Published: Aug. 22, 2023
ObjectivesIn clinical and laboratory settings, speech recognition is typically assessed in a way that cannot distinguish accurate auditory perception from misperception was mentally repaired or inferred context. Previous work showed the process of repairing misperceptions elicits greater listening effort, this elevated effort lingers well after sentence heard. That result suggests cognitive repair strategies might appear successful when testing single utterance but fail for everyday continuous conversational speech. The current study tested hypothesis has consequence carrying over to interfere with later words sentence. MethodStimuli were open-set coherent sentences presented intact word early replaced noise, forcing listener use context missing word. Sentences immediately followed by digit triplets, which served probe carryover Control conditions allowed comparison did not demand mental repair, as removed need attend post-sentence stimuli, digits altogether. Intelligibility scores accompanied time-series measurements pupil dilation assess load during task, subjective rating effort. Participants included adults cochlear implants (CIs), an age-matched group younger listeners typical hearing (TH) comparison. ResultsFor CI group, needing resulted more errors on sentence, especially sensible perception. needed also contained unmasked. All groups substantial increase required even successful. Younger TH clear differences moment-to-moment allocation different conditions, while other not. ConclusionFor listeners, can last long enough follow This pattern could pose serious problem regular communication would go overlooked utterances, where chance before responding. Carryover predictable basic intelligibility scores, be revealed behavioral data are extra such digits.
Language: Английский