Relationships between neuropsychological test performance and event-related potentials in alcoholic and nonalcoholic samples.

Are event-related potentials and nonconcurrently measured neuropsychological test performance correlated? Sober male and female middle-aged alcoholics and peer controls were administered an "oddball" event-related potential (ERP) task and several hours later, a battery of neuropsychological (NP) tests. Alcoholics performed significantly poorer than controls on NP tests. Male alcoholics had significantly altered ERP responses (N1, NdA, and P3 amplitudes) but female alcoholics did not differ on any ERP variables from controls. A number of significant correlations between ERP and NP measures were present. The most consistent findings were positive correlations between perceptual-motor (PM) tests and the P3 amplitude at Pz in both male and female alcoholics and in male alcoholics, a negative correlation between PM tests and P3 latency at Pz, findings similar to those seen in Parkinson patients. Significant correlations were most numerous in family history positive alcoholics. The results lead to two conclusions: first, Grant's postulation that sober alcoholics may manifest an intermediate duration organic mental disorder is supported; second, given the lack of ERP differences in the females, ERP measures should not be considered as being causally related to NP performance.