Publication Date


Document Type


Committee Members

Douglas Brungart (Committee Member), Robert Gilkey (Advisor), Michael Hennessy (Committee Member), Thomas Moore (Committee Member), Daniel Weber (Committee Member)

Degree Name

Doctor of Philosophy (PhD)


Several studies have shown that detection of a target in a concurrent noise masker improves when the masker onset occurs prior to that of the target (see, e.g., Zwicker, 1965; McFadden, 1966; Yost, 1985). Recent research indicates that this "masker fringe" (i.e., the portion of the masker occurring prior to the target onset) also facilitates sound localization in noise (Simpson, Gilkey, Brungart, Iyer, & Romigh, 2009a; Simpson, Gilkey, Brungart, Iyer, & Hamil, 2009b). However, these studies do not provide a clear indication of what information listeners are exploiting to obtain this improved localization performance. This dissertation was designed to determine the role of masker fringe in sound localization. Three masker configurations were examined: 1) A baseline condition in which the masker and target were pulsed on and off simultaneously (the Pulsed configuration); 2) A condition in which the masker was immediately preceded by a 500-ms masker fringe (the Fringe configuration); and 3) A condition in which a 500-ms silent interval was inserted between the 500-ms masker fringe and the masker (the Gap configuration). In addition, each of these configurations was examined under a low spatial uncertainty condition in which the masker, or fringe-plus-masker, was presented from a fixed and known location throughout a block of trials (the Fixed Masker condition) and a high spatial uncertainty condition in which the location of the masker, or fringe-plus-masker, varied randomly from trial to trial (the Variable Masker condition). The results suggest that listeners make use of at least two types of information in the masker fringe. The first is a spatial cue, which results from the fact that the fringe serves as a spatial "pointer" to the location of the subsequent masker. The benefit of spatial information from the fringe (in the Gap and Fringe configurations) can be as large as 8 dB. However, the spatial information provided by fixing the masker location (the Fixed Masker condition) is even greater (approximately 10 dB). In both cases, the benefit of spatial information appears to be greatest in the Left/Right dimension. The second type of information is related to the onset effects that occur as a result of temporally separating the activity associated with the masker onset and that associated with the target onset. The benefit of this onset effect appears to be approximately 4-6 dB (comparable to the size of the effect found in studies of binaural detection), and is equivalent across spatial dimensions. The relation of these results to previous work on auditory detection, sound localization, spatial attention, and other related phenomena are discussed, and future studies are described that will more fully characterize the role of masker fringe in sound localization.

Page Count


Department or Program

Department of Psychology

Year Degree Awarded