Tinnitus affects the relative roles of semantics and prosody in the perception of emotions in spoken language

Abstract

Objective: Understanding communication difficulties related to tinnitus, by identifying tinnitus-related differences in the perception of spoken emotions, focussing on the roles of semantics (words), prosody (tone of speech) and their interaction.

Study sample and design: Twenty-two people-with-tinnitus (PwT) and 24 people-without-tinnitus (PnT) listened to spoken sentences made of different combinations of four discrete emotions (anger, happiness, sadness, neutral) presented in the prosody and semantics (Test for Rating Emotions in Speech). In separate blocks, listeners were asked to attend to the sentence as a whole, integrating both speech channels (gauging integration), or to focus on one channel only (gauging identification and selective attention). Their task was to rate how much they agree the sentence conveys each of the predefined emotions.

Results: Both groups identified emotions similarly, and performed with similar failures of selective attention. Group differences were found in the integration of channels. PnT showed a bias towards prosody, whereas PwT weighed both channels equally.

Conclusions: Tinnitus appears to impact the integration of the prosodic and semantic channels. Three possible sources are suggested: (a) sensory: tinnitus may reduce prosodic cues. (b) Cognitive: tinnitus-related reduction in cognitive processing. (c) Affective: group differences were related to the existence of tinnitus, but not to the extent of tinnitus complaints and/or affective symptoms.

Keywords:

• Tinnitus
• speech
• speech perception
• emotion
• cognition

Acknowledgments

The authors would like to thank Ms. Maya Mentzel for her assistance on this project.

Declaration of interest

The authors declare that they have no conflict of interest and no interest to declare.

Notes

a Note, the original T-RES included the fear emotion as well. However, to shorten the test, we removed this emotion, as it was found to be the least reliable in a previous study (Ben-David et al., 2016).

Additional information

Funding

Research by the last author was partially supported by a research grant (I-1324-105.4/2015) from the German-Israeli Foundation for Scientific Research and Development.