The effects of scopolamine, lorazepam, and glycopyrrolate on classical conditioning of the human eyeblink response.

Abstract

Human eyeblink conditioning, a relatively simple form of learning and memory, has previously been shown to be impaired by the central and peripheral anticholinergic scopolamine. The present study compared the behavioral effects of scopolamine with the benzodiazepine lorazepam and a peripherally active anticholinergic, glycopyrrolate. Thirty-six healthy normal volunteers (mean age: 23.7 years) were studied with 12 assigned double-blind to each of three drug conditions (0.5 mg scopolamine IV, 2 mg lorazepam PO, or 0.2 mg glycopyrrolate IV). Subjects underwent classical conditioning of the eyeblink response in which the conditioned stimulus was an 80 dB binaural tone, and the unconditioned stimulus was a 2 psi airpuff to the right eye. Ten trials of unpaired stimulus presentations were followed by 60 paired trials and finally by an extinction period of five tone-alone presentations. An eyeblink response that occurred during the tone but before the airpuff was scored as a conditioned response (CR). Subjects treated with lorazepam (43% mean CRs) and scopolamine (51% mean CRs) exhibited a significantly lower asymptotic level of conditioning than those treated with glycopyrrolate (85% mean CRs; P < 0.01). However, during extinction, lorazepam-treated subjects (35% CRs) showed a lower overall level of responding to the tone than either scopolamine (60% CRs) or glycopyrrolate (62% CRs) treated subjects (P < 0.05). It seems unlikely that these differences could be accounted for by drug-induced alterations in motor responses because there were no significant differences between the three drug conditions in the frequency, latency, or amplitude of unconditioned responses to the airpuff. Overall, our data indicate that scopolamine and lorazepam impair eyeblink conditioning and suggest that some of the effects of benzodiazepines and anticholinergics on learning and memory can be differentiated using this paradigm.