ACQ Vol 10 No 2 2008

Work– l i f e balance : preserv i ng your soul

O utside the S quare The voice as a behavioural probe of emotional/neurophysiological disorders Adam Vogel

S peech pathologists commonly use the voice as a marker of the clinical disease state. In the case of Parkinson’s disease, for example, imprecise consonant and vowel production combined with changes in pitch variability often leads to a diagnosis of dysarthria. Moreover, changes in these speech characteristics tell us something about the progression of the disease. In a similar way, the voice has demonstrated its potential as a marker of central nervous

Adam Vogel

system functioning in several populations not considered part of our core business. The influence of depression, anxiety or fatigue on an individual’s functioning is difficult to quantitatively capture using existing psychometric assessments, as examination relies on a combination of subjective clinician/patient report and neuropsychological assessments. As neuropsychological tests have provided equivocal results in central nervous sys­ tem disorders that contain intrinsic emotional changes, the voice has been considered as an objective and non-invasive alternative. It is clear that the voice has strong face validity as a qualitative marker of neurophysiological functioning. Patients with depression can be recognised by their reduced rate of speech and diminished pitch variation. A similar vocal pattern manifests in populations undergoing extended periods of sustained wakefulness. These observable clinical disturbances in motor functioning combine with cognitive and emotional disturbances to provide the assessor with a psychopathological profile that reflects changes in the central nervous system. Aside from instrumental investigations, the majority of clinical evaluations continue to rely on subjective patient/ clinician report to determine the type and level of impairment. Perceptual or listener-based analysis of vocal changes related to emotion and physicality are important in the diagnosis and evaluation of pathological conditions. However, perceptual measurements are subjective, and have inherently poor intra- and inter-rater reliability. Having quantitative information about changes in the voice acoustic profile of a patient or participant can contribute to the accuracy of current subjective assessment protocols. Although the rationale for using the voice as a marker of clinical change has been established, capturing these changes on a large scale is challenging. Historically, voice studies have involved small sample sizes and idiosyncratic voice acquisition hardware/software configurations that lack utility and are labour intensive. This process has intrinsically higher costs related to personnel and equipment requirements. Further­ more, commercially available software and hardware designed to collect and analyse data can be cumbersome and com­ plicated, often requiring extensive user expertise, which can further drain the financial resources of a clinical trial or study. In this context, easy to use voice recording procedures and automated analysis needs to be developed and validated. The application of fully automated, fast and accurate voice acoustic regimes has the potential to extend voice assessment beyond speech pathology to a wider clinical and commercial audience. For example, the voice could be used as a marker of clinical change in pharmaceutical trials for depression, or, as it has in the past, as an indicator disease state in a pathological population. This stream of research offers a number of opportunities not previously available to speech pathologists through unique collaborations with big business and through

Adam Vogel completed a BA (Psychology) in 2000 and Masters of Speech Pathology Studies in 2003 at the University of Queensland. He spent the first few years after graduation working in London within the Neuro­ disability Service at Great Ormond Street Hospital. Since returning to Australia, Adam has been working as a speech pathologist and researcher in the Friedreich Ataxia Clinic at the Monash Medical Centre and as a clinical scientist for CogState Limited. He is currently completing a PhD at the Centre for Neuroscience, University of Melbourne. Within this framework, careful analysis of the voice can be considered a behavioural probe of emotional/neuro­ physiological disorders, with potentially greater sensitivity than existing neuropsychological approaches. the exploration of populations not typically under the care of our profession.

Correspondence to: Adam Vogel CogState Limited

Level 7, 21 Victoria Street, Melbourne, Victoria, 3000 email: avogel@cogstate.com

60

S peech P athology A ustralia