Patterns of memory failure after scopolamine treatment: implications for cholinergic hypotheses of dementia.

To test the idea that scopolamine provides a suitable pharmacological model of the memory defects associated with cortical or subcortical dementias, we assessed memory on a battery of tasks in healthy young normal subjects who received 0.5 mg scopolamine, 0.1-0.2 mg glycopyrrolate or physiological saline, once each on three separate occasions, and compared the pattern of memory failure induced by scopolamine to that observed on the same tasks in patients with Alzheimer's disease (AD) or Huntington's disease (HD). In agreement with previous reports, scopolamine impaired acquisition and delayed recall of a 14-word list and disrupted retention on the Brown-Peterson distractor task, whereas the peripherally active anticholinergic glycopyrrolate was without effect. However, under scopolamine the pattern of errors made on these memory tasks was quite different from that seen in patients with AD. Scopolamine did not increase the number of false positive errors on delayed recognition of the word list and also failed to increase the number of prior-item intrusions on the Brown-Peterson task. Also, scopolamine did not impair learning of a symbol-digit paired-associate task, and did not reduce the number of words retrieved or increase the number of words repeated on a standardized verbal fluency test. When the effects of scopolamine on memory were compared to the pattern of impairments observed in demented patients with HD, several differences were found. Although scopolamine clearly produces deficits on some measures of anterograde memory, the present findings question whether anticholinergic drugs adequately mimic the full range of memory impairments observed in cortical or subcortical dementias.