Age Differences in Hindsight Bias: The Role of Episodic Memory and Inhibition

Abstract

Background/Study Context: After learning an event’s outcome, people’s recollection of their former prediction of that event shifts towards the actual outcome. This hindsight bias (HB) phenomenon tends to be stronger in older compared with younger adults; however, it is unclear whether age-related changes in other cognitive abilities mediate this relationship.

Methods: Sixty-four younger adults (M_age = 20.1; range = 18–25) and 60 community-dwelling older adults (M_age = 72.5; range = 65–87) completed a memory design HB task. Two aspects of HB, its occurrence and magnitude, were examined. Multiple regression and mediation analyses were conducted to determine whether episodic memory and inhibition mediate age differences in the occurrence and magnitude of HB.

Results: Older adults exhibited a greater occurrence and magnitude of HB as compared with younger adults. The present findings revealed that episodic memory and inhibition mediated age-related increases in HB occurrence. Conversely, neither cognitive ability mediated age-related increases in HB magnitude.

Conclusion: Older adults’ susceptibility to the occurrence of HB is partly due to age-related declines in episodic memory and inhibition. Conversely, age differences in the magnitude of HB appear to be independent of episodic memory and inhibition. These findings have important implications for understanding the mechanisms by which susceptibility to HB changes across the adult life span.

Notes

¹We conducted two additional magnitude analyses, one using the other conventional magnitude index, Pohl’s Δz index (Pohl, 2007) after omitting perfect recollections, and another using Hell’s index including perfect recollections. With regard to the former analysis, we found that our magnitude findings held when using Pohl’s Δz index (i.e., significant age effect but neither cognitive variable predicted HB magnitude). Conversely, when using Hell’s index including perfect recollections, we found that older age predicted a greater magnitude of HB and that inhibition mediated this relationship, such that older age was associated with poorer inhibition, which subsequently predicted greater HB magnitude. Episodic memory was a marginal predictor of HB magnitude with perfect recollections included, and marginally mediated the relationship between age and HB magnitude.