Investigating the role of temporal lobe activation in speech perception accuracy with normal hearing adults: An event-related fNIRS study

- To our knowledge, this is the first study to employ an event-related design using fNIRS to investigate neural responses time-locked to a speech perception task to compare differences across conditions and trial types of speech perception and production.
- Compared to easier listening conditions, the temporal lobe exhibits significantly greater activity to extract speech in background noise.
- Neural activation is significantly higher when speech is accurately perceived compared to when speech perception is inaccurate.
- A reversal in the hemodynamic pattern was observed during the repetition phase of the task, demonstrating that speech perception and speech production evoke distinct responses in the temporal lobe.

Functional near infrared spectroscopy (fNIRS) is a safe, non-invasive, relatively quiet imaging technique that is tolerant of movement artifact making it uniquely ideal for the assessment of hearing mechanisms. Previous research demonstrates the capacity for fNIRS to detect cortical changes to varying speech intelligibility, revealing a positive relationship between cortical activation amplitude and speech perception score. In the present study, we use an event-related design to investigate the hemodynamic response in the temporal lobe across different listening conditions. We presented participants with a speech recognition task using sentences in quiet, sentences in noise, and vocoded sentences. Hemodynamic responses were examined across conditions and then compared when speech perception was accurate compared to when speech perception was inaccurate in the context of noisy speech. Repeated measures, two-way ANOVAs revealed that the speech in noise condition (−2.8 dB signal-to-noise ratio/SNR) demonstrated significantly greater activation than the easier listening conditions on multiple channels bilaterally. Further analyses comparing correct recognition trials to incorrect recognition trials (during the presentation phase of the trial) revealed that activation was significantly greater during correct trials. Lastly, during the repetition phase of the trial, where participants correctly repeated the sentence, the hemodynamic response demonstrated significantly higher deoxyhemoglobin than oxyhemoglobin, indicating a difference between the effects of perception and production on the cortical response. Using fNIRS, the present study adds meaningful evidence to the body of knowledge that describes the brain/behavior relationship related to speech perception.