Modelando procesos cognitivos de la lectura natural con GPT-2

TL;DR

This paper explores how GPT-2 models can simulate natural reading cognitive processes, showing improved results over previous models like Ngrams and LSTM in understanding eye movement predictability.

Contribution

It introduces GPT-2 based models for cognitive process modeling in reading, demonstrating their superior performance compared to earlier models.

Findings

GPT-2 models outperform Ngrams and LSTM in predicting eye movements.

GPT-2 captures more complex aspects of reading cognition.

Results support GPT-2's potential in cognitive neuroscience applications.

Abstract

The advancement of the Natural Language Processing field has enabled the development of language models with a great capacity for generating text. In recent years, Neuroscience has been using these models to better understand cognitive processes. In previous studies, we found that models like Ngrams and LSTM networks can partially model Predictability when used as a co-variable to explain readers' eye movements. In the present work, we further this line of research by using GPT-2 based models. The results show that this architecture achieves better outcomes than its predecessors.