GROUNDING — h02 piezoelectric literature audit
Weights: STATES = source asserts the claim · VARIANT = adjacent, bridging derivation given inline · ABSENCE = searched, not there.
G1 — OHC specific membrane resistance
Exact query run (WebSearch):
outer hair cell specific membrane resistance kOhm cm2 basolateral input resistance measurement Housley Ashmore
Results returned: Ashmore 2008 “Cochlear Outer Hair Cell Motility” review (UCLA open PDF); “Acetylcholine, OHC Electromotility, and the Cochlear Amplifier” (PMC6793750); “High-Frequency Force Generation in the Constrained Cochlear Outer Hair Cell: A Model Study” (PMC2504621); Dallos/Ashmore Springer chapter; two arXiv prestin-NLC preprints; ScienceDirect topic page; Frontiers 2025 in-vivo OHC length changes.
Verbatim from the search synthesis:
“Research by Housley and Ashmore showed that as the input conductance of the OHC cell increases, the receptor current-induced voltage drop on the basolateral (motor-bearing) membrane decreases, and the electromotile response is reduced.”
“the specific numerical values for membrane resistance in kOhm·cm² or detailed input resistance measurements appear to be contained in their primary research papers that are cited throughout these results rather than directly quoted in the accessible portions of these sources.”
Second attempt (WebFetch) on the one open-access candidate:
WebFetch https://www.uclahealth.org/sites/default/files/documents/OHC_Motility-Ashmore_Review08.pdf
prompt: quantitative values for OHC membrane resistance, input resistance, specific
membrane resistance, basolateral conductance, time constant, specific capacitance;
plus any prestin NLC / electromotility saturation voltage
Outcome: the fetch returned PDF structural metadata, not body text — no values
extracted. The binary was cached by the harness at
…/tool-results/webfetch-1787277624569-w3z5sj.pdf (harness scratch, not a vault
path). It was not MinerU-parsed this run: it is a review, so even a
successful parse would yield a secondary citation, and §0c wants the primary.
Evidential weight: INCONCLUSIVE — 2 attempts, not 5.
This is stated plainly rather than upgraded to a confirmed absence. The quantity
is measured in classical OHC electrophysiology (Housley & Ashmore 1992
J Physiol; Santos-Sacchi’s whole-cell work); it simply was not reached from the
two access vectors tried here. Recording this as “CONFIRMED NOT MEASURED” would
be false. It stays an open, retrievable blocker with a named next step:
retrieve Housley & Ashmore 1992 J Physiol 448:73–98 as a primary PDF
(Anna’s Archive → mineru -p <in> -o <out> -b vlm-engine), take the input
resistance / specific membrane resistance directly, and file a sources/lit/ note.
G2 — prestin V_saturation = 70 mV
Exact query run (WebSearch):
prestin nonlinear capacitance Boltzmann parameters Vpkcm valence z Qmax outer hair cell Santos-Sacchi measured values
Results returned: Santos-Sacchi et al., “Megahertz Sampling of Prestin (SLC26a5) Voltage-Sensor,” J Neurosci 43(14):2460; “Outer hair cell electromotility is low-pass filtered relative to the molecular conformational changes that produce nonlinear capacitance,” J Gen Physiol 151(12):1369; “Voltage-dependent changes in specific membrane capacitance caused by prestin”; “State dependent effects on the frequency response of prestin’s real and imaginary components of nonlinear capacitance,” Sci Rep (PMID 34373481); “Voltage and frequency dependence of prestin-associated charge transfer” (PMC2753271); AJP-Cell Tet-inducible prestin Boltzmann-characteristics paper.
Verbatim from the search synthesis:
“The nonlinear capacitance (NLC) of prestin is fit as a molecular capacitor obeying Boltzmann statistics according to Santos-Sacchi’s established approach.”
“The Boltzmann equation includes the following parameters: Qmax (the maximum nonlinear charge moved), Vh (voltage at peak capacitance or equivalently at half-maximum charge transfer), Vm (membrane potential), z (valence), Clin (linear membrane capacitance), e (electron charge), kB (Boltzmann’s constant), and T (absolute temperature).”
“NLC is bell-shaped, with its peak occurring at voltage Vh where sensor charge is equally distributed across the plasma membrane”
“in one model study, z = 0.92 for guinea pig outer hair cells and z = 1 for mouse outer hair cells.”
Evidential weight: STATES — for a reframing, not for the number.
The literature describes prestin’s voltage dependence as a Boltzmann, which
is asymptotic: charge transfer approaches Qmax but never “saturates” at a
discrete voltage. There is therefore no measurable quantity called
“V_saturation,” and no paper will ever report one. V_saturation = 70 mV is not
an unsourced measurement — it is a model ceiling with no physical referent.
This independently confirms what parameters/piezoelectric-materials.md:46
already suspected: “NOT in Santos-Sacchi 1991. The Boltzmann C-V function has no
discrete saturation voltage at 70 mV.” Two independent lines now agree.
The sourced replacement exists and is well characterised: the Boltzmann triple
(Vh, z, Qmax). One primary value surfaced this run — z = 0.92 (guinea pig OHC),
z = 1 (mouse OHC) — but via a modelling paper’s restatement, so it is a
VARIANT, adequate to establish that the parameters are measured and not
adequate to be pasted into a script. Retrieving Santos-Sacchi’s primary NLC
papers and filing them to sources/lit/ is the named follow-up.
G3 — TM Young’s modulus “210 kPa”
Not re-searched this run. The 2026-04-25 audit ran an exhaustive pass and the
result is recorded verbatim in parameters/tectorial-membrane.md:26:
”❌ NO PRIMARY SOURCE | Exhaustive search 2026-04-25: Teudt 2014 max 24.3 kPa (basal, ±25.2); Shoelson 2004 shear 2–6.5 kPa; Gavara 2009 fiber ~1 kPa; Gueta 2008 vertical 300 kPa (but that is vertical compressive modulus, not Young’s). The 210 kPa figure does not appear in any retrieved paper. Must be treated as in-house assumption or removed from scripts.”
Evidential weight: ABSENCE, and specifically a PHANTOM VALUE — a different
category from h09’s not-modelable gaps. TM Young’s modulus is measured;
sources/lit/2026-04-25-teudt-2014-tm-bm-stiffness-cba-mouse.md carries
24.3 ± 25.2 kPa basal / 5.1 kPa middle / 1.9 kPa apical for CBA/CaJ mouse. So
210 kPa is not an unobtainable parameter — it is a number with a measured
counterpart that disagrees with it by ~10×. The remedy is deletion from
models/piezo_voltage_budget.py:262,271 and substitution of the Teudt value,
not further searching.
The still-live phantom text at models/piezo_voltage_budget.py:262 reads:
"to 210 kPa (basal) per Masaki 2009 PLOS One 4:e4877; PVDF-TrFE "
“Masaki 2009 PLoS One 4:e4877” was identified as a phantom in the 2026-04-23 audit — that DOI is Gavara & Chadwick 2009. The citation was corrected in the parameter tables but the phantom string is still in the script, four months later. That is the finding worth surfacing: an audit that corrects the table and not the code leaves the fabrication in the executable path.
G4 — h02 values confirmed sourced (carried, spot-checked against the tables)
| constant | value | source note (all type: lit) | status |
|---|---|---|---|
| TM Young’s modulus, basal / middle / apical | 24.3 ± 25.2 / 5.1 / 1.9 kPa | 2026-04-25-teudt-2014-tm-bm-stiffness-cba-mouse | ✅ (basal, NOT apical — prior scripts read it backwards) |
| TM shear modulus G′ | 2.0–6.5 kPa | 2026-04-25-shoelson-2004-tm-elasticity-inhomogeneity-biophysj | ✅ context (guinea pig, shear not Young’s) |
| TM fiber modulus | ~1 kPa | 2026-04-25-gavara-2009-tm-collagen-anisotropy-afm | ✅ |
| PVDF-TrFE d33 (80/20) | 25–28 pC/N | 2026-04-25-hu-2021-pvdftrfe-insitu-poling-frontiers-energy | ✅ for 80/20; 🔶 for the 70/30 used in scripts |
| OHC specific membrane capacitance | 0.9 µF/cm² | Gentet 2000 Biophys J 79:314 (cited in script) | ✅ |
| TM displacement at 60 dB SPL | 5–30 nm | none — in-house table | 🔶 model assumption; closest anchor Lee 2015 PNAS PMID 25737536 (~20 nm @ 20 dB, ~90 nm @ 80 dB) |