Annals of Behavioral Medicine

, Volume 33, Issue 1, pp 112–116

Illness by suggestion: Expectancy, modeling, and gender in the production of psychosomatic symptoms

Rapid Communication

DOI: 10.1207/s15324796abm3301_13

Cite this article as:
Lorber, W., Mazzoni, G. & Kirsch, I. ann. behav. med. (2007) 33: 112. doi:10.1207/s15324796abm3301_13


Background: Expectancy and modeling have been cited as factors in mass psychogenic illness (MPI), which reportedly affects more women than men.Purpose: The purpose of the study is to assess the effects of expectancy and modeling in a controlled laboratory analogue of MPI.Methods: Students were randomly assigned to inhale or not inhale an inert placebo described as a suspected environmental toxin that had been linked to four symptoms typical of reported instances of MPI. Half of the students observed a female confederate inhale the substance and subsequently display the specified symptoms.Results: Students who inhaled the placebo reported greater increases in symptoms, and the increase was significantly greater for the specified symptoms than for other symptoms. Observation of the confederate displaying symptoms increased specified symptoms significantly among women but not among men. Changes in reported symptoms were significantly associated with changes in unobtrusively observed behavior.Conclusions: Symptoms typical of clinical reports of MPI can be induced by manipulating response expectancies, and the effects are specific rather than generalized. Among women, this effect is enhanced by observing another participant (who in this study is also female) display symptoms. This suggests that the preponderance of women showing symptoms in outbreaks of MPI may be due to gender-linked differences in the effects of modeling on psychogenic symptoms.

Copyright information

© Society of Behavioral Medicine 2003

Authors and Affiliations

  • William Lorber
    • 1
  • Giuliana Mazzoni
    • 2
  • Irving Kirsch
    • 2
  1. 1.University of ConnecticutUSA
  2. 2.Department of PsychologyUniversity of HullHullUK

Personalised recommendations