Collins, Ruth Elizabeth; (1997) Metamemory, aging, and the calibration of subjective probabilities. Doctoral thesis (Ph.D), UCL (University College London).

Text Metamemory,_aging,_and_the_cal.pdf Download (8MB)

Abstract

Traditionally, studies of changes in metacognition with age have employed self-assessment questionnaires, but this approach has been justly criticized on methodological grounds. In this thesis, the calibration of subjective probabilities is used to examine the effects of aging on metamemory performance. This radically different technique allows the 'realism' of confidence judgments to be determined, and the degree of over- or underconfidence to be quantified. The calibration performance of younger and older adults is investigated in both episodic and semantic memory tests, and with two types of stimulus sampling (random and selected). The memory tests are always two-alternative forced choice, and a confidence rating is elicited for each memory judgment made. The aim of the study is to determine whether or not realism of confidence judgments changes with age, but in addition, the relationships between crystallized and fluid intelligence, self-esteem, memory performance and calibration are examined. The calibration performance of both the younger and older subjects was found to vary systematically with task-difficulty and appeared to be independent of the stimulus sampling procedure employed. This result supports a 'pessimistic' view of calibration (e.g. Ferrell & McGoey) and is inconsistent with ecological models from the 'Optimistic' School (e.g., Gigerenzer, Hoffrage & Kleinbolting, 1991; Juslin, 1993, 1994). However, a meta-analysis revealed that the old were consistently more overconfident than the young—a result which cannot easily be explained by calibration models for either school—or indeed by an appeal to changes in intelligence or self-esteem with age. An alternative explanation is proposed, based on recent research into the effects of random error in judgmental processes (Erev, Wallsten & Budescu, 1994). It is concluded that older adults may in fact be no more biased than younger adults when making judgments concerning their memorial capabilities, but that increasing age leads to greater variance in the error component associated with their judgments, resulting in the apparently greater overconfidence they exhibit in comparison to the young.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as JSONCSVXML

Downloads by country - last 12 months

Export as CSVXMLJSON

Archive Staff Only

View Item