Date of Graduation


Document Type



Unlike dyadic conversation where the social burden of conversation can be shared, the quantity and quality of speech produced during a public speaking challenge is primarily dependent upon the speaker. This study examined whether differences in disclosure patterns reflect a form of subtle avoidance during public speaking challenges. Termed para-verbal avoidance, it was hypothesized individuals with increased anxiety would refrain from talking about themselves for fear of revealing potentially embarrassing personal information. Also, it was proposed individuals with increased anxiety would avoid discussing subjective information, as it can be debated by others and used as the basis for forming a negative impression. Undergraduate students were initially screened with an anxiety measure to identify 80 individuals high and 80 individuals low in social anxiety. Participants gave two brief, impromptu speeches in front of a small audience as speech content and subjective anxiety were recorded. For each speech, participants were randomly assigned to one of two experimental conditions. During the first speech condition, half were told their videotaped speeches would be critiqued by experts in the field of public speaking while remaining participants were told the content of their speeches would be ignored. During the second speech condition, the task demand of the public speaking challenge was manipulated through the assignment of speech topics. Although all speech topics presented speakers with an opportunity to discuss personal information, half of the participants received topics considered taboo and not typically discussed in the presence of strangers. Additional questionnaires assessed concerns related to social anxiety, public speaking and depression. Speeches were transcribed and coded for content. Differences between subjective anxiety and patterns of self-disclosure among anxious and low-anxious participants were examined. Analyses of variance and covariance failed to support hypothesized patterns of paraverbal avoidance as described above. However, this may be due to an observed difficulty manipulating experimental factors. Significant negative correlations were observed between anxiety and the discussion of subjective information. Implications are discussed.