An event-related potential study of semantic congruity and repetition in a sentence-reading task: Effects of context change

Authors


  • This research was supported by the National Health and Medical Research Council, the NSW Institute of Psychiatry, and the Rebecca I Cooper Medical Research Foundation.

  • The technical assistance of Anne-Marie Shelley and Carol Stevens is gratefully acknowledged.

Abstract

The experiment examined the joint effects of semantic congruity and repetition on event-related brain potentials (ERPs) elicited in a sentence priming task. In an initial training phase, subjects were familiarized with a list of 60 congruous and incongruous sentences. During the second phase, ERPs were recorded as subjects silently read a set of 180 unconnected sentences. One third of the sentences were presented exactly as they had been seen in training (Old sentences), one third were presented for the first time (Completely New sentences), and one third involved a re-pairing of the frames and completions of the congruous and incongruous Old sentences (New Pair sentences). The N400 congruity effect was reduced for Old as compared with Completely New and New Pair sentence completions. These results suggest that N400 reflects processes that are sensitive to both existing semantic associations and representations of previous episodes that include the context of the eliciting stimulus. A late positive component (LPC) involving a sustained positive shift in the waveform after 600 ms was largest for incon-grous completions and occurred somewhat earlier for Old sentences. This pattern of results is consistent with the notion that the LPC is an index of episodic retrieval and elaborative processes. The data also suggest the presence of an early onset slow positive shift that is only evident for New Pair congruous sentences.

Ancillary