Modeling Human Reading with Neural Attention

TL;DR

This paper introduces an unsupervised neural attention model that predicts human reading and skipping behavior by balancing understanding accuracy and attention economy, trained via reinforcement learning.

Contribution

It presents a novel neural attention and autoencoding framework that models both reading and skipping, capturing human behavior more comprehensively than prior models.

Findings

Accurately predicts human skipping behavior and reading times

Competitive with surprisal in modeling reading patterns

Captures qualitative features of human reading behavior

Abstract

When humans read text, they fixate some words and skip others. However, there have been few attempts to explain skipping behavior with computational models, as most existing work has focused on predicting reading times (e.g.,~using surprisal). In this paper, we propose a novel approach that models both skipping and reading, using an unsupervised architecture that combines a neural attention with autoencoding, trained on raw text using reinforcement learning. Our model explains human reading behavior as a tradeoff between precision of language understanding (encoding the input accurately) and economy of attention (fixating as few words as possible). We evaluate the model on the Dundee eye-tracking corpus, showing that it accurately predicts skipping behavior and reading times, is competitive with surprisal, and captures known qualitative features of human reading.