Research in Developmental Disabilities 35 (2014) 2849–2857

Contents lists available at ScienceDirect

Research in Developmental Disabilities

Deaf children’s use of clear visual cues in mindreading

Jian Hao a,b, Yanjie Su a,*a Department of Psychology, Peking University, No. 5 Yiheyuan Road, Haidian District, Beijing 100871, Chinab Beijing Key Laboratory of Learning and Cognition, Department of Psychology, College of Education, Capital Normal University, No. 105

North Xisanhuan Road, Haidian District, Beijing 100048, China

A R T I C L E I N F O

Article history:

Received 22 March 2014

Received in revised form 15 July 2014

Accepted 15 July 2014

Available online

Keywords:

Deaf children

Theory of mind

False belief

Visual cues

A B S T R A C T

Previous studies show that typically developing 4-year old children can understand other

people’s false beliefs but that deaf children of hearing families have difficulty in

understanding false beliefs until the age of approximately 13. Because false beliefs are

implicit mental states that are not expressed through clear visual cues in standard false

belief tasks, the present study examines the hypothesis that the deaf children’s

developmental delay in understanding false beliefs may reflect their difficulty in

understanding a spectrum of mental states that are not expressed through clear visual

cues. Nine- to 13-year-old deaf children of hearing families and 4–6-year-old typically

developing children completed false belief tasks and emotion recognition tasks under

different cue conditions. The results indicated that after controlling for the effect of the

children’s language abilities, the deaf children inferred other people’s false beliefs as

accurately as the typically developing children when other people’s false beliefs were

clearly expressed through their eye-gaze direction. However, the deaf children performed

worse than the typically developing children when asked to infer false beliefs with

ambiguous or no eye-gaze cues. Moreover, the deaf children were capable of recognizing

other people’s emotions that were clearly conveyed by their facial or body expressions.

The results suggest that although theory-based or simulation-based mental state

understanding is typical of hearing children’s theory of mind mechanism, for deaf

children of hearing families, clear cue-based mental state understanding may be their

specific theory of mind mechanism.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

It is difficult to understand other people’s mental states and to predict their behavior. This ability, namely ‘‘theory of mind’’(Premack & Woodruff, 1978), is difficult to acquire because mental states per se are invisible and must be inferred. By the age ofapproximately 4, typically developing children can infer other people’s false beliefs (Wellman, Cross, & Watson, 2001). Deafchildren of deaf families perform identically to hearing children on standard false belief tasks (Peterson & Siegal, 1999; Schick,de Villiers, de Villiers, & Hoffmeister, 2007), and thus their theory of mind normally develops. However, it is challenging for deafchildren of hearing families to gain insight into others’ mental worlds. Although deaf children of hearing families can freelyattribute mental states to others in storytelling tasks at the age of 9 (Marschark, Green, Hindmarsh, & Walker, 2000), they have

* Corresponding author. Tel.: +86 10 62751833; fax: +86 10 62756460.

E-mail addresses: janeh@pku.edu.cn (J. Hao), yjsu@pku.edu.cn (Y. Su).

http://dx.doi.org/10.1016/j.ridd.2014.07.034

0891-4222/� 2014 Elsevier Ltd. All rights reserved.

J. Hao, Y. Su / Research in Developmental Disabilities 35 (2014) 2849–28572850

difficulty in standard false belief tasks until the age of approximately 13 (O’Reilly, Peterson, & Wellman, 2014; Peterson & Siegal,1995; Russell et al., 1998; Wellman & Peterson, 2013). Thus, it is suggested that 9–13-year-old deaf children of hearing familieshave an awareness of mental states but still fail to infer mental states precisely. Their theory of mind has not matured as much asthat of typically developing children over 4.

Theory of mind delay in deaf children of hearing families is manifested mainly in the understanding of false beliefs.Numerous studies demonstrate that the deaf children perform poorly on standard false belief tasks (Courtin & Melot, 2005;de Villiers & de Villiers, 2012; Moeller & Schick, 2006; Peterson, 2002; Peterson & Siegal, 1999; Peterson & Slaughter, 2006;Schick et al., 2007). However, they perform relatively better on theory of mind tasks that assess the understanding of desiresand intentions (Rhys-Jones & Ellis, 2000; Scott, Russell, Gray, Hosie, & Hunter, 1999; Want & Gattis, 2005). Are false beliefsmore difficult to understand than desires or intentions for deaf children of hearing families? Wellman and Liu (2004) askedtypically developing preschoolers to complete different theory of mind tasks, and found that their understanding of desirespreceded their understanding of false beliefs. Peterson, Wellman, and Liu (2005) demonstrated a similar developmentalsequence of theory of mind in deaf children of hearing families. Moreover, studies on maternal talk show that mothers’mental state talk is consistent with children’s theory of mind developmental progression. Mothers first frequently talk aboutdesires and then their desire language decreases, whereas their belief language increases over time (Taumoepeau & Ruffman,2006, 2008). Previous studies provide solid evidence that understanding of desires is easier and develops beforeunderstanding of false beliefs. However, the reason why false beliefs are more difficult to understand needs to be furtherclarified. The discrepancies between understanding of different types of mental states lie not only in whether the mentalstates to be understood are beliefs, desires or intentions but also in whether these mental states are implicit or explicit. Instandard false belief tasks, the protagonist’s false belief is completely implicit and must be understood through reasoning orsimulation. In the desire or intention understanding task (Scott et al., 1999), for example, the protagonist’s desire or intentionis clearly expressed through his or her eye-gaze direction. Therefore, it is not false beliefs per se that are more difficult tounderstand than desires or intentions for deaf children of hearing families but mental states without clear cues that areharder to comprehend than those with clear cues. In other words, deaf children of hearing families seem incapable ofunderstanding a spectrum of mental states that are not expressed through clear cues.

It is theoretically important to investigate mental state understanding under different cue conditions in deaf children ofhearing families. With respect to the mechanisms of theory of mind, there have been some predominant perspectives.According to the ‘‘theory theory’’ (Gopnik & Meltzoff, 1997; Gopnik & Wellman, 1994), folk psychology as a theory is used formental state reasoning. The theory involves a body of knowledge about the mind, including the causal relationships amongvarious mental states and behavior. Another theory, the ‘‘simulation theory’’ (Harris, 1992), emphasizes that an individualgains insight into others’ mental states by imagining what he or she will think if placed in the situation of others. Althoughthe inferences are based on different processes according to the mechanisms above, both of these mechanisms areparticularly used to infer implicit mental states. Conversely, when mental states are expressed through clear cues, they arecompletely explicit and thus can be easily understood without complex reasoning or simulation. If 9–13-year-old deafchildren of hearing families show the competence to infer mental states with clear cues, it suggests that clear cue-basedmental state understanding may be another theory of mind mechanism that is dissociated from theory-based andsimulation-based mental state understanding. More importantly, for deaf children of hearing families, clear cue-basedmental state understanding may be their specific theory of mind mechanism.

Because language development is severely delayed in deaf children of hearing families (Lederberg, Schick, & Spencer,2013; Schick et al., 2007), the deaf children are likely to be more sensitive to visual cues than to linguistic cues of mentalstates. Few empirical data are available regarding deaf children’s understanding of mental states with visual cues. Pellicanoand Rhodes (2003) explored the role of clear visual cues in the understanding of false beliefs in typically developing children.In their experiment, the protagonist’s false belief was clearly expressed via her eye-gaze direction. The results showed that 3-year-olds performed significantly worse than 4-year-olds in the eye-gaze cue condition, which suggests no facilitative role ofclear visual cues in typically developing 3-year-olds’ false belief understanding. However, several explanations may beconsidered. First, the notably different performances of the two groups are most likely due to their language skill gap; thus,controlling for the effect of language abilities is a necessity. Second, it is not clear whether young children pay attention tothe protagonist’s eye-gaze direction. Therefore, control questions should be asked to ensure their attention to the visual cues.In addition, some studies have examined deaf children’s understanding of emotions with clear visual cues but have foundmixed results. For example, Hosie, Gray, Russell, Scott, and Hunter (1998) demonstrated that deaf children of elementaryschool years were able to recognize other people’s emotions that were expressed through their facial expressions. Dyck,Farrugia, Shochet, and Holmes-Brown (2004) found similar performance on the Emotion Recognition Scale in hearing-impaired children and hearing children matched for verbal ability. Nevertheless, poor performance of deaf children on afacial expression-based emotion recognition task was observed in Ludlow, Heaton, Rosset, Hills, and Deruelle’s (2010) study.More evidence is needed to indicate whether clear cue-based mental state understanding is the specific theory of mindmechanism for deaf children of hearing families.

The present study aimed to investigate mental states understanding under different cue conditions in deaf children ofhearing families. The deaf children’s performances on false belief tasks under different cue conditions were first examined toclarify whether they could understand exclusively false beliefs with clear visual cues. Pellicano and Rhodes’s (2003) falsebelief task with eye-gaze cues was used to assess their understanding of false beliefs with clear visual cues. A task adaptedfrom Pellicano and Rhodes’s (2003) experimental paradigm was used to explore their understanding of false beliefs with

J. Hao, Y. Su / Research in Developmental Disabilities 35 (2014) 2849–2857 2851

ambiguous visual cues. A standard false belief task (Wimmer & Perner, 1983) was used to indicate their understanding offalse beliefs without visual cues. Additionally, the deaf children’s performances on emotion recognition tasks with clearvisual cues (facial or body expression cues) were also measured to provide further evidence for their clear cue-based mentalstate understanding.

It was hypothesized that theory of mind delay in deaf children of hearing families existed in understanding a spectrum ofmental states that were not expressed through clear visual cues. Accordingly, the deaf children may experience difficulty instandard false belief tasks but perform as successfully as language-ability-matched hearing children on false belief taskswith clear eye-gaze cues. Moreover, deaf children of hearing families are also likely to perform well on emotion recognitiontasks with clear facial or body expression cues. Taken together, clear cue-based mental state understanding will be shown tobe their specific theory of mind mechanism.

2. Method

2.1. Participants

A total of 47 children participated in the present study. The sample consisted of 22 deaf children of hearing families and25 typically developing children with normal hearing. The deaf children ranged in age from 9 years, 0 months to 13 years, 11months (M = 11.32, SD = 1.34). This group consisted of 9 boys and 13 girls. They were profoundly deaf children with hearingloss in their left ears ranging from 90 dB to 113 dB (M = 100.41, SD = 7.85) and hearing loss in their right ears ranging from92 dB to 120 dB (M = 103.73, SD = 7.65). All the deaf children were from hearing families in which both of their parents hadnormal hearing. All the children became deaf before 12 months of age and before language acquisition. They used signlanguage as their predominant communication mode in everyday interaction. With the exception of deafness, they had noany known handicaps. Their scores on the Raven’s Standard Progressive Matrices (Chinese version) (Zhang & Wang, 1985)were within the normal range, indicating that their intelligence was normally developed. They were recruited from anelementary school for the deaf in Beijing.

The typically developing children with normal hearing ranged in age from 4 years, 0 month to 6 years, 11 months(M = 5.49, SD = 0.40). This group consisted of 12 boys and 13 girls. They had no known handicaps and were recruited from anordinary preschool in Beijing. The hearing children over 4 were selected as the control group for two reasons. First, hearingchildren pass standard false belief tasks at the age of approximately 4 (Wellman et al., 2001). Second, typically developingchildren over 4 are generally selected as the control group in previous studies on theory of mind development in deafchildren of hearing families (Moeller & Schick, 2006; Schick et al., 2007).

2.2. Tasks and scoring

The children’s general cognitive abilities were tested because theory of mind is closely related to language ability(Astington & Jenkins, 1999; Milligan, Astington, & Dack, 2007; Slade & Ruffman, 2005) and executive functions (Carlson,Moses, & Claxton, 2004; Gordon & Olson, 1998; Sabbagh, Xu, Carlson, Moses, & Lee, 2006). Three false belief tasks were usedto assess the children’s understanding of false beliefs under different cue conditions. Two emotion recognition tasks wereused to evaluate the children’s understanding of emotions with clear visual cues.

2.2.1. General cognitive ability tasks

2.2.1.1. Language ability. The vocabulary test in the Chinese Wechsler Young Children Scale of Intelligence(Gong & Dai, 1992) was used to test the children’s language abilities. The children were asked to interpret the meaningof each word according to the test’s word list. For the hearing children, the experimenter read each word to them andrecorded their answers. For the deaf children of hearing families, they were instructed to watch a video in which a teachersigned each word and asked them to give the corresponding meaning of the word. The teacher was a professionally trainedsign language translator in the deaf children’s school who was familiar with their communication mode andeveryday language. The teacher translated the deaf children’s answers to the experimenter, and their answers wererecorded. The vocabulary test consisted of 22 words, and the scores ranged from 0 to 44. Another rater, who was blinded tothe purpose of the study, rated the answers of 50% of the participants. The resulting inter-rater reliability was good(Kappa = 0.95).

2.2.1.2. Inhibitory control. Inhibitory control is one of the most important components of executive function and wasmeasured because successful mental state understanding requires inhibiting self-perspective (Apperly, Samson, &Humphreys, 2005). The tapping task (Luria, 1973) was used to test the children’s inhibitory control. The children were askedto tap the table with a pencil twice after the experimenter tapped once and to tap once after the experimenter tapped twice.The formal test began when the children succeeded in two consecutive practice trials. In the 14 test trials, the experimenterrandomly tapped the table once or twice, and the children’s responses were recorded. The inhibitory control scores rangedfrom 0 to 14.

J. Hao, Y. Su / Research in Developmental Disabilities 35 (2014) 2849–28572852

2.2.2. False belief tasks

All false belief tasks were presented through video clips. In all video clips, the protagonist placed an object into a box andthen covered the box.

2.2.2.1. False belief task without eye-gaze cues. The standard false belief task was based on Baron-Cohen, Leslie, and Frith’s(1985) adaption of Wimmer and Perner’s (1983) story task and served as the false belief task without eye-gaze cues. In thevideo, a female placed a toy biscuit into a red box and left the scene. In the absence of the female, a male transferred thebiscuit from the red box to a blue box. Then, the female returned and looked toward the camera. Children were then askedthe test question: ‘‘Where will the female look for the biscuit?’’ This question was intended to tap children’s understanding offalse beliefs. In addition, two control questions were presented to children in order to ensure that they memorized the storyinformation correctly. The two control questions included ‘‘Where is the biscuit really?’’ and ‘‘Where was the biscuit in thebeginning?’’ For each question, the corresponding pictures of the boxes were presented to the children, and they wererequired to make a choice. After they completed the first trial, the scenario was repeated with a new location for the biscuit.This resulted in three possible locations in the second trial. For each trial, the children were given a score of 1 only if theyanswered all the test and control questions correctly. Therefore, the scores for the understanding of false beliefs without eye-gaze cues ranged from 0 to 2.

2.2.2.2. False belief task with clear eye-gaze cues. This task was based on Pellicano and Rhodes’s (2003) experimentalparadigm and was used to assess the children’s understanding of false beliefs with clear eye-gaze cues. The scenario wassimilar to that of the standard false belief task except for its ending. In the video, a female placed a toy butterfly into a purplebox and left the scene. Subsequently, a male transferred the butterfly from the purple box to a red box. When the femalereturned, her eyes were directed to the purple box. The children were asked a test question and two control questions similarto those asked previously. In addition, they were asked about the visual cues of the protagonist’s false beliefs to guaranteetheir attention to the cues: ‘‘Which box did the female look at when she returned?’’ The children completed two trials withthe second one containing three possible locations. The scores for the understanding of false beliefs with clear eye-gaze cuesranged from 0 to 2.

2.2.2.3. False belief task with ambiguous eye-gaze cues. This task was adapted from Pellicano and Rhodes’s (2003)experimental paradigm and was used to examine the children’s understanding of false beliefs with ambiguous eye-gazecues. The scenario began with a male placing a toy banana into an orange box and leaving the scene. In the absence ofthe male, a female came in with a toy watermelon. She transferred the banana into a blue box and placed thewatermelon into the orange box. Then, the male returned with his eyes staring at the orange box. The male’s eye-gazewas an ambiguous cue of his false belief because he looked at the box where he believed the banana was but also thebox containing the undesired watermelon. In other words, he looked at the right location but at the wrong object. Afterwatching the video, the children were first asked the desire test question: ‘‘What will the male look for?’’ If the childrengave the correct answer of banana, they were then asked the false belief test question: ‘‘Where will the male look for thebanana?’’ If the children answered incorrectly with ‘‘watermelon,’’ they were then asked the false belief test question:‘‘If the male is asked to look for the banana, where will he look for it?’’ Moreover, a series of control questions wereasked to guarantee the children’s memories of the background information. The control questions included thefollowing: ‘‘What did the male place into a box,’’ ‘‘Where was the object in the beginning,’’ ‘‘Where is the object really,’’‘‘What did the female bring,’’ ‘‘Where did she place the object,’’ and ‘‘Which box did the male look at when he returned.’’There were two trials, and the second one involved three possible locations. The scores for the understanding of falsebeliefs with ambiguous eye-gaze cues and the scores for the understanding of desires with ambiguous eye-gaze cuesranged from 0 to 2.

2.2.3. Emotion recognition tasks

2.2.3.1. Emotion recognition task with clear facial expression cues. Based on Hosie et al.’s (1998) labeling task, the childrenwere required to recognize emotions from clear facial expressions. The stimuli consisted of four photographs of a female’sface from Wang and Markham’s (1999) study. The four photographs presented the facial expressions of happiness, anger,sadness, and fear. The experimenter presented each photograph to the children and asked them to name the emotionconveyed by the facial expression: ‘‘Is the female in the photo happy, angry, sad, or scared?’’ There were two test trials foreach facial expression. The scores for the recognition of emotions with clear facial expression cues ranged from 0 to 8.

2.2.3.2. Emotion recognition task with clear body expression cues. This task required children to recognize emotions from clearbody expressions. The stimuli consisted of four video clips of a female interacting with a balloon with her back to the camerato convey her emotions. Therefore, the female’s facial expressions were unavailable, and her body expressions were the onlycues for her emotions. For example, in one video clip, the female jumped up and threw a balloon into the air to express herhappiness. The four video clips presented the body expressions of happiness, anger, sadness, and fear. Each video clip lasted4 s. The validity of these body expressions was confirmed, with over 80% of raters judging that each of the body expressionsexpressed its corresponding emotion. Children were asked to name the emotion of the female after they watched each video

J. Hao, Y. Su / Research in Developmental Disabilities 35 (2014) 2849–2857 2853

clip: ‘‘Is the female in the video happy, angry, sad, or scared?’’ There were two test trials for each body expression. The scoresfor the recognition of emotions with clear body expression cues ranged from 0 to 8.

2.3. Procedure

Each child was tested individually in a quiet room. For the experimental sessions on the deaf children, a teacher in theschool assisted the experimenter in administrating all the tasks. The teacher was a professional sign language translatorfamiliar with the deaf children’s communication mode and everyday language in the school. The experimenter’s orallanguage was translated into sign language by the teacher. Therefore, the narrative and questions of each task wereunderstood well by the deaf children. The deaf children’s signed answers were also translated into oral language by theteacher.

The experiment was divided into two phases. The children were first tested for their general cognitive abilities. One weeklater, they were required to complete false belief tasks and emotion recognition tasks. The two groups of tasks wereadministered in randomized orders. Each child taking part in the experiment received a gift as acknowledgment.

3. Results

3.1. Children’s general cognitive abilities

The deaf and hearing children’s general cognitive abilities were compared. The group descriptive statistics of thelanguage ability and inhibitory control scores were summarized in Table 1. The deaf children’s language ability scores weresignificantly lower than those of the hearing children, t (45) = 3.41, p = 0.001. There were no significant differences ininhibitory control scores between the two groups, t (45) = 1.37, p = 0.178. These results showed that the deaf children wereseverely delayed in their language development. However, they were normally developed with respect to inhibitorycontrol.

3.2. Children’s understanding of false beliefs under different cue conditions

The full samples of deaf and hearing children were first compared in their understanding of false beliefs under differentcue conditions. The group descriptive statistics of the false belief understanding scores were shown in Table 1. A 2(group)� 3 (cue type) repeated measure ANOVA was conducted. There was no significant interaction effect of group� cuetype, F (2, 45) = 1.12, p = 0.330, h2 = 0.024. The main effect of cue type was also not significant, F (2, 45) = 0.89, p = 0.413,h2 = 0.019. However, a main effect of group was observed, F (1, 45) = 11.16, p = 0.002, h2 = 0.199. Overall, the deaf childrenscored significantly lower than the hearing children in the understanding of false beliefs, irrespective of the cue type. Inaddition, the deaf children performed worse than the hearing children when asked to infer desires with ambiguous eye-gazecues, t (45) = 2.89, p = 0.008. Taken together, these results indicated that the deaf children with language impairment haddifficulty in understanding false beliefs before controlling for the effect of language ability.

To match the two groups of children for language ability, Woolfe, Want, and Siegal’s (2002) procedure was used. Thechildren whose language ability scores were in the intersection of the original two groups’ scores were selected. Thelanguage-ability-matched samples consisted of 14 deaf children and 22 hearing children. The descriptive statistics of thelanguage ability and inhibitory control scores for the language-ability-matched samples were shown in Table 2. There wereno significant differences between the deaf and hearing children in language ability, t (34) = 0.65, p = 0.523. Their scores forinhibitory control were also similar, t (34) = 1.38, p = 0.186.

The children’s understanding of false beliefs under different cue conditions was analyzed again. The group descriptivestatistics of the false belief understanding scores for the language-ability-matched samples were shown in Table 2. A 2

Table 1

Mean scores and standard deviations (in parentheses) for each group on all the tasks: full samples.

Deaf children (n = 22) Hearing children (n = 25)

M (SD) M (SD)

General cognitive ability tasks

Language ability 8.45 (4.65) 12.68 (3.85)

Inhibitory control 12.36 (2.72) 13.20 (1.32)

False belief tasks

Understanding of false beliefs without eye-gaze cues 1.05 (0.90) 1.76 (0.66)

Understanding of false beliefs with clear eye-gaze cues 1.32 (0.84) 1.76 (0.66)

Understanding of false beliefs with ambiguous eye-gaze cues 1.14 (0.89) 1.84 (0.55)

Understanding of desires with ambiguous eye-gaze cues 1.45 (0.80) 1.96 (0.20)

Emotion recognition tasks

Recognition of emotions with clear facial expression cues 6.18 (1.82) 5.68 (1.89)

Recognition of emotions with clear body expression cues 7.50 (0.86) 7.40 (0.82)

Table 2

Mean scores and standard deviations (in parentheses) for each group on all the tasks: language-ability-matched samples.

Deaf children (n = 14) Hearing children (n = 22)

M (SD) M (SD)

General language ability tasks

Language ability 11.14 (3.55) 11.91 (3.42)

Inhibitory control 11.86 (3.28) 13.14 (1.39)

False belief tasks

Understanding of false beliefs without eye-gaze cues 0.93 (1.00) 1.73 (0.70)

Understanding of false beliefs with clear eye-gaze cues 1.50 (0.86) 1.73 (0.70)

Understanding of false beliefs with ambiguous eye-gaze cues 1.21 (0.89) 1.82 (0.59)

Understanding of desires with ambiguous eye-gaze cues 1.50 (0.76) 1.95 (0.21)

Emotion recognition tasks

Recognition of emotions with clear facial expression cues 6.29 (1.64) 5.45 (1.87)

Recognition of emotions with clear body expression cues 7.64 (0.63) 7.32 (0.84)

J. Hao, Y. Su / Research in Developmental Disabilities 35 (2014) 2849–28572854

(group)� 3 (cue type) repeated measure ANOVA was carried out. The results were shown in Fig. 1. It yielded a significantinteraction effect of group� cue type, F (2, 34) = 3.61, p = 0.032, h2 = 0.096. Simple effect analysis found that the deaf childrenscored significantly lower than the hearing children in understanding false beliefs without eye-gaze cues, F (1, 34) = 7.97,p = 0.008, h2 = 0.190. However, they were comparable with the hearing children in understanding false beliefs with clear eye-gaze cues, F (1, 34) = 0.76, p = 0.391, h2 = 0.022. When the eye-gaze cues were ambiguous, the deaf children performed poorlyagain, F (1, 34) = 6.02, p = 0.019, h2 = 0.150. The results showed that the deaf children were capable of understanding falsebeliefs when their visual cues were available and unambiguous. Moreover, the main effect of group was significant, F (1,34) = 5.46, p = 0.026, h2 = 0.138. A significant main effect of cue type was also found, F (2, 34) = 3.61, p = 0.032, h2 = 0.096. Inaddition to false beliefs, the language-ability-matched samples were compared in their understanding of desires withambiguous eye-gaze cues. The deaf children’s scores were still significantly lower than those of the hearing children, t

(34) = 2.19, p = 0.046. The results above demonstrated that after the effect of language ability was ruled out, the deaf childreninferred the clearly expressed false beliefs as accurately as the hearing children. However, they still had difficulty inunderstanding false beliefs with ambiguous or no visual cues.

3.3. Children’s recognition of emotions with clear visual cues

To confirm the clear cue-based mental state understanding in the deaf children, their recognition of emotions with clearvisual cues was compared to that of the hearing children. The descriptive statistics of the emotion recognition scores for thefull samples were shown in Table 1. A 2 (group)� 2 (cue type) repeated measure ANOVA was conducted. The results wereshown in Fig. 2. No significant interaction effect of group� cue type was found, F (1, 45) = 0.55, p = 0.464, h2 = 0.012. Therewas also no significant main effect of group, F (1, 45) = 0.88, p = 0.353, h2 = 0.019. The deaf and hearing children were similarin recognizing emotions with clear visual cues. However, the main effect of cue type was significant, F (1, 45) = 31.20,p< 0.001, h2 = 0.409. The children’s body expression-based emotion recognition scores were significantly higher than theirfacial expression-based emotion recognition scores. These results indicated that the deaf children were able to recognizeemotions with clear visual cues even if they were delayed in language ability. Consistent results were obtained for thelanguage-ability-matched samples.

[(Fig._1)TD$FIG]

Fig. 1. False belief understanding scores by cue type for the language-ability-matched samples. Error bars represent standard error.

[(Fig._2)TD$FIG]

Fig. 2. Emotion recognition scores by cue type for the full samples. Error bars represent standard error.

J. Hao, Y. Su / Research in Developmental Disabilities 35 (2014) 2849–2857 2855

4. Discussion

Theory of mind, as an important social cognitive ability, has received much attention. Deaf children of hearing familiesshow difficulty in standard false belief tasks until the age of approximately 13 (Peterson & Siegal, 1995; Russell et al., 1998),but typically developing 4-year-olds often succeed in this task (Wellman et al., 2001). The significant theory of mind delay indeaf children of hearing families is typically thought to exist mainly in understanding false beliefs (Courtin & Melot, 2005; deVilliers & de Villiers, 2012; Schick et al., 2007). However, the present study challenges this point of view and provides newfindings. After controlling for the effect of the children’s language abilities, the deaf children were capable of understandingfalse beliefs with clear eye-gaze cues but still had difficulty in understanding false beliefs with ambiguous or no eye-gazecues. It is accordingly implied that the deaf children’s theory of mind delay does not lie in understanding false beliefs per sebut in understanding a spectrum of mental states that are not expressed through clear visual cues. In other words, mentalstates are easily understood by the deaf children when they are expressed clearly through visual cues. Moreover, the deafchildren’s superb performance on emotion recognition tasks with clear facial or body expression cues confirms their clearcue-based mental state understanding in the present study.

Based on Pellicano and Rhodes’s (2003) experimental paradigm, the present study examined false belief understandingunder different cue conditions in deaf children of hearing families. The presentation of story scenarios through videosenacted by real people guaranteed the validity and salience of the eye-gaze cues. As the hypothesis predicted, the deafchildren were comparable to the hearing children in their performance on the false belief task with clear eye-gaze cues. Theyinclined to choose the box the protagonist looked at even though it did not contain an object. The deaf children’s goodperformance on this task indeed reflects their complete understanding of false beliefs rather than a mere associationbetween other people’s eye gaze and their behavior. If they simply formed this association, they would tend to choose thebox the protagonist looked at in all false belief tasks. However, the deaf children chose the box the protagonist looked at lessoften than the hearing children in the false belief task with ambiguous eye-gaze cues. The deaf children’s inconsistentperformance on the two tasks indicates that they do not simply associate other people’s eye gaze with the people’s behavior.In fact, under the clear eye-gaze cue condition, the deaf children first successfully inferred other people’s false beliefsthrough their clear eye-gaze direction and then predicted the people’s behavior. Under the ambiguous eye-gaze cuecondition, the deaf children were distracted by the ambiguous cues and failed to infer others’ false beliefs. Specifically,although the protagonist looked at the right box, the box contained an undesired object. Therefore, the deaf children tendedto think that the protagonist wanted the undesired object and thus looked at the box. When they were asked theprotagonist’s false belief about the location of the desired object, they incorrectly chose the box containing the desiredobject. As a result, only the clear eye-gaze cue helped the deaf children to infer the protagonist’s false belief. Consistent withthese results, Scott et al. (1999) found that most 9–12-year-old deaf children could infer others’ desires and intentions fromtheir clear line of regard.

The deaf children’s preference for visual cues of mental states may be related to their abilities to process visualinformation. Neville and Lawson (1987) showed that deaf adults detected the motion of visual stimuli in the peripheralvisual field better than hearing adults. Likewise, Stivalet, Moreno, Richard, Barraud, and Raphel (1998) found that deaf adultscould effectively detect the target stimuli from distracters when they were presented in the central visual field. Some studiesalso showed that deaf signers were more capable of accurately memorizing shapes and faces than hearing non-signers(Arnold & Mills, 2001; Cattani, Clibbens, & Perfect, 2007). Overall, deaf individuals’ excellent visual processing abilities mayaccount for their dependence on clear visual cues in mental state understanding. Deaf children of hearing families not onlyprefer visual cues of mental states, but also use clear visual cues effectively in inferring others’ false beliefs. In the presentstudy, they performed well on the false belief task with clear eye-gaze cues. It is because false beliefs with clear eye-gaze cuesare completely explicit. As long as the deaf children pay attention to the clear eye-gaze cues and grasp their meaning, the

J. Hao, Y. Su / Research in Developmental Disabilities 35 (2014) 2849–28572856

corresponding mental states are easily understood. However, in standard false belief tasks, false beliefs are not expressedand must be understood through complex reasoning. Previous studies demonstrate that children’s mental state reasoning instandard false belief tasks is related to their mental state discourse in families (Symons, 2004; Symons, Peterson, Slaughter,Roche, & Doyle, 2005). Deaf children of hearing families lack shared communication mode with their parents from earlyyears and they are delayed in language development (Schick et al., 2007). Thus, limited access to conversation about mentalstates may prevent deaf children of hearing families from developing mature theory of mind (Peterson & Siegal, 1995, 2000).

Another piece of evidence from emotion recognition in deaf children of hearing families further points to their use of clearvisual cues in mindreading. The deaf children performed well on the two emotion recognition tasks even though theirlanguage abilities were limited. Unlike clear cue-based false belief understanding, clear cue-based emotion recognitionseems to depend little on language ability. The differences may be explained by Tager-Flusberg and Sullivan’s (2000) view.These investigators proposed two components of theory of mind. The social-cognitive component involved ‘‘the conceptualunderstanding of the mind as a representational system’’ (Tager-Flusberg & Sullivan, 2000, p. 61) and was strongly related tolanguage ability. The social-perceptual component involved ‘‘on-line rapid judgments about people’s mental state from theirfacial and body expressions’’ (Tager-Flusberg & Sullivan, 2000, p. 62) and was less related to language ability. According tothis argument, false beliefs are a type of representative mental state and therefore cannot be understood without sufficientlanguage ability. Because emotion recognition is in the realm of person perception, it only requires limited language ability.In addition, the present study also found the effect of cue type in emotion recognition. The children performed better whenbody expression cues were presented. Body expressions convey emotions dynamically and thus may provide moreinformation of emotions than do facial expressions.

Theoretically, the two pieces of evidence together support a specific theory of mind mechanism for deaf children ofhearing families. Theory-based (Gopnik & Meltzoff, 1997; Gopnik & Wellman, 1994) or simulation-based (Harris, 1992)mental state understanding is typical of hearing children’s theory of mind mechanisms and are more applicable to implicitmental states. Specifically, the two mechanisms involve the inference of implicit mental states through a complex reasoningor simulation process with the ‘‘initial information about target other’’ (Apperly, 2008, p. 269). However, understanding ofimplicit and explicit mental states requires different theory of mind mechanisms (Hao, Su, & Chan, 2010). When mentalstates are completely explicit, clear cue-based mental state understanding is a more efficient way of mindreading. In thepresent study, clear cue-based mental state understanding is within the reach of deaf children of hearing families and hasbeen shown to be their considerable advantage. Therefore, deaf children of hearing families can be practically instructed topay careful attention to various visual cues of mental states, for instance, others people’ eye-gaze direction, pointingbehavior, facial expressions, etc. and to take full advantage of visual cues in understanding others’ mental states. In addition,the present study contributes to reconsideration of developmental progression in theory of mind. Wellman and Liu (2004)claimed that the progression was from understanding of diverse desires to understanding of false beliefs. The present studyclarifies why some of the mental states are more difficult to be understood than others, and suggests that children’sunderstanding of mental states with clear visual cues precedes their understanding of mental states with ambiguous or novisual cues.

5. Conclusion

In conclusion, the present study showed that deaf children of hearing families were capable of inferring false beliefs thatwere clearly expressed through eye-gaze cues after the effect of language ability was ruled out. Moreover, they were alsoable to recognize emotions that were clearly conveyed by facial or body expression cues. These findings suggest that for deafchildren of hearing families, clear cue-based mental state understanding may be their specific theory of mind mechanism.

Acknowledgements

This research was supported by the National Natural Science Foundation of China (Grant Nos. 31170995, 31371040) andNational Basic Research Program of Ministry of Science and Technology of China (973 Program: 2010CB833904). We aregrateful to the children who participated in our study and the teachers who provided assistance.

References

Apperly, I. A. (2008). Beyond Simulation-Theory and Theory-Theory: Why social cognitive neuroscience should use its own concepts to study theory of mind.Cognition, 107, 266–283.

Apperly, I. A., Samson, D., & Humphreys, G. W. (2005). Domain-specificity and theory of mind: Evaluating neuropsychological evidence. Trends in Cognitive Science,9, 572–577.

Arnold, P., & Mills, M. (2001). Memory for faces, shoes, and objects by deaf and hearing signers and hearing nonsigners. Journal of Psycholinguistic Research, 30, 185–195.

Astington, J. W., & Jenkins, J. M. (1999). A longitudinal study of the relation between language and theory-of-mind development. Developmental Psychology, 35,1311–1320.

Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a theory of mind? Cognition, 21, 37–46.Carlson, S. M., Moses, L. J., & Claxton, L. J. (2004). Individual differences in executive functioning and theory of mind: An investigation of inhibitory control and

planning ability. Journal of Experimental Child Psychology, 87, 299–319.Cattani, A., Clibbens, J., & Perfect, T. J. (2007). Visual memory for shapes in deaf signers and nonsigners and in hearing signers and nonsigners: Atypical

lateralization and enhancement. Neuropsychology, 21, 114–121.

J. Hao, Y. Su / Research in Developmental Disabilities 35 (2014) 2849–2857 2857

Courtin, C., & Melot, A.-M. (2005). Metacognitive development of deaf children: Lessons from the appearance-reality and false belief tasks. Developmental Science,8, 16–25.

de Villiers, P. A., & de Villiers, J. G. (2012). Deception dissociates from false belief reasoning in deaf children: Implications for the implicit versus explicit theory ofmind distinction. British Journal of Developmental Psychology, 30, 188–209.

Dyck, M. J., Farrugia, C., Shochet, I. M., & Holmes-Brown, M. (2004). Emotion recognition/understanding ability in hearing or vision-impaired children: Do sounds,sights, or words make the difference? Journal of Child Psychology and Psychiatry, 45, 789–800.

Gong, Y., & Dai, X. (1992). Chinese Wechsler Young Children Scale of Intelligence (C-WYCSI). Hunan: Map Press.Gopnik, A., & Meltzoff, A. N. (1997). Words, thoughts, and theories. Cambridge, MA: MIT Press.Gopnik, A., & Wellman, H. M. (1994). The theory theory. In L. A. Hirschfeld & S. A. Gelman (Eds.), Mapping the mind: Domain-specificity in cognition and culture (pp.

257–293). Cambridge, UK: Cambridge University Press.Gordon, A. C. L., & Olson, D. R. (1998). The relation between acquisition of a theory of mind and the capacity to hold in mind. Journal of Experimental Child

Psychology, 68, 70–83.Hao, J., Su, Y., & Chan, R. C. K. (2010). Do deaf adults with limited language have advanced theory of mind? Research in Developmental Disabilities, 31, 1491–1501.Harris, P. F. (1992). From simulation to folk psychology: The case for development. Mind and Language, 7, 120–144.Hosie, J. A., Gray, C. D., Russell, P. A., Scott, C., & Hunter, N. (1998). The matching of facial expressions by deaf and hearing children and their production and

comprehension of emotion labels. Motivation and Emotion, 22, 293–313.Lederberg, A. R., Schick, B., & Spencer, P. E. (2013). Language and literacy development of deaf and hard-of-hearing children: Successes and challenges.

Developmental Psychology, 49, 15–30.Ludlow, A., Heaton, P., Rosset, D., Hills, P., & Deruelle, C. (2010). Emotion recognition in children with profound and severe deafness: Do they have a deficit in

perceptual processing? Journal Of Clinical and Experimental Neuropsychology, 32, 923–928.Luria, A. R. (1973). Higher cortical functions in man. New York: Basic Books.Marschark, M., Green, V., Hindmarsh, G., & Walker, S. (2000). Understanding theory of mind in children who are deaf. Journal of Child Psychology and Psychiatry, 41,

1067–1073.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, 622–646.Moeller, M. P., & Schick, B. (2006). Relations between maternal input and theory of mind understanding in deaf children. Child Development, 77, 751–766.Neville, H. J., & Lawson, D. (1987). Attention to central and peripheral visual space in a movement detection task: An event-related potential and behavioral study.

II. Congenitally deaf adults. Brain Research, 405, 268–283.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.Pellicano, E., & Rhodes, G. (2003). The role of eye-gaze in understanding other minds. British Journal of Developmental Psychology, 21, 33–43.Peterson, C. C. (2002). Drawing insight from pictures: The development of concepts of false drawing and false belief in children with deafness, normal hearing, and

autism. Child Development, 73, 1442–1459.Peterson, C. C., & Siegal, M. (1995). Deafness, conversation and theory of mind. Journal of Child Psychology and Psychiatry, 36, 459–474.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., & Siegal, M. (2000). Insights into theory of mind from deafness and autism. Mind and Language, 15, 123–145.Peterson, C. C., & Slaughter, V. P. (2006). Telling the story of theory of mind: Deaf and hearing children’s narratives and mental state understanding. British Journal

of Developmental Psychology, 24, 151–179.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.Premack, D., & Woodruff, G. (1978). Does the chimpanzee have a theory of mind? Behavioral and Brain Sciences, 1, 515–526.Rhys-Jones, S. L., & Ellis, H. D. (2000). Theory of mind: Deaf and hearing children’s comprehension of picture stories and judgments of social situations. Journal of

Deaf Studies and Deaf Education, 5, 248–265.Russell, P. A., Hosie, J. A., Gray, C. D., Scott, C., Hunter, N., Banks, J. S., et al. (1998). The development of theory of mind in deaf children. Journal of Child Psychology

and Psychiatry, 39, 903–910.Sabbagh, M. A., Xu, F., Carlson, S. M., Moses, L. J., & Lee, K. (2006). The development of executive functioning and theory of mind: A comparison of Chinese and U.S.

preschoolers. Psychological Science, 17, 74–81.Schick, B., de Villiers, P., de Villiers, J., & Hoffmeister, R. (2007). Language and theory of mind: A study of deaf children. Child Development, 78, 376–396.Scott, C., Russell, P. A., Gray, C. D., Hosie, J. A., & Hunter, N. (1999). The interpretation of line of regard by prelingually deaf children. Social Development, 8, 412–426.Slade, L., & Ruffman, T. (2005). How language does (and does not) relate to theory of mind: A longitudinal study of syntax, semantics, working memory and false

belief. British Journal of Developmental Psychology, 23, 117–141.Stivalet, P., Moreno, Y., Richard, J., Barraud, P.-A., & Raphel, C. (1998). Differences in visual search tasks between congenitally deaf and normally hearing adults.

Cognitive Brain Research, 6, 227–232.Symons, D. K. (2004). Mental state discourse, theory of mind, and the internalization of self-other understanding. Developmental Review, 24, 159–188.Symons, D. K., Peterson, C. C., Slaughter, V., Roche, J., & Doyle, E. (2005). Theory of mind and mental state discourse during book reading and story-telling tasks.

British Journal of Developmental Psychology, 23, 81–102.Tager-Flusberg, H., & Sullivan, K. (2000). A componential view of theory of mind: Evidence from Williams syndrome. Cognition, 76, 59–89.Taumoepeau, M., & Ruffman, T. (2006). Mother and infant talk about mental states relates to desire language and emotion understanding. Child Development, 77,

465–481.Taumoepeau, M., & Ruffman, T. (2008). Stepping stones to others’ minds: Maternal talk relates to child mental state language and emotion understanding at 15,

24, and 33 months. Child Development, 79, 284–302.Wang, L., & Markham, R. (1999). The development of a series of photographs of Chinese facial expressions of emotion. Journal of Cross-cultural Psychology, 30, 397–

410.Want, S. C., & Gattis, M. (2005). Are late-signing deaf children mindblind? Understanding goal directedness in imitation. Cognitive Development, 20, 159–172.Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of theory-of-mind development: The truth about false belief. Child Development, 72, 655–684.Wellman, H. M., & Liu, D. (2004). Scaling of theory-of-mind tasks. Child Development, 75, 523–541.Wellman, H. M., & Peterson, C. C. (2013). Deafness, thought bubbles, and theory-of-mind development. Developmental Psychology, 49, 2357–2367.Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: Representation and constraining function of wrong beliefs in young children’s understanding of deception.

Cognition, 13, 103–128.Woolfe, T., Want, S. C., & Siegal, M. (2002). Signposts to development: Theory of mind in deaf children. Child Development, 73, 768–778.Zhang, H., & Wang, X. (1985). Raven standard progressive matrices: Chinese city revision. Beijing: Beijing Normal University Press.