Cross-modal enhancement of the MMN to phonemes indicates automatic processing of grapheme-phoneme correspondences.

Dries Froyen, university Maastricht psychology neurocognition

Abstract
Learning the correspondences between graphemes and phonemes is a crucial step in reading acquisition. The aim of our study is to investigate whether in literate adults, grapheme-phoneme pairs are processed automatically as compound stimuli. Van Atteveldt, et.al. (2004) where the first to demonstrate neural binding of letters and speech sounds in cross-modal superior temporal regions with fMRI. To investigate the time course and automatic nature of this neural binding we conducted an EEG experiment, focusing on the Mismatch Negativity component.
We compared MMN responses to passively perceived deviant phonemes in a single channel (speech sounds) and cross-modal (speech sounds - letters) context. If the same auditory deviant phoneme evokes a quantitatively different auditory MMN in both contexts, this would suggest that the presence of the grapheme has an influence on the auditory sensory-memory trace formed by the phoneme, i.e. on the neural processing of the speech sound.
We found enhanced MMN amplitude to auditory deviants in the audiovisual context. Considering the timing of this EEG-component, this finding suggests automatic processing of grapheme-phoneme pairs as integrated compound stimuli. Furthermore, we show that the width of the time window between the presentation of letters and speech sounds is critical for neural binding.

Not available

Back to Abstract