Several explanations for the weak relations between subjective memory judgments and objective memory performance were investigated in two groups of normal older adults. Group 1 sampled a general population (mean age 61.6 yr., range 46–89), while Group 2 sampled subjects who were on a waiting list for memory training (mean age 63.0 yr., range 45–85 years). In both groups, subjective memory judgments were assessed with global ratings of memory capacity and with ratings of frequency of forgetting in specific memory situations. Memory performance was assessed with several well-known tests and with recently developed tests for domain-specific aspects of memory. Most tests concerned episodic memory. Study 1 also included measures of semantic, incidental, and working memory. Study 2 further examined the influence of the domain-specificity of objective and subjective measures for remembering names, intentions, and texts. Relations between memory self-reports and performance were weak in both groups and for all kinds of tests. Against expectations, the low correlations could not be explained by differences between ecological and laboratory tests or incidentally and intentionally remembered information, or by differences between specific failures compared to global, stereotyped judgments. Surprisingly, correlations did not increase when subjective and objective measures assessed the same ability, like remembering names. Also noncognitive variables (mood and lifestyle) did not influence the relations. The (weak) relations between subjective and objective memory measures were comparable for subjects over and under 65 years of age. Furthermore, relations were comparable for the general population sample and the memory compliant group.
