On the role of phonological short-term memory in sentence processing: ERP single case evidence on modality-specific effects

Abstract

The present study explored a possible interaction between distinct language processes and components of phonological short-term memory (pSTM) in a patient with a pSTM profile. Event-related brain potentials (ERPs) were recorded while HG and age-matched controls engaged in auditory and visual sentence correctness tasks. Stimulus onset asynchrony (SOA) was varied in the visual modality. Controls showed an early anterior negativity followed by a P600 for syntactic violations and an N400 for semantic violations in the auditory and the short visual SOA condition. In the long visual SOA condition only a P600 and an N400 were observed. Across all tasks, HG displayed a comparable early anterior negativity and N400 pattern to controls. However, the P600 was replaced by a centro-parietal negativity (500–800 ms) that was followed by a very late positivity (900–1300 ms) in the visual modality, indicating that late syntactic processes are sensitive to SOA manipulation. This result implies that the cortical regions lesioned in HG may be part of a neural network that engages the pSTM system during “temporally variable” late syntactic processing in the visual modality. The combined results indicate that the pSTM system differentially impacts semantic and late syntactic processes.

Acknowledgments

The authors would like to thank HG, who supported us without hesitation in this endeavour. Many thanks to Anja Hahne for providing the stimulus material of her dissertation to conduct this study. We are grateful to Ina Koch for running both the patient and the control experiments.

Notes

Note, however, that in these studies no behavioural task was applied that would have allowed the analysis of correctly understood sentences from those that were not understood. The analyses were thus conducted on data averaged over correctly and incorrectly understood sentences.

Further evidence supporting the “automaticity” of the early anterior negativity comes from manipulations of proportion and instruction (Hahne & Friederici, 1999, 2002), load (Vos, Gunter, Schriefers, & Friederici, 2001), and proficiency (Hahne & Friederici, 2001).