Evaluation Model of Cognitive Distraction State Based on Eye Tracking Data Using Neural Networks

3. Distraction Model

The input data and output value of the distraction model that we propose are shown in Figure 1. The input data are eye-tracking data, and the output value is a degree of distraction.

Figure 1.

Input and output for distraction model

A simple recurrent neural network (Jeffrey, 1990) is used as the basic structure in the proposed model, such as for a selective attention model or a short-term memory model in the study (T. Harada et al., 2012). A simple recurrent neural network is shown in Figure 2, which presents the basic structure with two units for simplicity.

Figure 2.

Simple recurrent neural network