Numerous studies have established the prognostic validity of family at
titudes and interactional behaviors in psychiatric disorders. However,
most have relied on observer-rated rather than patient-rated measures
of the family context. Bipolar patients (N = 31) rated their relative
s' verbal behaviors in problem-solving discussions using the Patient's
Experiences of a Relative's Affective Style (PERAS) scale. The PERAS
had acceptable reliability and internal consistency. Patients' PERAS j
udgments of the frequencies of harsh and benign criticisms in relative
s correlated positively with an observer's affective style (AS) rating
s of these same behaviors in relatives. However, patients' ratings of
intrusiveness and support in relatives were unrelated to the observer'
s AS ratings. The PERAS is a promising measure of how patients experie
nce their family interactions.