The hearing system groups frequency bands together when their level variation is similar

Evidence

Co-modulation masking release is the decrease in the masking effect when the masker is amplitude-modulated with the same modulating function at all frequencies. Let us consider the case where the signal is a single sinusoid and the masker is a noise band around the frequency of the sinusoid. If the noise band is not modulated, the masking effect grows stronger when the noise bandwidth is extended. Interestingly, if the masker is amplitude-modulated coherently in all frequencies of the masker at the rate of 10–20 Hz, the masking effect deviates from the non-modulated case when the bandwidth of the masker exceeds 100 Hz (Hall et al., 1984). This can be seen to imply that the hearing system groups different frequency bands together based on the similarity of the variation of the level in the bands. — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 1013

Categorization

Category: Causal Relationships

• DEFINE: Defines co-modulation masking release and explains the mechanism behind it.
• XREF: Connects to auditory scene analysis / Bregman’s streaming literature and to perceptual grouping theory.

Related

• Hearing groups frequency bands by shared envelope modulation, reducing masking (similarity 0.95)
• Louder maskers increasingly mask high frequencies over low ones. (similarity 0.83)
• Auditory masking depends only on energy within a band surrounding the signal. (similarity 0.82)
• Upward masking makes intermodulation difference tones more audible than harmonics (similarity 0.82)
• Similar frequencies mask each other by competing for the same cochlear location. (similarity 0.82)
• Variable-frequency experiments reveal frequency effects that fixed-frequency masking hides. (similarity 0.81)
• Each masker dominates masking only within its own frequency region. (similarity 0.81)
• Fluctuating maskers produce masking release that Fletcher’s power spectrum model cannot explain. (similarity 0.81)
• Uniform masking noise can substitute for threshold-in-quiet across all frequencies. (similarity 0.81)
• Psychophysical tuning curves map masker level needed to just mask a fixed tone. (similarity 0.81)
• Masking acts within critical bandwidths, leaving distant-frequency sounds unaffected (similarity 0.81)
• Frequency masking is modelled as convolution with a spreading function (similarity 0.81)
• Auditory masking spread can be modelled by convolving with a spreading function (similarity 0.81)
• Tonal masker spread inverts from low to high frequencies as masker level rises (similarity 0.81)
• Frequency-specific hearing loss drives multi-band hearing aid processing (similarity 0.81)