Obituary – In Memory of Professor Ray Meddis (1944-2018)

Guy Brown and Wendy Lecluyse

It is with great sadness that the Society reports the passing of Ray Meddis, Emeritus Professor of Psychology at the University of Essex, who died suddenly at his home in Wivenhoe on 25th November 2018.

Ray was born in Sunderland on 23rd July 1944. He studied psychology at the University of London, before taking up a lectureship at Bedford College. His early work was on sleep, leading to the publication of a book in 1977 (“The Sleep Instinct”) which challenged the established notion that sleep has a recovery function. He also published a number of introductory books on statistical methods, which were used in numerous undergraduate psychology classes. Ray’s clarity of thought, and skill at explaining complex ideas to others, was very evident in these early works and remained a hallmark of his writing throughout his career.

Ray’s interests turned to hearing and speech science following his move to the Department of Human Sciences at Loughborough University, where he was Director of the Speech and Hearing Laboratory for a decade. During this time, he became a strong advocate of the use of computer models as a means of understanding auditory function. His elegant model of the inner hair cell – auditory nerve synapse (1986) became widely adopted, not least because Ray presented his work at interdisciplinary meetings (such as the Institute of Acoustics conferences at Windermere) which were a melting pot of speech scientists, audiologists, acousticians and engineers. As a result, Ray’s work had considerable impact on academics building computer systems for automatic speech recognition and sound separation, as well as those studying the perception of speech and auditory function. His subsequent work at Loughborough focused on models of the auditory periphery and brainstem, including a correlogram-based model of pitch perception (1991). This important paper demonstrated that a computational implementation of Licklider’s theory of pitch, based on across-frequency pooling of a time interval analysis, could account for many classical pitch phenomena including the missing fundamental, pitch of interrupted noise, and sensitivity to phase relationships between the components of a harmonic complex.

In 1996, Ray joined the Department of Psychology at the University of Essex and almost immediately became Head of Department. Under his leadership, the department developed as an internationally recognised research centre for psychology. He continued to make broad contributions to hearing science, continuously developing his models of the auditory periphery. His work during this period included an influential dual-resonance nonlinear (DRNL) model of cochlear filtering (2001), a model of auditory absolute threshold (2006) and a computational model of auditory efferent suppression (2007). The former was novel in its use of parallel linear and nonlinear pathways to model important characteristics of the cochlear response, achieving a close match to human psychophysical data. This work opened up the possibility to develop personalised models of hearing impairment. As a result, Ray’s work increasingly focused on the mechanisms, assessment and treatment of hearing disorders, leading to the concept of a ‘hearing dummy’ (2014) – a computer model of hearing impairment that could be rapidly fitted to an individual and then used ‘offline’ to optimise a hearing aid algorithm.

Building on insights from his computer models, including the DRNL and a revised model of auditory efferent suppression that included frequency-dependent feedback (2012), Ray’s team at Essex developed a novel hearing aid algorithm. This sought to compensate for two key components that can be absent in hearing loss, namely instantaneous compression of the basilar membrane response and a slower efferent suppression of basilar membrane activity driven by changes in environmental sound levels. The hearing aid algorithm was implemented in a smartphone app (BioAid) and was distributed free of charge. It has been downloaded many thousands of times, and was shortlisted for the 2014 Times Higher Education Awards in the category of Outstanding Contribution to Innovation and Technology.

Ray retired from the University of Essex in 2011 but remained active as Emeritus Professor. He firmly believed that his work should be reproducible, and encouraged others to use his program code, verify it and extend it. Indeed, one of the reasons that his work gained such significant impact is that he freely published the program code for his auditory models (starting back in 1990, when the Journal of the Acoustical Society of America took the unusual step of publishing the code for his hair cell model in a technical note). Ray’s Matlab Auditory Periphery (MAP) model was still actively being developed at the time of his death, and a community effort is underway to ensure that this important tool for hearing research remains widely available.

Those that had the honour of working with Ray will speak of his kindness, wit and generosity. He was an inspirational mentor, taking particular delight in developing the careers of his PhD students and other early career scientists. He was quick to identify and challenge lazy assumptions, even if that meant rattling a few cages; tellingly, in the preface to his 1977 book, he writes that “swimming against the scientific mainstream is, at the same time, thrilling and frustrating”. He leaves behind a significant body of work – nearly 150 scientific publications – that has inspired a generation of hearing scientists and audio engineers.

Outside of academic life, Ray divided his time between good food, sailing and his family. Post-retirement he continued to travel widely, visiting colleagues and friends in laboratories across Europe and North America. Sadly, Ray’s youngest son Christopher died in 2016. He is survived by his wife Valerie, their eldest son, William, and his sister Vera.

Selected publications

  1. Meddis (1986) Simulation of mechanical to neural transduction in the auditory receptor. Journal of the Acoustical Society of America 79 (3), 702-711.

 

  1. Meddis and M. J. Hewitt (1991) Virtual pitch and phase sensitivity of a computer model of the auditory periphery. I: Pitch identification. Journal of the Acoustical Society of America 89 (6), 2866-2882.

 

  1. A. Lopez-Poveda and R. Meddis (2001) A human nonlinear cochlear filterbank. Journal of the Acoustical Society of America 110 (6), 3107-3118.

 

  1. Meddis (2006) Auditory-nerve first-spike latency and auditory absolute threshold: a computer model. Journal of the Acoustical Society of America 119 (1), 406-417.

 

  1. T. Ferry and R. Meddis (2007) A computer model of medial efferent suppression in the mammalian auditory system. Journal of the Acoustical Society of America 122 (6), 3519-3526.

 

  1. R. Clark, G. J. Brown, T. Jürgens and R. Meddis (2012) A frequency-selective feedback model of auditory efferent suppression and its implications for the recognition of speech in noise. Journal of the Acoustical Society of America 132 (3), 1535-1541.

 

  1. R. Panda, W. Lecluyse, C. M. Tan, T. Jürgens and R Meddis (2014) Hearing dummies: Individualized computer models of hearing impairment. International journal of audiology 53 (10), 699-709.