CONFABULATION EXTENDING ACROSS EPISODIC, PERSONAL, AND GENERAL SEMANTIC MEMORY

This paper compares a patient, AB, who showed florid confabulation wit h two other patients, one of whom had bilateral frontal lesions, and t he other bilateral temporal lobe pathology. AB's confabulation followe d a Wernicke episode, and she had a clinical diagnosis of an alcoholic Korsakoff syndrome, implying diencephalic pathology. However, cervica l carcinoma was also found, and the persistence of her confabulation w as attributed to metabolic or other non-metastatic complications of ca rcinoma, affecting cortical function. In the late stages of her illnes s, she also developed bilateral posterior (parietal-occipital) metasta ses. AB was assessed on three measures of confabulation-Dalla Barba's (1993a) Confabulation Battery, the Autobiographical Memory Interview, and an Informal Interview. She showed ''spontaneous'' confabulation ex tending across episodic, personal semantic, and general semantic memor y. In this, she contrasted with the frontal lobe lesion patient, who s howed confabulation only in personal semantic memory, and the temporal lobe lesion patient, who did not confabulate. From an account of her confabulations given by her brother, as well as our own observations o f her pattern of responses, three factors were identified as contribut ing to the confabulation-confusions in the context of memories, involv ing both time and place; a high rate of perseverations, particularly i n semantic memory; and a tendency to respond indiscriminately to the i mmediate social and environmental context. It is concluded that fronta l pathology and a disorder in context memory may be necessary, but not sufficient, conditions for spontaneous confabulations to occur, and a ny theoretical description of confabulation may need to take account o f a number of contributing factors.