Human ageing is associated with more rigid concept spaces

Abstract

Prevalence-induced concept change describes a cognitive mechanism by which someone's definition of a concept shifts as the prevalence of instances of that concept changes. The phenomenon has real-world implications because this sensitivity to environmental characteristics may lead to substantial biases in judgements. While prevalence-induced concept change has been established in young adults, it is unclear how it changes as a function of human ageing. In this cross-sectional study, we explore how prevalence-induced concept change affects older adults' lower-level, perceptual, and higher-order, ethical, judgements. We find that older adults are less sensitive to prevalence-induced concept change than younger adults across domains. Using a combination of computational and experimental approaches, we demonstrate that these changes in judgements are sensitive to the pace with which the stimuli occur in the environment and are affected by the effort that subjects invest in order to make accurate decisions. Based on findings from three experiments we argue that older adults' concept spaces are more rigid than those of younger adults. However, what appear as an age-related cognitive "deficit" may turn out to be beneficial because it makes older adults less susceptible to biases in judgments.

Changes in the prevalence of muscular, but not thin, bodies bias young men’s judgments about body size.

“…3 Here, we find that participants had narrower decision thresholds when judging muscular stimuli than when judging overweight stimuli (MAP difference = 0.16, CI [0.06, 0.25], p = .001), suggesting that participants required less evidence to make decisions about muscular stimuli than overweight stimuli (Figure 6C). We also observed statistically robust posterior differences in nondecision times between overweight and muscular judgments (MAP difference = −0.007, CI [−0.001, 0.00], p = .03), such that men were slightly slower at encoding stimuli and executing responses when judging muscular stimuli, but the size of this difference was small (on the order of ∼1 ms.) and thus negligible (see also Devine et al, 2023 for a similar pattern). Critically, response times in the task were, on average, between 200 and 300 ms (Figure 6A), and thus encoding time differences between stimulus types on this scale can be deemed unimportant.…”

“…This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. To help explain the descriptive differences we observed in responses and response times between stimulus types, and to probe the computational mechanisms underpinning sensitivity to concept change (Devine et al, 2023), we turned to DDM (Ratcliff & McKoon, 2008;Wiecki et al, 2013; see the Method and Analysis sections). All estimated fixed effects are summarized in Table 1, and intercepts per condition are shown in Figure 6B.…”

“…research proposals (see also Devine et al, 2023). In a recent study, we showed the PICC effect in young women's judgments about body image: When the prevalence of images of thin women in the environment (the experimental task) increased, women were more likely to judge objectively average bodies as overweight compared with when there was an equal proportion of thin and overweight women's images in the environment (Devine et al, 2022).…”

“…PICC is a phenomenon where, as the numbers of exemplars of a given concept increase in the environment, the boundaries of that concept shrink to exclude exemplars that they would otherwise include (Levari et al, 2018). Levari et al (2018) showed these prevalence-induced changes across concepts of differing complexity, ranging from color judgments to decisions about the ethicality of research proposals (see also Devine et al, 2023). In a recent study, we showed the PICC effect in young women’s judgments about body image: When the prevalence of images of thin women in the environment (the experimental task) increased, women were more likely to judge objectively average bodies as overweight compared with when there was an equal proportion of thin and overweight women’s images in the environment (Devine et al, 2022).…”

Diminished state space theory of human aging

“…Another reason for a reduced updating of transition probabilities in older adults could be that suboptimal transition learning rates in older adults reflect a greater tendency to perseverate on previous predictions in the elderly (Nassar et al, 2016). A greater rigidity of internal state spaces might be beneficial in some contexts (Devine et al, 2022) but is mal-adaptive in environments in which performance depends on a flexible adjustment of state space representations. To summarize, each of these computational explanations seems consistent with what is seen behaviorally (Bolenz et al, 2019;Hämmerer et al, 2018;Ruel, Bolenz, et al, 2021).…”