Fourteen psychiatric patients were evaluated by two psychiatrists before and after treatment with either chlorpromazine or placebo. The rating interviews were tape-recorded and the recordings were rated by two other psychiatrists who did not see the patients.There was significant agreement between and among ratings based on the face-to-face interviews and those based on the tape-recorded interviews with regard to global ratings of severity of illness, incapacitation, clinical change following treatment and the "nonvisual" elements of a 26-item symptom scale.Tape recording of clinical interviews is a promising tool for the evaluation of treatments and for attempts to objectify and quantify clinical impressions. With tape recordings the investigator gains control of the interview situation without distorting it beyond recognition. He can repeat fleeting events as often as necessary, immediately contrast events separated widely in time and space and eliminate some of the factors which presently appear as "error" in his investigations. Further, since the information on a tape recording can be described physically, the use of tapes may make it possible to isolate and characterize the cues which subserve clinical impressions.Abstract Teaser