PAPERS W. Howie, D. Martin, S. Kim, T. Kamekawa and R. King, “Effect of Audio Production Experience, Musical Training, and Age on Listener Performance in 3D Audio Evaluation” J. Audio Eng. Soc., vol. 67, no. 10, pp. 782–794, (2019 October.). DOI: https://doi.org/10.17743/jaes.2019.0031 Effect of Audio Production Experience, Musical Training, and Age on Listener Performance in 3D Audio Evaluation WILL HOWIE, 1 AES Associate Member, DENIS MARTIN, 2 AES Student Member, SUNGYOUNG KIM, 3 AES Member, TORU KAMEKAWA, 4 AES Member, AND RICHARD KING, 2 AES Member 1 CBC/Radio-Canada, Vancouver, Canada 2 The Graduate Program in Sound Recording, McGill University, Montr´ eal, Quebec, Canada 3 College of Engineering Technology, Rochester Institute of Technology, Rochester, NY, USA 4 Department of Musical Creativity and the Environment, Tokyo University of the Arts, Tokyo, Japan A study was undertaken to determine the effect of audio production experience, musical training and technical ear training, age, and previous experience listening to 3D music record- ings on listener performance within the context of 3D audio evaluation. Subjects ranging from highly experienced to na¨ ıve listeners evaluated an excerpt of orchestral music captured by three different 3D music recording techniques. By combining the results with those of a previous study, it was found that both audio production experience and musical training are signifi- cant predictors of listener performance and that listener performance increases with age until 36 years. 0 INTRODUCTION Well executed listening tests often require a great deal of time and resources for the creation or acquisition of ap- propriate stimuli, design of a testing interface, selection of subjects, and implementation of the experiment in an environment that is acoustically and technologically ap- propriate. Previous research discussing the selection and training of listening test subjects has primarily taken place within the context of 2-ch stereo audio reproduction [1–5]. This study attempts to uncover the relative importance of audio production experience, musical training, and age for the purpose of pre-screening and selection of listening test subjects for 3D audio evaluation. 0.1 Experienced, Trained, and Practiced Listeners for Audio Evaluation Bech [3] suggests that classification of listening test sub- jects as “experienced” should be based on a number of fac- tors, including familiarity with live musical performance as a performer or audience member, experience in critical listening, and an aptitude for “detecting sonic differences in reproduced sound. [3]” Experience in critical listening is also seen as a primary factor for choosing “expert” listen- ers in earlier studies by Toole [2], Gabrielsson and Sj¨ ogren [6], and Killian and Tillman [7]. These studies all provide evidence that results obtained from experienced listeners are more consistent than those of untrained or “na¨ ıve” lis- teners, i.e., ranking or preference data from these listen- ers has less error variance than that of less experienced listeners. Greater consistency of subject ratings leads to greater statistical power within the data requiring less sub- jects to achieve meaningful results. ITU-R BS.1116 [8] rec- ommends the use of “expert” listeners as testing subjects and recommends selection be based on “previous experi- ence and performance in previous tests.” “Trained” listeners are those who have been trained specifically for the task of performing listening tests. Stud- ies by Olive [4] and Schinkel-Bielefeld [9] found that a greater number of na¨ ıve listeners will be needed to ob- tain results with the same statistical power as compared with trained listeners. Bech found that up to seven times as many untrained subjects were needed to produce the same confidence interval possible as when using highly trained 782 J. Audio Eng. Soc., Vol. 67, No. 10, 2019 October