HE FACT that most schizophrenic patients manifest marked difficulty in establishing and maintaining satisfactory relationships is well known. More recently, attempts have been made to quantify and elucidate these dif- ficulties so as to establish their significance for clinical practice. One line of inquiry has concentrated particularly on family relationships and their influence on the patient’s clinical course. Thus, a series of studies1-3 has shown that the likelihood of relapse in schizophrenia can be predicted by the amount of emotion expressed by patients’ relatives and, in particular, the number of critical comments made by a key relative about the patient and the illness, in the course of a standardized interview. In addition, it has been found that the probability of relapse increases if the amount of face-to-face contact with the key relative is greater. While these findings have obvious implications for the management of schizophrenia, they also raise interesting questions concerning the processes whereby the relative’s emotional state and attitudes come to influence the behavior of the schizophrenic patient. Clearly these attitudes and emotions may be conveyed through a number of channels-some more and some less readily studied. Since, however, the authors referred to above have particularly emphasized the importance of face-to-face contact, it is of interest to examine the ways in which schizophrenic patients respond to facial emo- tions, especially as such stimuli can be presented in a reasonable standardized and controlled fashion. In recent years increasing attention has been focused on the facial expression of emotion, a topic first discussed systematically (with the use of photography) by Charles Darwin4 Research in this general area, which has been comprehen- sively reviewed by Bruner and Tagi& and Izard,6 indicates that certain facial displays are universally associated with particular emotional states7 and are recognized and labeled with a high degree of interobserver agreement. Thus, “primary” emotions such as anger, fear, joy and sadness, if presented in a well characterized manner, are similarly described by independent observers. Of these, fear has been found” to be the “most dreaded” and least experienced of the emotions, across a number of cultures. The purpose of this article is to report a study in which the responses of schizophrenic patients to photographs depicting primary emotions have been compared to those of controls and nonpsychotic psychiatric patients. Based on the work referred to earlier, the hypotheses being tested are (1) that schizo-