Auditory Time-Frequency Masking for Spectrally and Temporally Maximally-Compact Stimuli
Level 4 - case-series / case-control
Non-clinical human psychophysical experiment and computational modeling (graded by design analogy, not clinical CEBM)
OpenAlex W2555290060 · doi:10.1371/journal.pone.0166937
What was done
Auditory time-frequency (TF) masking was measured in humans using spectrally and temporally compact stimuli, where both masker and target were 10-ms Gaussian-shaped sinusoids with a bandwidth of approximately one critical band. A computational temporal window model was evaluated under two configurations—standard parameters without efferent activation and cochlear gain reduction simulating efferent activation—to predict both the new compact-stimulus data and existing long-masker data.
What was found
The abstract reports no numerical values, thresholds, or statistical effect sizes. Qualitatively, the TF masking pattern for compact stimuli was similar to historical data from long maskers. The standard temporal window model accurately predicted the compact-stimulus data but performed poorly when cochlear gain reduction was added, whereas long-masker data were better predicted with gain reduction enabled.
Why it matters
Provides psychoacoustic masking data and computational validation for mathematically well-localized signals, improving perception-based time-frequency analysis and audio coding algorithms.
Limits
The abstract does not disclose the sample size, subject characteristics, or quantitative findings. Physiological inferences about efferent mechanisms rely entirely on computational modeling rather than direct in vivo measurements.
Cited by
- supports Perceptual lossy audio compression formats like MP3 utilize psychoacoustic models of auditory masking, discarding sounds that the human auditory system cannot perceive when played simultaneously with louder sounds at adjacent frequencies.