Connection between control perceptions and the tendency to moralize and hold people responsible for their actions: Conservatives around the world perceived they had more control over their lives relative to liberals

Age Differences in Free Will and Control Perceptions Across the Lifespan and Around the World. William J. Chopik, Joshua A. Confer, Matt Motyl. Current Research in Ecological and Social Psychology, March 15 2023, 100093. https://doi.org/10.1016/j.cresp.2023.100093

Highlights

• Free will and control perceptions vary across development and cultures.

• Few studies examined cultural variation in age differences in these characteristics.

• We found that older adults perceived less control across cultures.

• Cultural and individual characteristics occasionally moderated age differences.

Abstract: Variation in free will and control perceptions has been examined across the development of young children, adults, and in several different countries. In two studies (three samples; total N = 492,134), older adults believed less in free will, fatalistic determinism, and perceived less control over their lives compared to younger adults. In Study 2 (Samples 1 [48 countries] and 2 [99 countries]), control perceptions were highest among individuals who lived in countries that were more indulgent (versus restricted). Country-level characteristics often moderated the link between age and control perceptions, although variation in age differences was relatively small. The current studies are the largest and most comprehensive investigations of demographic and cultural differences in free will and control perceptions. The findings are discussed in the context of the mechanisms that drive changes in free will and control perceptions across the lifespan and across cultures.

General Discussion

Although most people believe in free will and perceive that they have control over their lives, these perceptions likely vary across several dimensions. Across three samples (N = 492,134), we found that free will and control perceptions were higher among young adults and lower among older adults. Men, conservatives, educated individuals, and those with higher religious attendance tended to report having more control over their lives. In Study 2, people from countries higher in indulgence reported higher control perceptions. Country-level characteristics often moderated the link between age and free will beliefs, but these associations were often very small. These studies represent the largest and most comprehensive examinations of lifespan and cultural differences in free will and control perceptions conducted to date.

Demographic Differences in Free Will and Control Perceptions

Across the three samples, older adults perceived less free will and control over their lives. This finding could be attributable to the fact that as people age, they begin to learn and acknowledge the wide-ranging constraints on one's mind and behavior (see Chernyak et al., 2013; Gergely et al., 2002). There is a continuation of learning about constraints placed on one's behavior, such as those imposed by social institutions (i.e., workplace rules and norms), and that these, in turn, drive individual and social development (Roberts, Wood, et al., 2005). However, it was an open question about whether perceptions of control and free will also mapped on to age differences in these perceived constraints across life.

We also investigated how demographic characteristics were associated with variation in free will and control perceptions. Many findings from our studies support the connection between control perceptions and the tendency to moralize and hold people responsible for their actions. For example, we find that conservatives around the world perceived they had more control over their lives relative to liberals. This is likely connected to conservatives’ emphasizing the role of personal responsibility and downplaying the role of external constraints on one's behavior (Eidelman et al., 2012; Skitka & Tetlock, 1992, 1993). Indeed, Everett et al. (2020)’s recent research demonstrates how conservatives’ moralizing attitudes heightens their free will perceptions relative to liberals. In line with this idea, we also found that, compared to women, men also reported higher control perceptions on average. This may likewise be attributed to the tendency of men to moralize and support retribution (Helgeson & Fritz, 1999), which is associated with a higher perception of moral responsibility (Caspar et al., 2017; Krueger et al., 2014; Shariff et al., 2014). Lastly, higher religious attendance was associated with greater perceptions of control/free will in Study 1 and Sample 2 of Study 2. This supports the idea that many religions promote free will and control through the encouragement of choice and taking responsibility for decisions (and their outcomes) (see Baumeister et al., 2010; McCullough & Willoughby, 2009).

As for education, our results were not as clear. Study 1 suggested that educated adults believe less in free will, while Study 2’s samples showed highly educated people perceive more control over their lives. In this specific case, it is worth noting that there are likely plenty of people who can hold disbelief in free will but nevertheless think they have control over the outcomes of their lives (the two are only modestly positively correlated). In Study 1, although older adults believed less in free will, they were also less likely to endorse beliefs about fatalistic determinism and that the universe is a random place. Although highly educated adults may be more likely to endorse a deterministic-scientific worldview (and therefore believe less in free will), they may also be in a position to overcome various constraints (i.e., financial) and therefore have a higher perception of how much freedom and control they have over their lives. However, the current studies cannot answer these questions definitively.

Cultural Differences in Control Perceptions

The analysis of cross-cultural variation in control perceptions in Study 2 provided several surprising findings. The most counter-intuitive was that individualism/collectivism was not a consistent predictor of control perceptions (although it showed the opposite pattern than would be expected—people living in individualistic societies reported lower perceptions of control over their lives). This was a puzzling result as we expected individualistic cultures, who are thought to emphasize individual autonomy, would stress individual autonomy and engender people with perceptions that they have control over their lives. Indeed, these results seem to contrast with Chernyak et al. (2013)’s finding that older children from the United States (an ‘individualistic’ culture) believe more in their freedom of choice relative to their counterparts in Nepal (a ‘collectivist’ culture). It appears it is still up for debate whether focusing more on one's individual autonomy and choices translates into feeling a freer will. Over time, individuals may realize that they possess (or lack) conscious influence on their decisions due to overcoming (or submitting to) constraints. This violation of expectations may lead to particularly low levels of free will among people from individualistic cultures. Nonetheless, because this is only speculation (and the effect was not present in Sample 1), and we encourage future researchers to further examine this finding.

Importantly, it should also be noted that the individualistic/collectivist dichotomy is often an oversimplification (or, according to some, a misrepresentation) of cultures (Talhelm, 2019). For example, regions around the world generally contain areas that have both individualistic/collectivistic influences (Vignoles et al., 2016). Some values serve both the individual and the collective (Schwartz, 1990), and collectivist cultures display a vigilance of in-group members that is indicative of attributing individual responsibility to people's actions (Liu et al., 2019). Given these critiques of individualism/collectivism, we also encourage researchers to examine different conceptualizations and taxonomies of cultural differences (e.g., Schwartz taxonomies, the other GLOBE characteristics) and how they might be associated with free will and control perceptions.

In Study 2, people from more restrained countries possessed lower perceptions of control. This finding seems to contrast with Baumeister et al. (2010)’s suggestion that belief in free will and control represent the ability to avoid behaviors and temptations that are deemed unacceptable by society. One might expect that more restrained countries think they have more control over their actions and the consequences of those actions. However, our results suggest the opposite. This may be interpreted through how citizens perceive these societies to be constraining their individual freedom. More restrained countries likely have many customs, norms, and laws that condemn or commend various desires, impulses, and behaviors. Perhaps individuals from restrained countries feel the weight of these regulations or are more often reminded of the impulses pulling on their decisions. An individual living in a more indulgent society may not feel compelled by any of these forces, and hence, may feel freer to do as they please.

In addition to overall cultural differences in free will belief, several cultural variables moderated associations between age and control perceptions in Study 2. The sharper age differences could result from people in these cultures reflecting more on constraints (e.g., countries high in restraint and femininity) or being confronted with constraints that violate their perceptions of agency over time (e.g., countries high in individualism). It should be noted that these variables rarely and inconsistently moderated age differences. Because the link between age and control perceptions was largely consistent across cultural contexts in adulthood, this may suggest that cultures engender specific beliefs in control earlier in life. Moreover, as people from different cultures age, they may universally experience the force of constraints placed upon their lives and their control perceptions could decline in a largely homogenous way. Although people around the world may face different constraints, the result appears to be that any constraints—irrespective of their cultural specificity—likely lowers perceptions of control as people grow into adulthood and old age.

Limitations and Future Directions

The studies had many strengths. For example, we analyzed three large data sets from relatively diverse participants and countries using multiple measures of our umbrella construct—free will and control perceptions. We also integrated data from multiple sources, including information on how countries varied economically and socially.

Nevertheless, there are some limitations worth acknowledging. First, the data from the three samples were cross-sectional. This limitation leaves open the possibility that we captured differences in free will and control perceptions between members of different birth cohorts rather than lifespan differences. In other words, it would be the difference between (a) concluding that people born more recently in history have higher free will or control perceptions and (b) concluding that these beliefs and perceptions decline across the lifespan. The danger in this ambiguity is that different birth cohorts are exposed to different sociocultural norms that might influence their perceptions or any other psychological characteristic (Roberts et al., 2010; Stewart & Healy, 1989). Worth noting, because the age differences were so consistent across data sets, cultures, and cultural variables, our data are more likely to lend itself to the developmental interpretation—that free will and control perceptions decline across age and that they are relatively resilient to the modeling of cultural factors (Bleidorn et al., 2013; McCrae et al., 2000; McCrae et al., 1999). Formally modeling year of data collection in the analyses yielded conflicting results—free will beliefs tended to be lower in more recent years in Study 1 but control perceptions tended to higher in more recent years in Study 2 (a replication and extension of Inglehart et al., 2008).

The pattern seen in Study 2 more squarely aligns with studies examining birth cohort differences in perceived control. For example, in cohort sequential studies, more recent cohorts in the U.S. and Germany tend to perceive fewer constraints, and this is particularly true for people's perceptions of the degree to which luck and fate govern their lives (Drewelies et al., 2018; Gerstorf et al., 2019). However, there is some evidence to suggest that more recent cohorts of U.S. college students might be shifting toward more external evaluations of control (i.e., less internal perceptions of control; Twenge et al., 2004). The exact pattern of cohort differences might also be moderated by age, such that more recent cohorts of older adults perceive fewer constraints but more recent cohorts of younger adults perceive less control over their lives, some of which might be attributable to economic differences between the cohorts and across the lifespan (Drewelies et al., 2018).5 Ultimately, formal tests of these questions in the current data was not possible and beyond the current scope of the paper. Future research should follow different cohorts of individuals over time to appropriately separate cohort and developmental effects on free will and control perceptions. A more direct test might be to experimentally manipulate many of the processes we have proposed to see if they affect free will and control perceptions, whether these processes involve culture (Oyserman & Lee, 2008) or lifespan development (Fung et al., 1999).

Second, cultures are not static in their characteristics and can change considerably over time. Relevant to the current studies, this means that cultures are influencing people differently as they—both cultures and individuals—age. This makes it unclear if differences in free will and control perceptions can be just attributed to individuals aging across time, or our cultures becoming more similar/different across time (Chopik, 2020; Grossmann & Varnum, 2015; Santos et al., 2017; Varnum & Grossmann, 2017). Again, we were only able to superficially model this possibility by including the year of data collection in our analyses, but cultures likely change at much slower rates that what was captured in these samples. Future research can model broader, macro-level changes in cultures to see how temporal variation at a region level might affect individual decision making and perceptions.

Finally, the age-related differences in free will and control perceptions varied dramatically across samples, ranging from moderate in size to relatively small. This can also be said of the moderating role of individual and cultural characteristics. It was generally the case that, despite significant moderation being present, decomposing the interactions revealed relatively similar effects across different levels of the moderating variables. Although the effects were relatively small, we thought it was important to provide some reasonable expectations about effect sizes for future research. It is also worth noting that we chose age as a relatively imperfect measure of a process we thought unfolds across life (e.g., that people witness events or have experiences that challenge their existing thoughts about control and free will). Examining these questions using more proximal measures or even under experimental conditions might more carefully test the processes we outlined or establish a clearer causal chain. Whether or not manipulating free will and control perceptions on their own translates into judgments of others and pro(anti)-social behavior is another question entirely—a possibility that researchers have been critical of recently (Crone & Levy, 2019; Monroe et al., 2017; Nadelhoffer et al., 2020). In the meantime, evaluating whether personally witnessing or experiencing exogenous forces put upon one's will and control affects perceptions is an important issue to examine. For example, does experiencing a life event not directly under one's control affect how they think about control and free will? Some evidence suggests that it may be possible (Luhmann et al., 2020), but it has largely been untested so far.