Neural encoding of speech in noise: Effects of background noise and behavioral correlates

Humans often communicate in noisy environments where speech-in-noise (SIN) perception is challenged by spectral masking that degrades speech signals. Although most healthy individuals perform well in SIN tasks, there is significant variability in abilities across the population. Research in the field has considered the Frequency Following Response (FFR) as a valuable tool to assess the auditory system's neural ability to track acoustic stimuli, as it mimics their spectro-temporal features. While significant associations have been found between FFR parameters and SIN performance, inconsistent results across studies raise questions about the exact relationship between FFR and SIN perception. To clarify these relationships, we recorded the neural signals of 116 young healthy adults in response to a two-vowel /oa/ syllable, with a rising pitch ending, in two different background conditions: in quiet and in noise. In addition, the behavioral performance in a SIN perception task was assessed. Results showed that noise significantly impacts the FFR, enhancing responses to the fundamental frequency and improving pitch contour tracking, while simultaneously degrading responses to formants and harmonics and increasing neural transmission delay. Among the FFR measures, only the response to the first formant of /a/ showed some partial association with performance in the SIN perception task. Our results point to the FFR as a reliable biomarker of SIN auditory processing, but limited support as a relevant biomarker of SIN perception, at least among healthy young adults. These results underscore the need for further research to clarify the specific neural mechanisms underlying SIN perception.