Negativity bias in infants’ expectations about agents’ dispositions

Abstract

This study investigated 6- and 10-month-old infants' abilities to infer others' preferences based on social interactions using looking time and choice measures. Infants were randomly assigned to either a helping/neutral or hindering/neutral condition. Those in the helping/neutral condition were first familiarized with a helping event, in which an agent helped a circle climb a hill, and a neutral event, in which another agent followed the same path as the helping agent but had no interaction with the circle. During the test phase, the circle approached either the helper or the neutral agent. In the hindering/neutral condition, the infants were familiarized with a hindering event, in which an agent hindered the circle from reaching the top of the hill, and a neutral event, in which another agent followed the same path as the hindering agent but had no interaction with the circle. During the test phase, the circle approached either the hinderer or the neutral agent. For the looking-time measure, infants in the hindering/neutral condition looked reliably longer at the approach-hinderer than at the approach-neutral agent event, whereas those in the helping/neutral condition looked for equal amounts of time at both test events. These results suggest that the infants expected the circle to avoid the hinderer but did not expect it to approach the helper. In the choice task, infants chose the helper more often than the neutral agent and the neutral agent more often than the hinderer, suggesting an ability to generate their own preferences for a particular agent based on the valence of helping and hindering actions. This research demonstrates infants' sensitivity to the moral valence of agents' social interactions, which may serve as a foundation for advanced socio-moral reasoning. Statement of contribution What is already known on this subject? Research on infants' ability in social evaluation has established that even preverbal infants can distinguish between positive and negative social interactions. Infants as young as 6 months of age can distinguish between helping and hindering actions and can generate their own preference towards helpful agents. What does this study add? The present study sheds light on infants' ability to infer a third-party's preference, which is a more challenging task for the infants than generating their own preference. Specifically, 6- and 10-month-old infants could infer others' preference for the neutral agent over the hinderer. Such results demonstrate infants' sensitivity to the moral valence of agents' social interactions and provide an evidence of negativity bias in social evaluation.

How do toddlers evaluate defensive actions toward third parties?

“…These findings revealed that 10‐month‐olds are capable to understand the intention behind intervening to prevent someone being harmed. These results are in line with previous findings, suggesting that infants can infer or predict an agent's disposition toward another agent, based on whether a previous social interaction between the agents was positive or negative (Chae & Song, 2018; Fawcett & Liszkowski, 2012; Hamlin et al, 2007; Kuhlmeier, Wynn, & Bloom, 2003; Lee, Yun, Kim, & Song, 2015). Taken together, these previous findings suggest that infants can evaluate prosocial or antisocial actions and can generate their own preferences or infer others’ preference for positive agents.…”

Social evaluation and imitation of prosocial and antisocial agents in infants, children, and adults

“…A number of studies have tried to replicate and extend the findings of Hamlin and colleagues [ 7 , 8 , 10 ]. The results were often in line with the original findings (e.g., [ 13 – 24 ]) with seven studies not conducted by Hamlin and colleagues [ 13 , 14 , 19 – 22 ]. This overall impression was confirmed by a recent meta-analysis that focused on young children’s manual choice between a prosocial and an antisocial agent [ 11 ].…”

Use of Repeated Within-Subject Measures to Assess Infants’ Preference for Similar Others

“…The current study included a total of 44 infant/parent dyads (20 in Experiment 1 and 24 in Experiment 2), and these numbers were determined prior to the start of data collection. Although recommended (Simmons et al, 2011), a description for pre-determining sample size is rarely provided in published studies using the single-choice infant paradigm (e.g., Hamlin et al, 2007, 2010, 2013; Mahajan and Wynn, 2012; Gerson et al, 2017; Chae and Song, 2018; Margoni and Surian, 2018) and at times is not included as part of the study protocol (Eason et al, 2017). Small sample sizes are problematic, as they increase the likelihood of spurious findings (Oakes, 2017).…”

“…Infants as young as 5 months (Hamlin et al, 2007, 2010, 2011; Hamlin and Wynn, 2011, 2012) and older infants and toddlers (Geraci and Surian, 2011; Buon et al, 2014; Scola et al, 2015; Woo et al, 2017; Chae and Song, 2018) seem capable of socially evaluating the behavior of others and appear to show a preference for prosocial others (for reviews, see Martin and Olson, 2015; Holvoet et al, 2016; Van de Vondervoort and Hamlin, 2018) as well as those who are similar to themselves along some dimension (e.g., Hamlin and Wynn, 2012; Mahajan and Wynn, 2012; Hamlin et al, 2013; Burns and Sommerville, 2014; Gerson et al, 2017). These findings have led researchers to hypothesize that we may be born with something akin to an innate moral core (Cook, 2013; Hamlin, 2013) or early strong tendencies (Martin and Olson, 2015; Holvoet et al, 2016; Hare, 2017), which include a preference for similar others (Hamlin and Wynn, 2012; Mahajan and Wynn, 2012; Hamlin et al, 2013) and that these tendencies observed during infancy may predict social and behavioral adjustment at 4 years of age (Tan et al, 2018).…”