Design and evaluation of the effectiveness of a corpus of congruent and incongruent English sentences for the study of event related potentials

Abstract

Objective

To design and evaluate the effectiveness of a stimulus material in eliciting the N400 event related potential (ERP).

Design

A set of 700 semantically congruent and incongruent sentences was developed in accordance with current linguistic norms, and validated with an electroencephalography (EEG) study, in which the influence of age and gender on the N400 ERP magnitude was analysed.

Study sample

Forty-five normal-hearing subjects (19–57 years, 21 females) participated in the EEG study.

Results

The stimulus material used in the EEG study elicited a robust N400 ERP, with a morphology consistent with the literature. Results also showed no statistically significant effect of age or gender on the N400 magnitude.

Conclusions

The material presented in this paper constitutes the largest complete stimulus set suitable for both auditory and text-based N400 experiments. This material may help facilitate the efficient implementation of future N400 ERP studies, as well as promote standardisation and consistency across studies.

Keywords:

• N400
• speech perception
• language-related ERPs
• semantic violation

Acknowledgements

The authors gratefully acknowledge Ms Lorna Betts for her help with the audio recording of the sentences; Mr Greg Stewart (NAL: National Acoustic Laboratories, Sydney, Australia) for his help with the calibration of the stimuli; and the Macquarie University students who participated in the subjective evaluation of the sentences.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The semantic violation that elicits the N400 ERP can also be primed by a list of words that do not form sentences or discourse (Titone and Salisbury, 2004; Romei et al. 2011; Brown and Hagoort, 1993).

2 The Zipf scale takes the log10 of the frequency per billion words (Van Heuven et al., 2014). Thus a Zipf value equal to 4 indicates a frequency per million words (fpmw) of 10; and a Zipf equal to 7 corresponds to a fpmw of 10,000. This scale comes from the American linguist George Kingsley Zipf, who was the first to formulate a law about the regularities of word frequency distribution (Zipf, 1949).

Additional information

Funding

This work was supported by the Australian Government Department of Health.