“Do I sound happy?” Acoustic characteristics of affective prosody in older adults

Abstract

It is known that older adults have difficulty comprehending affective prosody. However, it is unclear how well they manipulate affective prosody to express emotion. Five older adults were recorded when they completed three speech tasks: (1) Talking about happy and sad life events; (2) Describing the Refused Umbrella pictures; (3) Explaining how to make a peanut butter and jelly sandwich. Two independent investigators listened to the recordings and parsed them into utterances considering pauses and connectivity. Afterwards, the investigators identified the primary emotion (i.e., happy, sad, neutral) in each utterance. The recordings were acoustically analyzed using Pratt for rate, pitch, and amplitude measures. The rate measures included speech and articulation rates and the pitch/amplitude measures contained initial, final, minimum, maximum f0/dB and the difference between the maximum and minimum f0/dB. Multivariate ANOVAs with Tukey post-hoc test were conducted to analyze the acoustic data. Among the acoustic measures, all rate and pitch measures were significantly different across the three emotions. However, none of the amplitude measures differed significantly. Research on production of affective prosody in healthy older adults has been scarce and this study may serve as a catalyst for further exploration in this area.