Understanding the complex nature of pain perception requires the ability to separately analyze its psychological dimensions and their interaction, and relate them to specific variables and responses. The present study, therefore, attempted to selectively modulate the sensory and affective dimensions of pain, using a cognitive intervention, and to assess the possible relationship between these psychological dimensions of pain and changes in physiological responses to the noxious stimuli. In three experiments, normal subjects trained in hypnosis rated pain intensity and pain unpleasantness produced by a tonic heat-pain stimulus (1-min immersion of the hand in 45.0-47.5 degrees C water). Two experiments were designed to test hypnotic suggestions to decrease (Experiment one (Section 2.5.1)), or increase and decrease (Experiment two (Section 2.5.2)) pain affect. Suggestions in Experiment three (Section 2.5.3) were directed towards an increase or decrease in pain sensation. In Experiments one and two (Sections 2.5.1 and 2.5.2), the significant modulation in pain unpleasantness ratings was largely independent of variations in perceived pain intensity. Moreover, in Experiment two (Section 2.5.2), there was a significant correlation between the stimulus-evoked heart-rate increase and ratings of pain unpleasantness, but not of pain intensity, suggesting a direct functional interaction between pain affect and autonomic activation. In Experiment three (Section 2.5.3), suggestions to modulate the sensory aspect of pain produced significant modulation of pain intensity ratings, with secondary changes in pain unpleasantness ratings. Hypnotic susceptibility (Stanford Hypnotic Susceptibility Scale form A) was specifically correlated to pain unpleasantness modulation in Experiment two (Section 2.5.2) and to pain intensity modulation in Experiment three (Section 2.5.3), suggesting that this factor relates to the primary process toward which hypnotic suggestions are directed. The specific pain dimension on which hypnotic suggestions act depends on the content of the instructions and is not a characteristic of hypnosis itself. Results are consistent with a successive-stage model of pain perception (e.g. Wade JB, Dougherty LM, Archer CR, Price DD. Assessing the stages of pain processing: a multivariate analytical approach. Pain 1996;68:157-167) which provides a conceptual framework necessary to study the cerebral representation of pain perception.