Abstract
Scan pattern analysis has been discussed as a promising tool in the context of real-time gaze-based applications. In particular, information-theoretic measures of scan path predictability, such as the gaze transition entropy (GTE), have been proposed for detecting relevant changes in user state or task demand. These measures model scan patterns as first-order Markov chains, assuming that only the location of the previous fixation is predictive of the next fixation in time. However, this assumption may not be sufficient in general, as recent research has shown that scan patterns may also exhibit more long-range temporal correlations. Thus, we here evaluate the active information storage (AIS) as a novel information-theoretic approach to quantifying scan path predictability in a dynamic task. In contrast to the GTE, the AIS provides means to statistically test and account for temporal correlations in scan path data beyond the previous last fixation. We compare AIS to GTE in a driving simulator experiment, in which participants drove in a highway scenario, where trials were defined based on an experimental manipulation that encouraged the driver to start an overtaking maneuver. Two levels of difficulty were realized by varying the time left to complete the task. We found that individual observers indeed showed temporal correlations beyond a single past fixation and that the length of the correlation varied between observers. No effect of task difficulty was observed on scan path predictability for either AIS or GTE, but we found a significant increase in predictability during overtaking. Importantly, for participants for which the first-order Markov chain assumption did not hold, this was only shown using AIS but not GTE. We conclude that accounting for longer time horizons in scan paths in a personalized fashion is beneficial for interpreting gaze pattern in dynamic tasks.
Highlights
In the context of driving, visual scanning behavior must be organized such as to obtain relevant information for the driving task just in time
For group l > 1 we did not find a significant effect of trial period on normalized local gaze transition entropy (LGTE) (Fig 6A), while we found a significant decrease in the normalized local active information storage (LAIS) in the after lane change interval compared to the previous intervals (t(36.9) = 2.481, p = 0.0178, Fig 6B)
We found evidence supporting three hypotheses raised in the beginning: 1) we showed that in a dynamic task the assumption that scan path data are sufficiently modeled by first-order Markov chains does not hold in general, 2) the time horizon of past fixations being predictive for the fixation varies between participants, and 3) accounting for detected long-range dependencies between fixations in a personalized way is beneficial for the meaningful quantification of the predictability of scan paths
Summary
In the context of driving, visual scanning behavior must be organized such as to obtain relevant information for the driving task just in time. In the context of human-vehicle cooperation, the development of driving assistance systems that can adapt its functionalities in order to support the driver in demanding situations or according to the driver’s individual needs has been put forward as an important topic, lately [2] For such adaptations to be useful one important prerequisite is usually to identify relevant changes in user states or user behavior and to evaluate the effects of the adaptation on the user. In contrast to the GTE, the AIS optimizes the time span of past fixations being informative for the predictability of the fixation and entering the computation of scan path regularity To evaluate whether this representation of scan paths is beneficial in the context of human state estimation in a dynamic task like driving, we compare the two measures on data from a driving simulator experiment
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
