Context specificity in adolescents’ implicit theories of morality

Highlights

• Adolescents endorse moral essentialism and incrementalism in context-specific ways.

• There are between and within person differences associated with moral mindset.

• Moral essentialism is endorsed more in prosocial contexts.

• Moral incrementalism is endorsed more in antisocial contexts.

• Moral mindset predicts character and situation judgments across contexts.

Abstract

This study investigated between and within person differences in adolescents’ endorsement of moral essentialism and moral incrementalism across various types of morally-relevant situations. The sample included 97 adolescents (age range = 12–15 years). Adolescents responded to vignettes depicting recurrent and nonrecurrent actions in prosocial and antisocial contexts, with questions assessing moral essentialism, incrementalism, and other character and situation judgments. The majority of the variance for essentialism and incrementalism was linked to within-person differences across contexts. Findings revealed between-person associations between moral mindset, implicit theories of personality, and externalization of blame. Adolescents endorsed moral essentialism more in prosocial contexts and incrementalism more in antisocial contexts. Within-person variations in essentialism were linked to likability, acceptability, person attributions and consequence judgments in particular contexts. Results corroborate research and theorizing about the flexible endorsement of moral essentialism and incrementalism, and document the ways in which adolescents’ judgments are responsive to the unique features of events. Findings are discussed in terms of their contributions to scholarship on moral mindsets during adolescence.

Introduction

As they navigate through everyday experiences, adolescents engage with the complexities of moral life. In their day to day interactions, adolescents reason about morally-laden situations, weigh conflicting considerations, and apply their moral concepts flexibly (Recchia, Wainryb, & Pasupathi, 2013; Turiel & Wainryb, 2000; Wainryb, 2006). In this regard, the mindsets that adolescents have developed may be linked to how they navigate their moral experiences. That is, adolescents may process and respond to situations differently depending on their endorsement of moral essentialism (i.e., that moral attributes are stable and due to internal factors) and incrementalism (i.e., that attributes are unstable, change with effort, and due to external factors; Huang, Zuo, Wang, Cai, & Wang, 2017). However, little is known about how morality and these social information processing patterns intersect, and whether these associations are similar or different across prosocial and antisocial moral contexts. In other words, while mindsets in the moral domain are typically considered with respect to between-person differences, it is also important to capture within-person variations in adolescents’ essentialist and incremental attributions (e.g., Hughes, 2015). Furthermore, given evidence suggesting people can hold both incremental and essentialist beliefs simultaneously (e.g, Dweck, Chiu, & Hong, 1995; Molden, Plaks, & Dweck, 2006), it is useful to consider moral essentialism and incrementalism as distinct processes. Examining adolescents’ essentialist and incremental beliefs in the moral domain is important, inasmuch as they increasingly develop a sense of their moral identities during this period (Krettenauer & Hertz, 2015) as well as advancing in their thinking about whether and how others can change (Killen, Crystal, & Watanabe, 2002; Yeager, Trzesniewski, Tirri, Nokelainen, & Dweck, 2011). Yet this age range is understudied in that most research on moral mindsets has focused on either childhood or adulthood. Therefore, the purpose of this study was to extend a growing body of research on moral mindsets by investigating between and within person differences in adolescents’ endorsement of moral essentialism and incrementalism across various types of morally-relevant situations.

Children formulate theories about their social worlds that guide the ways in which they understand and evaluate themselves and their experiences (Dweck & Leggett, 1988; Levy & Dweck, 1998). Specifically, there are two overarching types of mindsets that inform how children think about and react to social events. The first involves an understanding that qualities, aptitudes, and traits are fixed (i.e., fixed mindset; essentialist view). In this mindset, individuals believe that intelligence, personality, and moral character are decided by inheritance, are innate, and do not change with effort or time (Dweck, Hong, & Chiu, 1993). By contrast, incremental theorists understand that qualities, traits, and aptitudes are malleable, in that they are shaped by the environment and personal experiences (Chiu, Hong, & Dweck, 1997; Dweck, 2006; Levy & Dweck, 1998). Importantly, distinct implicit theories can be held across different domains (Dweck et al., 1995).

Cognitive-developmental scholarship suggests that essentialism is a cognitive bias that emerges early on (e.g., Eidson & Coley, 2014; Gelman, 2003; Gil-White, 2001). For instance, describing behavior using trait labels (e.g., “smart”) influences 4-years-olds’ predictions of future behavior (Liu, Gelman, & Wellman, 2007). More specific to the moral domain, kindergarteners expect a character’s future antisocial behaviors to match the valence of their previous actions (Cain, Heyman, & Walker, 1997). There is also emerging evidence that moral essentialism and incrementalism are influential in informing prosocial behaviors, such as sharing resources, in early childhood (i.e., 4–5 years-old; Rhodes, Leslie, Saunders, Dunham, & Cimpian, 2018).

Across the elementary school years, children increasingly expect people to behave in consistent ways, such as anticipating that a child who shared with a hungry peer would also behave generously in the future by helping someone perform a chore (Rholes & Ruble, 1984). One interpretation of these findings is that older children make more trait-based inferences; in this sense, they appear to increasingly understand morality in essentialist terms whereby morally relevant behaviors stem from internal, unchanging characteristics (Heiphetz, 2020; Heyman, 2009). Indeed, moral characteristics may be perceived as especially fundamental to one’s “essence”: school aged children were more likely to believe that a change of identity occurred if one’s moral beliefs changed as compared to non-morally relevant characteristics, such as preferences (Heiphetz, Strohminger, Gelman, & Young, 2018). Further, mindset has been shown to influence how elementary school aged children make social judgments, whereby essentialist views are linked to more generalized, global negative evaluations and trait labels such as “bad”, less empathy, and more endorsement of punishment (Erdley & Dweck, 1993; Heyman & Dweck, 1998).

Early to middle adolescence is a crucial period for the emergence of a moral identity (Krettenauer & Hertz, 2015). In this age range, morality and identity become increasingly connected and ideologically-based, and adolescents begin to define themselves in light of their moral ideals and commitments (Hardy & Carlo, 2011; Hardy, Bean, & Olsen, 2015). Moreover, adolescents’ self-understandings become more nuanced, differentiated and complex as they become increasingly adept at drawing self-connections and exploring the psychological facets of their morally laden experiences (McLean & Pasupathi, 2012). Alongside self development, adolescents also develop in their understandings of whether and how others can change (Yeager et al., 2011). Nevertheless, it is unclear whether essentialism is likely to increase or decrease from middle childhood to adolescence. As compared to younger children, adolescents may hold more fixed views regarding a peer’s ability to change their aggressive behavior (Killen et al., 2002). Moreover, adolescents increasingly use previous information regarding a peer’s prior transgressions in judging their current actions (Killen, Kelly, Richardson, & Jampol, 2010). Conversely, there is other evidence suggesting that essentialism might decline across this age range; elementary school aged children have been found to hold more essentialist beliefs than adults and to perceive less environmental influence (e.g., Cimpian & Steinberg, 2014; Heiphetz, 2019). With respect to individual differences, adolescents who hold more essentialist views of personality are more likely to experience negative feelings such as shame and sadness as well as endorse vengeful responses such as punishment and retaliation (Yeager et al., 2011). As such, though there is emerging evidence to suggest the predictive significance of mindset in adolescence, little research has examined moral mindsets during this developmental period.

Among adults, research suggests that individuals who endorse essentialist beliefs tend to be more concerned with duty-based morality than those who endorse incremental beliefs, as they are more likely to support the status quo and focus on sanctioning and punishing deviance. On the other hand, those who endorse moral incrementalism are more concerned with rights-based morality as they tend to focus on social change and support people’s rights to equality (Chiu et al., 1997; Dworkin, 1978; Hughes, 2015; Wurthmann, 2017). Additionally, moral incrementalism and essentialism are linked to distinct moral character attributions such that adults who hold a more essentialist view use less information to cast judgments and make moral decisions (Chiu et al., 1997; Haselhuhn, Schweitzer, & Wood, 2010). Moreover, adults who hold an incremental view of moral character are more likely to trust and forgive after a received apology and to engage in voluntary service behavior (Han, Choi, Dawson, Jeong, & Lamm, 2018; Haselhuhn et al., 2010). Nevertheless, as noted above, these associations have not been widely examined in adolescence, and thus it is important to address whether these patterns are evident across development.

Overall associations with moral mindsets, however, should be considered in conjunction with evidence suggesting context-specificity in incrementalism and essentialism, both generally and more specifically in the domain of morality. That is, even though overall levels of incrementalism and essentialism tend to be negatively correlated with one another, adults have been shown to hold distinct beliefs depending on the domain (i.e., intelligence, personal attributes, morality; Hughes, 2015). For instance, someone with a view that intelligence is malleable and can be changed through effort might also simultaneously believe that others are born with a stable personality that cannot change regardless of how much effort is applied. Moreover, even within a given domain such as intelligence, there is evidence to suggest that pre-adolescents (i.e., 7-9th grade students) can hold distinct implicit theories depending on the particular focus of consideration (i.e., mathematics versus other academic domains; Tarbetsky, Collie, & Martin, 2016).

There is also recent evidence pointing to context-specificity of implicit theories within the domain of morality. Findings by Heiphetz (2019) indicated that goodness is viewed in more essentialist terms than badness, by both 5- to 8-year-old children and adults, implying that moral essentialism should be considered in relation to specific domain components (e.g., goodness/badness). This is in line with work suggesting a person-positivity bias, whereby adults tend to see human beings in a positive light and tend to regard people’s “true selves” as fundamentally good (Newman, Bloom, & Knobe, 2014; Newman, De Freitas, & Knobe, 2015). Conversely, other studies suggest negativity and extremity biases in impression formation, such that negative moral behaviors are viewed as more diagnostic (Skowronski & Carlston, 1987, 1989). It is worth noting that recent reviews on diagnosticity suggest that these processes are primarily at play when there are alternative categories to choose from (e.g., consistent vs. inconsistent behavior) and also that the effect might be particularly robust when applied to extreme behaviors (Rusconi, Sacchi, Brambilla, Capellini, & Cherubini, 2020). With respect to age-related change, Lockhart, Chang, and Story (2002) found that the tendency to essentialize positive qualities more readily than negative qualities may be particularly evident in early childhood (as compared to middle childhood and adulthood), although these patterns have not been examined in adolescence. In sum, these findings suggest that it is useful to extend research on the context-specificity of moral mindsets into the adolescent period.

A second important feature of the context to consider involves behavioral repetition. Both moral transgressions and prosocial actions tend to be understood differently when they occur repetitively. For instance, Heiphetz (2019) notes that, in a legal context, three strike laws tend to suggest that multiple offences are indicative of bad moral character and that it is especially important to punish multiple offenders harshly. Further, adults who are induced with an incremental mindset tend to blame others significantly more after they show continual (but not single) failure as compared to participants induced with an essentialist mindset, suggesting increased blame when participants were primed with perceived control over one’s abilities (Ryazanov & Christenfeld, 2018a). As such, both valence and recurrence of moral behaviors are relevant contexts to explore.

Some scholars have also pointed to context-specificity in considering the outcomes associated with mindset. Overall, as noted above, incrementalism tends to be associated with positive outcomes and essentialism with negative outcomes among both children and adults (e.g., Erdley & Dweck, 1993; Haselhuhn et al., 2010). However, it has also been suggested that essentialized thinking is not always detrimental; rather, the effects associated with essentialized thinking depend on motivation and context (Ryazanov & Christenfeld, 2018a). For instance, essentializing the good aspects of oneself or of one’s group can be beneficial (Ryazanov & Christenfeld, 2018b). In prosocial situations, characterological (i.e., “helper”) rather than behavioral attributions (i.e., “helping”) has been shown to encourage 3- to 6-year-olds’ prosocial behavior, and similarly, asking people to not be a “cheater” rather than to not “cheat” has been shown to discourage cheating (Bryan, Master, & Walton, 2014), although these attributions may also backfire if children experience setbacks (Foster-Hanson, Cimpian, Leshin, & Rhodes, 2020). Thus, it is important to go beyond characterizing individuals as moral essentialists or incrementalists and to consider how patterns and outcomes of social information processing vary depending on the unique features of the moral situation under consideration.

The present study aimed to assess both between- and within-person variations in adolescents’ endorsement of moral mindsets. To address this goal, we developed a new vignette-based measure of incrementalism and essentialism across various morally-laden contexts. We considered global between-person differences by examining associations with a measure of implicit theories of personality (Dweck et al., 1995) as well as affective reactions to moral events (Tangney, Wagner, Gavlas, & Gramzow, 1991), based on established links to these factors and other definitional elements associated with moral incrementalism and essentialism in the literature (e.g., Dweck et al., 1993; Han et al., 2018). That is, adolescents who hold more essentialist views of personality tend to experience more negative feelings about themselves and are more likely to experience shame or depressive symptoms after victimization (Erdley, Cain, Loomis, Dumas-Hines, & Dweck, 1997; Graham & Juvonen, 1998); as such, we expected to find that moral essentialism would be linked to shame- and guilt-proneness.

We also assessed whether and how adolescents’ moral mindsets varied across contexts depicting others’ prosocial and antisocial actions that were described as recurrent or nonrecurrent, as well as the judgments of others associated with adolescents’ situation-specific endorsement of essentialist and incremental attributions. Based on past theory and research (e.g., Dweck et al., 1993; Heiphetz, 2019; Ryazanov & Christenfeld, 2018b; Yeager et al., 2011) we expected adolescents to endorse moral essentialism more when stories depicted a hypothetical character acting prosocially, whereas we expected adolescents to endorse moral incrementalism more when the stories depicted antisocial behavior. We further expected that adolescents would endorse moral incrementalism more than moral essentialism in one-time incidents since individuals who hold a more incremental mindset tent to draw less abrupt conclusions (Yeager et al., 2011). Conversely, we expected adolescents to endorse moral essentialism more than moral incrementalism when actions were described as recurrent. We anticipated that this might be particularly the case for prosocial actions described as recurrent, given research suggesting that children and adults have predispositions to essentialize goodness (Heiphetz, 2019).

Regarding associations between adolescents’ moral mindsets and their character and situation judgments in response to hypothetical moral situations, on the basis of past research (e.g., Erdley & Dweck, 1993; Heiphetz, 2019; Yeager et al., 2011; Yeager & Dweck, 2012) we expected that within-person variations in essentialist and incremental attributions across contexts would be linked to predictable differences in judgments of action acceptability (whether the behavior was okay or not okay), character likability (whether the protagonist was likable or unlikable), consequences of actions (whether the behavior was deserving of praise/punishment), and person attributions (whether the protagonist was a good/bad person). Specifically, we hypothesized that greater endorsement of moral incrementalism and less endorsement of essentialism would be linked to less extreme judgments of (un)acceptability and character likability, less endorsement of praise and punishment, and fewer person attributions. We also considered whether these associations differed across contexts; given the limited literature on moral mindsets among adolescents, we tested the variations in the magnitude of these associations between prosocial/antisocial contexts and recurrent/nonrecurrent events on an exploratory basis.

Finally, we considered associations with age, especially considering the substantive ways in which adolescents’ understandings of themselves and others develop across early adolescence (Krettenauer & Hertz, 2015). These associations were exploratory, given inconsistencies in the current literature. For example, some research suggests that 5–8-year-olds hold more essentialist beliefs than adults (Cimpian & Steinberg, 2014; Gelman, Heyman, & Legare, 2007; Heiphetz, 2019). Conversely, other evidence suggests that older children expect more consistent moral behavior and make more trait-based inferences (Heiphetz, 2020; Rholes & Ruble, 1984). Gender differences were also examined in an exploratory way, given that gender differences in mindset have not been widely investigated in the adolescent years.