The purpose of this study was to determine if physical restraint is a major factor in the elicitation of the hallucination phenomena associated with perceptual (or sensory) deprivation studies. Experimental subjects were exposed to three one-hour sessions of perceptual deprivation one week apart, with physical restraint being used during the second session. A group of control subjects was used to determine t he effects of three unrestrained sessions of perceptual deprivation. No significant differences were found between sessions for the experimental group in terms of number of reports or the cumulative duration of the reports. There was also no difference found between the two groups for any session. The data and experiences of the individual subjects are discussed at length with particular attention to the effects of the restraint procedure on indications of stress or anxiety levels (expressed in terms of time estimation and subjective reports). Recommendations for the use of more objective measures of anxiety such as biofeedback and electroencephalographic equipment are made as well as better defining procedures for the measurement of the hallucination phenomena. It is further suggested that the group design is not suited to the study of hallucinations due to the great degree of variability. A new procedure for the unobtrusive measure of the duration of the hallucinations is used successfully in this study and may prove to be a useful tool for future studies in this area.