No abstract

PROBLEM Spielberger(lO) has recently suggested that the term anxiety is used more or less indiscriminately to refer to two very different constructs. It is perhaps most conimonly used in an empirical sense to denote a more or less transitory state or condition of the organism that varies in intensity and fluctuates over time. The term is also used to refer to a relatively stable personality trait denoting individual differences in anxiety-proneness. As an organismic state, anxiety (A-state) is characterized by subjective, consciously perceived feelings of apprehension and tension, together with activation of the autonomic nervous system. As a personality trait, anxiety (A-trait) refers to the degree to which individuals are disposed to manifest A-state in response to various forms of stress.In this study, the effects of a muscle relaxation training procedure and the passage of time on empirical measures of A-state and A-trait were investigated. On the bask of Spielberger's (lo) state-trait conception of anxiety, nieasures of A-state were expected to decline following relaxation training and to fluctuate nonsystematically over time, while measures of A-trait were expected to be impervious to relaxation training and to be stable over time. METHOD

The initial therapy interviews of eight psychotherapists with 54 inpatients were recorded and rated on the accurate empathy scale by two raters. Estimates of the consistency of therapists from one segment to another within the same session and from one patient to the next suggested that accurate empathy may not be a stable quality of the therapist as is usually assumed, but instead may reflect a dyadic or relationship variable. Thus, the usual method of assessing the interrater reliability of the scale using the number of patient-therapist combinations rather than the number of therapists alone may be defensible in spite of recent criticisms.

Measures of state anxiety (A state) and trait anxiety (A trait) were obtained from 48 VA psychiatric in-patients before and after either a stressful or nonstressful (control) interview. Scores on 2 of 3 A-state measures were significantly increased by the stress interview; none of the A-state measures was influenced by the control interview, Neither of the 2 A-trait measures was affected by either type of interview. These results demonstrated the meaningfulness of a conceptual distinction between trait and state anxiety and the differential influence of stressful conditions on empirical measures of these concepts. 1 This paper is based upon a dissertation submitted to the Graduate School of Arts and Sciences of Vanderbilt University in partial fulfillment of the requirements for the doctoral degree. This investigation was supported in part by Grant HD-947 from the National Institute of Child Health and Human Development (C. D, Spielberger, principal investigator) and Grant NIH-P07Fr0016-04 from the National Institutes of Health, The author expresses his deepest appreciation to Charles D. Spielberger, dissertation chairman, for his encouragement and guidance during all stages of the investigation.

B-type therapists with neurotic patients and A-type therapists with schizophrenic patients were compared with the opposite patient-therapist pairings for two criterion measures of "improvement" and a measure of "accurate empathy." Neither judged improvement nor length of hospitalization was significantly related to either high-versus low-empathy ratings or to therapist-patient-type pairings. As predicted, however, B-type therapists with neurotic patients and A-type therapists with schizophrenic patients displayed more empathy than A-type therapists with neurotics and B-type therapists with schizophrenics. The results were interpreted as demonstrating a relationship between the findings of two independent lines of research on therapy outcome, one basing outcome predictions on client-centered theoretical constructs, the other on the empirical value of the A-B scale.