Neural encoding of sound duration persists in older adults Bernhard Ross a,b, ⁎, Joel S. Snyder c , Meaghan Aalto a , Kelly L. McDonald a , Benjamin J. Dyson d , Bruce Schneider e , Claude Alain a,e a Rotman Research Institute, Baycrest Centre, Toronto, Ontario, Canada M6A 2E1 b Department of Medical Biophysics, University of Toronto, Ontario, Canada M5G 2M9 c Department of Psychology, University of Nevada, Las Vegas, NV 89154, USA d Department of Psychology, Ryerson University, Toronto, Ontario, Canada M5B 2K3 e Department of Psychology, University of Toronto, Ontario, Canada M8V 2S4 abstract article info Article history: Received 4 February 2009 Revised 9 April 2009 Accepted 10 April 2009 Available online 22 April 2009 Speech perception depends strongly on precise encoding of the temporal structure of sound. Although behavioural studies suggest that communication problems experienced by older adults may entail deficits in temporal acuity, much is unknown about the effects of age on the neural mechanisms underlying the encoding of sound duration. In this study, we measured neuromagnetic auditory evoked responses in young, middle-aged and older healthy participants listening to sounds of various durations. The time courses of cortical activity from bilateral sources in superior temporal planes showed specific differences related to the sound offsets indicating the neural representation of onset and offset markers as one dimension of the neural code for sound duration. Model free MEG source analysis identified brain areas specifically responding with an increase in activity to increases in sound duration in the left anterior insula, right inferior frontal, right middle temporal, and right post-central gyri in addition to bilateral supra-temporal gyri. Sound duration- related changes in cortical responses were comparable in all three age groups despite age-related changes in absolute response magnitudes. The results demonstrated that early cortical encoding of the temporal structure of sound presented in silence is little or not affected by normal aging. © 2009 Elsevier Inc. All rights reserved. Introduction The ability to perceive the duration of sound is important for understanding speech. For instance, in many languages differences in the duration of a vowel on the order of tens of milliseconds may distinct between words of different meaning. Although it is well recognized that speech perception depends on the processing of the fine temporal structure of sounds (e.g., envelope), little is known about how sound duration itself is encoded in human cortex. Studies measuring neuroelectric brain responses to changes in sound duration have been shown as mismatch negativity (MMN) in auditory evoked potentials (AEP) and most importantly the size of the response determined the accuracy of perceptual performance (Amenedo and Escera, 2000). Other studies have shown that the amplitude of the N1 wave, in EEG a negative deflection with maximum at vertex electrodes, peaking at about 100 ms after sound onset, increases as sound duration increases up to 40 ms (Alain et al., 1997; Forss et al., 1993; Gage and Roberts, 2000; Joutsiniemi et al., 1989; Onishi and Davis, 1968). These observations support the hypothesis that neurons can act as a linear integrator of the acoustical energy with their output being proportional to the sound duration. Furthermore, neurons tuned to a specific sound duration between 30 and 300 ms have been found in auditory cortex of the cat (He et al., 1997). In human, Alain et al. (1997) showed that the N1c subcomponent of the N1, recorded at mid temporal electrode sites with peak latency of about 130 ms, increased strongly with duration increase up to 24 ms, suggesting a short integration time and triggering by the sound onset. In contrast, the amplitude of the N1a subcomponent, recorded from same electrodes with peak latency around 90 ms, increased more for longer sound durations, indicating integration of ongoing sound. In addition to integration mechanisms complex networks have been proposed in which the onset and offset of sound is marked and a counting mechanism estimates how many time units elapsed between both markers (Creelman, 1962). The integrator model would appear to be suitable for short sound durations only and unsuitable across longer time intervals that become more and more vulnerable to noise and stimulus intensity. In these cases the latter stopwatch-like mechanism would be more effective. However, in principle, the marking of sound onsets and offsets could serve as an important encoding principle even for short tone durations. Related onset and offset markers have been identified in the auditory thalamus of the guinea pig for sound duration as short as 50 ms (He, 2001). NeuroImage 47 (2009) 678–687 ⁎ Corresponding author. Baycrest Centre, Rotman Research Institute, 3560 Bathurst Street, Toronto, Ontario, Canada M6A 2E1. Fax: +1417 785 2862. E-mail address: bross@rotman-baycrest.on.ca (B. Ross). 1053-8119/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2009.04.051 Contents lists available at ScienceDirect NeuroImage journal homepage: www.elsevier.com/locate/ynimg