Moral reinforcement in children 1 Running head : Moral reinforcement in children

Does considering one’s past prosociality affect future behavior? Prior research has revealed instances in which adults engage in additional prosocial behavior––moral reinforcement––as well as instances in which adults engage in worse behavior––moral licensing. The current study examined the developmental origins of these effects by testing whether sixto eight-year-olds (N=225) are more or less generous after recalling their own good deeds. Children were asked to recount a time when they were nice, mean, or watched a movie. Children behaved more generously after recalling a time when they were nice. We show that this boost in generosity was not simply the result of instructing children to consider nice behavior––children’s giving did not increase after recalling others’ good deeds. We also show that, even after recounting multiple instances of their past goodness, children continue to behave more generously. These findings suggest that doing good leads to more good in children. Moral reinforcement in children 3 Memories of good deeds past: The reinforcing power of prosocial behavior in children In recent years, the question of why people behave prosocially has been of much interest to psychologists; indeed, a burgeoning literature has investigated the role of one’s past good deeds in promoting future good deeds. A number of studies have demonstrated that instructing adults to recount their prior good deeds increases their subsequent prosocial behavior, an effect known as moral reinforcement (Nelson & Norton, 2005; Stone & Cooper, 2001; Young, Chakroff, & Tom, 2012). These studies suggest that reflecting on one’s past positive behavior leads to further positive behavior, perhaps because people perceive themselves as “do-gooders”. However, a parallel line of research has uncovered instances of moral licensing––in some cases, adults feel licensed to behave badly when reminded of their previous good deeds (Mazar & Zhong, 2010; Monin & Miller, 2001; Sachdeva, Iliev, & Medin, 2009). Specifically, this work indicates that people show less moral striving after they confirm their goodness through recalling past good deeds. Several accounts have been offered for when and why reinforcement versus licensing occurs (Conway & Peetz, 2012; Cornelissen, Bashshur, Rode, Le Menestrel, 2013; Gneezy, Imas, Brown, Nelson, & Norton, 2012; Mullen & Monin, 2016), but surprisingly little is known about the developmental origins of these effects. The study of children provides an important window into this debate for several reasons. As previous researchers have argued, understanding children’s behavior could constrain theorizing concerning adults’ behavior (Dunham & Olson, 2008; Olson & Dunham, 2010). Moreover, work in this domain has important, everyday implications. Understanding how children behave after recalling their past good deeds will not only contribute to current theorizing and research, but also inform approaches for eliciting moral behavior from an early age. Moral reinforcement in children 4 Is there any reason why children would behave more or less generously after recalling past good deeds? Previous research indicates that children actively manage their moral identities (Bryan, Master, & Walton, 2014). In particular, children were introduced to the idea of “being a helper” or “helping” following previous work documenting that a noun wording (e.g., helper) compared to a verb wording (e.g., helping) invokes a person’s identity (Gelman & Heyman, 1999; Walton & Banaji, 2004). In this research, activating children’s moral identity by exposing them to the idea of “being a helper” led to greater moral motivation. Given these findings, it may be that reflecting on one’s past good deeds leads children to perceive themselves as “dogooders”, as has been argued with adults, which in turn motivates them to behave more generously. Alternatively, children may behave more selfishly after reflecting on their past good deeds and the “moral currency” they have accrued over the years. From an early age, children routinely engage in positive behaviors. For example, in the first few years of life, children share their toys (Schmidt & Sommerville, 2011) and help others achieve their goals (Warneken & Tomasello, 2006). Given the early emergence of human prosociality, it may be that individual children are able to reflect on the moral credit they have gained, feel confident in their moral goodness, and, consequently, use their past good deeds to justify selfish behavior, as in previous work in adults (Merritt, Effron, & Monin, 2010; Miller & Effron, 2010). Here, we focused on sixto eight-year-olds given that children of these ages show flexibility in their moral behavior. For example, children between the ages of six to eight voluntarily incur costs to avoid unfairness (Blake & McAuliffe, 2011; Shaw & Olson, 2012), yet they will behave unfairly when no one is watching (Shaw, Montinari, Piovesan, Olson, Gino, & Norton, 2013). Thus, sixto eight-year-olds are motivated to do good but are willing to do Moral reinforcement in children 5 otherwise if given the opportunity, inviting the question of whether children at this age engage in additional good behavior or bad behavior after reflecting on their past prosociality. The Current Study In the current study, children were assigned to one of five conditions. One of these conditions (baseline) served as a baseline measure of children’s generosity, in which children were asked to recount a time in the past when they watched a movie. Three of these conditions, in which children were instructed to recall moral memories, served as key tests of our hypotheses; children were asked to recall (1) a time when they were nice to someone (nice), (2) a time when they were mean to someone (mean), or (3) three different times when they were nice to someone (moral credit). We included the moral credit condition following work in adults showing that people behave particularly stingily after they accrue a surplus of “moral credit” (Merritt, Effron, & Monin, 2010; Miller & Effron, 2010). Thus, if children were to show effects of moral licensing, then they should be especially likely to do so in the moral credit condition, as they would accumulate three times the amount of moral credit as in the nice condition. Finally, we included a condition in which children were asked to recount a time in the past when someone was nice to them (other-nice). This condition allowed us to distinguish whether any boost in giving following the recall of past good deeds was the specific result of reinforcing children’s own prosocial behavior, or simply reminding children to consider prosocial behavior in general. Although the primary focus of the current study was to examine the impact of recalling past good deeds on children’s future behavior, our design also allowed us to investigate the nature of children’s moral memories. First, prior work examining children’s accounts of their immoral behavior indicates that they focus on others’ provocations when explaining their reasons Moral reinforcement in children 6 for transgressing (Wainryb, Brehl, & Matwin, 2005). For example, in one study, when asked to describe sibling conflicts, children were more likely to claim that they were innocent, whereas their siblings were guilty (Wilson, Smith, Ross, & Ross, 2004). Based on this research, we predicted that children would be more likely to describe their good versus bad deeds as the product of their own volition. Second, we examined the types of memories produced and whether children identified a specific nice or mean action. Recent theorizing and research suggests that abstractly thinking about one’s moral behavior activates one’s moral identity and leads to reinforcement, whereas concretely thinking about one’s moral behavior activates self-regulatory behavior and leads to licensing (Conway & Peetz, 2012; Mullen & Monin, 2016). This account would predict that children who did not identify a specific action in the nice condition (e.g., “I was nice to my sister”) would be thinking more abstractly about their moral identity and therefore give more than children who did identify a specific action (e.g., “I gave my sister a toy”). Similarly, children who did not identify an action in the mean condition (e.g., “I was mean to my sister”) would give less than children who did (e.g., “I took a toy away from my sister”). Participants 225 sixto eight-year-old children (124 girls; M = 7.10 years; range = 6.03-8.54 years) were recruited from the greater New Haven, Connecticut area and were tested individually in a quiet room at their elementary school. Parents of participating children gave written informed consent; children also provided oral assent. All experiments were conducted in elementary schools in suburban Connecticut representing largely White, middle-class, and educated