Plasticity and transfer in the sound system: exposure to syllables in production or perception changes their subsequent production

ABSTRACT

This study focuses on the ability of the adult sound system to reorganise as a result of experience. Participants were exposed to existing and novel syllables in either a listening task or a production task over the course of two days. On the third day, they named disyllabic pseudowords while their electroencephalogram was recorded. The first syllable of these pseudowords had either been trained in the auditory modality, trained in production or had not been trained. The EEG response differed between existing and novel syllables for untrained but not for trained syllables, indicating that training novel sound sequences modifies the processes involved in the production of these sequences to make them more similar to those underlying the production of existing sound sequences. Effects of training on the EEG response were observed both after production training and mere auditory exposure.

KEYWORDS:

• Language production
• EEG
• syllables
• phonetic encoding
• transfer

Acknowledgements

This research was funded by the Swiss National Science foundation (grant n° 100014_159374 to A. Bürki). The first and third authors were funded by the Deutsche Forschungsgemeinschaft (DFG), project number 317633480 – SFB 1287 (Projects B05 and C04) during the preparation of the article. The authors wish to thank Samuel Schmid for his help during data collection / pre-processing and Daniel Schad and the Q project of the DFG-SFB 1287 for their help in designing the contrast matrices for the statistical analyses.

Disclosure statement

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

Data availability statement

The data used in the present paper can be accessed here: https://osf.io/97pyj/

Notes

1 Results for the model that includes low frequency syllables are presented in Appendix 10.

2 In a factorial design, the number of possible contrasts corresponds to the number of levels minus 1. We selected the two contrasts involving the no training condition (rather than the contrast between auditory and production training) for our main analysis as these contrasts directly test our hypotheses.

3 General arguments against baseline corrections can be found, for instance, in Michel et al. (2009).

4 The contrast between the no training and auditory training conditions is significant in some but not in all analyses. For instance, it is significant when low frequency syllables are included in the model, but only when we consider the dataset trimmed of contaminated epochs. As a consequence, we consider that the evidence in favour of this effect is unreliable.

Additional information

Funding

This research was funded by the Swiss National Science foundation [grant n° 100014_159374 to A. Bürki]. The first and third authors were funded by the Deutsche Forschungsgemeinschaft (DFG), project number 317633480 – SFB 1287 (Projects B05 and C04) during the preparation of the article.