Neighbourhood distribution interacts with orthographic priming in the lexical decision task

Abstract

Lexical decision tasks (LDTs) were used with a masked priming procedure to test whether neighbourhood distribution interacts with orthographic priming. Word targets had either ‘single’ neighbours when their two higher frequency orthographic neighbours were spread over letter positions (e.g., neighbours of LOBE: robe-loge) or ‘twin’ neighbours when they were concentrated on a single letter position (e.g., neighbours of FARD: lard-tard). All word targets were preceded by their highest frequency orthographic neighbour or by a control prime. An inhibitory priming effect was found for words with single neighbours, but not for words with twin neighbours, in both a yes/no LDT (Experiment 1a) and a go/no-go LDT (Experiment 1b). This interaction was replicated in a go/no-go LDT when the position of the letter yielding the neighbour prime was controlled (Experiment 2). Simulations run on the word materials revealed that the interactive activation model captures the inhibitory priming effect in the single-neighbour condition but fails to capture the loss of priming in the twin-neighbour condition.