We know that reading involves a coordination between textual characteristics and visual attention, but what does eye gazeduring reading tell us about comprehension? We addressed this question by training random forest models (a machinelearning technique) to predict reading comprehension from ensembles of interacting global gaze features in a person-generalizable manner. We used data from two prior studies in which readers (Ns = 104, 130) answered multiple-choicecomprehension questions during and/or shortly after ( 30 mins) reading a 6500-word text. The models were highly accurateat predicting reading comprehension assessed during reading at both the page- (AUROC = .882) and participant- level (r= .671; computed by aggregating page-level predictions). Accuracy for the post-reading models was lower (AUROCsbetween .538 and .552; rs between .343 and .373), but significantly above chance baselines. Collectively, these findingsconfirm a link between global eye movement behavior and higher-order outcomes of reading.

How much do we mind wander during conversations, and how does that affect objective outcomes and subjective per-ceptions of the conversation? We studied computed-mediated dyadic negotiations during which participants (N = 144)discreetly reported whenever they were thinking about something else, and whenever they thought their partner was not at-tending. Participants mind wandered around 19% of the time. Surprisingly, the number of times that a participant thoughtthat their counterpart was not attending correlated almost perfectly with the first participants own number of mind wander-ing reports (r-partial = .941), but very poorly with the other participants number of reports (r-partial = .004) (controlled fortime until agreement). Mind wandering negatively affected subjective (F(1, 57) = 6.48, p = .014) but not objective (F(1,57) = .089, p = .766) outcomes. These findings suggest that mind wandering, and the attribution of mind wandering toothers, leads to worse social psychological outcomes.