The neighborhood distribution effect in visual word recognition: Words with single and twin neighbors

Lexical-decision tasks were used to test the role of neighborhood distribut ion in visual word recognition. Predictions based on the interactive activa tion model were generated by running simulations. The data were compared fo r words with 2 higher frequency neighbors that differed in their neighborho od distribution. The neighbors were "single" when they did not share a neig hborhood relationship (e.g., neighbors of flanc:franc-blanc) or "twin" when they shared a neighborhood relationship (e.g., neighbors of firme: ferme-f onne). Results show a facilitatory neighborhood distribution effect on word s in Experiments 1 (easy pseudowords) and 3 (difficult pseudowords and easy pseudowords) and on pseudowords in Experiment 2. These data can be account ed for in terms of lexical inhibition in the interactive activation framewo rk.