Thirteen patients with idiopathic Parkinson's disease of recent onset (mean age 63·2 years) and a group of 10 young healthy volunteers (mean age 26·1 years) underwent a series of neuropsychological tests for assessment of memory, learning ability and mental processing speed before and during treatment with trihexyphenidyl. Retesting after anticholinergic exposure (mean of 2 weeks for patients and 1 week for controls) revealed in young healthy controls the same pattern and magnitude of decline in memory function as in Parkinson patients. Non-demented subjects with Parkinson's disease of recent onset thus do riot seem to be selectively vulnerable to cognitive side-effects of anticholinergic treatment.