Authors:

DAVIS, M. H., Coleman, M. R., Absalom, A., Rodd, J. M., Johnsrude, I. S., Matta, B., Menon, D.K. & OWEN, A. M.

Reference:

Year of publication:

2006

CBU number:

6329

Abstract:

To investigate the relationship between speech perception, comprehension, and fMRI responses at different levels of awareness, twelve medically qualified anaesthetists were scanned across three sessions while listening to sentences containing ambiguous words (e.g., "there were dates and pears in the fruit bowl"), matched sentences without ambiguous words ("there was beer and cider on the kitchen shelf") and signal correlated noise (SCN). During the first session, participants were unsedated (awake), during the second session they were sedated using Propofol to Ramsey level 2 (relaxed and tranquil), and during the third session they were sedated to Ramsey level 3 (sleepy only responding to command). Large, bilateral temporal-lobe activations were observed for SCN compared to silence, and for speech (both sentence types) compared to SCN, at all three levels of sedation. Consistent with previous observations (Rodd, Davis & Johnsrude, 2005), bilateral inferior frontal activation was observed for the comparison of ambiguous and unambiguous sentences in the awake condition. However, this activation was markedly absent in both sedation conditions. Furthermore, activity in inferior frontal regions predicted post-scan recognition memory at moderate levels of sedation (Ramsey 2). These results suggest a graded degradation of linguistic functions in response to sedation such that ‘higher-level’ semantic processes can be impaired at relatively low levels of sedation, while perceptual processing of speech remains relatively resilient even at higher levels of sedation. These findings may have important implications for interpreting fMRI responses to speech in vegetative patients, where evidence of complex cognitive function can alter diagnosis and prognosis.