On the impact of adjacency on transposed-word effects under serial presentation

In two experiments we examined the impact of adjacency (adjacent vs. nonadjacent words) on transposed-word (TW) effects. The nonadjacent words were separated by a single intervening word. Both experiments involved the serial presentation of words in a sequence of words that could be grammatically correct or not. Participants performed a speeded grammatical decision task. The critical ungrammatical sequences were either TW sequences for which transposing two words would result in a grammatically correct sentence (e.g., the white was cat big) or the corresponding control sequences for which transposing any two words could not generate a grammatically correct sentence (e.g., the white was cat slowly). Experiment 1 presented word sequences auditorily, and Experiment 2 used the rapid serial visual presentation procedure. In both experiments there was clear evidence for an impact of adjacency on TW effects, with the effects being greater in the adjacent condition. However, this pattern was only robust in the analysis of reaction times under auditory presentation and only in the analysis of accuracy rates under rapid serial visual presentation. We discuss these results in terms of the relative role played by bottom-up word position uncertainty and sentence-level constraints in driving TW effects. Our results also provide a further dissociation between TW effects observed with auditory sequences and serially presented written words, and we discuss possible reasons for this dissociation in terms of differences in how participants make speeded grammatical decisions to sequences of auditory stimuli compared with serially presented visual stimuli. (PsycInfo Database Record (c) 2026 APA, all rights reserved).