This study investigated the influences of specific elements of surgery students' verbal and nonverbal communication on evaluators' "objective" ratings of several categories of the students' performances on oral examinations. Three actors and two actresses, dressed as surgery students in a wide range of attire, were videotaped as they reenacted five transcripts of actual students' responses in their oral examinations. For each examination, the actors portrayed the students' responses to the same examining surgeon in two formats, one using direct eye contact with a moderate response rate (Style A) and the other using indirect eye contact with a slower response rate (Style B). All transcripts were taped at least twice. The resulting 255 videotaped "examinations" were randomly distributed in 1988 to 78 clinical surgery faculty representing 46 institutions throughout the United States and Canada. These faculty viewed the reenactments (under the impression they were actual examinations) and rated the "students" performances overall and in ten categories concerning different aspects of the students' knowledge, clinical decision-making skills, and personal characteristics. The performances done in Style A were rated significantly higher than those done in Style B (1) in every performance category except decision making and (2) when the scores were classified by the content of the responses and how professionally dressed the students were. There were also a significant relationship between scores on communication skills and the overall all scores on examinations. These findings suggest that regardless of the content of a student's responses on an oral examination, evaluators are strongly influenced by how well the student communicates.