Effects of audio modeling versus no modeling and of experiential supervision versus didactic supervision were examined in brief empathy training. Three supervisors had six trainees in each of four groups: modeling-didactic supervision, no-modeling-didactic supervision, modeling-experiential supervision, no-modeling-experiential supervision. Two groups of 12 each received no supervision: modeling-controls and no-modeling-controls. Subject responses to recorded client statements indicated significant training effects from audio modeling as well as from didactic supervision on rated level of empathy. No improvement was found for the experiential supervision.

Effects of supervisor style (techniques‐oriented, counseling‐oriented, and placebo‐oriented) and the presupervision conditions of audio modeling versus no modeling were examined in a supervision analogue. Six groups of 18 trainees heard three sets of recorded client statements and responded as counselors. Supervised groups had two 15‐minute conferences with supervisors between sets 1 and 2 and between sets 2 and 3. Control groups received no supervision. Among supervised groups only those trainees receiving the techniques‐oriented supervision showed significant improvement in rated level of empathy. Significant learning effects from the presupervision use of audio modeling were also found.

Effects of supervisor style (techniques type and counseling type) and supervisor empathy upon counselor learning of empathy in a supervision analogue were examined. 3 groups of 14 Ss heard 2 sets of recorded client statements and responded as counselors. The 2 experimental groups had 20-minute conferences with supervisors (N = 14) between Sets 1 and 2. The control group received no supervision. An analysis of recorded responses indicated that 8a in the techniques and control groups were higher in the learning of empathy than Ss receiving counseling-oriented supervision. Supervisor perceptions of improvement in their own supervisees showed considerable distortion when compared with ratings by independent judges. Supervisor empathy was not related to counselor learning.

The effects of training method (programmed vs. individual supervision) and supervisor experience (peer vs. professional) upon the learning of empathy were examined. Eight supervisors (four professional and four peer) were assigned three subjects in each of three groups: individually supervised, programmed, and control. Subjects listened to three sets of recorded client statements and responded as counselors. The two experimental groups received training after the first and second sets. The control group received no training. An analysis of recorded responses indicated that the individually supervised subjects achieved a significantly higher level of empathy than the controls. Subjects receiving programmed training were intermediate in learning between individually supervised and control subjects. Supervisor experience was not found to be a significant factor as no posttraining differences were noted between professional and peer-trained groups. This result was interpreted as supporting the use of peers in such interpersonal skills training.