Meta-analysis on age-related differences in mentalizing: role of task characteristics

As people age, declines in social cognition and more specifically mentalizing, the ability to attribute mental states to others, have been frequently reported. More than a decade ago, Henry et al. (2013) conducted a foundational meta-analysis on age-related mentalizing differences, reporting consistent declines in older adults regardless of task features. Since then, the literature has expanded substantially, extending their conclusions while also revealing inconsistencies that highlight the complexity of age-related mentalizing differences. Considering this, we conducted an updated meta-analysis to reassess these differences and examine whether task-related characteristics moderate the effects, directly building on the work of Henry et al. The analysis included 72 studies and tested various (and novel) moderators such as task domain (cognitive, affective, mixed), modality (verbal, visual-static, visual-dynamic), task type (e.g., interpretation of actions, shared world knowledge), ecological validity, and respondent perspective (first-, second-, third-person). Results confirmed and extended previous findings, showing that older adults exhibit significant declines in mentalizing compared to younger adults (g=0.75), and that these deficits are specific to mentalizing rather than general cognitive performance. However, none of the examined moderators significantly influenced this effect, suggesting that the decline reflects a general difficulty with mentalizing in older adults rather than one driven by specific task characteristics. Notably, most tasks relied on third-person perspectives and lacked ecological validity, highlighting the need for more ecologically valid, interactive, and second-person paradigms. Such approaches may help clarify whether tasks that are more engaging, socially meaningful, and reflective of real-world interactions could positively affect older adults’ mentalizing performance.