文章基本信息

标题：A Recurrent Neural Network for Attenuating Non-cognitive Components of Pupil Dynamics
本地全文：下载
作者：Koorathota, Sharath ; Thakoor, Kaveri ; Hong, Linbi 等
期刊名称：Frontiers in Psychology
电子版ISSN：1664-1078
出版年度：2021
卷号：12
页码：12
DOI：10.3389/fpsyg.2021.604522
出版社：Frontiers Media
摘要：There is increasing interest in how the pupil dynamics of the eye reflect underlying cognitive processes and brain states. Problematic, however, is that pupil changes can be due to non-cognitive factors, for example luminance changes in the environment, accommodation and movement. Such factors are ever present in natural scenes and non-laboratory conditions and thus make it challenging to establish pupil measures as correlates of cognition which relate to real-world behavior and task performance. In this paper we consider how by modeling the response of the pupil in real-world environments we can capture the non-cognitive related changes and remove these to extract a residual signal which is a better index of cognition and performance. Specifically, we utilize sequence measures such as fixation position, duration, saccades and blink-related information as inputs to a deep recurrent neural network (RNN) model for predicting subsequent pupil diameter. We build and evaluate the model for a task where subjects are watching education videos and subsequently asked questions based on the content. Compared to commonly-used models for this task, the RNN had the lowest errors rates in predicting subsequent pupil dilation given sequence data. Most importantly, and consistent with our hypothesis, the model performance negatively correlated with participant performance on a post-viewing test. This suggests that deep learning sequence models may be good for separating components of pupil responses that are linked to luminance and accommodation from those that are linked to cognition and arousal.
关键词：recurrent neural network (RNN); Pupil diameter (PD); eye tracking; Video viewing; Pupil response