicate with others by any means necessary.” For the deaf group only, the latter item continued ‘‘. . . including through sign, speech, gesture, pantomime, or in writing.” Each item was scored from 4 = very true to 1 = very untrue, and item totals were summed to produce a total score from 3 to 12. Cronbach’s alpha was .62, revealing adequate internal consistency.

Language ability Because it was conceivable that some deaf children might be suffering language delays (unlike

their hearing peers who were selected prior to sampling on the basis of teachers’ reports that they were free of disabilities and delays, including linguistic ones), we assessed language ability for the deaf children only. If deaf children were seriously limited linguistically, this could confound any observed associations between ToM and other focal variables. We measured language production and compre- hension abilities in Auslan given that all deaf children in the sample (including those with cochlear implants and those in mainstream-only schools) were currently learning and using Auslan. We tested both expressive (productive) and receptive (comprehension) Auslan language ability. The production test we used was the same for all deaf children in the sample, but owing to testing practicalities unre- lated to our experimental design (see below), we used two different receptive language measures, each with roughly half of the sample.

The expressive signing test we used was the Sign Language Production Skill test, a 22-item Auslan adaptation of the production subscale of the British Sign Language Test (Herman, Holmes, & Woll, 1999). For each item, the tester showed the child a picture (e.g., of a stamped envelope) and asked the child to express the sign for it. If a child finger-spelled, the tester was encouraging but repeated the request for a sign (all children knew the Auslan word ‘‘sign”). If the child then said or signed ‘‘don’t know” or continued to finger-spell, the answer was accepted but was scored as incorrect. Even when signs were produced, scoring was strict. For example, in response to the envelope picture, if the child signed ‘‘postcard,” this was deemed incorrect, as were idiosyncratic iconic gestures. Only standard Auslan signs for ‘‘letter” or ‘‘envelope” were accepted.

To assess receptive language comprehension, children took either the Auslan version of the BSL (British Sign Language): Reception Test (from Herman et al., 1999) or the CELF-P (the 22-item syntax scale of the Clinical Evaluation of Language Fundamentals–Preschool test; Wiig, Secord, & Semel, 1992). Both tests had been used effectively in prior research with deaf Auslan users, and both evaluate a range of lexical, morphological, and syntactic concepts (including verb tense, plurality, negation, and relative clauses). Essentially random factors (e.g., timing of testing sessions and availability of test materials) determined which of the two tests any given child received. Because of school absences or interfering activities, 2 children had missing expressive language data and 1 child had missing receptive language data. To ensure that our receptive language scores were comparable across the full sample, raw scores on each test were standardized and we then used z scores only in statistical analyses.

Testing procedure

Each child was tested individually in a quiet school area by a highly experienced graduate exper- imenter. For the deaf group, the mode of task and question presentation was fully bilingual. In approx- imately half of the cases, the main experimenter had a native language level of proficiency in Auslan, being a CODA (adult child of deaf adults). For the children’s testing, she first stated each question, story segment, or instruction orally and then immediately followed with the signed version of the same dis- crete statement, using a bilingual communication mode highly familiar to each child as part of ordi- nary classroom routines in school. In the remaining cases, the main experimenter (male) did not have

152 C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158

a native language level of proficiency in sign. In these cases, the mode of presentation was parallel to the first except that two adults took part. The main experimenter was assisted by one of several highly professionally qualified female sign language interpreters. For each discrete question, story event, or instruction, the main experimenter first spoke the utterance with lips clearly visible. Next, he paused and the interpreter immediately translated the experimenter’s spoken statement into the child’s pre- ferred signing modality (either Auslan or Natural Sign System [NSS]). The latter involves ‘‘signing in English” via Auslan signs presented in spoken English word order (Johnston & Schembri, 2007). Fur- ther attesting to their familiarity and comfort with Auslan, only 2 children in the current sample pre- ferred NSS to Auslan.

The bilingual experimenter was a hearing native signer. The other interpreters were professionally accredited at the highest level of Auslan signing skill by the peak national accreditation body in Aus- tralia, the National Accreditation Authority for Translators and Interpreters (National Accreditation Authority for Translators, 2011). All were likewise extensively experienced in working with deaf chil- dren in schools. The hearing children were tested orally and individually. Each child’s main teacher was given the six-item rating measure to privately complete for the child.

Results

Preliminary analyses

There was no significant age difference between the deaf and hearing groups, t(55) = 1.65, p = .104, nor did gender balance differ, v2(1, N = 57) = 2.18, p = .139. Because gender was also not significantly associated with any other variable in either group (all ps > .190), it was not considered further in sta- tistical analyses. In terms of language ability, deaf children’s scores on the 22-item Auslan test of pro- ductive (expressive) language ability ranged from 9 to 22 with a mean of 18.08 (SD = 3.60), indicating just over 80% average accuracy, consistent with their sound overall performance on the control ques- tions accompanying the false belief tests. On the receptive language (Auslan comprehension ability) variable, their z scores ranged from �1.94 to +1.75 with a mean of 0.00.

Type of schooling (on the scale described above of increasingly frequent access to social contact with signing deaf peers) was likewise not significantly related either to ToM or to any of the social variables (see Table 3 below). Recall, however, that only a small number of children in our sample (n = 5) had no contact at all with signing deaf peers at school.

Main analysis: Group differences

Table 1 shows children’s mean scores on focal variables, and Table 2 shows performance on indi- vidual scale tasks. In terms of overall ToM performance, there was a statistically significant difference between deaf and hearing groups in total steps passed on the 6-step ToM Scale, t(55) = 5.88, p < .001. The hearing children passed an average of 2 more scale steps than the deaf children. Yet, they were not at ceiling on the full 6 steps. Indeed, as Table 2 shows, only roughly half of them understood hidden emotion, and only one third passed sarcasm. Furthermore, only 7 of the 21 hearing children (33%) scored perfectly on all 6 steps. These data parallel Peterson et al. (2012) findings and replicate that this extension of Wellman and Liu’s (2004) original 5-step ToM Scale to 6 steps clearly challenges even TD

Table 1 Deaf and hearing children’s mean scores (and standard deviations) on key variables.

Variable Deaf (n = 36) Hearing (n = 21)

Age (years) 9.76 (2.28) 8.87 (1.21) ToM Scale (out of 6) 2.72 (1.23) 4.67 (1.07) Peer popularity (out of 4) 2.89 (0.91) 3.48 (0.68) Social isolation (out of 4) 2.15 (0.87) 1.67 (0.79) Leadership (out of 4) 1.75 (0.84) 2.95 (1.02) Positive social dispositions (out of 12) 9.42 (1.48) 10.52 (1.83)

Table 2 Percentages (and numbers) of children in each group passing each successive (left to right) step on the ToM Scale and displaying perfect scale consistency with this ordering.

Diverse Desires

Diverse Beliefs

Knowledge Access

False belief

Hidden emotion

Sarcasm Perfectly scale consistent

Deaf 92 (33) 80 (29) 58 (15) 28 (10) 11 (4) 3 (1) 83 (30) Hearing 95 (20) 90 (19) 95 (20) 90 (19) 57 (12) 33 (7) 86 (18)

C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158 153

children and is sensitive to individual differences among them in ToM even during middle and late childhood (i.e., for children who are 7–13 years old).

In the deaf group, there was also considerable individual variability. No deaf child scored perfectly on the ToM Scale, and only 1 passed the final (sarcasm) step. Yet, 9 deaf children (25%) managed to pass 4 or 5 scale tasks, and the vast majority passed the first 2 scale tasks (see Table 2). Only 1 child scored zero. Table 2 also shows the numbers and percentages of children in each group who con- formed perfectly to the expected Guttman scale ordering. Consistent with prior findings (Peterson, Wellman, & Liu, 2005; Peterson et al., 2012), the vast majority of children in both the deaf and hearing groups passed the tasks in the same sequence as one another, exactly in the order consistently observed among hearing and deaf groups from Western cultures in numerous prior scaling studies.

In terms of peer popularity (see Table 1), there was also a significant difference between the deaf and hearing groups, t(55) = 2.55, p = .014. Hearing children were perceived by their teachers as more popular with peers than were deaf children their age. Conversely, on social isolation, significantly more deaf than hearing children were rated as isolated or neglected by the peer group, t(55) = 2.04, p = .046. In fact, for only 4 hearing children (19%) but 14 deaf children (39%), the descriptor ‘‘isolated with few or no friends” was seen as more true than false.

Leadership likewise differed between groups, t(55) = 3.97, p = .001, with hearing children scoring higher. However, even in the deaf group there was individual variability. One deaf child received the maximum rating (4: very true), and 5 other children scored 3. In other words, for 23% of deaf chil- dren, the ‘‘leader” label was seen as more true than untrue. Among the hearing children, however, the corresponding percentage was noticeably greater (67%).

Correlational analyses for deaf children

Table 3 shows the bivariate correlations of deaf children’s ToM scores with their peer popularity, social isolation, and positive social dispositions. Links among these focal variables and the control vari- ables of age, language ability, school type, and CI are also shown. Peer popularity and social isolation were both significantly correlated with ToM understanding, as was the disposition toward positive social interaction. However, because receptive language skill was also a significant ToM correlate, par- tial correlations were also computed to check on the independence of these associations. All three focal ToM correlations (with popularity, isolation, and positive dispositions) remained statistically sig- nificant even with language ability controlled, rs(32) = .49, �.57, and .54, all ps < .01.

Consistent with past evidence that peer popularity is not simply either the mirror image of peer group isolation or completely synonymous with leadership, the correlations among these variables in the deaf group (see Table 3), although uniformly significant, were not overwhelming. Most fell within the range of medium effect size (i.e., r = .30–.50; Cohen, 1992). This supported our methodolog- ical decision (made on conceptual grounds, as noted earlier) to measure leadership, popularity, and isolation as separate variables, just as Slaughter et al. (2015) did for the latter two in their meta- analysis for similar reasons. The motivational dispositions for positive interaction also correlated sig- nificantly with these group variables of leadership, popularity, and isolation (see Table 3).

The link of ToM with peer popularity for deaf children is consistent with what has been shown pre- viously for hearing children (Slaughter et al., 2015). The fact that the motivational disposition toward positive social interaction (via eagerness to communicate and to be conciliatory and tolerant) was also linked to ToM is novel but plausible because these dispositional proclivities would be likely to foster social experiences with peers that could enhance ToM understandings.

Table 3 Correlations (Pearson r values) among key variables for deaf children.

ToM Scale

Popularity Isolation Leadership Positive disposition

Age Language ability (receptive)

Language ability (productive)

School type

Cochlear implant

ToM Scale – .46** �.56** .24 .46** .29 .54** .34 .28 �.05 Popularity – – �.65*** .48** .48** .16 .16 .18 .25 �.06 Isolation – – – �.44* �.47* �.03 �.15 �.15 �.33 .10 Leadership – – – – .45** .02 .02 .05 �.03 �.03 Positive

disposition – – – – – .22 .08 .16 �.17 �.17

* p < .05. ** p < .01.

*** p < .001.

154 C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158

As Table 3 shows, there was no significant link between having versus not having a cochlear implant and either ToM understanding or any of the social variables. Clearly, at least for our current deaf sample, this technological prosthesis was less important than other variables we studied in influ- encing outcomes such as ToM development, peer popularity, and leadership. However, the CI variable certainly warrants continued investigation in other populations of deaf children. All of those in the current sample had been taught to sign and were continuing to receive sign language instruction irre- spective of whether or not they had a cochlear implant. Most also had frequent contact at school with signing deaf peers. Furthermore, given the ages of our sample (all born in 2008 or earlier) relative to the very recent changes that have arisen for earlier implantation and in the quality of CI technology, few if any deaf children in our sample would have been likely either to have received latest generation implants or to have had them installed before 2.5 years of age. (Unfortunately, details of exact ages at implantation were not available to us.) Yet, recent research suggests that when deaf children receive their implants by 2.5 years of age or earlier, they often score higher in ToM understanding than those receiving them at 3 years of age or later (e.g., Sundqvist, Lyxell, Jönsson, & Heimann, 2014). Thus, fur- ther study of the CI variable in newer cohorts of younger deaf children in different and varied educa- tional environments is clearly warranted.

Correlational analyses for hearing children

Table 4 shows correlation coefficients for the hearing group. ToM Scale scores were not signifi- cantly related to peer popularity or to social isolation in this relatively small sample of hearing chil- dren. However, there was a significant ToM link for leadership. Plausibly, a child who understands peers’ minds exceptionally well is more likely to be selected as a leader in the peer group. Leadership experience might likewise help children to understand others’ thoughts and feelings still better. How- ever, in the current sample, the link was a weak one and fell to marginal significance (p = .085) with age statistically controlled. The correlation between popularity and leadership (see Table 4) was of only medium effect size (Cohen, 1992), and social isolation was not even significantly correlated with

Table 4 Correlations (Pearson bivariate) among key variables for hearing children.

Variable ToM Scale Popularity Isolation Leadership Positive disposition Age

ToM Scale – .33 �.33 .44* .29 .61**

Popularity – – �.71*** .54* .75*** .17 Isolation – – – �.39 �.49* �.28 Leadership – – – – .55* .21 Positive disposition – – – – – .26

* p < .05. ** p < .01.

*** p < .001.

C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158 155

leadership in the hearing group. Thus, data for the hearing group reinforced earlier reported evidence from the correlations for deaf children suggesting that these three aspects of a child’s social status in the peer group are best conceptualized and measured separately as distinct entities.

Discussion

This study had two main aims. One was to compare deaf children with their hearing peers during middle childhood in terms of ToM understanding, peer group popularity, social isolation, peer leader- ship, and their motivational dispositions to interact positively with peers. Second, we aimed to dis- cover, for the first time, whether peer group popularity correlated with ToM understanding for deaf children in the manner previously found for hearing children (e.g., Slaughter et al., 2015) and, relat- edly, whether leadership (an arguably more demanding social status within the peer group than pop- ularity) correlated with ToM in the hearing and/or deaf groups.

With respect to ToM, the current findings for both deaf and hearing children are in line with past cross-sectional and longitudinal evidence (e.g., O’Reilly, Peterson, & Wellman, 2014; Wellman et al., 2011) that ToM understanding continues to develop in both of these groups during late childhood and beyond. These results supply data inconsistent with a strictly nativist–maturational view of ToM development and, in particular, the hypothesis of an early critical period for ToM mastery. Instead, during middle childhood, some deaf children of hearing parents, like their hearing peers, dis- played mastery of steps beyond false belief on the 6-step ToM Scale, including mastery or progress on the advanced ToM skills of HE and SARC in our scale. Conversely, despite the fact that, in accord with past evidence (Wellman et al., 2001), the vast majority (90%; see Table 2) of children in the hearing group (all 7 years of age or older) had already mastered false belief, many of them had not yet achieved subsequent scale steps of HE and (especially) sarcasm understanding (see Table 2). At the very least, these data demonstrate the possibility of extended developmental progress during middle childhood for hearing children and even into adolescence in the critical test case of deaf children of hearing parents. Of course, our data do not definitely rule out critical period proposals altogether if the critical period for ToM is conceived as extending past early adolescence. However, they are incon- sistent with proposals for a preschool critical period for ToM. Moreover, with regard to ‘‘advanced” ToM, they add fresh evidence to past findings. For example, O’Reilly et al. (2014) discovered that many deaf adults in their 30s and 40s outperformed older deaf children and adolescents on ToM tests of sar- casm understanding, and Pyers and Senghas (2009) observed longitudinal advances in ToM perfor- mance on false belief tests among ToM-delayed Nicaraguan deaf adults in their 20s.

Turning to social development, our findings from comparisons of levels of popularity and social iso- lation between deaf children and age-matched hearing groups reinforce and extend past evidence (see Batten, Oakes, & Alexander, 2014, for a review). Deaf children in our own and previous studies are found to have greater peer group difficulties than hearing children (including social isolation and lack of friends) and are less likely to be either leaders or even popular members of their classroom peer groups.

These findings help to set up the examination of one of our key aims—providing needed data about how individual differences in ToM understanding among children with and without ToM delays relate to peer group relations during middle childhood. Focally, we found significant links for deaf children between ToM cognitions and their status either as popular members of the peer group or as socially isolated outsiders. Leadership (arguably an even more socially demanding peer group status than pop- ularity: Hartup, 1970) was linked to ToM for hearing children, although not fully independent of age. No such link appeared for the deaf children, perhaps because so few deaf children were perceived by their teachers as leaders (e.g., for only 1 child in the sample (3%) was this descriptor judged very true).

These novel findings of links between popularity and ToM for deaf children complement and extend a small but growing body of studies of hearing children showing links between ToM and peer popularity, social isolation, friendship, and loneliness in hearing groups not only during preschool (Slaughter et al., 2015) but also during middle and late childhood. For example, when Devine and Hughes (2013) studied TD children from 8 to 13 years of age using a Silent Film test of children’s ToM-based appreciation of the mental states of characters in dramas resembling Happé’s (1994)

156 C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158

Strange Stories, they found that self-reported loneliness and peer rejection were significant negative correlates of ToM. Similarly, in a prospective longitudinal study, Fink et al. (2015) found not only that friendlessness correlated negatively with ToM at 5 years of age but also that TD 7-year-olds who had been chronically friendless since age 5 (i.e., lacking any reciprocated mutual friendship irrespective of their overall levels of popularity in the peer group as a whole) had scored lower on a battery of basic and advanced false belief tests at age 5 than children who had had at least one reciprocated mutual friendship at some point during this 2-year longitudinal time period.

For deaf children, our discovery of a similar pattern of correlations between ToM scores and peer group relations (e.g., popularity, isolation, leadership) is important. As noted earlier, deaf children are at greater risk for social problems connected with popularity and isolation than are hearing children (e.g., Batten et al., 2014). Furthermore, the possible ramifications of this are far-reaching. Social isola- tion and low popularity status during childhood have been shown to predict not only poor emotional well-being and reduced academic achievement in the short term but also future adverse consequences for mental health (e.g., Rubin et al., 2006). Thus, interventions on behalf of deaf children who are unpopular or socially isolated are clearly needed. One promising avenue worthy of trial could be the implementation of systematic ToM training in an effort to boost deaf children’s ToM understand- ing. A number of training strategies have proven to be effective for enhancing ToM in both hearing (e.g., Lohmann & Tomasello, 2003) and deaf groups (Wellman & Peterson, 2013). Therefore, it could prove to be both theoretically useful and practically valuable to discover whether training-induced gains in ToM understanding might transfer socially to boost deaf children’s popularity status and reduce social isolation from the peer group.

Any study has limitations, and possible limitations of ours were methodological. For logistic rea- sons (amount of time the school could make available for research participation), we were unable to obtain verbal ability data for our TD control group. Thus, the variables of receptive and expressive general language ability were examined in relation to focal social and ToM variables for our deaf group only. In future research on popularity and ToM in TD groups, it would be useful to include language ability as a control variable, especially because Slaughter et al. (2015) reported that they were unable to consider verbal ability in their meta-analysis owing to lack of data in the past studies of TD children that they reviewed. A second possible limitation of our study was the use of single items to measure each of the focal peer group status variables (popularity, social isolation, and leadership). Although previous research with hearing children has successfully employed teachers’ single-item ratings like these and demonstrated their validity in relation to multi-item teacher- and peer-report measures of popularity and unpopularity (e.g., French & Waas, 1985), it would be useful in future studies of deaf and hearing children to include comparisons with multi-item measures of the same constructs.

A third possible limitation was the fact that we assessed peer popularity and other social attributes only indirectly via teacher-observers. For popularity, the acknowledged ‘‘gold standard” of assessment is widely deemed to be the sociometric peer nomination technique in which every child names three other children in the peer group whom they like and three whom they dislike (e.g., Rubin et al., 2006). Future research could profitably test the replicability of our findings using this approach. Nonetheless, at least for TD children, Slaughter et al. (2015) meta-analysis revealed that patterns of correlations and effect sizes for the link between ToM and popularity were similar regardless of whether teachers’ rat- ings or peers’ sociometric reports were used. Furthermore, teachers’ ratings in general have the acknowledged advantages over several other modes of social assessment in that they (a) provide com- prehensive summative evaluation, (b) use teachers’ long-standing privileged acquaintance with their students, and (c) sample behavior observed in a variety of informative settings. Of course, in the case of our deaf group, teachers were inevitably aware of children’s hearing status. Teachers might have underestimated or overestimated deaf children’s peer competence owing to preconceived ideas about the social consequences of deafness and/or specialized background experiences of teaching primarily in deafness units rather than in mainstream hearing classrooms. Empirically, however, our teacher ratings had sufficient variability to informatively relate to children’s ToM achievements, and this was our primary aim for using such ratings in our particular project.

Our findings nonetheless warrant replication with different measures of popularity, leadership, and other techniques, including direct observation. In addition, our results set up the need for (a) further study of still more in-depth measures of deaf school-age children’s social functioning along with their

C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158 157

social dispositions and their popularity and leadership status in the peer group, (b) further study of links between these variables and ToM, and (c) recruiting of samples of deaf children more widely so as to include groups with different educational and communication backgrounds than ours (e.g., children in oral-only schools who, unlike all children in our sample, have never learned sign language). For obvious reasons, sample characteristics such as type of schooling, presence versus absence of a cochlear implant, and the child’s degree of social access to other deaf pupils cannot be subjected to experimental manipulation or control. Thus, further study of deaf children in other cultures, commu- nities, and school settings where these factors vary naturally is a clear future priority.

Yet another limitation of our study is that the data are cross-sectional and concurrent. In the future, the study of longitudinal prediction of later peer skills from earlier ToM understandings (and vice versa) would enhance understanding of causal mechanisms involved. In advance of such data, we pre- dict bidirectional pathways. This is because at the same time that advanced ToM cognitions can supply the impetus for gaining new social skills, so too can advanced social skills open up new and richer opportunities for everyday social interactions with peers, which can then trigger new cognitive insights into others’ minds. Arguably, once children master even the most basic social skills required for joining in and engaging with peers, much then depends on the quality of their social environment and their own social dispositions. We have shown here that children’s dispositions toward coopera- tion, open communication, and tolerance of difference are all positively linked with their status in the peer group (as popular vs. isolated) and their leadership roles. Furthermore, this is true for deaf and hearing children alike. Further examination of these inter-relationships is now called for, given our findings, and is a ripe area for future comprehensive and longitudinal research.

In summary, then, we have shown that ToM concepts continue to evolve after the preschool years and that these extended ToM developments interconnect with aspects of children’s everyday social lives such as peer popularity and leadership versus social isolation from the peer group. Our findings, therefore, show that the continued study of these developments and peer group relations are likely to be fruitful domains for future investigation. We propose that comparative studies like this one, where on-time ToM mastery is contrasted against development in atypical groups whose ToM growth is delayed by factors such as growing up deaf in a hearing family, would be especially useful additions to this emerging and promising research endeavor.

References

Astington, J. W., & Jenkins, J. M. (1995). Theory of mind development and social understanding. Cognition and Emotion, 9, 151–165.

Banerjee, R., Watling, D., & Caputi, M. (2011). Peer relations and understanding of faux pas. Child Development, 8, 1887–1905. http://dx.doi.org/10.1111/j1467-8264.2011.01699.x.

Baron-Cohen, S. (1995). Mindblindness: An essay on autism and theory of mind. Cambridge, MA: MIT Press. Batten, G., Oakes, P., & Alexander, T. (2014). Factors associated with social interaction between deaf children and their hearing

peers: A review. Journal of Deaf Studies and Deaf Education, 19, 285–302. http://dx.doi.org/10.1093/deafed/ent052. Cohen, J. (1992). A power primer. Psychological Bulletin, 135, 749–773. http://dx.doi.org/10.1037/0033:2909.112.1.155. Devine, R., & Hughes, C. (2013). Silent films and strange stories. Child Development, 84, 989–1003. http://dx.doi.org/10.1111/

cd12017. Fink, E., Begeer, S., Peterson, C., Slaughter, V., & de Rosnay, M. (2015). Friendlessness and theory of mind: A prospective

longitudinal study. British Journal of Developmental Psychology, 33, 1–17. http://dx.doi.org/10.1111/bjdp.12060. French, D., & Waas, G. (1985). Teachers’ ability to identify peer-rejected children. Journal of School Psychology, 23, 347–353. Happé, F. (1994). An advanced test of theory of mind: Understanding of story characters’ thoughts and feelings by able autistic,

mentally handicapped, and normal children and adults. Journal of Autism and Developmental Disorders, 24, 129–154. http:// dx.doi.org/10.1007/bf02172093.

Happé, F. G. E. (1995). The role of age and verbal ability in the theory of mind task performance of subjects with autism. Child Development, 66, 843–855. http://dx.doi.org/10.1111/j.1467-8624.1995.tb00909.x.

Hartup, W. W. (1992). Peer relations in early and middle childhood. In V. B. Van Hasselt & M. Hersen (Eds.). Handbook of social development: A lifespan perspective (Vol. 1, pp. 257–281). New York: Springer.

Hartup, W. W. (1970). Peer interaction and social organization. In P. H. Mussen (Ed.), Carmichael’s manual of child psychology (pp. 361–456). New York: John Wiley.

Herman, R., Holmes, S., & Woll, B. (1999). Assessing British Sign Language. Gloustershire, UK: Forest Bookshop. Johnston, T., & Schembri, A. (2007). Australian Sign Language (Auslan): An introduction to sign language linguistics. Cambridge, UK:

Cambridge University Press. Lalonde, C. E., & Chandler, M. J. (1995). False belief understanding goes to school: On the social–emotional consequences of

coming early or late to a first theory of mind. Cognition and Emotion, 9, 167–185.

158 C.C. Peterson et al. / Journal of Experimental Child Psychology 149 (2016) 146–158

Leslie, A. M., & Thaiss, L. (1992). Domain specificity in conceptual development: Neuropsychological evidence from autism. Cognition, 43, 225–251. http://dx.doi.org/10.1016/0010-0277(92)90013-8.

Lohmann, H., & Tomasello, M. (2003). The role of language in the development of false belief understanding: A training study. Child Development, 74, 1130–1144. http://dx.doi.org/10.1111/1467-8624.00597.

Martin, D., Bat-Chava, Y., Lalwani, A., & Waltzmann, S. B. (2011). Peer relationships of deaf children with cochlear implants: Predictors of peer entry and peer interaction success. Journal of Deaf Studies and Deaf Education, 16, 108–120.

Meadow, K. P. (1983). The Meadow-Kendall Social–Emotional Assessment Inventory for Deaf and Hearing-Impaired Students. Washington, DC: Gallaudet University.

Miller, S. A. (2009). Children’s understanding of second-order mental states. Psychological Bulletin, 135, 749–773. http://dx.doi. org/10.1037/a0016854.

Morgan, G., & Kegl, J. (2006). Nicaraguan sign language and theory of mind: The issue of critical periods and abilities. Journal of Child Psychology and Psychiatry, 47, 811–819. http://dx.doi.org/10.1111/j.1469-7610.2006.01621.x.

National Accreditation Authority for Translators and Interpreters. (2011). Accreditation standards for Auslan interpreters. Retrieved from http://www.naati.com.au.

Newport, E. L. (1991). Contrasting concepts of the critical period for language. In S. Carey & R. Gelman (Eds.), The epigenesis of mind: Essays on biology and cognition (pp. 111–130). Hillsdale, NJ: Lawrence Erlbaum.

Nunes, T., Pretzlik, U., & Olsson, J. (2001). Deaf children’s social relationships in mainstream schools. Deafness & Education International, 3(3), 123–136.

O’Reilly, K., Peterson, C. C., & Wellman, H. M. (2014). Sarcasm and advanced theory of mind understanding in children and adults with prelingual deafness. Developmental Psychology, 50, 1862–1877. http://dx.doi.org/10.1037/a0036654.

Peskin, J., & Ardino, V. (2003). Representing the mental world in children’s social behavior: Playing hide-and-seek and keeping a secret. Social Development, 12, 496–512.

Peterson, C. C. (2009). Development of social cognitive and communication skills in children born deaf. Scandinavian Journal of Psychology, 50, 475–483.

Peterson, C. C., & Siegal, M. (1995). Deafness, conversation, and theory of mind. Journal of Child Psychology and Psychiatry, 36, 459–474. http://dx.doi.org/10.1111/j.1469-7610.1995.tb01303.x.

Peterson, C. C., & Siegal, M. (1999). Representing inner worlds: Theory of mind in autistic, deaf, and normal hearing children. Psychological Science, 10, 126–129.

Peterson, C. C., Slaughter, V. P., & Paynter, J. (2007). Social maturity and theory of mind in typically developing children and those on the autism spectrum. Journal of Child Psychology and Psychiatry, 48, 1243–1250. http://dx.doi.org/10.1111/j.1469- 7610.2007.01810.x.

Peterson, C. C., Wellman, H. M., & Liu, D. (2005). Steps in theory-of-mind development for children with deafness or autism. Child Development, 76, 502–517.

Peterson, C. C., Wellman, H. M., & Slaughter, V. (2012). The mind behind the message: Advancing theory of mind scales for typically developing children and those with deafness, autism, or Asperger syndrome. Child Development, 83, 469–485. http://dx.doi.org/10.1111/j.1467-8624.2011.01728.x.

Pyers, J. E., & Senghas, A. (2009). Language promotes false-belief understanding: Evidence from learners of a new sign language. Psychological Science, 20, 805–812.

Rubin, K. H., Bukowski, W. M., & Parker, J. G. (2006). Peer interactions, relationships, and groups. Hoboken, NJ: John Wiley. Scholl, B. J., & Leslie, A. M. (2001). Minds, modules, and meta-analysis. Child Development, 72, 696–701. Siegal, M., & Peterson, C. C. (2008). Language and theory of mind in atypically developing children: Evidence from studies of

deafness, blindness, and autism. In C. Sharp, P. Fonagy, & I. M. Goodyer (Eds.), Social cognition and developmental psychopathology (pp. 81–112). New York: Oxford University Press.

Slaughter, V., Imuta, K., Peterson, C. C., & Henry, J. D. (2015). Meta-analysis of theory of mind and peer popularity in the preschool and early school years. Child Development, 86, 1159–1184. http://dx.doi.org/10.1111/cdev.12372.

Sundqvist, A., Lyxell, B., Jönsson, R., & Heimann, M. (2014). Understanding minds: Early cochlear implantation and the development of theory of mind in children with profound hearing impairment. International Journal of Pediatric Otorhinolaryngology, 78, 538–544. http://dx.doi.org/10.1016/j.ijporl.2013.12.039.

Sutton, J., Smith, P., & Swettenham, J. (1999). Social cognition and bullying. British Journal of Developmental Psychology, 33, 1–17. http://dx.doi.org/10.1348/0261510099165384.

Wellman, H. M., Cross, D., & Watson, J. (2001). A meta-analysis of theory-of-mind development: The truth about false belief. Child Development, 72, 655–684. http://dx.doi.org/10.1111/1467-8624.00304.

Wellman, H. M., Fang, F., & Peterson, C. C. (2011). Sequential progressions in a theory of mind scale: Longitudinal perspectives. Child Development, 82, 780–792. http://dx.doi.org/10.1111/j.1467-8624.2011.01583.x.

Wellman, H. M., & Liu, D. (2004). Scaling of theory-of-mind tasks. Child Development, 75, 523–541. http://dx.doi.org/10.1111/ j.1467-8624.2004.00691.x.

Wellman, H. M., & Peterson, C. C. (2013). Deafness, thought bubbles, and theory-of-mind development. Developmental Psychology, 49, 2357–2367.

Wiig, E. H., Secord, W. A., & Semel, E. M. (1992). Clinical evaluation of language fundamentals-preschool. San Antonio, TX: Harcourt/PsychCorp.