Probing auditory scene analysis

Deike, Susann; Denham, Susan; Sussman, Elyse

doi:10.3389/fnins.2014.00293

Cited by 3 publications

(2 citation statements)

References 23 publications

(25 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Neural hyperactivity alone might not be sufficient to generate a persistent tinnitus, but in conjunction with unbalanced lateral inhibition components, it might be able to evoke a salient percept, as a coherent input can effectively inhibit less coherent competing inputs via GABA mediated inhibitory interneurons [6]. Cortical top-down projections to thalamic areas, i.e., attentional feedback loops, and even to hierarchically lower stages of sensory processing, might promote synchrony in excess of the bottom-up self-organization dynamics by a further increase of neural response reliability and cross fiber synchrony [2], [8], [9]. The pathogenic chronification of auditory phantom perceptions might be coupled to plastic processes in areas responsible for the allocation of attention resources and may thus originate in brain areas only indirectly involved in auditory processing, such as limbic structures [10].…”

Section: Introductionmentioning

confidence: 99%

Implementation and Long-Term Evaluation of a Hearing Aid Supported Tinnitus Treatment Using Notched Environmental Sounds

Haab

Lehser

Corona-Strauss³

et al. 2019

IEEE J. Transl. Eng. Health Med.

View full text Add to dashboard Cite

Recent work has shown that sharp spectral edges in acoustic stimuli might have advantageous effects in the treatment of tonal tinnitus. In the course of this paper, we evaluate the long-term effects of spectrally notched hearing aids on the subjective tinnitus distress. By merging recent experimental work with a computational tinnitus model, we modified the commercially available behind-the-ear hearing aids so that a frequency band of 0.5 octaves, centered on the patient’s individual tinnitus frequency, was blocked out. Those hearing aids employ a steep notch filter that filters environmental sounds to suppress the tinnitus-related changes in neural firing by lateral inhibition. The computational model reveals a renormalization of pathologically increased neural response reliability and synchrony in response to spectrally modified input. The target group, fitted with spectrally notched hearing aids, was matched with a comparable control group, fitted with standard hearing aids of the same type but without a notch filter. We analyze the subjective self-assessment by tinnitus questionnaires, and we monitor the objective distress correlates in auditory evoked response phase data. Both, subjective and objective results show a noticeable trend of a larger therapeutic benefit for notched hearing correction.

show abstract

Section: Introductionmentioning

confidence: 99%

Implementation and Long-Term Evaluation of a Hearing Aid Supported Tinnitus Treatment Using Notched Environmental Sounds

Haab

Lehser

Corona-Strauss³

et al. 2019

IEEE J. Transl. Eng. Health Med.

View full text Add to dashboard Cite

show abstract

“…Because of this, we have lacked a comprehensive account of key phenomena in auditory scene analysis, despite many proposed conceptual approaches (43). It is also unclear whether putative principles for auditory perceptual organization which apply to synthetic sounds could also extend to explaining auditory scene analysis in everyday sounds (44).…”

Section: Introductionmentioning

confidence: 99%

Listening with generative models

Cusimano

Hewitt

McDermott

2023

Preprint

View full text Add to dashboard Cite

Perception has long been envisioned to use an internal model of the world to infer the causes of sensory signals. However, tests of inferential accounts of perception have been limited by computational intractability, as inference requires searching through complex hypothesis spaces. Here we revisit the idea of perception as inference in a world model, using auditory scene analysis as a case study. We applied contemporary computational tools to enable Bayesian inference in a structured generative model of auditory scenes. Model inferences accounted for many classic illusions. Unlike most previous accounts of auditory illusions, our model can be evaluated on any sound, and exhibited human-like perceptual organization for real-world sound mixtures. The combination of stimulus-computability and interpretable structure enable 'rich falsification', revealing additional assumptions about sound generation needed to explain perception. The results show how a single generative theory can account for the perception of both classic illusions and everyday sensory signals.

show abstract