Controversy surrounds the existence and nature of reaction time deficits in Parkinson's disease. Three areas of research are reviewed: the use of precues to speed movement (motor preprogramming), the effects of medication on reaction time, and simple reaction times. No evidence is found for a motor preprogramming deficit, and the presence of a parkinsonian reaction time deficit after medication withdrawal is found to be dependent upon experimental design and the withdrawal method used. Parkinson's disease is found to cause a consistent deficit in simple reaction time. A quantitative analysis of past studies reveals that a parkinsonian reaction time deficit is more likely to be present in tasks that controls can perform with a fast reaction time. This relationship between deficit and control group reaction time applies to choice, but not simple, reaction time tasks. Many studies compare patient and control choice reaction times across experimental conditions that cause control reaction time to vary. The authors of these studies should consider whether their results can be explained in terms of the simple relationship between patient reaction time deficit and control reaction time before drawing more complex conclusions from their data.