A Study on Verbal Memory Functions in Temporal Lobe Epilepsy with the Rey Auditory‐Verbal Learning Test

Abstract

Purpose: Verbal memory functions have often been reported to be impaired, at least in some patients with temporal lobe epilepsy (TLE). One of the factors responsible for the verbal memory impairments has now been clarified. The presence of co‐morbid psychiatric symptoms in patients with TLE is one factor that may be responsible because patients with schizophrenia have consistent verbal memory impairments, but the relation between the memory impairments and psychiatric symptoms in TLE has not been examined. We studied memory functions in patients with TLE without and with psychotic symptoms and compared them with those of patients with schizophrenia by using the Rey Auditory–Verbal Learning Test (RAVLT). Methods: The subjects included nine patients with TLE without a history of hallucinatory‐paranoid episodes (E group), five patients with TLE with a history of interictal hallucinatory‐paranoid episodes (P group), 26 patients with schizophrenia (S group), and 22 normal control subjects (N group). All the patients received the same medication, anticonvulsive or antipsychotic drugs or both, during at least the 3 months before the examination. The N group was significantly superior to the other three groups with respect to years of education and their scores on the Mini‐Mental State Examination (MMSE), and the vocabulary and block‐design subtests in the revised Wechsler Adult Intelligence Scale (WAIS‐R). Written consent was obtained from all the subjects. Verbal memory functions were assessed by using the RAVLT. In the RAVLT, the subjects were presented 6 times with an auditory list of 15 words and were required to recall the list for each trial. Indices used for the verbal memory functions in the RAVLT were (a) the number of correctly recalled words during each trial, (b) the decline in the number of correctly recalled words after a trial of recalling a different list of words (interference score), (c) the recognition score (d’) calculated by using the number of hits (correct “yes” responses) and false alarms (incorrect “yes” responses) in a recognition test conducted 5 min after the sixth trial, (d) two types of error responses [i.e., the words that were absent in the original word list (falsely recalled words) and the words recalled erroneously more than once (repeatedly recalled words]. Results: In each trial, the E and N groups recalled nearly an equal number of words and a significantly greater number of words than either the P or the S groups (p < 0.0001). This significant difference was maintained even when the scores on the vocabulary and block‐design subtests in the WAIS‐R were incorporated into the analysis as covariance (p < 0.0007). The interference scores were in the order of P > S > N > E groups. The d'scores were larger in the E and N groups than in the P and S groups (p = 0.0037). The number of falsely recalled words was significantly greater in the P and S groups than in the N group (p = 0.004), although there was no significant difference between the groups in the number of repeatedly recalled words. Conclusion: This study demonstrated that verbal memory functions in patients with TLE were impaired in those with a history of psychotic episodes but not in those without such a history. This suggests that verbal memory impairments in patients with TLE are related to the presence of co‐morbid psychotic symptoms.