Abstracts

Self-reports of memory functioning are widely used in epilepsy populations, despite a lack of established reliability and validity. Past findings suggest that higher intellectual ability, intact objective memory performance, and few psychological symptoms and psychosocial stressors may increase accuracy of self-reported cognitive change. Based on the ubiquitous nature of these measures, it is important to identify patient variables that increase the likelihood of reliable reporting.
Eighty-seven adult patients referred to a regional epilepsy center underwent neuropsychological testing on two occasions, including intellectual and memory tests, a Minnesota Multiphasic Personality Inventory (MMPI), and a Washington Psychosocial Seizure Inventory (WPSI). The majority of these patients (84%) underwent surgery to improve seizure control between the two evaluations. At the second evaluation, patients responded to structured interviews regarding their subjective assessment of memory change. These reports were compared to objective change in memory performance using reliable change indices.
Intellectual ability and objective memory performance were not related to patients[apos] accuracy of self-reported changes in memory functioning. Poor overall psychosocial functioning (WPSI) was related to underestimation of memory performance (p=.046). Elevations on MMPI scales indicative of preoccupation with somatic complaints (Hs; p=.036) and increased anxiety (Pt; p=.008) were also associated with a pessimistic assessment of change in memory performance. Interestingly, female patients were more accurate in their self-report of memory change than male patients (p=.050). When the sample was divided into men (n=40) and women (n=47), variables associated with an accurate self-assessment differed between the genders. For women, elevations on MMPI scales indicative of preoccupation with health concerns (Hs; p=.037) and a tendency to translate psychological stressors into somatic complaints (Hy; p=.028) were related to a pessimistic report of memory change. For men, increased anxiety (Pt; p=.031) was related to a negative assessment of change in memory performance.
Contrary to previous hypotheses, general intellectual ability and objective performance on formal memory tests were not related to self-report of memory change. However, consistent with other findings, poor psychosocial functioning and increased psychological distress were related to a negative self-assessment of change in memory performance. In the current sample, women were more accurate overall, and differed from men in the psychological variables associated with self-report. Men were more likely to over-report memory problems in the presence of increased anxiety symptoms, while women with symptoms consistent with somatizational tendencies gave a more negative assessment of their memory. These findings may help care providers identify patients who are more accurate at reporting changes in their memory.