How do preschoolers’ sharing behaviors relate to their theory of mind understanding?

Journal of Experimental Child Psychology 120 (2014) 73–86

Contents lists available at ScienceDirect

Journal of Experimental ChildPsychology

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

How do preschoolers’ sharing behaviors relate totheir theory of mind understanding?

0022-0965/$ - see front matter � 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.jecp.2013.11.007

⇑ Corresponding author. Fax: +86 10 62756460.E-mail address: [email protected] (Y. Su).

1 Current address: Department of Psychology, University of Iowa, Iowa City, IA 52242, USA.

Zhen Wu 1, Yanjie Su ⇑Department of Psychology, Peking University, Beijing 100871, China

a r t i c l e i n f o

Article history:Received 22 March 2013Revised 14 November 2013Available online 31 December 2013

Keywords:PreschoolersSharing behaviorTheory of mind understandingScaffoldingDiverse beliefsKnowledge access

a b s t r a c t

This study aimed to explore the development of resource sharingin preschoolers and its relationship with children’s theory of mind(ToM) understanding. A total of 74 2- to 4-year-old Chinese chil-dren participated in three tasks with toys that could be shared witha puppet (animated by a female experimenter). In each sharingtask, the puppet communicated her desire for children’s items witha series of progressively more explicit cues. Results showed that 2-and 3-year-olds relied on more explicit communicative cues toshare resources with others, whereas 4-year-olds shared morespontaneously. In addition, children’s ToM understanding was pos-itively correlated with their sharing behavior independent of theirage. Specifically, children who had acquired the ability to under-stand that people could have different beliefs about the same thingand that people were ignorant if they had not seen the fact sharedmore spontaneously and shared more items with their playmatethan children who had not acquired these two abilities. Findingssuggest that preschoolers’ sharing behavior is enhanced by theirToM understanding and explicit communicative cues provided bythe playmate.

� 2013 Elsevier Inc. All rights reserved.

Introduction

Sharing is defined as a social interaction in which ‘‘the individual holder of a thing grants to an-other the partial use, enjoyment, or possession of a thing, though it may merely imply the mutualuse, enjoyment, or possession’’ (Merriam–Webster, 1969, p. 1082). Sharing seems to have its roots

http://crossmark.crossref.org/dialog/?doi=10.1016/j.jecp.2013.11.007&domain=pdf

http://dx.doi.org/10.1016/j.jecp.2013.11.007

mailto:[email protected]

http://dx.doi.org/10.1016/j.jecp.2013.11.007

http://www.sciencedirect.com/science/journal/00220965

http://www.elsevier.com/locate/jecp

74 Z. Wu, Y. Su / Journal of Experimental Child Psychology 120 (2014) 73–86

very early in human life, with studies indicating that infants as young as 8 months showed sponta-neous offering of resources to parents (e.g., Hay, 1979; Hay & Murray, 1982; Rheingold, Hay, & West,1976). Although it emerges early, it appears to be a unique challenge for young children. Sharing isinfrequent unless it is requested by a potential recipient (Birch & Billman, 1986; Brownell, Iesue,Nichols, & Svetlova, 2013; Brownell, Svetlova, & Nichols, 2009; Dunfield, Kuhlmeier, O’Connell, &Kelley, 2011; Hay, Castle, Davies, Demetriou, & Stimson, 1999; Levitt, Weber, Clark, & McDonnell,1985). Even during preschool years, spontaneous sharing of valued resources remains difficult forchildren, and it occurs much less frequently compared with other cooperative activities such asempathy-related responding, helping, and instrumental collaboration (Eisenberg, 2005; Grusec,1991; Warneken & Tomasello, 2007). Children younger than 5 years tend to behave selfishly, sharingvery little with others while keeping the majority of the resources for themselves (e.g., Benenson,Pascoe, & Radmore, 2007; Birch & Billman, 1986; Blake & Rand, 2010; Fehr, Bernhard, & Rockenbach,2008; Rochat et al., 2009; Shaw & Olson, 2012; Thompson, Barresi, & Moore, 1997). As they approachschool years, children are more willing to share resources with others (Benenson et al., 2007; Fehret al., 2008; Gummerum, Hanoch, Keller, Parsons, & Hummel, 2010; Rochat et al., 2009; Thompsonet al., 1997). In fact, by around 7 years of age, children of Western middle-class families share half oftheir resources with peers (Blake & McAuliffe, 2011; Fehr et al., 2008). Even more impressive, chil-dren in societies that promote traditional collective values, such as China, start sharing resourcesspontaneously and equally with peers by around 5 years of age (Rao & Stewart, 1999; Rochatet al., 2009).

Numerous studies have been conducted on developmental changes in children’s sharing behavior,yet we know little about the underlying mechanisms of sharing. Current theories have proposed var-ious underlying processes such as the basic affiliative and imitative tendencies toward people (Grusec,2006; Hay & Cook, 2007; Rheingold, 1982), the ability to differentiate one’s own and others’ internalstates (e.g., Moore, 2007), the sympathetic ability to relate one’s own emotions and feelings to others(e.g., Eisenberg, 2007; Zahn-Waxler & Radke-Yarrow, 1990), the understanding of ownership (Brow-nell, Iesue, et al., 2013), and an innate biological predisposition for empathy and altruism in infants(Tomasello, 2008; Warneken & Tomasello, 2009; Zahn-Waxler, Robinson, & Emde, 1992). These differ-ent theoretical perspectives emphasize social–cognitive and motivational components of early proso-cial responding at different levels. Yet they agree, to different extents, that the origin of humans’altruistic prosocial behavior is in compassion—that is, understanding and caring about others’ needsor emotions (e.g., Batson, Duncan, Ackerman, Buckley, & Birch, 1981)—in combination with universalnorms of reciprocity and fairness (e.g., Fehr & Fischbacher, 2004; Lane, Wellman, Olson, LaBounty, &Kerr, 2010; Singer & Steinbeis, 2009).

This understanding of others’ needs and emotions may be manifested in theory of mind (ToM)understanding, that is, attributing mental states such as desires, intentions, beliefs, and needs tooneself and others (e.g., Premack & Woodruff, 1978). Sharing increases dramatically in preschoolers(e.g., Benenson et al., 2007; Blake & Rand, 2010; Fehr et al., 2008; Rochat et al., 2009) at the sametime that ToM understanding is also undergoing significant developmental change (e.g., Wellman,Fang, Liu, Zhu, & Liu, 2006; Wellman & Liu, 2004). With the development of the ability to monitorand understand each other’s intentions, desires, beliefs, emotions, and other mental states, onemay become progressively better at taking others’ perspectives into account and regulating one’sactivities in concert with others’ activities; hence, one may grow to be more concerned with others’welfare (Fehr et al., 2008; Tomasello & Carpenter, 2007). The main goal of the current study was totest this hypothesis by investigating how children’s ToM understanding associates with their shar-ing behavior.

Previous studies have shown that preschoolers’ scores on ToM tasks negatively predict subsequentaggressive or disruptive behavior for boys and positively predict prosocial behavior for girls after con-trolling for age (Walker, 2005). In addition, more advanced ToM understanding related to more fairoffers in the Ultimatum Game (Takagishi, Kameshima, Schug, Koizumi, & Yamagishi, 2010). In the Ulti-matum Game, a participant is given a monetary endowment and is asked to make a decision regardinghow to divide the endowment between himself or herself and a responder. If the responder acceptsthe offer, each one will receive the payment according to the proposer’s offer; if the responder rejects

Z. Wu, Y. Su / Journal of Experimental Child Psychology 120 (2014) 73–86 75

the offer, they both get nothing. Five-year-old children who passed a ToM task were more likely topropose fair offers than children who failed a ToM task (Takagishi et al., 2010).

These prior studies suggest that ToM understanding relates to prosocial behavior (Sally & Hill,2006), and they mainly used a false belief paradigm to assess children’s ToM. However, besides falsebelief, achieving a ToM requires the understanding of multiple other concepts such as intentions,desires, true beliefs, and hidden emotions (Harris, de Rosnay, & Pons, 2005; Wellman & Liu, 2004;Wellman et al., 2006). Yet we know little about how younger children’s understanding of these mul-tiple mental concepts relates to their prosocial behavior. There is a robust finding that children donot pass false belief tasks until around 4 years of age (for a review, see Baillargeon, Scott, & He,2010), and the understanding of intentions, desires, and true beliefs is shown to be acquired priorto the understanding of false beliefs (Harris et al., 2005; Wellman & Liu, 2004; Wellman et al.,2006). Therefore, in the current study, we investigated the relationship between sharing and ToMin children between 2 and 4 years of age and measured children’s multiple aspects of ToM under-standing, including desires, knowledge states, beliefs, and real–apparent emotions (Wellman &Liu, 2004; Wellman et al., 2006).

To measure children’s sharing behavior, we varied the expressive cues provided by a partner suchthat the partner made her wants of children’s toys progressively more explicit using a fixed sequenceof cues. We studied at what point children shared items and how many items they shared. This vary-ing cues method allowed us to explore how others’ expression of desires, wants, and needs influenceschildren’s sharing (see also Brownell, Iesue, et al., 2013; Brownell et al., 2009). Previous studies haveshown that the majority of Chinese children at around 5 years of age share nearly half of their re-sources spontaneously without any cues from the recipient (Rao & Stewart, 1999; Rochat et al.,2009). Thus, to avoid reaching a ceiling effect of sharing behavior in the ‘‘varied cues’’ paradigm, wefocused on 2- to 4-year-olds with the aim of investigating how the early sharing behavior was relatedto the development of ToM understanding.

By varying the communications from a social partner, previous studies have shown that 18-month-old infants shared only when the recipient directly asked for items and when they saw a demonstra-tion of sharing (Brownell, Iesue, et al., 2013). Likewise, 2-year-old children were likely to shareresources with an adult when the adult partner communicated directly about what she wanted(Brownell, Iesue, et al., 2013; Brownell et al., 2009) or when the adult partner reached toward the childwith her palm up while alternating her gaze between the child and the desired item (Dunfield,Kuhlmeier, O’Connell, & Kelley, 2011). Similarly, 29- to 36-month-old children did not share with apeer who had no toys unless their mothers exhorted them to do so (Levitt et al., 1985). These studiessuggest that explicit communication from the adult partner may draw young children’s attention to,and/or increase interest in, others’ needs or wants, thereby facilitating children’s sharing. Therefore,we hypothesized that children with more advanced ToM understanding would rely on less explicitsupports or scaffolding from the social partner to perform sharing behavior compared with childrenwith less advanced ToM understanding.

In addition to the cues provided for the child, we also varied the number of items available to thechild. This was done to examine the extent to which children’s sharing behavior is affected by theamount of resources available and to measure how ToM relates to sharing behaviors in these variouscontexts. There are two possible results regarding how the amount of resources would influence chil-dren’s sharing. One is that children might be less likely to share when there are fewer objects presentbecause there are more potential costs for sharing when resources are fewer (Svetlova, Nichols, &Brownell, 2010; Thompson et al., 1997). However, another possibility is that young children may beconcerned only with the absolute number of items shared and not the proportion. A recent studyfound that 4-year-old children judged the puppet that gave a greater absolute amount of resourcesas being nicer both when it gave the same proportion as and when it gave a smaller proportion thanthe other puppet (McCrink, Bloom, & Santos, 2010). These results implied that 4-year-olds might careonly about absolute amounts and not proportions in resource distribution (McCrink et al., 2010). Inthis sense, children may share the same amount with others no matter how many items they havein total. These two contrasting possibilities were tested by varying the number of toys given to chil-dren in the current study.


Method

Participants

A total of 74 Chinese children from two kindergartens in Beijing, China, participated in the study.Both kindergartens largely served children of university staff and faculty in urban Beijing. Based ontheir kindergarten classes, there were 25 2-year-olds (M = 28.86 months, SD = 3.16, range = 24.46–34.43, 14 girls and 11 boys), 25 3-year-olds (M = 39.49 months, SD = 2.21, range = 34.92–42.69, 12girls and 13 boys), and 24 4-year-olds (M = 47.30 months, SD = 2.51, range = 43.02–51.08, 14 girlsand 10 boys). An additional 38-month-old boy was excluded from the study due to experimentererror.

Materials and procedure

Each child participated in both sharing tasks and ToM tasks in a quiet room adjacent to the child’sclassroom. Whether the sharing tasks or the ToM tasks were first was counterbalanced across chil-dren. All of the sessions were video-recorded.

Sharing tasksTesting was conducted by a female experimenter (E) animating a hand puppet (a brown bear)

named ‘‘Maomao’’ with the help of a female assistant (AE). Three sharing tasks were administeredwith the order counterbalanced across participants. Each task featured a different number of itemsthat could be shared (items were shown to be equally liked by children in a pilot study): a sticker task(two different stickers of similar patterns), a watermelon task (two slices of toy watermelon and twotoy knives), and a bead task (four colorful beads and two strings). These resources were chosen to befamiliar to the children and to be similar to resources they would typically play with. The variety ofresources was used so as to elicit variation in sharing behavior, maintain children’s interest acrossthese three tasks, and provide as comprehensive a picture of sharing behavior as possible.

During the test, E sat across a child-sized table from the child, and AE sat by one side of the table ata 90 degree angle to the child and the puppet. After a familiarization phase to ensure that the childwas comfortable, AE brought out toys (e.g., four beads and two strings in the bead task), introducedthem to the child and Maomao, and showed them how to play with these toys (e.g., put a stringthrough beads to make a necklace). Then, AE said that she needed to leave because she had to dosomething. She said to the child, ‘‘Could you please take care of these toys while I am gone? I’ll be backsoon. You can play with them by yourself or with your new friend Maomao. Thanks. Bye!’’ She thenleft the room.

After AE left, E provided three progressively more explicit cues about her wants and desires. Thecues were presented during three phases as follows. First, during the commenting phase, when thechild was exploring toys, E positively commented on the toys, for example, ‘‘These are so cool!’’ Sherepeated this comment twice, pausing for approximately 10 s between each comment. Second, duringthe desiring phase, E expressed her desires for these toys, for example, ‘‘I like these toys! I have none. Iwant to play with them.’’ This was also repeated two times. Third, during the requesting phase, E madean explicit request by asking the child for the toys, for example, ‘‘Would you please give me some toplay with?’’ Again this was repeated two times with 10 s between each comment. If the child shared atany point, E discontinued the cues, thanked the child, played with the toy(s) for approximately 10 s,and then signaled AE that the task ended by knocking the table (pretending that it was an accident). Ifthe child did not share in the final phase, E signaled AE in the same way.

ToM understanding tasksThese tasks were modeled after the Chinese version of the five core ToM understanding tasks

(Wellman & Liu, 2004; Wellman et al., 2006). They were used to measure children’s understandingthat two people (the child and someone else) may have different desires (Diverse Desires task), beliefs(Diverse Beliefs task), and knowledge about the same objects (Knowledge Access task); that another


person might have a false belief (Contents False Belief task); and that a person can feel one thing butdisplay a different emotion (Real–Apparent Emotion task). Although formats and general ideas werenot different from those in the tasks used by Wellman and colleagues (2006), task materials weremodified in several places so that they would be familiar and appropriate for our sample. The DiverseDesires task compared preferences for an apple versus a pear (rather than ice cream vs. an egg), theDiverse Beliefs task compared beliefs about a car in a schoolbag versus a drawer (rather than a bedvs. a cupboard), and the Contents False Belief task used a yolk pie box familiar to children in our sam-ple (rather than a potato chip tube). See the task descriptions below for more details.

A small toy figure with Chinese visages and dark hair, whose name was ‘‘Feifei,’’ served as the tar-get protagonist for the tasks.

Diverse Desires. The participant was presented with pictures of an apple and a pear and was asked topick the one that he or she liked better. Then the participant was told that Feifei likes X better, where Xwas always the opposite of the child’s answer. The subsequent question for the child was which oneFeifei would choose if she was hungry.

Diverse Beliefs. The participant was presented with pictures of a schoolbag and a drawer along with apicture of a car and was told that Feifei was looking for her car. Then the child was asked to choosewhere (schoolbag vs. drawer) he or she believed the car was. E then stated that Feifei thinks thatthe car is in X, where again X was always the opposite of the child’s answer. The subsequent targetquestion asked of the child was where Feifei was going to search for her car.

Knowledge Access. The participant saw a box with a fork inside. E then told the child that Feifei hadnever opened this box before. The target question was whether Feifei knew what was inside the box.

Contents False Belief. E presented the participant with a box with pictures of cookies on it. E then askedthe child what he or she thought was inside the box. E then showed the child that it was actually asmall pencil inside. After showing the child the real contents of the box, she told the child that Feifeihad never opened this box. The target question was what Feifei thought was inside the box.

Real–Apparent Emotion. The participant saw a sheet of paper with three faces on it: a happy face, aneutral face, and a sad face. After ensuring that the child understood these emotional expressions, Etold the child a story about a boy who was expecting a toy gun as his birthday gift but actually gota boring book instead. The boy did not want to behave impolitely, so he decided to hide his feelings.The participant was then asked how this boy really felt and how he tried to appear to others by point-ing to the pictures with faces.

As in previous studies (Wellman & Liu, 2004; Wellman et al., 2006), each child participated in allfive tasks presented in one of two orders. In both orders, the Diverse Desires task was presented first(to help children warm up with a task that has been shown to be the easiest), the Contents False Belieftask was presented next to last, and the Real–Apparent Emotion task (shown to be the hardest) waspresented last. The order of the Diverse Beliefs task and the Knowledge Access task wascounterbalanced.

Coding and scoring

Sharing was scored if the child actively gave an item to Maomao by pushing it toward her and plac-ing it within her reach or by holding it out to Maomao and depositing it in her hand (see also Brownell,Iesue, et al., 2013). Each child received a sharing score from 0 to 3 for each task, corresponding to thephase during which sharing occurred (0 = did not share at all, 1 = shared during the requesting phase,2 = shared during the desiring phase, 3 = shared during the commenting phase. Thus, higher scores indi-cated more spontaneous sharing with less explicit cues from the recipient. Moreover, because less ex-plicit cues came first, higher scores also indicated quicker sharing behavior. In addition to the sharingscore, the number of items shared by each child in each task was also recorded, and the proportion ofitems shared was used as a second dependent variable. For ToM understanding tasks, children got 1


point if they passed one ToM understanding task, so that the range for scores on the ToM understand-ing scale was 0 to 5.

The tasks were video-recorded and coded by the first author. Another coder who was blind to theresearch goal coded 25% of the videos. Cohen’s kappa was computed to measure interrater reliability.Values for Cohen’s kappa were 1.00 for the sticker and watermelon tasks, as well as for the ToM task,and .98 for the bead task.

Results

Preliminary analysis found that the performance on the tasks was not influenced by gender or taskorder. Therefore, data were collapsed across these factors in the following analyses. Below we first re-port the age and task differences in resource sharing and then present how sharing behavior relates toToM understanding.

Age and task differences in sharing behavior

When do children share?The descriptive data for children’s sharing behavior is shown in Table 1. A 3 (Age: 2, 3, or 4 years;

between-participant) � 3 (Task: watermelon, sticker, or bead; within-participant) mixed-design anal-ysis of variance (ANOVA) was conducted with children’s sharing score as the dependent variable. Re-sults showed a significant main effect of age, F(2,71) = 6.38, p = .003, partial g2 = .15, and a significantmain effect of task, F(2,142) = 20.36, p < .001, partial g2 = .22, but not a significant interactive effect ofage and task, F(4,142) = 1.84, p = .13, partial g2 = .05. Bonferroni-corrected post hoc tests showed that4-year-olds (M = 2.40, SD = 0.52) had significantly higher sharing scores than 2-year-olds (M = 1.73,SD = 0.84) and 3-year-olds (M = 1.77, SD = 0.79), ps = .006 and .011, respectively, yet there was no sig-nificant difference between children of the two younger groups, p = 1.00. In addition, children had sig-nificantly lower sharing scores in the bead task (M = 1.53, SD = 0.10) than in the watermelon task(M = 2.07, SD = 0.88) and the sticker task (M = 2.30, SD = 1.11), ps < .001, with no significant differencesfound between the latter two tasks, p = .21.

In addition, chi-square analyses were conducted to detect age differences in how many childrenshared at each cue level. Significant age differences were found for communicative cues in the beadtask, v2(6) = 21.20, p = .002, but not in the sticker or watermelon tasks, v2(6) = 2.94 and 9.26,ps = .82 and .16, respectively. Specifically, in the bead task, 33% of 4-year-olds shared immediatelyat the first cue (commenting), whereas only 8% of 2-year-olds and 4% of 3-year-olds shared at the firstcue, v2(2) = 9.73, p = .008; moreover, all 4-year-olds shared in the bead task, whereas only 72% of 2-year-olds and 64% of 3-year-olds shared, v2(2) = 10.27, p = .006 (see Table 1).

In sum, the results suggested that older children required fewer explicit cues from the social part-ner to perform sharing behavior; this age effect was most apparent in the bead task compared with theother two tasks.

How much do children share?In the sticker task, 13.51% of children shared none, 74.32% of children shared a half, and 12.16% of

children shared more than a half. In the watermelon task, 5.4% of children shared none, 31.1% of chil-dren shared less than a half, 62.2% children shared a half, and 1.4% of children shared more than a half.In the bead task, 21.6% of children shared none, 47.3% of children shared less than a half, 24.3% of chil-dren shared a half, and 6.8% of children shared more than a half. Thus, in each task, children were cat-egorized into two groups: (a) shared none or less than a half and (b) shared a half or more than a half.Chi-square tests showed that there were no significant age differences in the proportions of childrenwho fell into these two groups in any task, v2(2) = 1.50, 1.19, and 1.95, ps = .47, .55, and .38, for thesticker, watermelon, and bead tasks, respectively. In addition, a related-samples Cochran’s Q testshowed that the more resources children had (two, four, or six items), the more children shared lessthan a half, Q(2) = 53.04, p < .001.

Table 1Means (and standard deviations) for sharing as a function of age and task.

Sticker Watermelon Bead

Age (years): 2 3 4 2 3 4 2 3 4

Sharing score (0–3) 2.16(1.25)

2.24(1.09)

2.50(0.98)

1.84(0.94)

1.88(0.93)

2.50(0.59)

1.20(0.96)

1.20(1.00)

2.21(0.66)

Number of items shared 0.92(0.57)

1.00(0.50)

1.04(0.46)

1.56(0.71)

1.64(0.64)

1.58(0.50)

1.84(1.93)

1.48(1.50)

2.17(0.82)

Proportion items shared .46(.29)

.50(.25)

.52(.23)

.39(.18)

.41(.16)

.40(.13)

.31(.32)

.25(.25)

.36(.14)

Proportion children shared duringcommenting phase

.64 .60 .75 .28 .28 .54 .08 .04 .33

Proportion children shared duringdesiring phase

.08 .16 .08 .36 .40 .42 .32 .48 .54

Proportion children shared duringrequesting phase

.08 .12 .08 .28 .24 .04 .32 .12 .04

Proportion children shared none .20 .12 .08 .08 .08 0 .28 .36 0

Note: Standard deviations are in parentheses.


In addition, a mixed-design ANOVA was conducted with age and task as the independent variablesand the proportion of children’s shared items as the dependent variable. The age effect was not signif-icant, F(2,71) = 0.46, p = .63, partial g2 = .01, nor was the interactive effect of age and task,F(4,142) = 0.10, p = .41, partial g2 = .03; however, there was a significant main effect of task,F(2,142) = 19.65, p < .001, partial g2 = .22. Bonferroni-corrected post hoc tests showed that the propor-tion of children’s shared items was smallest in the bead task (M = .30, SD = .25), followed by the water-melon task (M = .40, SD = .15), and was highest in the sticker task (M = .49, SD = .25), ps < .01. Furtheranalyses showed that, overall, the mean proportion of items shared across these three tasks (M = .40,SD = .17) was significantly below the level of equal split of resources (50/50), t(73) = �5.20, p < .001.Specifically, the proportion of items shared in the sticker task was not significantly different froman equal split of resources (50/50), t(73) = 0.23, p = .82, whereas the proportions of items shared inthe other two tasks were significantly below the equal split level, t(73) values = �5.65 and �6.73for the watermelon and bead tasks, respectively, ps < .001.

Together, these results suggest that there were no significant age differences in the amount ofitems shared by children. Overall, 2- to 4-year-olds tended to behave selfishly and shared less thana half with others (except when they were given two stickers). In addition, children shared a smallerproportion of items when there were more items available to them.

What do children share?Because there were two watermelons and two knives in the watermelon task and there were four

beads and two strings in the bead task, further analyses were conducted to see whether childrenshared the items in groups (e.g., one watermelon with a knife). Children were categorized into threegroups in the watermelon task and then the bead task separately: (a) did not share at all, (b) sharedonly one kind of item (e.g., watermelon or knife), and (c) shared both kinds of items (see Table 2). Chi-square tests showed that there were no age differences in whether children shared one or two kinds ofitems in the watermelon task, v2(2) = 4.36, p = .11, nor in the bead task, v2(2) = 1.23, p = .54; this

Table 2Percentages of children shared none, one type, or two types of items as a function of age.

Age (years): Watermelon Bead

2 3 4 2 3 4

Proportion children shared none .08 .08 .00 .28 .36 .00Proportion children shared one type of item .32 .16 .46 .36 .20 .42Proportion children shared two types of items .60 .76 .54 .36 .44 .58


suggested that, across all age groups, children did not always share items in groups even after seeingthe experimenter play with two types of toys together (e.g., cutting a watermelon with a knife). Nota-bly, in the watermelon task, 22 of 23 children who shared one type of object actually shared just anindividual item, and 46 of 47 children who shared two items actually shared one watermelon and oneknife. In the bead task, 17 of 24 children who shared one type of object actually shared just an indi-vidual item, 12 of 19 children who shared two items actually shared one bead with one string, and 17of 17 children who shared three items actually shared two beads with one string. These results suggestthat 2- to 4-year-old children rarely gave out all of the individual items of a kind (e.g., all of the beads,keeping the string); either they gave just one individual item or, if they decided to share more thanone item, they shared items of different kinds. This means that for each kind of toy, they kept at leastone to play with.

Age differences in ToM understanding

The age effect was significant in all ToM measures except scores of the Diverse Desires and Real–Apparent Emotion tasks (see Table 3 for descriptive data and ANOVA test results). The results suggestedthat there might be a ceiling effect in the performance on the Diverse Desires task and a floor effect in theperformance on the Contents False Belief and Real–Apparent Emotion tasks in the current sample.

Relations between ToM understanding and the growth of sharing

Because the age effect was found in both the sharing and ToM understanding tasks, partial corre-lation analyses were conducted to test whether sharing was associated with ToM understanding aftercontrolling for age. As presented in Table 4, the total ToM score was positively correlated to children’s

Table 3Mean scores (and standard deviations) and ANOVA tests on the ToM tasks.

Age(years)

DiverseDesires

DiverseBeliefs

KnowledgeAccess

Contents FalseBelief

Real–ApparentEmotion

TotalToM

2 0.92 (0.28) 0.52 (0.51) 0.08 (0.28) 0 (0) 0 (0) 1.52 (0.77)3 0.96 (0.20) 0.76 (0.44) 0.40 (0.50) 0.04 (0.20) 0 (0) 2.16 (0.80)4 0.96 (0.20) 0.83 (0.38) 0.67 (0.48) 0.33 (0.48) 0.08 (0.28) 2.88 (1.03)F(2, 71)a 0.24 3.33 11.41 9.09 2.18 14.73Partial g2 .01 .09 .24 .20 .06 .29p .79 .04 .00 .00 .12 .00

Note: Standard deviations are in parentheses.a ANOVA tests with scores on ToM tasks as dependent variables and age as the independent variable.

Table 4Relationships between ToM and sharing after controlling for age (N = 74).

Sharing score Proportion of items shared

Watermelon Sticker Bead Mean Watermelon Sticker Bead Mean

Total ToM .26* .28* .30** .36** .24* .21+ .31** .34**

Diverse Desires �.07 .11 .04 .04 .13 .11 .17 .18Diverse Beliefs .30** .15 .24* .29* .15 .27* .22+ .29*

Knowledge Access .15 .25* .20+ .26* .23* .11 .22* .23*

Contents False Belief .12 .09 .13 .14 .04 �.02 .09 .05Real–Apparent Emotion .03 .09 .09 .09 �.03 �.01 .07 .02

+ p < .10.* p < .05.** p < .01.


sharing score in each individual sharing task, as well as the mean sharing score across three sharingtasks, with rs ranging from .26 to .36 and all ps < .05. In addition, the total ToM score was also posi-tively associated with the proportion of items shared in each individual task (except the sticker task,which was marginally significant, r = .21, p = .07) and the mean proportion of items shared across allthree tasks, with rs ranging from .24 to .34 and ps < .05. These results suggest that children’s ToMunderstanding relates to their sharing behaviors in various contexts.

Moreover, the correlational results suggested that sharing measures had robust associations withchildren’s performance on the Diverse Beliefs and Knowledge Access tasks but not on the other threeToM tasks (see Table 4). To further examine the effect of children’s performance on the Diverse Beliefsand Knowledge Access tasks on sharing, analysis of covariance (ANCOVA) tests were conducted withthe mean sharing score or mean proportion of items shared over three sharing tasks as the dependentvariable and age as the covariate. After confirming the homogeneity-of-regression assumption, theANCOVA analyses showed that, after controlling for age, children who passed the Diverse Beliefs task(npass = 52) shared with fewer cues than children who failed (nfail = 22), F(1,71) = 6.39, p < .01, partialg2 = .08, as did children who passed the Knowledge Access task (npass = 28) compared with childrenwho failed (nfail = 46), F(1,71) = 5.34, p < .05, partial g2 = .07 (see Fig. 1A). In addition, children whopassed the Knowledge Access task shared more items overall across the three sharing tasks than chil-dren who failed, F(1,71) = 4.06, p < .05, partial g2 = .05, and the difference between children whopassed the Diverse Beliefs task and those who failed was also significant, F(1,71) = 6.55, p < .05, partialg2 = .08 (see Fig. 1B). By contrast, none of the age effects was significant, ps > .05.

To further examine the effect of ToM understanding on children’s sharing behavior, we conducted aseries of multiple regression analyses. We first regressed age on the mean sharing score and found thatage had a significant effect on the sharing score (b = .30, p = .01), but the model explained only 8.7% of thevariance in the sharing score, F(1,72) = 6.87, p = .01. This age effect became nonsignificant (b = .06,p = .65) when the total ToM score was added as another independent variable. The effect of ToM was sig-nificant (b = .42, p = .002) and increased the variance accounted for by 11.8%, Fchange(1,71) = 10.55,p = .002. After adding ToM as the predictor, the full model explained 20.5% of the variance in sharingscore, F(2,71) = 9.17, p < .001.

0

0.5

1

1.5

2

2.5

3

Diverse Beliefs Knowledge Access

Mea

n sh

arin

g sc

ore

PassFail

***

00.10.20.30.40.50.60.70.80.9

1

Diverse Beliefs Knowledge Access

Mea

n pr

opor

tion

of it

ems s

hare

d

PassFail

* *

A

B

Fig. 1. Mean sharing scores (A) and proportions of shared items (B) over three sharing tasks by children who passed or failed inthe Diverse Beliefs and Knowledge Access tasks. ⁄p < .05; ⁄⁄p < .01.


We conducted another regression analysis with the mean proportion of items shared as the depen-dent variable. Results showed that age alone had no significant effect on the number of items shared(b = .06, p = .62), and the model explained only 0.3% of the variance in the proportion of shared items,F(1,72) = 0.25, p = .62. After adding ToM as the independent variable, the full model explained 11.5% ofthe variance in the proportion of items shared, F(2,71) = 4.62, p = .01. The effect of ToM was significant(b = .41, p = .004) and increased the variance accounted for by 11.2%, Fchange(1,71) = 8.97, p = .004. Theresults of the above regression analyses indicate that preschoolers’ performance on the ToM tasks,rather than age per se, was clearly related to both the mean sharing score and the mean proportionof items shared over three sharing tasks.

Discussion

In this study, we examined developmental differences in preschoolers’ resource sharing and theirrelation to children’s theory of mind understanding. We found that children who were around 4 yearsof age required fewer explicit communicative cues to share than 2- and 3-year-olds. The 4-year-oldsgenerally shared when the partner was just commenting on the toys, whereas the majority of theother two groups of children shared only when the partner verbally requested the items. This resultsuggests that older children needed less scaffolding from a social partner to perform sharing behavior(Brownell, Iesue, et al., 2013; Brownell et al., 2009; Svetlova et al., 2010). It also implies that older chil-dren might have a clearer intention to benefit others because they shared spontaneously and quicklywith a playmate who had none, whereas younger children might be likely to share under pressure orto comply with others’ requests (e.g., Hay, Caplan, Castle, & Stimson, 1991).

Notably, there were no significant age differences, but there were task differences in the proportion ofitems shared by children of different ages. Across three age groups, children shared a smaller proportionof items as the total amount of resources increased (two, four, or six items in three tasks), suggesting that2- to 4-year-olds might be likely to focus on the absolute number of items in distributing resources (seealso McCrink et al., 2010). Consistent with previous studies (e.g., Birch & Billman, 1986), children shared,on average, roughly one item with others while keeping most of the resources for themselves. The aver-age level of children’s sharing was significantly below the equal split level, suggesting that childrenyounger than 5 years tend to be selfish (e.g., Benenson et al., 2007; Blake & Rand, 2010; Fehr et al.,2008; Rochat et al., 2009; Thompson et al., 1997). The only exception was in the sticker task, where chil-dren were given two stickers; most children shared equally, giving one sticker to the partner. Yet thismight again confirm our speculation that children were willing to share one individual item, especiallyin such a context where one item corresponds exactly to one person (two items and two people).

Interestingly, we found that children who shared more than one item actually shared one item ofeach kind but rarely gave out all toys of a certain type. For example, 98% of children who shared twoitems in the watermelon task actually shared one watermelon with one knife, only two children gaveout two watermelons, and no children gave out two knives. This behavioral pattern was consistentacross the three age groups. One possible explanation for these results is that children were very con-cerned about the items left for them to play with. When facing a social partner, they were willing toshare roughly one item with others and, even when some of them shared two items, they shared oneof each type so that they could still keep a set of toys to play with. An alternative explanation would bethat after seeing the experimenter playing with both types of toys (e.g., cutting a watermelon with aknife), children were likely to share a set of toys with others. However, the second possibility is hard toreconcile with the finding that approximately one third of children shared only one type of item.

In sum, we found that there were age differences in the communicative cues required for childrento share but not in how much children shared. Although older children needed fewer explicit cues togenerate a sharing response than younger children, they shared the same amount of resources. Thisresult suggests that 2- to 4-year-olds need substantive communicative scaffolding from the socialpartner to process sharing; if enough support is given, younger children may share as much as olderchildren. Different developmental trajectories of these two measures imply that children may engagein two separate decisions when interacting with a social partner exhibiting various communicativecues: (a) whether to share given communicative cue and (b) how much to share (see also Blake &Rand, 2010). More studies are required to identify these differential underlying processes.


In addition to the developmental changes in children’s sharing behavior, we also found that children’ssharing behaviors related to their ToM understanding. Children with more advanced ToM understandingshared more spontaneously, shared more quickly, and shared more items than children with less ad-vanced ToM understanding. Compared with older children, younger children had more difficulty inunderstanding others’ mental states such as needs and desires; thus, they may need more scaffoldingin interpreting another’s communication about his or her desires (e.g., Smiley, 2001). Our results con-firmed this hypothesis by showing that advances in children’s ToM understanding were associated withchildren’s sharing behavior after controlling for age. Moreover, our regression analyses showed that ToMwas a significant predictor of how spontaneously and quickly children shared, as well as how many itemschildren shared, whereas age did not predict these sharing measures. Specifically, children who under-stood that people could have different beliefs about the same object (the Diverse Beliefs task) and thatpeople might not know the truth (the Knowledge Access task) shared more spontaneously and morequickly and shared more items with their playmate than children who failed in these two tasks. By con-trast, the age effect in children’s sharing behavior disappeared after controlling for children’s perfor-mance on these two tasks. Together, our results suggest that ToM understanding might be a potentialunderlying mechanism of children’s age-related increase in sharing.

This finding is consistent with and extends previous research demonstrating associations betweenToM ability and prosocial behavior (e.g., Lane et al., 2010; Ruffman, Slade, Devitt, & Crowe, 2006; Sally& Hill, 2006; Walker, 2005). Evidence suggests that these two abilities may share the same underlyingneural processes (McCabe, Houser, Ryan, Smith, & Trouard, 2001). For example, studies using func-tional magnetic resonance imaging (fMRI) found that when adults participated cooperatively in a trustgame, the brain areas related to ToM (medial prefrontal cortex) were activated (McCabe et al., 2001).Furthermore, children who passed a false belief task proposed higher offers in the Ultimatum Game,which requires consideration of another person’s possible behavior, than children who failed (Takag-ishi et al., 2010). In addition, studies have shown that parents’ elicitation of their children’s talk aboutinternal states had a robust correlation with the development of ToM (for a review, see Symons, 2004),and this internal state discourse is also positively related to the rate and spontaneity of children’ssharing (Brownell, Svetlova, Anderson, Nichols, & Drummond, 2013). These results suggest that pro-social behavior is associated with the ability to infer the mental states of others. Our findings furthersupport this hypothesis by showing that more advanced ToM ability was positively associated withchildren’s spontaneous sharing and the amount of items shared with others.

In addition to studying the association between sharing behavior and ToM ability in general, we alsoinvestigated whether sharing behavior relates to each specific aspect of ToM abilities. Understanding de-sires, beliefs, knowledge states, and emotions are all important components of ToM abilities (Wellman &Liu, 2004; Wellman et al., 2006). The development of prosocial behaviors has been attributed to the ad-vances in the understanding of others’ desires and emotions (e.g., Eisenberg, 2005; Mascolo & Fischer,2010). If a child attends only to his or her own desires and emotions and has no understanding of the oth-ers’ desires and emotions, sharing processes are not likely to be activated. Previous studies have alsoshown that making the social partner’s desires more apparent by verbally or gesturally requesting theitem dramatically increased young children’s sharing behavior (Brownell, Iesue, et al., 2013; Brownellet al., 2009). Based on these studies, we expected to find associations between children’s performanceon each ToM task and their sharing behavior. However, our results partly supported the hypothesis byshowing that, among the multiple aspects of ToM we tested, the ability to understand diverse beliefsand knowledge/ignorance states was especially related to sharing responses, but the ability to under-stand diverse desires, false belief, and real–apparent emotion was not related to sharing behavior. Onepossible explanation of the current results may be that the ages used in the current study (2–4 years)did not fully capture the complete development of ToM. The ability to understand diverse beliefs andknowledge states is an important part of inferring others’ internal states and is undergoing significantchanges during the period from 2 to 4 years of age; thus, the variability in these two capacities may dra-matically influence the development of sharing behaviors. By contrast, there seemed to be a ceiling effectin children’s performance on the Diverse Desires task and floor effects in the Contents False Belief andReal–Apparent Emotions tasks. These ceiling and floor effects may be responsible for the nonsignificantcorrelations between children’s sharing behavior and their performance on the Diverse Desires, ContentsFalse Belief, and Real–Apparent Emotions tasks.


An alternative explanation, however, may be that children might not need the ability to understandfalse beliefs to perform sharing behavior in the current tasks. The progressively more explicit cues sig-naled by the social partner aimed to draw children’s attention to the partner’s interests, desires, wants,needs, and true beliefs, but not necessarily their false beliefs or hidden emotions. Therefore, children’sperformance on the Contents False Belief and Real–Apparent Emotion tasks were not correlated to theirsharing behavior measured by the current tasks, consistent with previous findings (Rochat et al., 2009;Zhang, 2008). Rochat and colleagues (2009) and Zhang (2008) used similar sharing tasks as in our studyand found no clear evidence of a positive correlation between the passing of a false belief task and sharingmore items with others in 3- to 5-year-old Chinese children (Rochat et al., 2009; Zhang, 2008). Thesestudies suggested that even when there were no floor effects in children’s false belief understanding,no significant correlations were found between false belief understanding and resource distribution inthe similar sharing tasks. However, a more complicated sharing task, such as the Ultimatum Game usedby Takagishi and colleagues (2010), might demand higher ToM abilities such as false belief understand-ing. In the Ultimatum Game, failure to attribute others’ mental states would potentially result in receiv-ing zero tokens; thus, fairness-related behavior was closely related to ToM ability. By contrast, in ourstudy, as well as in the study of Rochat and colleagues (2009), sacrificing one’s welfare might competewith the desire of self-gains, and no punishment existed if children chose not to share; thus, a weakercorrelation between sharing and false belief understanding was found.

Nevertheless, it would be beneficial for future work to test the above two possible explanations byexpanding the current study to children younger than 2 years and older than 4 years and by usingmore age-appropriate measures of ToM abilities as well as comparing children’s performance on dif-ferent sharing tasks. This may contribute to a more comprehensive picture of the development ofsharing and its relation to ToM understanding.

Limits and future directions

Several features of this study limit its generalization and require further investigation. First, the dif-ferent kinds of items may be confounded with the number of items used in sharing tasks in influencingchildren’s sharing behavior. Our results showed that children required more explicit cues from the re-cipient to share when their resources were ample (four items in the watermelon task and six items inthe bead task) than when their resources were relatively scarce (two items in the sticker task). Onepossibility is that it took more time for children to explore ample resources than to explore scarce re-sources. For example, children were usually busy putting the string through beads, but it took lesstime to play with the stickers; thus, they shared more quickly in the sticker task than in the bead task.Therefore, individual differences in the amount of time needed to explore toys may have existed in thecurrent research.

Second, inferences about the causal directions of the relations between ToM understanding andsharing behavior were not ascertained because the current study is cross-sectional and correlational.Experimental studies manipulating the requirement of ToM understanding in sharing tasks will behelpful in clarifying the causal effect of ToM understanding in sharing behavior. Future work expand-ing the current study to children younger than 2 years and older than 4 years would be beneficial. Inaddition, longitudinal studies tracking the emergence of ToM abilities and changes in sharing behaviorcould elucidate the relationship between these behaviors.

Finally, the sample in the current study was drawn largely from urban middle-class neighborhoodsin Beijing. Given the finding that cultural and socialization practices shape prosocial behavior in pre-school-age children (e.g., Rochat et al., 2009), future studies with more diverse samples can furtherinform questions about the development and underlying mechanisms of sharing behavior.

Conclusion

Our results showed that sharing behavior may be more likely to occur when the partner makes hisor her wants, needs, desires, and emotions more apparent, thereby reducing the requirement for com-plex inferences about others’ mental states, especially for young children whose ability to infer others’


psychological states is immature. For older children who have more advanced ToM understanding, theprovision of ostensive cues from the recipient might not be necessary to elicit spontaneous sharing ofitems. We propose that the developmental differences in children’s sharing behavior typically thoughtto be associated with age may be partly driven by the growth in ToM understanding.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Projects 31170995,31371040, and J1103602) and the National Basic Research Program (973 Program: 2010CB833904).We thank Larissa Samuelson, Julie Gros-Louis, Daryl Cameron, Zhen Zhang, and two anonymousreviewers for helpful comments on earlier versions of the manuscript; teachers and staff at preschoolsfor their help with recruiting participants; and children who participated in the research.

References

Baillargeon, R., Scott, R. M., & He, Z. (2010). False-belief understanding in infants. Trends in Cognitive Sciences, 14, 110–118.Batson, D. C., Duncan, B. D., Ackerman, P., Buckley, T., & Birch, K. (1981). Is empathic emotion a source of altruistic motivation?

Journal of Personality and Social Psychology, 40, 290–302.Benenson, J. F., Pascoe, J., & Radmore, N. (2007). Children’s altruistic behavior in the Dictator Game. Evolution and Human

Behavior, 28, 168–175.Birch, L. L., & Billman, J. (1986). Preschool children’s food sharing with friends and acquaintances. Child Development, 57,

387–395.Blake, P. R., & McAuliffe, K. (2011). ‘‘I had so much it didn’t seem fair’’: Eight-year-olds reject two forms of inequity. Cognition,

120, 215–224.Blake, P. R., & Rand, D. G. (2010). Currency value moderates equity preference among young children. Evolution and Human

Behavior, 31, 210–218.Brownell, C. A., Iesue, S. S., Nichols, S. R., & Svetlova, M. (2013). Mine or yours? Development of sharing in toddlers in relation to

ownership understanding. Child Development, 84, 906–920.Brownell, C. A., Svetlova, M., Anderson, R., Nichols, S. R., & Drummond, J. (2013). Socialization of early prosocial behavior:

Parents’ talk about emotions is associated with sharing and helping in toddlers. Infancy, 18, 91–119.Brownell, C. A., Svetlova, M., & Nichols, S. (2009). To share or not to share: When do toddlers respond to another’s needs?

Infancy, 14, 117–130.Dunfield, K., Kuhlmeier, V. A., O’Connell, L., & Kelley, E. (2011). Examining the diversity of prosocial behavior: Helping, sharing,

and comforting in infancy. Infancy, 16, 227–247.Eisenberg, N. (2005). The development of empathy-related responding. Nebraska Symposium on Motivation, 51, 73–117.Eisenberg, N. (2007). Empathy-related responding and prosocial behaviour. Novartis Foundation Symposium, 278, 71–80;

discussion: 80–96, 216–221.Fehr, E., Bernhard, H., & Rockenbach, B. (2008). Egalitarianism in young children. Nature, 454, 1079–1083.Fehr, E., & Fischbacher, U. (2004). Social norms and human cooperation. Trends in Cognitive Sciences, 8, 185–190.Grusec, J. E. (1991). Socializing concern for others in the home. Developmental Psychology, 27, 338–342.Grusec, J. E. (2006). The development of moral behavior and conscience from a socialization perspective. In M. Killen & J. G.

Smetana (Eds.), Handbook of moral development (pp. 243–265). Mahwah, NJ: Lawrence Erlbaum.Gummerum, M., Hanoch, Y., Keller, M., Parsons, K., & Hummel, A. (2010). Preschoolers’ allocations in the Dictator Game: The

role of moral emotions. Journal of Economic Psychology, 31, 25–34.Harris, P. L., de Rosnay, M., & Pons, F. (2005). Language and children’s understanding of mental states. Current Directions in

Psychological Science, 14, 69–73.Hay, D. F. (1979). Cooperative interactions and sharing between very young children and their parents. Developmental

Psychology, 15, 647–653.Hay, D. F., Caplan, M., Castle, J., & Stimson, C. A. (1991). Does sharing become increasingly ‘‘rational’’ in the second year of life?

Developmental Psychology, 27, 987–993.Hay, D. F., Castle, J., Davies, L., Demetriou, H., & Stimson, C. A. (1999). Prosocial action in very early childhood. Journal of Child

Psychology and Psychiatry and Allied Disciplines, 40, 905–916.Hay, D. F., & Cook, K. V. (2007). The transformation of prosocial behavior from infancy to childhood. In C. A. Brownell & C. B.

Kopp (Eds.), Socioemotional development in the toddler years: Transitions and transformations (pp. 100–131). New York:Guilford.

Hay, D. F., & Murray, P. (1982). Giving and requesting: Social facilitation of infants’ offers to adults. Infant Behavior andDevelopment, 5, 301–310.

Lane, J. D., Wellman, H. M., Olson, S. L., LaBounty, J., & Kerr, D. C. R. (2010). Theory of mind and emotion understanding predictmoral development in early childhood. British Journal of Developmental Psychology, 28, 871–889.

Levitt, M. J., Weber, R. A., Clark, M. C., & McDonnell, P. (1985). Reciprocity of exchange in toddler sharing behavior.Developmental Psychology, 21, 122–123.

Mascolo, M., & Fischer, K. W. (2010). The codevelopment of self and sociomoral emotions during the toddler years. In C. A.Brownell & C. B. Kopp (Eds.), Socioemotional development in the toddler years: Transitions and transformations (pp. 66–99).New York: Guilford.

McCabe, K., Houser, D., Ryan, L., Smith, V., & Trouard, T. (2001). A functional imaging study of cooperation in two-personreciprocal exchange. Proceedings of the National Academy of Sciences of the United States of America, 98, 11832–11835.

http://refhub.elsevier.com/S0022-0965(13)00244-0/h0005


















































McCrink, K., Bloom, P., & Santos, L. R. (2010). Children’s and adults’ judgments of equitable resource distributions. DevelopmentalScience, 13, 37–45.

Merriam–Webster (1969). Webster’s ninth new collegiate dictionary. Springfield, MA: Author.Moore, C. (2007). Understanding self and others in the second year. In C. A. Brownell & C. B. Kopp (Eds.), Socioemotional

development in the toddler years: Transitions and transformations (pp. 43–65). New York: Guilford.Premack, D., & Woodruff, G. (1978). Does the chimpanzee have a theory of mind? Behavioral and Brain Sciences, 1, 515–526.Rao, N., & Stewart, S. M. (1999). Cultural influences on sharer and recipient behavior sharing in Chinese and Indian preschool

children. Journal of Cross-Cultural Psychology, 30, 219–241.Rheingold, H. L. (1982). Little children’s participation in the work of adults, a nascent prosocial behavior. Child Development, 53,

114–125.Rheingold, H. L., Hay, D. F., & West, M. J. (1976). Sharing in the second year of life. Child Development, 47, 1148–1158.Rochat, P., Dias, M. D. G., Liping, G., Broesch, T., Passos-Ferreira, C., Winning, A., et al (2009). Fairness in distributive justice by 3-

and 5-year-olds across seven cultures. Journal of Cross-Cultural Psychology, 40, 416–442.Ruffman, T., Slade, L., Devitt, K., & Crowe, E. (2006). What mothers say and what they do: The relation between parenting, theory

of mind, language, and conflict/cooperation. British Journal of Developmental Psychology, 24, 105–124.Sally, D., & Hill, E. (2006). The development of interpersonal strategy: Autism, theory-of-mind, cooperation, and fairness. Journal

of Economic Psychology, 27, 73–97.Shaw, A., & Olson, K. R. (2012). Children discard a resource to avoid inequity. Journal of Experimental Psychology: General, 141,

382–395.Singer, T., & Steinbeis, N. (2009). Differential roles of fairness- and compassion-based motivations for cooperation, defection,

and punishment. Annals of the New York Academy of Sciences, 1167, 41–50.Smiley, P. A. (2001). Intention understanding and partner-sensitive behaviors in young children’s peer interactions. Social

Development, 10, 330–354.Svetlova, M., Nichols, S. R., & Brownell, C. A. (2010). Toddlers’ prosocial behavior: From instrumental to empathic to altruistic

helping. Child Development, 81, 1814–1827.Symons, D. K. (2004). Mental state discourse, theory of mind, and the internalization of self–other understanding. Developmental

Review, 24, 159–188.Takagishi, H., Kameshima, S., Schug, J., Koizumi, M., & Yamagishi, T. (2010). Theory of mind enhances preference for fairness.

Journal of Experimental Child Psychology, 105, 130–137.Thompson, C., Barresi, J., & Moore, C. (1997). The development of future-oriented prudence and altruism in preschoolers.

Cognitive Development, 12, 199–212.Tomasello, M. (2008). Origins of human communication. Cambridge, MA: MIT Press.Tomasello, M., & Carpenter, M. (2007). Shared intentionality. Developmental Science, 10, 121–125.Walker, S. (2005). Gender differences in the relationship between young children’s peer-related social competence and

individual differences in theory of mind. Journal of Genetic Psychology, 166, 297–312.Warneken, F., & Tomasello, M. (2007). Helping and cooperation at 14 months of age. Infancy, 11, 271–294.Warneken, F., & Tomasello, M. (2009). The roots of human altruism. British Journal of Psychology, 100, 445–471.Wellman, H. M., Fang, F., Liu, D., Zhu, L., & Liu, G. (2006). Scaling of theory-of-mind understandings in Chinese children.

Psychological Science, 17, 1075–1081.Wellman, H. M., & Liu, D. (2004). Scaling of theory-of-mind tasks. Child Development, 75, 523–541.Zahn-Waxler, C., & Radke-Yarrow, M. (1990). The origins of empathic concern. Motivation and Emotion, 14, 107–130.Zahn-Waxler, C., Robinson, J. L., & Emde, R. N. (1992). The development of empathy in twins. Developmental Psychology, 28,

1038–1047.Zhang, Z. (2008). One for you and one for me: Other-regarding preferences in young children. Unpublished doctoral dissertation,

Peking University.












































Documents

How do preschoolers’ sharing behaviors relate to their theory of mind understanding?