The Impact of Imaginary Companions on Moral Behaviors: An Experimental Study in Early Childhood

• In Event: 1-142 - A Multicontextual Examination of Imagination in Development: Exploring Culture, Social Environment, Cognition, and Morality

Integrative Statement

Today, the creation of an imaginary companion (IC) is regarded as a normal aspect of development, with about 30-40 % of children creating an IC (Moriguchi & Todo, 2018). Two types of ICs are generally observed: a personified object (PO) and an invisible friend (IF). In fact, engaging in play with ICs is associated with various cognitive, emotional, and social benefits, such as advanced theory of mind, narrative skills, emotional coping and competence, and the ability to get along well with others (Gleason & Kalpidou, 2014; Hoff, 2005; Taylor, 1999; Trionfi & Reese, 2009). However, no work has specifically explored whether interactions with an IC affect children’s moral development. Previous studies have shown that people engage in more moral behaviors when they believe someone else is watching them (Fujii, Takagishi, Koizumi, & Okada, 2015). Based on this research and previous findings that children are sensitive to the perceptual properties of ICs (e.g., children think that ICs watch them; Moriguchi & Shinohara, 2012), we hypothesized that children’s prosocial behaviors would be improved by the presence of their IC.
Forty 4- to 6-year-old healthy Japanese children participated in the present study (M ± SD = 65.53 ± 9.93 months, 28 girls). All participants had at least one PO and came our laboratory with their IC. To confirm the existence of an PO, each child was interviewed following Taylor and Carlson’s (1997) Imaginary Companion Interview. No children were excluded by this confirmation. Afterward, children completed a prosocial behavior sticker task. In this task, the children were allowed to take some stickers from an anonymous child’s 10 stickers, while researchers measured the number of stickers taken. More stickers taken was indicative of less prosocial behavior. All children completed the sticker task three times with the following manipulations: (1) alone, (2) in front of a female assistant (human condition), and (3) in front of their IC (IC condition). The order in which these manipulations were presented was counterbalanced.
The number of stickers children took during each condition was examined using a repeated-measures ANOVA with manipulation (alone vs human vs IC) as the within-subject variable. Results indicated a significant main effect of manipulation, F(2, 39) = 3.81, p = .027, ηp2 = .089. Post-hoc analyses revealed that children tend to take more stickers when in the presence of their IC than when in the presence of another human or when alone, t(39) = 2.47, p = .054, r = .33 and t(39) = 1.80, p = .080, r = .28, respectively. The difference between alone manipulation and human manipulation was not significant, t(39) = 1.24, p =.221, r = .20. (See Figure 1.) Thus, children showed lower prosocial behaviors during the IC manipulation compared to the human and the alone manipulations, perhaps because ICs functioned as a motivator for mischievous deeds (Hoff, 2005) and their presence encouraged children to take more stickers from anonymous children.

Authors

• Masanori Yamaguchi, Kyoto University
  Presenting Author

• Yusuke Moriguchi, Kyoto University
  Non-Presenting Author