Abstract
Contextual clues aid in speech perception, especially when the signal is degraded by speech disorders or background noise. This study examined whether different types of degradation affect how listeners use contextual predictability. Two groups of 50 listeners were tested across three conditions: dysarthric speech, neurotypical speech masked by noise, and dysarthric speech masked by noise. Listeners relied on semantic context similarly for dysarthric speech in quiet and neurotypical speech in noise (single degradations). However, when dysarthric speech was masked by noise (concurrent degradation), contextual benefit was greatly reduced. Findings highlight the communication burden noise adds for understanding dysarthric speech.