Development of a Model of Multisource Sound Localization

Report No. ARL-TR-5223
Authors: Paula P. Henry and Justin A. MacDonald
Date/Pages: June 2010; 20 pages
Abstract: The time needed to test the effects of sequential changes in head-borne equipment on auditory localization in humans can be extensive. Rather than test the effects of these changes on humans, a model could be used to predict human performance. This report summarizes work that was completed as part of a Director's Research Initiative to enhance an existing model of basic human auditory localization. Using human performance data and head-related transfer functions (HRTFs) collected from listeners, the model was enhanced to identify the location of two sounds that were presented to listeners simultaneously from different locations. Following adaptations to the model, predictions were made based on HRTFs collected in the Knowles Electronics Manikin for Acoustic Research (KEMAR). The model was able to replicate the pattern of human performance results to a moderate extent. Error rates were generally comparable between the experimental and simulated data, and the model accurately predicted the effect of headgear on corrected error.
Distribution: Approved for public release
  Download Report ( 0.393 MBytes )
If you are visually impaired or need a physical copy of this report, please visit and contact DTIC.
 

Last Update / Reviewed: June 1, 2010