Is lorazepam-induced amnesia specific to the type of memory or to the task used to assess it?

Retrieval tasks can be classified along a continuum from conceptually driven (relying on the encoded meaning of the material) to data driven (relying on the perceptual record and surface features of the material). Since most explicit memory tests are conceptually driven and most implicit memory tests are data driven there has been considerable confounding of the memory system being assessed and the processing required by the retrieval task. The purpose of the present experiment was to investigate the effects of lorazepam on explicit memory, using both types of retrieval task. Lorazepam (2.5 mg) or matched placebo was administered to healthy volunteers and changes in subjective mood ratings and in performance in tests of memory were measured. Lorazepam made subjects significantly more drowsy, feeble, clumsy, muzzy, lethargic and mentally slow. Lorazepam significantly impaired recognition memory for slides, impaired the number of words remembered when the retrieval was cued by the first two letters and reduced the number of pictures remembered when retention was cued with picture fragments. Thus episodic memory was impaired whether the task used was conceptually driven (as in slide recognition) or data driven, as in the other two tasks. Analyses of covariance indicated that the memory impairments were independent of increased sedation, as assessed by self-ratings. In contrast to the deficits in episodic memory, there were no lorazepam-induced impairments in tests of semantic memory, whether this was measured in the conceptually driven task of category generation or in the data-driven task of wordstem completion.