Abstract Emotional information in speech is conveyed through prosodic cues, including variations in fundamental frequency (F0 contour), intensity, and duration. Emotion perception has been largely studied behaviorally but the neural mechanisms underlying it are not well understood. The Speech-evoked Frequency Following Response (FFR) is a non-invasive neural measure reflecting the encoding of speech acoustics in the auditory system. This study investigated the extent to which the FFR can represent prosody-related F0 contours and compared neural responses between male and female listeners. Sixteen normal-hearing adults underwent FFR recording in response to the word “balloon” spoken with sad and happy emotion by a male and a female talker. Using a pitch-tracking algorithm, F0 tracking accuracy was quantified via Root Mean Squared Error (RMSE) and 5% accuracy. The results showed that the FFR can track emotional F0 contours; however, the degree of accuracy is modulated by emotion type and talker voice characteristics. Enhanced F0 tracking was observed for the sad male voice, followed by the happy male, sad female, and happy female voices. On the other hand, listener sex did not influence the results. These findings provide new insights into the use of FFR as a neural measure for prosody assessment.
Karimi-Boroujeni et al. (Mon,) studied this question.