Eye-movements can help disentangle mechanisms underlying disfluency
To reveal the underlying cause of disfluency, several authors related the pattern of disfluencies to difficulties at specific levels of production, using a Network Task. However, disfluencies are arguably multifactorial. To disentangle disfluency related to word preparation from other factors, we combined this task with eye-tracking. We manipulated lexical and grammatical selection difficulty. In Experiment 1, lines connecting the pictures varied in length. Experiment 2 only used short lines. In both experiments, lexical selection difficulty promoted self- corrections and pauses and longer viewing times. Multivariate pattern analyses (MVPA) demonstrated that lexical selection difficulty is predictable from disfluency and eye-movement data patterns. In Experiment 1, participants spent less time gazing at pictures preceded by long lines, while producing more pauses. This suggests they used a strategy to inspect other areas than the upcoming picture. We conclude that eye-tracking is informative about mechanisms underlying disfluency related to speech encoding, self-monitoring, and stalling strategies.