Do filled pauses serve a …um… communicative function? A comparison of self-directed and social speech

Some authors argue that filled pauses serve a communicative function in speech. The current study aims to test this function by analyzing the difference in filler-use in a self-directed and social speech condition. Additionally, the influence of Autism Spectrum Disorder (ASD) traits and stress on the difference in filler-use between the two conditions were examined. In the first condition, participants described a series of different tangrams to themselves. In the second condition, participants described tangrams to another participant. More words and filled pauses were used when talking to someone else, but there were no more filled pauses per word. This suggests that filled pauses do not primarily serve as communicative signals. Instead they indicate that at least to some extent, they appear involuntary and automatically as a result of errors in the language production system. Furthermore, no significant correlation was found between the score on the ASD questionnaire or stress and the difference in fillers used in both speech conditions. Future research should consider experimental designs that more clearly differentiate between the presence or absence of a communicative goal, as the context of a monologue may still provide speakers with reasons to use communicative signals.