Extracting blinks from continuous eye-tracking data in a mind wandering paradigm.

• Blink extraction methods are inconsistently reported in attention research. • Selecting suboptimal blink parameters can affect results. • Consulting eye-tracking data can help justify optimal blink parameters. • Blink durations are positively associated with mind wandering. • Blink durations can help detect mind wandering in multiple modalities. Mind wandering is a universal phenomenon in which a person's attention decouples from stimuli within their current environment. Researchers have sought objective, less disruptive indicators of cognitive disengagement, resulting in a focus eye tracking and blink characteristics. Such research has found positive associations between mind wandering and blink characteristics, typically in reading tasks. However, extracting blinks accurately from continuous eye-tracking data is complex, and the literature contains inconsistently reported data processing methods, some of which may have an elevated risk of identifying noise as signal. Further, the relationship between attentional disengagement and blink durations has not been fully explored in multiple task modalities. We conducted three modality-specific experiments while recording eye movements. Blink durations varied as a function of stimulus/task engagingness; less engaging tasks yielded longer blink durations, suggesting a link between blinking and mind wandering. Recommendations are provided for researchers seeking to accurately derive blink events from continuous, binocular, eye-tracking data. [ABSTRACT FROM AUTHOR]