Understanding how target-word associates shape lexical-semantic processing and representation

Lexical-semantic processing underlies how people comprehend, retrieve, and produce language. Research has traditionally emphasized properties of individual words, such as frequency, concreteness, orthographic neighborhood density, and treated recognition and meaning access as largely self-contained. However, a growing body of work suggests that the lexical and semantic properties of a word’s associates (the items activated alongside it) also shape processing of the word in both recognition and production tasks. This network-based perspective aligns with distributed and interactive models of semantic representation, but the scope, automaticity, and boundary conditions of associate recruitment remain unclear. A complementary framing goes further: associative networks may function as a source of prior knowledge that the language system exploits to predict and integrate relations between words, positioning semantic prediction and integration as processes that are just as fundamental to lexical-semantic processing as recognition and meaning access.

An associative neighborhood comprises the words semantically, lexically, or contextually linked to the target word. Recent studies indicate that properties of associates contribute to performance in both recognition and production tasks, and that their recruitment varies with task demands. Open questions include which types of associate information (semantic, lexical, contextual) are recruited, under which task conditions, and for which kinds of target words—particularly across the concrete–abstract continuum. Systematic investigation is needed to refine models of lexical-semantic representation and to specify the mechanisms by which associates exert their influence.

This Research Topic gathers work that clarifies how the lexical and semantic properties of target-word associates shape the lexical and semantic processing of the target. Priority questions include: when associative information is recruited during lexical and semantic decision tasks, how such recruitment depends on the target properties, how such recruitment depends on the task, and what these patterns imply for theories of semantic representation. Beyond isolated word processing, this topic explicitly treats semantic integration and prediction as core scientific questions: how do associative networks supply prior knowledge that constrains or facilitates the prediction of upcoming words and the integration of relational meaning across words? By integrating behavioral, computational, and neurocognitive approaches, this topic aims to move beyond isolated word models and toward an account that better captures the dynamic interplay between a word and its associative network.

We welcome papers addressing (but not limited to) the following themes:

- Effects of associates’ lexical properties (e.g., frequency, length, orthographic or phonological neighborhood) on target-word processing
- Effects of associates’ semantic properties (e.g., concreteness, typicality, semantic diversity, valence) on target-word processing
- How task demands (lexical decision, semantic decision, naming, production) modulate the recruitment of associate information
- Computational accounts (spreading activation, distributional semantics, word embeddings) that predict associate effects
- Methodological advances in identifying, quantifying, or manipulating associative neighborhoods
- The role of associative networks as prior knowledge in semantic prediction and integration — including how the structural and statistical properties of a word’s neighborhood shape anticipatory processing and the resolution of meaning across multi-word contexts