Comparing predictors of sentence self-paced reading times: Syntactic complexity versus transitional probability metrics

When estimating the influence of sentence complexity on reading, researchers typically opt for one two main approaches: Measuring syntactic (SC) or transitional probability (TP). Comparisons predictive power both approaches have yielded mixed results. To address this inconsistency, we conducted a self-paced reading experiment. Participants read sentences varying complexity. From alternatives, selected set SC and TP measures, respectively, that provided best fit to data. We then compared contributions measures times when entered into same model. Our results showed while explained significant portions variance in (over above control variables: word/sentence length, word frequency position) included independent models, their changed drastically were Specifically, only observed effects TP. conclude our experiment variables bulk variance. comparing small TP, appear be more robust.