Localization in Real and Virtual Rooms

I n t r o d u c t i o n Elderly listeners experience more difficulty understanding spoken language in non-ideal lis ten ing conditions than do younger adults. Specifically, they have more trouble in many everyday listening situations where background noise or reverberation conditions are unfavourable. Despite their difficulty understanding what is said in non-ideal conditions, elderly listeners often perform like younger adults in ideal listening conditions, such as when they are talking to one familiar person in a small, quiet room. Furthermore, many do not have clinically significant elevations in pure-tone thresholds, and existing clinical tests conducted in the artificial conditions of soundbooths are not useful in predicting an individual’s performance in real-world com munication situations. Therefore, it is im portant to devise new methods to allow researchers and clinicians to better evaluate how listeners perform in non-ideal, real-world conditions. Testing listeners in actual acoustic conditions would be the most ecologically valid approach; however, precise control of the test stimuli would be jeopardized. A uralization[l], or the simulation of acoustic environments, is another approach that permits more realistic conditions to be created while maintaining precise control of test stimuli. As a First step in adopting the latter approach, we tested and compared the abilities of listeners to localize speech signals in a real room and in simulations of the same room