Suppr超能文献

辨别训练引起的双耳线索重新加权

Binaural-cue reweighting induced by discrimination training.

作者信息

Klingel Maike, Singhal Udbhav, Seitz Aaron R, Kopčo Norbert

机构信息

Acoustics Research Institute, Austrian Academy of Sciences, Vienna, Austria.

Pavol Jozef Šafárik University, Košice, Slovakia.

出版信息

Atten Percept Psychophys. 2025 May 14. doi: 10.3758/s13414-025-03082-x.

Abstract

When localizing sounds, listeners combine the two binaural cues interaural time and level difference (ITD and ILD). The relative weight assigned to each cue is frequency dependent, with ITDs dominating at low and ILDs at high frequencies. However, this weighting changes (e.g., depending on room reverberation or cue reliability). To achieve better spatial hearing in various listener populations, changing the weighting might be advantageous. Previous studies showed that such changes can be induced (e.g., using a lateralization training with visual reinforcement in virtual reality). Here, a new training procedure is introduced, based on a simple auditory-only discrimination task. An experiment evaluated the procedure, consisting of a pretest, three training sessions, and a posttest. Subjects were divided into three groups-one trained by reinforcing the ILDs, one by reinforcing the ITDs, and one no-training control. The training consisted of an adaptive staircase of relative discrimination trials. Stimuli were two consecutive narrow-band noise bursts (2-4 kHz), each presented with a different combination of ITD and ILD. Participants' task was to indicate the perceived location of the second noise burst versus the first. During training, feedback was provided requiring the subject to imagine the sound moving in the trained cue's direction. We observed an increase in reinforced-cue weight for both training groups, but not in the control group, that continued during all three training sessions. Thus, this training method is effective for reweighting in both directions. Moreover, it is individualized, and, since it does not rely on sophisticated equipment, it can be easily accessible for a range of listeners.

摘要

在进行声音定位时,听众会综合利用双耳时间差和强度差这两种双耳线索(ITD和ILD)。分配给每个线索的相对权重取决于频率,低频时ITD起主导作用,高频时ILD起主导作用。然而,这种权重会发生变化(例如,取决于房间混响或线索可靠性)。为了在不同听众群体中实现更好的空间听觉,改变权重可能会有好处。先前的研究表明,可以诱导这种变化(例如,在虚拟现实中使用视觉强化的偏侧化训练)。在此,引入了一种基于简单纯听觉辨别任务的新训练程序。一项实验对该程序进行了评估,实验包括一次预测试、三次训练课程和一次后测试。受试者被分为三组——一组通过强化ILD进行训练,一组通过强化ITD进行训练,一组为无训练对照组。训练由相对辨别试验的自适应阶梯组成。刺激是两个连续的窄带噪声脉冲(2 - 4千赫),每个脉冲呈现不同的ITD和ILD组合。参与者的任务是指出第二个噪声脉冲相对于第一个的感知位置。在训练过程中,会提供反馈,要求受试者想象声音朝着训练线索的方向移动。我们观察到,两个训练组的强化线索权重都有所增加,而对照组没有,这种增加在所有三次训练课程中都持续存在。因此,这种训练方法在两个方向上进行权重调整都是有效的。此外,它是个性化的,并且由于不依赖复杂设备,所以对于一系列听众来说都很容易采用。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验