A signal detectability analysis of the effect of physostigmine on memory in patients with Alzheimer's disease.

Ten patients with clinically diagnosed Alzheimer's disease were given tests of recognition memory while receiving placebo or physostigmine, a cholinesterase inhibitor, given intravenously over 30 min. Doses of 0.0, 0.125, 0.25, and 0.5 mg were given in random order on 4 separate days. All patients had their best performance following one of the doses of physostigmine rather than saline although dose-response curves varied from one patient to another. The dose associated with the best performance in each patient was compared with saline in a replication study using a double-blind cross-over design. In the replication study the total number of correct responses on the recognition memory task was significantly greater following physostigmine than following saline. To determine whether this improvement in performance was a result of an increase in the amount of information stored in memory or was secondary to a change in the patients' criteria for saying that they recognized an item, the results were subjected to a signal detectability analysis. This analysis indicated that studied items and new items were more discriminable following physostigmine, as evidenced by an increase in d', and that the criteria for saying which items had been studied also changed following physostigmine, as evidenced by an increase in C. The increase in d' indicates that physostigmine enhanced storage of information in memory while the change in C is consistent with the view that patients altered their decision criteria to maximize the number of correct responses in both conditions.

RCT Entities:   Population Interventions Outcomes