Temporal jitter disrupts speech intelligibility: A simulation of auditory aging

• Published in: Hearing research Vol. 223; no. 1; pp. 114 - 121
• Main Authors: Pichora-Fuller, M. Kathleen, Schneider, Bruce A., MacDonald, Ewen, Pass, Hollis E., Brown, Sasha
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 2007; Elsevier
• Subjects: Acoustic Stimulation; Adult; Aging; Aging - physiology; Aging - psychology; Biological and medical sciences; Ear and associated structures. Auditory pathways and centers. Hearing. Vocal organ. Phonation. Sound production. Echolocation; Ear, auditive nerve, cochleovestibular tract, facial nerve: diseases, semeiology; Fundamental and applied biological sciences. Psychology; Hearing Loss - physiopathology; Hearing Loss - psychology; Humans; Medical sciences; Models, Psychological; Neural synchrony; Noise; Non tumoral diseases; Otorhinolaryngology. Stomatology; Speech intelligibility; Speech Intelligibility - physiology; Temporal jitter; Vertebrates: nervous system and sense organs; Word identification; DL; IFFT; M; SD; SNK; ANOVA; BW; FFT; Word identification; Aging; dB HL; dB SPL; kHz; Neural synchrony; D-to-A; ms; S:N; Speech intelligibility; STEP; SPIN-R; Hz; dB S:N; RMS; Temporal jitter; DC; Noise; Internal ear disease; Auditory disorder; Alteration; Distortion; Low frequency; Long term; Hearing; ENT disease; Perception hearing loss; Speech; Sensory hearing loss
• ISSN: 0378-5955; 1878-5891
• DOI: 10.1016/j.heares.2006.10.009

We disrupted periodicity cues by temporally jittering the speech signal to explore how such distortion might affect word identification. Jittering distorts the fine structure of the speech signal with negligible alteration of either its long-term spectral or amplitude envelope characteristics. In Experiment 1, word identification in noise was significantly reduced in young, normal-hearing adults when sentences were temporally jittered at frequencies below 1.2 kHz. The accuracy of the younger adults in identifying jittered speech in noise was similar to that found previously for older adults with good audiograms when they listened to intact speech in noise. In Experiment 2, to rule out the possibility that the reductions in word identification were due to spectral distortion, we also tested a simulation of cochlear hearing loss that produced spectral distortion equivalent to that produced by jittering, but this simulation had significantly less temporal distortion than was produced by jittering. There was no significant reduction in the accuracy of word identification when only the frequency region below 1.2 kHz was spectrally distorted. Hence, it is the temporal distortion rather than the spectral distortion of the low-frequency components that disrupts word identification.