Background: Human beings possess an extraordinary ability to adapt their voice quality based on the communicative context, leveraging this vocal flexibility to convey diverse information about the message and the speaker, such as emotions. One of the goals of voice therapy is to achieve this flexibility. Understanding the cognitive demands of adjusting phonatory modes for vocal flexibility is essential for designing therapies that facilitate effective learning and skill acquisition. Toward this goal, this study examined the effect of phonatory and articulatory modes on cognitive load using a dual-task paradigm, the auditory Stroop task.

Methods: Our preliminary participants included ten females, aged 19-23 years, all native speakers of American English with no known voice, speech, or language disorders, participated in the study. They performed an auditory Stroop task in silence, as well as while counting numbers aloud in three distinct phonatory and articulatory modes: habitual, whisper, and clear speech modes. Reaction times and accuracy were measured across different speech styles and congruency conditions. Congruent trials were those where the gender of the spoken word matched the task (e.g., the word "man" spoken by a male voice), while incongruent trials featured mismatches (e.g., the word "man" spoken by a female voice). The effects of speech styles and congruency were tested with repeated measures ANOVA and pairwise t-tests.

Results: Reaction times in congruent trials were significantly longer for both habitual and clear speech compared to silence. In incongruent trials, the only significant difference was between silent and clear speech.

Conclusion: Our preliminary results indicate that producing clear speech imposes a higher cognitive load compared to habitual and whispered speech. Our findings underscore the importance of considering cognitive load when implementing speech modification techniques in voice therapy. Additionally, the complexity of the dual task used to measure cognitive load may also influence detection sensitivity, highlighting the need for carefully designed tasks that accurately reflect the cognitive demands of speech production. Future research should explore why whispering did not significantly increase cognitive load, despite requiring intentional changes in phonatory modes, to further elucidate the complexities of voice modulation and its cognitive implications.