A Neural Model of Auditory Space Compatible with Human Perception under Simulated Echoic Conditions

Area of Science:

• Auditory Neuroscience
• Psychoacoustics
• Computational Auditory Neuroscience

Background:

• Auditory scenes involve complex sound summation, leading to fluctuating spatial cues.
• Current models of spatial encoding often overlook these fluctuations, requiring additional mechanisms.
• The precedence or Haas effect, where direct sound and reflections interact, highlights cue instability.

Purpose of the Study:

• To test a novel hypothesis where spatial perception relates to intermediate neural firing probabilities.
• To investigate how the Haas effect influences spatial cue representation and perception.
• To explore the conditions under which auditory objects acquire perceptual 'edges'.

Main Methods:

• Examined the precedence or Haas effect as a model for fluctuating spatial cues.
• Proposed a hypothesis linking spatial perception to sub-maximal firing probabilities in spatially selective neurons.
• Analyzed how interaural time differences (ITDs) are represented under varying acoustic conditions.

Main Results:

• Subjective spatial experiences appear to correlate with momentary spatial cues arising from acoustic conditions.
• Neural representations may reflect these fluctuating cues rather than requiring separate detection mechanisms.
• Broadly distributed interaural time differences are associated with the emergence of auditory object 'edges'.

Conclusions:

• Spatial perception may be an emergent property of dynamic neural activity patterns.
• The proposed hypothesis offers a parsimonious explanation for spatial encoding in complex auditory environments.
• Understanding cue fluctuations and ITD distributions is crucial for explaining auditory object formation.