Auditory Accommodation to Poorly Matched Non-Individual Spectral Localization Cues Through Active Learning

Abstract

This study examines the effect of adaptation to non-ideal auditory localization cues represented by the Head-Related Transfer Function (HRTF) and the retention of training for up to three months after the last session. Continuing from a previous study on rapid non-individual HRTF learning, subjects using non-individual HRTFs were tested alongside control subjects using their own measured HRTFs. Perceptually worst-rated non-individual HRTFs were chosen to represent the worst-case scenario in practice and to allow for maximum potential for improvement. The methodology consisted of a training game and a localization test to evaluate performance carried out over 10 sessions. Sessions 1–4 occurred at 1 week intervals, performed by all subjects. During initial sessions, subjects showed improvement in localization performance for polar error. Following this, half of the subjects stopped the training game element, continuing with only the localization task. The group that continued to train showed improvement, with 3 of 8 subjects achieving group mean polar errors comparable to the control group. The majority of the group that stopped the training game retained their performance attained at the end of session 4. In general, adaptation was found to be quite subject dependent, highlighting the limits of HRTF adaptation in the case of poor HRTF matches. No identifier to predict learning ability was observed.

Re-weighting of Sound Localization Cues by Audiovisual Training

“…Because the virtual auditory space stimuli were presented using non-individualized HRTFs, we would expect the listeners’ localization judgments to be less accurate than if the stimuli had been based on measurements from their own ears (Wenzel et al, 1993; Middlebrooks, 1999; Mrsic-Flogel et al, 2001). Our results confirm findings from previous work showing that training with spatio-temporally congruent visual and auditory cues in listeners with abnormal or non-individualized auditory inputs produces improvements in sound localization (Isaiah et al, 2014; Trapeau and Schönwiesner, 2015; Stitt et al, 2019), and that these improvements can occur following limited exposure to multisensory inputs (Berger et al, 2018).…”

Short-term effects of sound localization training in virtual reality

“…Typically, such adaptation has been measured over the course of hours, days and weeks (for a review see Mendonça et al 19 ). In agreement with the few studies that have also focused on short-term changes, we show that significant changes occur after a total training duration of the order of an hour spaced over three days 20,[22][23][24]37 . This is an encouraging result if future systems using generic, non-individualized HRTFs are to find broader application, since it seems likely only a minority of use cases will justify longer adaptation periods.…”

“…All participants underwent localization training. Despite participants undergoing training of a similar total duration (10 sessions of 12 minutes), Stitt et al 37 report larger changes on average in PAE than observed here or in the study described above 18 . One participant group showed a 15 • improvement on PAE, using an identical measure.…”

“…Most notably, they used real, rather than virtual, sound sources and were interested in changes over a considerably longer timescale. Stitt et al 37 employed a more similar approach, whereby the ability of participants to localize virtual sounds, spatialized using non-individualized HRTFs, was trained and tested. Their study also included a group using individualized HRTFs.…”

Re-weighting of Sound Localization Cues by Audiovisual Training

“…Because the VAS stimuli were presented using non-individualized HRTFs, we would expect the listeners’ localization judgments to be less accurate than if the stimuli had been based on measurements from their own ears (Wenzel et al, 1993; Middlebrooks, 1999; Mrsic-Flogel et al, 2001). Other studies have shown that human listeners can, at least to some extent, learn to accommodate the inappropriate auditory spatial cues provided by non-individualized HRTFs (Zahorik et al, 2006; Mendonça et al, 2013; Berger et al, 2018; Stitt et al, 2019). Although the participants in our study were told to ignore the sounds, some subjects in the AV− group, who did not have the benefit of spatially informative visual cues, showed a reduction in localization bias, particularly if the bias was relatively high prior to training.…”