Data files and R scripts for the study described in Gambi, Pickering, and Rabagliati. Beyond Associations: Sensitivity to structure in pre-schoolers' linguistic predictions, Cognition
Please read the following notes on the structure of this project. We conducted two experiments as part of this project. In Experiment 1, we tested both adults and children (3-to-5 yo). In Experiment 2, we tested only children (3-to-5 yo). Please refer to the published article for details on the number of participants and their demographics.
Both experiments used eye-tracking while listening. In Experiment 1, participants heard a spoken sentence that contained either a Predictive or a Non-Predictive verb (this is the first factor used in our analyses). Sentences could also be Long or Short. Short sentences were simple active transitive sentences such as "Pingu will ride/pull the horse". Long sentences had additional modifiers inserted before the final noun (e.g., "Pingu will ride/pull the very tired horse"). This was the second factor in our analyses. While listening, participants observed 4 entities: Subject (always the well-known cartoon character Pingu), Agent (a prototypical agent of the predictive verb; e.g., cowboy), Patient (a protoypical patient of the predictive verb; e.g., horse), and a Distractor. In Experiment 2, participants listened to sentences like "Now, you can show Pingu...riding/pulling!", and Verb Type (Predictive/Non-Predictive was the only factor).
For each group of participants, we conducted two types of analyses: Snapshot analyses and Growth Curve Analyses.
Snapshot Analyses. In Experiment 1, these analyses included both the factor Length (Long/Short) and the factor Verb Type (Predictive/Non-Predictive). They compared the likelihood of fixations to the Patient (or Agent) across conditions. They were restricted to a window ranging from 200 ms before to 100 ms after the onset of the critical final word of the sentence. In Experiment 2, the time window ranged from 200 ms before to 100 ms after the offset of the sentence final verb, and Verb Type was the only factor.
Growth Curve Analyses. These analyses only included the factor Verb Type and compared fixation proportions to the Patiemt (or Agent) across conditions over time. They were restricted to a window ranging between 500 ms before and 1700 ms after the offset of the critical verb (Experiment 1), or between 500 ms before and 200 ms after verb offset (Experiment 2).