Children are expected to listen and learn in very complex acoustic environments. Nonetheless, little is known about how young listeners are affected by the 'cocktail party' phenomenon. An important component of the cocktail party problem is the advantage created by spatial separation of the target and interfering sounds. One way to measure this effect, known as spatial release from masking (SRM), is to compare speech intelligibility when the target and interferers are coincident in space versus spatially separated. Although SRM in children has been recently demonstrated (Litovsky, 2003; Johnstone and Litovsky, 2003), the acoustic and spatial environments were highly predictable. The proposed work will investigate masking and SRM under more realistic scenarios. Speech intelligibility will be studied when the spatial location of the interferers is fixed and the type of interferers varies probabilistically. It is hypothesized that masking and SRM will increase when informational masking is maximized.
