Increased vocal intensity due to the Lombard effect in speakers with Parkinson's disease: simultaneous laryngeal and respiratory strategies

Abstract

Purpose: The objective of the present study was to investigate whether speakers with hypophonia, secondary to Parkinson's disease (PD), would increases their vocal intensity when speaking in a noisy environment (Lombard effect). The other objective was to examine the underlying laryngeal and respiratory strategies used to increase vocal intensity.

Methods: Thirty-three participants with PD were included for study. Each participant was fitted with the SpeechVive™ device that played multi-talker babble noise into one ear during speech. Using acoustic, aerodynamic and respiratory kinematic techniques, the simultaneous laryngeal and respiratory mechanisms used to regulate vocal intensity were examined.

Results: Significant group results showed that most speakers with PD (26/33) were successful at increasing their vocal intensity when speaking in the condition of multi-talker babble noise. They were able to support their increased vocal intensity and subglottal pressure with combined strategies from both the laryngeal and respiratory mechanisms. Individual speaker analysis indicated that the particular laryngeal and respiratory interactions differed among speakers.

Conclusions: The SpeechVive™ device elicited higher vocal intensities from patients with PD. Speakers used different combinations of laryngeal and respiratory physiologic mechanisms to increase vocal intensity, thus suggesting that disease process does not uniformly affect the speech subsystems.

Learning outcomes: Readers will be able to: (1) identify speech characteristics of people with Parkinson's disease (PD), (2) identify typical respiratory strategies for increasing sound pressure level (SPL), (3) identify typical laryngeal strategies for increasing SPL, (4) define the Lombard effect.

Keywords: Laryngeal; Lombard effect; Parkinson's disease; Respiratory; Vocal intensity.

Figure 1

Schematic of equipment and measurements for laryngeal aerodynamic, respiratory kinematic and acoustic recordings.

Figure 2

Change in Sound Pressure Level (speech-in-noise minus speech-in-quiet) for each participant in the monologue. Bars denote the mean change, lines denote the standard error. Black bars follow the significant condition effect. White bars do not follow the condition effect.

Figure 3

Change in Subglottal Air Pressure and Peak to Peak Glottal Airflow (speech-in-noise minus speech-in-quiet) for each participant in the sentence task. Bars denote the mean change, lines denote the standard error. Black bars follow the significant condition effect. White bars do not follow the condition effect.

Figure 4

Change in Open Quotient and Maximum Flow Declination Rate (speech-in-noise minus speech-in-quiet) for each participant in the sentence task. Bars denote the mean change, lines denote the standard error. Black bars follow the significant condition effect. White bars do not follow the condition effect.

Figure 5

Change in Lung Volume Initiation and Termination (speech-in-noise minus speech-in-quiet) for each participant in the monologue. Bars denote the mean change, lines denote the standard error. Black bars follow the significant condition effect. White bars do not follow the condition effect.

Figure 6

Rib Cage Volume (y-axis) by Abdominal Volume (x-axis) average motion-motion plots for five participants in the monologue. White symbols reflect speech-in-quiet (SQ). Dark symbols reflect speech-in-noise (SN). Larger symbols indicate the mean initiation of utterance. Smaller symbols indicate the mean termination of utterance. Vertical 0 represents rib cage EEL. Horizontal 0 represents abdominal EEL.