Inferring mind wandering from perceptual decision making

People need to sustain focused attention to achieve goals. Yet, attention often lapses, as minds wander towards task-unrelated thoughts. The conventional way to study temporal dynamics in mental states is through intermittent thought probes, which explicitly ask if thoughts are task-related. However, probes are rare and interrupt behavior. We designed a novel paradigm aiming to infer mind wandering (MW) from performance alone. On each trial, participants see a random dot kinematogram with varying evidence and indicate the coherent direction. Importantly, responses are repetitive: 90% of trials exhibit the same direction. To validate our task, participants respond to occasional thought probes. When they reported being off-task, accuracy was higher and response time (RT) was lower, suggesting less stimulus processing and more reliance on bias. To classify internal states for individual trials from performance, we fit a Hidden Markov Model with Generalized Linear Models (HMM-GLM) for each state to responses. In a two-state HMM-GLM, RT was lower on off-task trials. Latent states are also aligned with self-reported focus. This shows that attentional states can be measured on a trial-to-trial basis without thought probes, paving the way for future MW research.