Tracking - uh - attention: Examining the influence of filled pauses during speech comprehension

In this study, we investigated the temporal dynamics of attention during comprehension and the subsequent effects of changes in attention on speech production manipulated via the filled pause uh. Participants orally described a series of pictures (pretest description phase), then listened to a short informational passage and answered comprehension questions about the passage, and then orally described another series of pictures (posttest description phase). Participants completed this twice: the first passage included a dative structure alternation, and the second passage included a passive structure alternation. For each passage, half of the participants heard a passage with filled pauses included (present) or with no filled pauses (absent). We measured participants' pupil diameter as they listened to the passages to index online attention and measured pupil size synchrony - an index of moment-to-moment shared attention across observers. We also measured downstream influences on processing via tendency to reuse the prevalent structure in the passage, and via comprehension accuracy of the passages. For speech with filled pauses compared to without, we found that pupil size was larger overall, peaking after the filled pause appeared, and that listeners' synchrony was higher. Regarding subsequent behavior, we found no evidence of structural alignment in any condition, nor any influence from filled pauses. We also did not find effects of filled pauses on comprehension accuracy. Taken together, these results suggest that filled pauses influence attention in the moment and do not interfere with downstream processing.