Two experiments demonstrated that the experience of physical sensations is the result of hypothesis-guided selective search and encoding. In the first experiment, subjects listened to a stimulus that was said to increase, decrease, or have no effect on skin temperature. The results indicated that subjects selectively monitored only those changes in skin temperature congruent with the experimentally induced hypotheses, thus resulting in self-reports of change in skin temperature in the suggested directions. Two replications using heart rate and nasal congestion-which imposed search strategies-produced comparable results. In Experiment 2, subjects viewed a heartbeat chart and assessed the degree to which a hypothetical drug influenced a target person's heart rate. Perceptions of change in heart rate were influenced by both experimentally induced hypotheses as well as subjects' perceptions of their own heart rate. That is, if prior to the experiment subjects believed their own pulse was accelerated, they were more likely to perceive the graphed heartbeat accelerating as well. The findings point to parallels between perception of internal state and the external environment. Finally, the roles of hypothesis-guided selective search are extended to placebo effectiveness.