A methodological study of the effect of experimentally induced demand characteristics in research on nocturnal dreams.

Abstract

This experiment tested the hypothesis that demand characteristics, as outlined by Orne (1962), may affect dream content. Twelve subjects, allocated to two matched groups balanced for age and sex, were selected on the basis of good dream recall from 94 volunteer psychology I students. Subjects were involved in the study for 15 consecutive nights, three of which were laboratory recording nights, when subjects were woken to report their laboratory dreams during rapid eye movement (stage REM) sleep. Night 1 served as an habituation night (Hall, 1967). On nights 2-7 subjects recorded their home dreams on blank diary forms, from which baseline measures were derived. Night 8 served as the laboratory baseline night prior to the introduction of the independent variable on the following morning. The independent variable was a simple form administered to each group requesting that the subjects in the respective groups pay special attention to either the outdoor/nature or urban settings of their nocturnal dreams. The forms were administered in sealed envelopes and the experimenter was blind as to which form and thus to which group subjects were assigned. The remaining nights served as the treatment condition. Home dreams were again recorded on nights 9-14 whilst night 15 served as the third laboratory night. A post experimental enquiry was conducted in the week following night 15. Subjects did not consciously construe the forms as representing an attempt to influence their dreams. Nevertheless in the treatment condition, the dream settings of both groups changed significantly in the predicted direction as assessed by two quantitative indices; a scale giving a global impression of the dream and a content analysis of words in the dream text. The implications of these findings are discussed with reference to both laboratory dream research and psychotherapy.