Doll Play Improves Social Processing: Evidence from a Randomized-Control Trial

doi:10.21203/rs.3.rs-4794246/v1

Play is often described as a child’s “occupation,” both because young children spend the majority of their time playing when given the option, and because it is a critical mechanism through which children learn both cognitive and socio-emotional skills. In a randomized control trial (N = 73), we found the first causal evidence that doll play, more so than creative tablet play, improves theory of mind in 4- to 8-year-old children following a month-long play intervention. Improvements in theory of mind were particularly strong for children who had more parent-reported peer problems. Diaries of play during the intervention indicated that children were more likely to play socially with dolls than socially with tablets. Together, this shows that doll play may be a particularly suitable play pattern for practicing and improving social skills like theory of mind by both encouraging social interaction and practicing these skills outside of peer interactions.

Social science/Psychology/Human behaviour

Biological sciences/Psychology/Human behaviour

Theory of mind (ToM), or the ability to understand and differentiate the internal states of both oneself and others (i.e., thoughts, beliefs, and desires), is a foundational skill for interacting with, learning from, and developing relationships with peers, teachers, and parents, with eventual implications for successfully engaging with colleagues, romantic partners, and business partners (Abdullah et al., 2021; Caputi et al., 2012, 2021; Galinsky et al., 2008; Lecce & Devine, 2022; Ramezani et al., 2020). Research suggests that acquiring a rich understanding of others’ minds is a complex and prolonged endeavor, with some going so far as to suggest that the capacity to understand others’ minds and the propensity to use this skill in social interactions is uniquely human (Buttelmann, 2022; MacLean, 2016; Tomasello et al., 2005; Tomasello & Rakoczy, 2003). One of the reasons that the development of these skills in human children is thought to be an enduring process is because of the rich, complex, and bidirectional nature of factors associated with theory of mind skills, including language, executive function, and social experiences (Hughes & Leekam, 2004; Weimer et al., 2021). In the current research, we focus specifically on the role of social experiences and interactions, both real and imagined.

Research examining social influences on the development of theory of mind have focused on interactions with caregivers (e.g. maternal mind mindedness; e.g. Hughes et al., 2018; Kirk et al., 2015; Lundy, 2013; Meins et al., 2002), siblings (Hughes & Devine, 2016; Hughes & Ensor, 2005; McAlister & Peterson, 2006), or peers (Bosacki, 2015; Kokkinos et al., 2016; Lecce et al., 2024; Suway et al., 2012). Within these social interactions, some have recognized the unique role that play may afford for learning about and practicing social cognitive skills (e.g. McAlister & Peterson, 2006; Meins et al., 2013; Newton & Jenvey, 2011; Schwebel et al., 1999). Play is an important part of children’s development as it is a critical mechanism through which they learn both cognitive and socio-emotional skills (Lillard, 2015). In particular, pretend play has been proposed to be related to theory of mind acquisition due to their reliance on similar skills: 1) the ability to think of one object as being two things at once, 2) the ability to think of one object as representing another, and 3) the ability to have mental representations (Lillard, 1993). When children play with dolls, for example, they may create fantasy worlds, narratives, and characters which offer them the opportunity to think about other’s beliefs and perspectives (Harris, 2000), regardless of whether they are interacting with a friend, sibling, or parent, or playing on their own. In this way, doll play has the potential to encourage both real and imagined interactions with others.

Previous research has provided preliminary support for the notion that doll play may be a promising avenue for promoting social processing skills like theory of mind. A recent neuroimaging study measured activity in an area of the brain associated with social processing, theory of mind reasoning, and social engagement (the posterior superior temporal sulcus, pSTS; Hakuno et al., 2018; Lloyd-Fox et al., 2015) while children spontaneously played with dolls and tablets both alone and with a social partner (an experimenter). This research found similar pSTS activity when children played with a partner regardless of the types of toys with which they played (Hashmi et al., 2020; Keating et al., 2023). This was not particularly surprising given that the pSTS is active when both adults and children engage in social interactions (Lloyd-Fox et al., 2015; Redcay et al., 2010), but it was the first evidence that this brain region is activated when children engage in semi-naturalistic interactions with others. More importantly for the current research, this same region was also active when children played alone with dolls, more so than when they played alone with tablets (Hashmi et al., 2020; Keating et al., 2023) highlighting the social nature of dolls. In a follow-up study, Hashmi et al. (2022) found that children used more internal state language – verbalizations that refer to the thoughts, feelings, and desires of others – when playing with a partner and when playing alone with dolls more so than playing alone with tablets, thus mirroring the pSTS findings. Individual differences in the use of internal state language about others was also associated with pSTS activation (Hashmi et al., 2022). Taken together, these findings suggest that doll play affords children the opportunity to practice social skills, both with others and alone using their imagination, which may in turn improve their ability to engage in successful relationships with peers.

The previous studies identified an association between engaging in doll play and social processing skills with regards to language and brain activity. It is, as yet, unknown, however, whether doll play could actually improve social processing abilities. Thus, the current study aims to examine whether there is a causal effect of doll play on ToM skills. In this study, children’s ToM was measured before and after they completed a novel randomized-control home intervention involving either doll or tablet play. As pretend play involves similar skills to ToM, we expected that children who played with dolls during the home intervention would show greater improvement on the ToM task during their return visit, relative to children who played with tablets at home.

Intervention Compliance

The duration of the intervention was 7.20 weeks on average (SD = 3.01), with no significant difference in duration between the doll and tablet intervention groups (p = .144). Parents submitted an average of 8.88 (SD = 3.34) play diary entries over the intervention period (difference between groups not significant; p = .608). Although frequency of reported doll play within these diary entries did not significantly differ between the two intervention groups (p = .853), children in the tablet condition played with tablets more frequently than children in the doll condition (p = .023). Additionally, a one-sample t-test comparing the intervention ratio score to 1 (i.e. equal time playing with assigned and non-assigned toy-type), revealed that children spent more time playing with the assigned toy-type than the unassigned toy-type within each condition (doll: p = .018, Cohen’s d = .426; tablet: p = .006, Cohen’s d = .516). Children tended to use the study materials more than their own doll and tablet toys across both conditions (doll: p = .019, Cohen’s d = .416; tablet: p < .001, Cohen’s d = 1.070). Across groups, children were more likely to play alone with tablets than with dolls (p = .009, Cohen’s d = .327) and more likely to play with sibling(s), parent(s), or friend(s) with dolls than with tablets (ps < .03, Cohen’s d’s ≥ .27).

Theory of Mind Improvement

In order to assess the extent to which children’s performance on the sandbox task changed from pre- to post-intervention, we calculated a difference in magnitude of bias score (Change in Bias Score) separately for both false belief trials and memory trials by subtracting the magnitude of bias at Time 2 from the magnitude of bias at Time 1. Given that higher scores at each time point indicate more bias (i.e., more influenced by the child’s knowledge about the actual location of the object than by the protagonist’s false belief about the actual location), a positive difference score indicates a decrease in bias from Time 1 (pre-intervention) to Time 2 (post-intervention). This False Belief Change in Bias Score was entered into a Univariate Analysis of Variance as the dependent measure, with intervention group (2: doll, tablet) and sex (2: male, female) as between-subjects factors and age (at Time 2) as a covariate. This revealed a significant effect of group (F(1,68) = 5.20, p = .026, h_p² = .071; see Figure 1), a marginal effect of age (F(1,68) = 2.89, p = .094, h_p² = .041), and no other main effects or interactions (ps > .17).

In order to confirm that the effects were specific to false belief rather than general improvement on the task, we ran the above analysis with Memory Change in Bias Score as the dependent measure instead of False Belief Change in Bias Score. No significant effects emerged in this model (ps > .27). Additionally, when Memory Change in Bias Score was added as a covariate to the above model with False Belief Change in Bias Score as the outcome, no effect of Memory Change in Bias Score was detected (p = .20) and the effect of group remained (p = .040, h_p² = .062).

Individual Differences in Theory of Mind Improvement

In order to investigate whether individual differences in terms of peer relationship problems related to improvement in theory of mind, we conducted separate correlations within each condition (doll vs tablet intervention) between False Belief Change in Bias Score (note: higher scores indicate more improvement from time 1 to time 2) and peer problems as parent-reported in the SDQ. Due to skew in the data (one-sample Kolmogorov-Smirnov for peer problem scores, ps < .001), Spearman’s correlations are reported. In the doll intervention condition, a positive correlation between False Belief Change in Bias Score and peer problems was revealed (r_s(37) = .474, p = .003). For the tablet intervention condition, no relation was found (r_s(35) = -.253, p = .14). These correlations are significantly different from one another (Z = 3.05, p = .002).

In this study, we found the first casual evidence that doll play uniquely improves theory of mind reasoning above and beyond creative tablet play. The fact that this effect was consistent for both boys and girls and across ages over a developmental period when theory of mind is being refined suggests that it is a robust effect. Interestingly, given the bidirectional relations between peer relations and social cognition found in the literature (Bosacki & Wilde Astington, 1999; Fink et al., 2015; Lalonde & Chandler, 1995), we also found evidence that doll play was particularly beneficial for improving theory of mind in children with more reported peer problems. We speculate that the mechanisms underlying the marked improvement in children with peer problems may be two-fold: (1) doll play may represent a “safe” environment in which to practice social interactions when playing alone and (2) doll play may encourage children to engage in social interactions more than other kinds of play.

When playing with dolls, children have the opportunity to role play characters, create narratives, and act out scenarios (Harris, 2000), all of which rely on, but may also foster, the ability to imagine other’s thoughts, feelings, and intentions (Lillard, 1993). Through these pretend play scenarios, children may practice social skills, emotion processing, and emotion regulation, within a safe, controlled environment (see Kidd & Costano, 2018 for similarities about reading literary fiction). Previous research confirms this notion in that children both engage the pSTS, a social processing brain region, and use more internal state language about others when playing alone with dolls than with tablets (Hashmi et al., 2020, 2022; Keating et al., 2023). In this research, solo doll play mimicked social interactions (i.e., when children played with a social partner with either dolls or tablets) in terms of both linguistic and neural activity associated with mentalizing. This opportunity for rehearsed social interactions without another person present may be particularly helpful for children with peer problems, as children with peer problems may find peer interactions challenging or unpleasant.

In addition to solo doll play providing an opportunity to rehearse social skills and mentalization, doll play may also encourage children to engage in social interactions more than other kinds of play. In previous research investigating doll and tablet play in a neurodiverse sample (Keating et al., 2023), children higher in autistic traits tended to try to initiate talk with an adult in the room when asked to play alone with dolls more so than when asked to play alone with tablets. The current study mirrors this tendency more broadly. Reports from parent diaries of play at home during the intervention period found that children were more likely to play with dolls with a social partner (sibling, parent, friend, or other) and more likely to play with tablets by themselves. This was true regardless of intervention condition, gender, or reported peer problems, suggesting that it may reflect a broader play tendency in everyday life. Based on this, dolls might encourage even more solitary children to engage in social interactions that encourage them to think and talk about others’ mental states.

In order to ensure strict adherence to the randomized-control groups, we embraced the variability that presented itself in terms of compliance and a variety of other practical and logistical issues. Importantly, intervention groups were evenly matched in terms of age, sex, ethnicity, household income, and parent education. Due to the small numbers of attrition between time 1 and time 2, our groups at time 2 were still matched on all demographic variables. Despite this, we saw variability in terms of adherence in completing the diary entries during the intervention, with some parents only completing a small number of diaries – 15.8% of parents completed three or fewer entries. Additionally, we did not restrict children’s exposure to their unassigned play form (i.e., children in the doll condition were not told to avoid playing with tablets and vice versa). Because of this, not every child played with their assigned toy more than the unassigned toy, though we found a group effect of more play with the assigned than unassigned toy overall. This suggests the condition effects identified in the current study are most likely a conservative estimate of the overall intervention effects.

These findings raise new and important questions regarding the scalability of this intervention across school and community settings. A recent pretend-play training program improved 5-6-year-olds' emotion comprehension and found a decrease in aggressive behavioral responses (Richard et al., 2023). However, Richard and colleagues (2023) study required 20 hours of teacher training. Simply giving children dolls to play with would be a more cost-effective strategy for improving social processing, but before investing in this kind of intervention, replication and extension would be fruitful. In the current research, we only collected data on changes to theory of mind at one time point immediately post-intervention. Whether the effect of doll play would be maintained over a longer period is currently unknown. Additionally, questions regarding the extent to which the observed improvements in theory of mind transfer to social interactions in an iterative process should be thoroughly investigated.

The current findings are groundbreaking in identifying the first evidence of a causal effect of doll play on theory of mind. Given the importance of social cognition for long-term outcomes spanning social, academic, and emotional domains (Abdullah et al., 2021; Caputi et al., 2012, 2021; Galinsky et al., 2008; Lecce & Devine, 2022; Ramezani et al., 2020), identifying new and innovative ways to improve these skills could have profound consequences. As noted in the introduction, by no means do we suggest that doll play is the only avenue to improving theory of mind. Instead, theory of mind development is a highly complex and prolonged skillset, with dynamic factors both contributing to and being influenced by its development (Hughes & Leekam, 2004; Weimer et al., 2021). Partly due to the integration of these skills in everyday interactions and the potential far-reaching consequences of them, interest in programming social cognitive skills like theory of mind via artificial intelligence and social robotics has become a topic of increasing interest across computer science, engineering, cognitive neuroscience, and beyond (e.g., Langley et al., 2022; Moghaddam & Honey, 2023; Patacchiola & Cangelosi, 2022). Thus, this research opens new pathways to embracing and investigating play as a mechanism of social learning across humans and robots alike.

In this study, we found the first casual evidence that doll play uniquely improves theory of mind reasoning above and beyond creative tablet play. The fact that this effect was consistent for both boys and girls and across ages over a developmental period when theory of mind is being refined suggests that it is a robust effect. Interestingly, given the bidirectional relations between peer relations and social cognition found in the literature (Bosacki & Wilde Astington, 1999; Fink et al., 2015; Lalonde & Chandler, 1995), we also found evidence that doll play was particularly beneficial for improving theory of mind in children with more reported peer problems. We speculate that the mechanisms underlying the marked improvement in children with peer problems may be two-fold: (1) doll play may represent a “safe” environment in which to practice social interactions when playing alone and (2) doll play may encourage children to engage in social interactions more than other kinds of play.

When playing with dolls, children have the opportunity to role play characters, create narratives, and act out scenarios (Harris, 2000), all of which rely on, but may also foster, the ability to imagine other’s thoughts, feelings, and intentions (Lillard, 1993). Through these pretend play scenarios, children may practice social skills, emotion processing, and emotion regulation, within a safe, controlled environment (see Kidd & Costano, 2018 for similarities about reading literary fiction). Previous research confirms this notion in that children both engage the pSTS, a social processing brain region, and use more internal state language about others when playing alone with dolls than with tablets (Hashmi et al., 2020, 2022; Keating et al., 2023). In this research, solo doll play mimicked social interactions (i.e., when children played with a social partner with either dolls or tablets) in terms of both linguistic and neural activity associated with mentalizing. This opportunity for rehearsed social interactions without another person present may be particularly helpful for children with peer problems, as children with peer problems may find peer interactions challenging or unpleasant.

In addition to solo doll play providing an opportunity to rehearse social skills and mentalization, doll play may also encourage children to engage in social interactions more than other kinds of play. In previous research investigating doll and tablet play in a neurodiverse sample (Keating et al., 2023), children higher in autistic traits tended to try to initiate talk with an adult in the room when asked to play alone with dolls more so than when asked to play alone with tablets. The current study mirrors this tendency more broadly. Reports from parent diaries of play at home during the intervention period found that children were more likely to play with dolls with a social partner (sibling, parent, friend, or other) and more likely to play with tablets by themselves. This was true regardless of intervention condition, gender, or reported peer problems, suggesting that it may reflect a broader play tendency in everyday life. Based on this, dolls might encourage even more solitary children to engage in social interactions that encourage them to think and talk about others’ mental states.

In order to ensure strict adherence to the randomized-control groups, we embraced the variability that presented itself in terms of compliance and a variety of other practical and logistical issues. Importantly, intervention groups were evenly matched in terms of age, sex, ethnicity, household income, and parent education. Due to the small numbers of attrition between time 1 and time 2, our groups at time 2 were still matched on all demographic variables. Despite this, we saw variability in terms of adherence in completing the diary entries during the intervention, with some parents only completing a small number of diaries – 15.8% of parents completed three or fewer entries. Additionally, we did not restrict children’s exposure to their unassigned play form (i.e., children in the doll condition were not told to avoid playing with tablets and vice versa). Because of this, not every child played with their assigned toy more than the unassigned toy, though we found a group effect of more play with the assigned than unassigned toy overall. This suggests the condition effects identified in the current study are most likely a conservative estimate of the overall intervention effects.

These findings raise new and important questions regarding the scalability of this intervention across school and community settings. A recent pretend-play training program improved 5-6-year-olds' emotion comprehension and found a decrease in aggressive behavioral responses (Richard et al., 2023). However, Richard and colleagues (2023) study required 20 hours of teacher training. Simply giving children dolls to play with would be a more cost-effective strategy for improving social processing, but before investing in this kind of intervention, replication and extension would be fruitful. In the current research, we only collected data on changes to theory of mind at one time point immediately post-intervention. Whether the effect of doll play would be maintained over a longer period is currently unknown. Additionally, questions regarding the extent to which the observed improvements in theory of mind transfer to social interactions in an iterative process should be thoroughly investigated.

The current findings are groundbreaking in identifying the first evidence of a causal effect of doll play on theory of mind. Given the importance of social cognition for long-term outcomes spanning social, academic, and emotional domains (Abdullah et al., 2021; Caputi et al., 2012, 2021; Galinsky et al., 2008; Lecce & Devine, 2022; Ramezani et al., 2020), identifying new and innovative ways to improve these skills could have profound consequences. As noted in the introduction, by no means do we suggest that doll play is the only avenue to improving theory of mind. Instead, theory of mind development is a highly complex and prolonged skillset, with dynamic factors both contributing to and being influenced by its development (Hughes & Leekam, 2004; Weimer et al., 2021). Partly due to the integration of these skills in everyday interactions and the potential far-reaching consequences of them, interest in programming social cognitive skills like theory of mind via artificial intelligence and social robotics has become a topic of increasing interest across computer science, engineering, cognitive neuroscience, and beyond (e.g., Langley et al., 2022; Moghaddam & Honey, 2023; Patacchiola & Cangelosi, 2022). Thus, this research opens new pathways to embracing and investigating play as a mechanism of social learning across humans and robots alike.

Participants

Eighty-one children aged 4-8-years-old were recruited via social media and a database of local families interested in research to take part in this study (Mean age = 6.09 years, SD = 1.34, 42 Females). One child did not complete their first session, leaving a sample of 80 children. Parents confirmed that their child did not have any diagnosis of a neurodevelopmental condition or any developmental delays. Parents who did not want their children playing with either dolls or tablets were excluded.

Upon enrolling in the study, children were randomly assigned to one of two conditions: doll or tablet. Children assigned to the doll condition would choose three Barbie/Ken dolls of their choice to play with at home after their first visit. In the tablet condition, children would be given a tablet with three pre-loaded games (see Stimuli section for more details) to bring home and play with after their first visit. Of the 80 children who completed the first session, 42 were in the doll condition, and 38 were in the tablet condition. The groups did not differ in terms of age (p = .729) or sex (p =.832). 73 children completed their Time 2 visit (Mean age = 6.26 years, SD = 1.33, 39 Females). Of these, 38 were in the doll condition and 35 were in the tablet condition. The groups did not differ in terms of age (p = .954) or sex (p = .425). Parents completed written consent forms and children provided verbal assent at the start of both sessions. The study was reviewed and approved by the University Ethics Committee.

Table 1 Demographic information of children who provided data at both time points.

	Whole sample (n = 73)	Doll (n = 38)	Tablet (n = 35)	p
Age at T2 (years)	6.26 (1.33)	6.26 (1.35)	6.25 (1.32)	.954
Sex				.425
Male	34	16	18
Female	39	22	17
Ethnicity				.571
White	66	35	31
Asian/Asian British	1	0	1
Mixed Race	6	3	3
Annual household income	£58,271 (£28,870)	£57,500 (£31,652)	£59,188 (£25,650)	.810
Highest level of education				.789
≤ Secondary School	9	5	4
≥ Bachelor's Degree	63	32	31

Note: Values in parentheses are the standard deviations. p = p-value when calculating difference between Doll and Tablet groups. Not all parents completed information about their education level.

Measures

Parent Questionnaires

Play diaries

Parents were sent a short play diary to fill out every 3-4 days during the play at home period of the study. This short questionnaire consisted of 6 questions, asking parents how often their child played with dolls/tablets (information on both dolls and tablets were collected regardless of the condition the child was assigned to) over the past three days. Parents also reported who their child played with, if they used the materials provided to them or played with their own toys (e.g. using their own tablet/doll sets). A ratio score was calculated (Time spent playing with condition toy-type [i.e., dolls in doll condition and tablets in tablet condition]/time spent playing with other condition toy-type) to use as a measure of adherence to the intervention.

Strengths and Difficulties Questionnaire (SDQ; Goodman, 1997)

Parents completed the SDQ, a parent-report measure of child mental health for 4–17-year-olds. Parents rate their endorsement for each item on a Likert scale from 0 (not true) to 2 (certainly true). The SDQ consists of 5 subscales, each with 5 items: hyperactivity problems, conduct problems, peer problems, emotional problems, and prosocial behaviors. Higher scores indicate more difficulties, except for the prosocial behaviors scale which is inversed. In the current study, only the peer problems scale is analyzed, given our interest in links between social processing and peer interactions, as discussed in the introduction.

Behavioral Assessment

Sandbox Task

The Sandbox Task used in the current study was adapted from Bernstein et al. (2011) and used to measure theory of mind. Unlike conventional false belief tasks, this task assesses a continuous measure of ToM and results in variable performance across the full age range in the current study. In the task, an experimenter introduces two characters, one of whom buries an object in a sandbox (Location A). This character then leaves and the other moves the object to a new location (Location B). The participant is asked to mark where the first character will look for the object (i.e., a false belief question) and the distance between their mark and the correct answer (Location A) is calculated. This yields a continuous measure of how much children are influenced by their own knowledge of the object’s location versus the character’s false belief. The task also includes control memory trials, which simply require remembering where the object was originally hidden without accounting for beliefs .

In the current study, the story was presented on a screen, with each location marked as an X in a 2-D picture of a Sandbox (see Figure 3). Children completed four trials – two memory and two false belief trials using a mouse to indicate their answer, with experimenter assistance dragging the mouse when necessary. The order of trials was counterbalanced across participants such that there were four possible versions of the task. The stories and characters were adapted from Sommerville et al. (2013). Children were shown one set of stories during Time 1 and a separate set of stories during Time 2. A proximity score was calculated as the difference in pixels between the child’s response and the correct location (Location A) averaged for each trial type at both time points. In order to take into account the direction of their response and whether they demonstrated an egocentric bias, we also calculated a magnitude of bias score for each trial. For the magnitude of bias score, children’s proximity scores were coded as positive if their response was in the direction of Location B relative to Location A. In contrast, if children’s responses were in the opposite direction of Location B, their proximity scores were coded negatively. These scores were then averaged across the two trials for both false belief and memory trials, creating a magnitude of bias score for each type of task on average for each child at each time point of the intervention.

Procedure

During their first visit to the lab, children completed the Sandbox task after a 35-minute free play task with dolls and tablets measuring brain activity using fNIRS (see Keating et al., 2023 for details of the play task). The fNIRS cap was removed before participants took part in the task. The stories for the Sandbox task were presented on an iiyama ProLite 2400 LCD monitor using E-Prime 3.0 (Psychology Software Tools, PA, USA) and children sat at a round table in front of the monitor. Before each trial, the child watched a 10-s baseline of five pseudorandom images of clipart fruits presented on a black background in the center of the screen for 1.5 s each, interspersed with a white fixation cross displayed for .5 s. An experimenter narrated each story for the child, beginning with an introduction to the characters, introducing a location change, and then asking them either the memory or false belief prompt, depending on the trial type. A second experimenter changed the screen to the next part of the story through a button press once the previous part had been read to the child. The child answered each question by dragging the small black rectangle along the Sandbox to their desired location with a computer mouse.

Upon completion of the first session, each child was randomly assigned a tablet with three preloaded games on it or to choose 3 Barbie and Ken dolls (Mattel Co., El Segundo, CA, USA, more information in supplementary materials). The tablet games included Hoopa City 2 (Dr. Panda, Chengdu, China), in which players build roads, buildings, lakes, etc. to create a city for animated characters; Toca Hair Salon 3 (Toca Boca, Stockholm, Sweden), in which children can choose characters and wash, cut, and style their hair; and Toca Kitchen 2 (Toca Boca, Stockholm, Sweden), in which players can create their own recipes and dishes. Parents were instructed to allow their children to play with their assigned toy at least three times a week for an intervention period of five to seven weeks. Parents were also sent a play diary, via email, to complete every three to four days.

On their second visit (approximately 5-8 weeks later; range 4-24 weeks), children took part in the Sandbox task using the same procedure described for Time 1. During this session, children also wore a fNIRS cap to measure their brain activity. Calibration was performed to ensure good signal quality and an appropriate amplification factor for each source-detector combination. The lighting in the testing room was then dimmed. Parents/caregivers were invited to observe from an adjoining room. The current study only includes the behavioral data, no neuroimaging data analysis is included. Upon completing their second visit, children in the doll condition were offered a small prize worth approximately £5 and children in the tablet condition were offered a small prize or to choose a doll to take home.

Abdullah, M., Rehman, S., Nawaz, D. S., Asad, D. S., & Khalid, S. (2021). Association between Theory of Mind and Peer Problems. Journal of Professional & Applied Psychology, 2(2), Article 2. https://doi.org/10.52053/jpap.v2i2.51
Bernstein, D. M., Thornton, W. L., & Sommerville, J. A. (2011). Theory of Mind Through the Ages: Older and Middle-Aged Adults Exhibit More Errors Than Do Younger Adults on a Continuous False Belief Task. Experimental Aging Research, 37(5), 481–502. https://doi.org/10.1080/0361073X.2011.619466
Bosacki, S. L. (2015). Children’s theory of mind, self‐perceptions, and peer relations: A longitudinal study. Infant and Child Development, 24(2), 175–188. https://doi.org/10.1002/icd.1878
Bosacki, S., & Wilde Astington, J. (1999). Theory of Mind in Preadolescence: Relations Between Social Understanding and Social Competence. Social Development, 8(2), 237–255. https://doi.org/10.1111/1467-9507.00093
Buttelmann, D. (2022). What is unique in infant thinking about others? Infant social cognition from an evolutionary perspective. In Evolutionary perspectives on infancy (pp. 277–302). Springer Nature Switzerland AG. https://doi.org/10.1007/978-3-030-76000-7_13
Caputi, M., Cugnata, F., & Brombin, C. (2021). Theory of mind and loneliness: Effects of a conversation-based training at school. International Journal of Psychology, 56(2), 257–265. https://doi.org/10.1002/ijop.12707
Caputi, M., Lecce, S., Pagnin, A., & Banerjee, R. (2012). Longitudinal effects of theory of mind on later peer relations: The role of prosocial behavior. Developmental Psychology, 48(1), 257–270. https://doi.org/10.1037/a0025402
Fink, E., Begeer, S., Peterson, C. C., Slaughter, V., & de Rosnay, M. (2015). Friendlessness and theory of mind: A prospective longitudinal study. The British Journal of Developmental Psychology, 33(1), 1–17. https://doi.org/10.1111/bjdp.12060
Galinsky, A. D., Maddux, W. W., Gilin, D., & White, J. B. (2008). Why It Pays to Get Inside the Head of Your Opponent: The Differential Effects of Perspective Taking and Empathy in Negotiations. Psychological Science, 19(4), 378–384. https://doi.org/10.1111/j.1467-9280.2008.02096.x
Goodman, R. (1997). The Strengths and Difficulties Questionnaire: A Research Note. Journal of Child Psychology and Psychiatry, 38(5), 581–586. https://doi.org/10.1111/j.1469-7610.1997.tb01545.x
Hakuno, Y., Pirazzoli, L., Blasi, A., Johnson, M. H., & Lloyd-Fox, S. (2018). Optical imaging during toddlerhood: Brain responses during naturalistic social interactions. Neurophotonics, 5(1), 011020. https://doi.org/10.1117/1.NPh.5.1.011020
Harris, P. L. (2000). The work of the imagination (pp. xii, 222). Blackwell Publishing.
Hashmi, S., Vanderwert, R. E., Paine, A. L., & Gerson, S. A. (2022). Doll play prompts social thinking and social talking: Representations of internal state language in the brain. Developmental Science, 25(2), e13163. https://doi.org/10.1111/desc.13163
Hashmi, S., Vanderwert, R. E., Price, H. A., & Gerson, S. A. (2020). Exploring the Benefits of Doll Play Through Neuroscience. Frontiers in Human Neuroscience, 14. https://doi.org/10.3389/fnhum.2020.560176
Hughes, C., & Devine, R. (2016). Family influences on theory of mind: A review. In V. Slaughter & M. de Rosnay (Eds.), Theory of Mind Development in Context (pp. 41–56). Taylor & Francis. https://doi.org/10.4324/9781315749181
Hughes, C., Devine, R. T., & Wang, Z. (2018). Does Parental Mind-Mindedness Account for Cross-Cultural Differences in Preschoolers’ Theory of Mind? Child Development, 89(4), 1296–1310. https://doi.org/10.1111/cdev.12746
Hughes, C., & Ensor, R. (2005). Executive function and theory of mind in 2 year olds: A family affair? Developmental Neuropsychology, 28(2), 645–668. https://doi.org/10.1207/s15326942dn2802_5
Hughes, C., & Leekam, S. (2004). What are the Links Between Theory of Mind and Social Relations? Review, Reflections and New Directions for Studies of Typical and Atypical Development. Social Development, 13(4), 590–619. https://doi.org/10.1111/j.1467-9507.2004.00285.x
Keating, J., Hashmi, S., Vanderwert, R. E., Davies, R. M., Jones, C. R. G., & Gerson, S. A. (n.d.). Embracing neurodiversity in doll play: Investigating neural and language correlates of doll play in a neurodiverse sample. European Journal of Neuroscience, n/a(n/a). https://doi.org/10.1111/ejn.16144
Kidd, D., & Castano, E. (2018). Reading literary fiction can improve theory of mind. Nature Human Behaviour, 2, 604-604.
Kirk, E., Pine, K., Wheatley, L., Howlett, N., Schulz, J., & Fletcher, B. C. (2015). A longitudinal investigation of the relationship between maternal mind-mindedness and theory of mind. The British Journal of Developmental Psychology, 33(4), 434–445. https://doi.org/10.1111/bjdp.12104
Kokkinos, C. M., Kakarani, S., & Kolovou, D. (2016). Relationships among shyness, social competence, peer relations, and theory of mind among pre-adolescents. Social Psychology of Education: An International Journal, 19(1), 117–133. https://doi.org/10.1007/s11218-015-9317-7
Lalonde, C. E., & Chandler, M. J. (1995). False belief understanding goes to school: On the social-emotional consequences of coming early or late to a first theory of mind. Cognition and Emotion, 9(2–3), 167–185. https://doi.org/10.1080/02699939508409007
Langley, C., Cirstea, B. I., Cuzzolin, F., & Sahakian, B. J. (2022). Theory of Mind and Preference Learning at the Interface of Cognitive Science, Neuroscience, and AI: A Review. Frontiers in Artificial Intelligence, 5. https://doi.org/10.3389/frai.2022.778852
Lecce, S., & Devine, R. T. (2022). Theory of mind at school: Academic outcomes and the influence of the school context. Infant and Child Development, 31(1), e2274. https://doi.org/10.1002/icd.2274
Lecce, S., Ronchi, L., & Devine, R. T. (2024). The effect of peers’ theory of mind on children’s own theory of mind development: A longitudinal study in middle childhood and early adolescence. Developmental Psychology, No Pagination Specified-No Pagination Specified. https://doi.org/10.1037/dev0001758
Lillard, A. S. (1993). Pretend play skills and the child’s theory of mind. Child Development, 64(2), 348–371. https://doi.org/10.2307/1131255
Lillard, A. S. (2015). The development of play. In Handbook of child psychology and developmental science: Cognitive processes, Vol. 2, 7th ed (pp. 425–468). John Wiley & Sons, Inc. https://doi.org/10.1002/9781118963418.childpsy211
Lloyd-Fox, S., Széplaki-Köllőd, B., Yin, J., & Csibra, G. (2015). Are you talking to me? Neural activations in 6-month-old infants in response to being addressed during natural interactions. Cortex, 70, 35–48. https://doi.org/10.1016/j.cortex.2015.02.005
Lundy, B. L. (2013). Paternal and Maternal Mind-mindedness and Preschoolers’ Theory of Mind: The Mediating Role of Interactional Attunement. Social Development, 22(1), 58–74. https://doi.org/10.1111/sode.12009
MacLean, E. L. (2016). Unraveling the evolution of uniquely human cognition. Proceedings of the National Academy of Sciences, 113(23), 6348–6354. https://doi.org/10.1073/pnas.1521270113
McAlister, A., & Peterson, C. C. (2006). Mental playmates: Siblings, executive functioning and theory of mind. British Journal of Developmental Psychology, 24(4), 733–751. https://doi.org/10.1348/026151005X70094
Meins, E., Fernyhough, C., Arnott, B., Leekam, S. R., & de Rosnay, M. (2013). Mind-mindedness and theory of mind: Mediating roles of language and perspectival symbolic play. Child Development, 84(5), 1777–1790. https://doi.org/10.1111/cdev.12061
Meins, E., Fernyhough, C., Wainwright, R., Das Gupta, M., Fradley, E., & Tuckey, M. (2002). Maternal mind-mindedness and attachment security as predictors of theory of mind understanding. Child Development, 73(6), 1715–1726. https://doi.org/10.1111/1467-8624.00501
Moghaddam, S. R., & Honey, C. J. (2023). Boosting Theory-of-Mind Performance in Large Language Models via Prompting (arXiv:2304.11490). arXiv. https://doi.org/10.48550/arXiv.2304.11490
Newton, E., & Jenvey, V. (2011). Play and theory of mind: Associations with social competence in young children. Early Child Development and Care, 181(6), 761–773. https://doi.org/10.1080/03004430.2010.486898
Patacchiola, M., & Cangelosi, A. (2022). A Developmental Cognitive Architecture for Trust and Theory of Mind in Humanoid Robots. IEEE Transactions on Cybernetics, 52(3), 1947–1959. https://doi.org/10.1109/TCYB.2020.3002892
Ramezani, A., Ghamari, M., Jafari, A., & Aghdam, G. F. (2020). The effectiveness of a ToM training program in promoting empathy between married couples. Journal of Couple & Relationship Therapy, 19(1), 1–25. https://doi.org/10.1080/15332691.2019.1620145
Redcay, E., Dodell-Feder, D., Pearrow, M. J., Mavros, P. L., Kleiner, M., Gabrieli, J. D. E., & Saxe, R. (2010). Live face-to-face interaction during fMRI: A new tool for social cognitive neuroscience. NeuroImage, 50(4), 1639–1647. https://doi.org/10.1016/j.neuroimage.2010.01.052
Richard, S., Clerc-Georgy, A., & Gentaz, E. (2023). Pretend play-based training improves some socio-emotional competences in 5–6-year-old children: A large-scale study assessing implementation. Acta Psychologica, 238, 103961. https://doi.org/10.1016/j.actpsy.2023.103961
Schwebel, D. C., Rosen, C. S., & Singer, J. L. (1999). Preschoolers’ pretend play and theory of mind: The role of jointly constructed pretence. British Journal of Developmental Psychology, 17(3), 333–348. https://doi.org/10.1348/026151099165320
Sommerville, J. A., Bernstein, D. M., & Meltzoff, A. N. (2013). Measuring Beliefs in Centimeters: Private Knowledge Biases Preschoolers’ and Adults’ Representation of Others’ Beliefs. Child Development, 84(6), 1846–1854. https://doi.org/10.1111/cdev.12110
Suway, J. G., Degnan, K. A., Sussman, A. L., & Fox, N. A. (2012). The relations among theory of mind, behavioral inhibition, and peer interactions in early childhood. Social Development (Oxford, England), 21(2), 331–342. https://doi.org/10.1111/j.1467-9507.2011.00634.x
Tomasello, M., Carpenter, M., Call, J., Behne, T., & Moll, H. (2005). Understanding and sharing intentions: The origins of cultural cognition. The Behavioral and Brain Sciences, 28(5), 675–691; discussion 691-735. https://doi.org/10.1017/S0140525X05000129
Tomasello, M., & Rakoczy, H. (2003). What makes human cognition unique? From individual to shared to collective intentionality. Mind & Language, 18(2), 121–147. https://doi.org/10.1111/1468-0017.00217
Weimer, A. A., Warnell, K. R., Ettekal, I., Cartwright, K. B., Guajardo, N. R., & Liew, J. (2021). Correlates and antecedents of theory of mind development during middle childhood and adolescence: An integrated model. Developmental Review, 59. https://doi.org/10.1016/j.dr.2020.100945

Yes there is potential Competing Interest. Funding for this research was provided by Mattel Inc. They were involved in initial discussions regarding the study design. Mattel did not have a say in the final study design nor played any role in the collection or analysis of the data; interpretation of the results; or writing of the manuscript.

Doll Play Improves Social Processing: Evidence from a Randomized-Control Trial

Status:

Version 1

Abstract

Figures

Introduction

Results

Discussion

Methods

References

Additional Declarations

Status:

Version 1