Verbal Perception and the Word Length Effect

TL;DR

This paper presents a theoretical framework explaining the word length effect in verbal perception and short-term memory, emphasizing semantic space localization and phase space clustering, supported by experimental data and proposing new tests.

Contribution

It introduces a novel theory linking word length effects to semantic localization and phase space clustering, providing explanations for both standard and inverse effects, validated by experimental data.

Findings

Short words are recalled better than long words due to semantic localization.

Inverse word length effect occurs when the usual clustering property breaks down.

Experimental data from PEERS supports the theory's main predictions.

Abstract

A theoretical framework is proposed for the understanding of verbal perception -- the conversion of words into meaning, modeled as a compromise between lexical demands and contextual constraints -- and the theory is tested against experiments on short-term memory. The observation that lists of short words are recalled better than lists of long ones has been a long-standing subject of controversy, further complicated by the apparent inversion of the effect for mixed lists. In the framework here proposed, these behaviors emerge as an effect of the different level of localization of short and long words in semantic space. Events corresponding to the recognition of a nonlocal word have a clustering property in phase space, which facilitates associative retrieval. The standard word-length effect arises directly from this property, and the inverse effect from its breakdown. An analysis of data from the PEERS experiments (Healey and Kahana, 2016) confirms the main predictions of the theory. Further predictions are listed and new experiments are proposed. Finally, an interpretation of the above results is presented.