Probabilistic Predictions of People Perusing: Evaluating Metrics of Language Model Performance for Psycholinguistic Modeling (2009.03954v1)

Abstract: By positing a relationship between naturalistic reading times and information-theoretic surprisal, surprisal theory (Hale, 2001; Levy, 2008) provides a natural interface between LLMs and psycholinguistic models. This paper re-evaluates a claim due to Goodkind and Bicknell (2018) that a LLM's ability to model reading times is a linear function of its perplexity. By extending Goodkind and Bicknell's analysis to modern neural architectures, we show that the proposed relation does not always hold for Long Short-Term Memory networks, Transformers, and pre-trained models. We introduce an alternate measure of LLMing performance called predictability norm correlation based on Cloze probabilities measured from human subjects. Our new metric yields a more robust relationship between LLM quality and psycholinguistic modeling performance that allows for comparison between models with different training configurations.