Learning to have psychosomatic complaints: conditioning of respiratory behavior and somatic complaints in psychosomatic patients

Psychosom Med. Jan-Feb 1997;59(1):13-23. doi: 10.1097/00006842-199701000-00003.


Objective: Assuming a subjective similarity between the experience of a hyperventilation episode and inhaling CO2-enriched air, we tested whether a respiratory challenge in association with a particular stimulus could result in altered respiratory behavior and associated somatic complaints upon presenting the stimulus only.

Method: Psychosomatic patients (N = 28) reporting hyperventilation complaints participated in a differential conditioning paradigm using odors with a positive or negative valence as conditioned stimuli (CS+ or CS-) and 7.4% CO2-enriched air as the unconditioned stimulus (US). Three CS+ and three CS-acquisition trials were run. During the test phase, two CS(+)- and two CS(-)-only trials were run, followed by two new test odors (with a positive or negative valence). Respiratory frequency, tidal volume, end-tidal fractional concentration of CO2, and heart rate were measured throughout the experiment. Somatic complaints were registered after each trial.

Results: We observed a) increased respiratory frequency and an elevated level of somatic complaints upon presenting the CS+ only; b) a selective association effect: conditioning was only apparent with the negatively valenced CS+ odor; (c) no generalization of respiratory responses and complaints to the new odors; (d) no conditioning effect on dummy complaints that are usually not reported when inhaling CO2; (e) in exploratory comparisons with normal subjects, stronger conditioning effects on typical hyperventilation complaints in patients, and, in female subjects, on respiratory frequency.

Conclusion: Respiratory responses and psychosomatic complaints can be elicited by conditioned stimuli in a highly specific way. The findings are relevant for disorders in which respiratory abnormalities and/or psychosomatic complaints may play a role and for multiple chemical sensitivity.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Analysis of Variance
  • Carbon Dioxide / adverse effects
  • Case-Control Studies
  • Conditioning, Classical*
  • Cross-Sectional Studies
  • Female
  • Humans
  • Hyperventilation / physiopathology*
  • Hyperventilation / psychology*
  • Male
  • Middle Aged
  • Odorants
  • Psychophysiologic Disorders / etiology*
  • Regression Analysis
  • Respiratory Function Tests
  • Sex Factors
  • Sick Role*


  • Carbon Dioxide