Previous studies of the neuropsychiatric aspects of Parkinson’s disease were frequently methodologically inadequate. Small sample sizes, selection bias, lack of diagnostic criteria of Parkinson’s disease, different definitions and assessment of neuropsychiatric symptoms, and lack of control groups seriously questioned the validity of and ability to generalize the results from many studies. During the past decade, however, several of these methodological issues have been addressed. Recent studies have found that mild cognitive impairment is very common, and dementia, depression, and psychotic symptoms develop in a large proportion of patients. Neuropsychiatric symptoms are important determinants of mortality and disease progression, as well as of the patients quality of life and course of disease, caregiver distress, and nursing home admission. Few adequately designed treatment trials have been published, but available evidence suggests that depression and hallucinations may be effectively treated using new antidepressants and atypical antipsychotic agents without worsening of parkinsonism.