Effects of vocoding and intelligibility on the cerebral response to speech

Abstract

Background: Degrading speech through an electronic synthesis technique called vocoding has been shown to affect cerebral processing of speech in several cortical areas. However, it is not clear whether the effects of speech degradation by vocoding are related to acoustical degradation or by the associated loss in intelligibility. Using vocoding and a parametric variation of the number of frequency bands used for the encoding, we investigated the effects of the degradation of auditory spectral content on cerebral processing of intelligible speech (words), unintelligible speech (words in a foreign language), and complex environmental sounds.

Results: Vocoding was found to decrease activity to a comparable degree for intelligible and unintelligible speech in most of the temporal lobe. Only the bilateral posterior temporal areas showed a significant interaction between vocoding and intelligibility, with a stronger vocoding-induced decrease in activity for intelligible speech. Comparisons to responses elicited by environmental sounds showed that portions of the temporal voice areas (TVA) retained their greater responses to voice even under adverse listening conditions. The recruitment of specific networks in temporal regions during exposure to degraded speech follows a radial and anterior-posterior topography compared to the networks recruited by exposure to speech that is not degraded.

Conclusions: Different brain networks are involved in vocoded sound processing of intelligible speech, unintelligible speech, and non-vocal sounds. The greatest differences are between speech and environmental sounds, which could be related to the distinctive temporal structure of speech sounds.

Figure 1

Behavioral performance as a function of vocoder degradation.

Figure 2

Effect of stimulus Category. A. Main effect of stimulus category. Vocoder degraded speech (degraded IS and US) vs. vocoder degraded E (IS+US vs. 2E). The horizontal slice is at the coordinate z = -3 (p < 0.0001). IS-Intelligible Speech, US-Unintelligible Speech, ES-Environmental Sounds. In each group, the darker the color in the bar graphs of BOLD signal variation, the higher the level of vocoder degradation. In blue, areas from TVA localizer are presented at p < 0.05, FWE corrected. B. Degradation effect for speech vs. degradation effect for environmental sounds as defined by the expression (Speech_degraded-Speech_normal)-(Sounds_degraded-Sounds_normal). It can be algebraically rewritten as (Sounds_normal-Sounds_degraded)-(Speech_normal-Speech_degraded). The latter formula makes clear that the difference between normal and degraded stimuli in the right STS is smaller for speech compared with non-speech sounds. In each group, the darker the color in the bar graphs of BOLD signal variation, the higher the level of vocoder degradation.

Figure 3

Effect of vocoder degradation on BOLD signal variation. A. General regression with vocoder degradation. Plots are presented for the peaks in the left and right temporal regions: L STG/STS at (-57 -15 6) and R STG/STS at (60 -24 -3), the horizontal slice at the coordinate z = -3 (p < 0.0001). IS-Intelligible Speech, US-Unintelligible Speech, ES-Environmental Sounds. In each group, the darker the color, the higher the level of vocoder degradation. The level of degradation in this analysis was transformed to the logarithmic scale being the logarithm of the inverse number of channels n: log(1/n). In each group, the darker the color in the bar graphs of BOLD signal variation, the higher the level of vocoder degradation. B. Interaction between Category and Degradation (factorial analysis of IS, US and ES, contrast weights restricted to IS and US) in the regions of interest from the "All effects" contrast. Plots are presented for the peaks in the left and right temporal regions: L STG at (-57-18 15) and R STG at (57-27 15), the horizontal slice at the coordinate z = 15. In each group, the darker the color in the bar graphs of BOLD signal variation, the higher the level of vocoder degradation.