Journal Club: Study in single-side cochlear implant patients show close alignment between perceived pitch and the Greenwood function.

on

Today's journal article

Büchner A, Weller T, Penninger RT, Helpard L, Ladak HM, Agrawal S, Lenarz T, Schurzig D. The Greenwood function shows close alignment with pitch perceived by cochlear implant patients with long, flexible electrode arrays and fine-structure stimulation. 

Why I picked this article

The mammalian cochlea has a spiral shape, and each part of the spiral responds to different frequencies. Higher frequency sounds are detected near the base of the cochlear spiral, and the lower frequency sounds are detected at the top of the spiral. This arrangement is termed the tonotopic representation of sounds in the cochlea. 

The Greenwood function is one of the most famous equations in auditory science. It describes how frequencies are distributed along the cochlea in a logarithmic way, linking a physical location in the inner ear to the pitch we perceive. Greenwood function is below: 

Where A = 165.4, a = 2.1 and k = 0.88 for the human cochlea to cover the length of the basilar membrane, the portion that oscillates in response to sound. 

The concept of tonotopic representation and prediction by Greenwood function is very important to enhance the pitch discrimination capability by technologies like cochlear implants. While the Greenwood function has long been used in models and cochlear implant programming, there hasn’t been much direct experimental evidence showing that cochlear implant users perceive pitch according to this same tonotopic rule. This research aims to directly test whether the Greenwood function can accurately represent pitch perception in cochlear implant users. They do so based on the experience by people with single-sided deafness, a group that can directly compare their electrical hearing (in the implanted ear) to acoustic hearing (in the normal ear).

Some of the research findings

Participants and method
  • 12 total cochlear implant users with single-sided deafness took part in the study. 
    • mean AGE: 52.9 years; range: 35.5–80.6 years
    • implanted side hearing threshold of ≥ 65 dB HL at 250 Hz
    • non-implanted side, contralateral four-frequency pure-tone average, 4-PTA, of ≤ 30 dB HL
  • Each participant attended four test sessions and used two CI fitting maps: 
    • The standard clinical fitting
    • An anatomy-based fitting, which assigns sound frequencies to electrodes based on each person’s cochlear geometry (following the Greenwood function). Their cochlear anatomy was reconstructed from clinical CT scans, allowing researchers to calculate the physical electrode positions relative to the basilar membrane.
  • During testing, participants performed a pitch-matching task: 
    • They heard tones through a loudspeaker in the normal ear. 
    • They adjusted the frequency of those tones to match the perceived pitch when stimulating specific electrode contacts in the implanted ear. 
    • Pitch matching ability was quantified by the amount of deviation from the pitch ΔfPM, and the total root-mean-square deviation for an individual subject was calculated as ΔfPMS.
Findings: 
  • The subjective pitch percepts from single-side cochlear implant users matched closely with the frequencies predicted by the Greenwood function, across subjects and electrode positions.
  • No significant differences were found between the pitch matches and the Greenwood-predicted frequencies.
  • Whether participants used the standard fitting or anatomy-based fitting did not significantly affect this overall relationship.
  • The study also noted that some previous reports of “basal frequency shifts” (where perceived pitch is higher than expected) might have been due to differences in how tonotopic mapping was modelled in earlier research.
Figure 5A. perceived frequency plotted against Greenwood function. Buchner 2025.

Haruna's takeaway

This is such educational research!! These findings provide strong evidence that the Greenwood function accurately reflects the tonotopic frequency distribution. I also really like the simplicity of the Greenwood function - wouldn't we wish we could do a similar study in animals, to test how closely the Greenwood function remains accurate in the cochlea with different sizes in different animals! (Like Beluga that can hear 120kHz!) 

Also wonder - how much of the perceived frequency is influenced by correction from the central auditory system. The single-sided deafness patients have one-sided hearing, and they are likely able to fine-tune perception from both ears at the central auditory system level in the brain. Either way, huge thanks to the participants of this study who have done this research not for their benefit but for the benefit of our understanding of the cochlea. 

 ------- 

This is Haruna's 48/100 of the 100-day challenge to post a science blog article every day! I love inner ear biology & cochlear physiology.