Abstract Is a person's response to one noxious stimulus similar to his/her responses to other noxious stimuli? This long-investigated topic in pain research has provided inconclusive results. In the present study, 2 samples were studied: one using 60 healthy volunteers and the other using 29 patients with coronary artery disease. Results showed near-zero correlations between measures of heat, cold, ischemic, and electrical laboratory pains, as well as between these laboratory pains and an idiopathic pain, the latency to exercise-induced angina in the patients. Power analyses showed that the sample sizes were sufficient to detect a correlation of 0.50 or greater at the 0.05 level 99% of the time in the healthy volunteers, and between 80 and 85% of the time in the patients. Reliability analyses indicated retest correlations on the order of 0.60 for these measures, indicating that the lack of correlation between modalities was not due to unreliability within a measure. These studies fail to demonstrate alternate-forms reliability among these tests, and also fail to support the notion that a person can be characterized as generally stoical or generally complaining to any painful stimulus. In practice, this implies that a battery of tests should generally be used to assess pain sensitivity and also that assessments of one pain modality are not generally useful for making inferences about another.