Whether the effects of exposure to 1 movement generalize to another dissimilar movement was investigated in 37 patients with low back pain (15 men, 22 women). Two movements were executed twice: bending forward while standing and lifting 1 leg while lying down. During each trial, baseline pain, expected pain, and experienced pain were recorded. Similar ratings for perceived harm were obtained. Analyses revealed an initial overprediction of pain, but after exposure the overprediction was readily corrected. This exposure effect did not generalize toward another dissimilar movement. These results were only characteristic for patients with catastrophic thinking about pain. Low pain catastrophizers did not overpredict pain. There were no effects of exposure on perceived harm. Exposure may profitably be conceived of as the learning of exceptions to a general rule.