On the significance of speech pauses in depressive disorders: Results on read and spontaneous narratives

This paper investigates whether and how depressive disorders affect speech and in particular timing strategies for speech pauses (empty and filled pauses, as well as, phoneme lengthening). The investigation is made exploiting read and spontaneous narratives. The collected data are from 24 subjects, divided into two groups (depressed and control) asked to read a tale, as well as, spontaneously report on their daily activities. Ten different frequency and duration measures for pauses and clauses are proposed and have been collected using the PRAAT software on the speech recordings produced by the participants. A T-Student test for independent samples was applied on the collected frequency and duration measures in order to ascertain whether significant differences between healthy and depressed speech measures are observed. In the “spontaneous narrative” condition, depressed patients exhibited significant differences in: the average duration of their empty pauses, the average frequency, and the average duration of their clauses. In the read narratives, only the average pause’s frequency of the clauses was significantly lower in the depressed subjects with respect to the healthy ones. The results suggest that depressive disorders affect speech quality and speech production through pause and clause durations, as well as, clause quantities. In particular, the significant differences in clause quantities (observed both in the read and spontaneous narratives), suggest a strong general effect of depressive symptoms on cognitive and psychomotor functions. Depressive symptoms produce changes in the planned timing of pauses, even when reading, modifying the timing of pausing strategies.