@inproceedings {Altoe’1814_2016,
	year = {2016},
	author = {Altoe’, Alessandro and Verhulst, Sarah and Pulkki, Ville},
	title = {Resolving the Discrepancies between Basilar Membrane, Reticular Lamina and Inner Hair Cell Tuning},
	booktitle = {Assoc. Res. Otolaryng. MidWinter Meeting (ARO)},
	URL = {http://c.ymcdn.com/sites/www.aro.org/resource/resmgr/Abstract_Archives/UPDATED_2016_ARO_Abstract_Bo.pdf},
	abstract = {Classical recordings from the basal region of the chinchilla
cochlea (~8 kHz) show that neural tuning curves lie about
halfway between basilar-membrane (BM) constant-velocity
and constant-displacement tuning curves. This suggests
that inner-hair cells (IHCs) respond to a combination of BM
velocity and displacement; a statement that finds support in
classical guinea-pig IHC recordings.
In this study, we offer an explanation for the well-documented
relationship between neural and mechanical tuning that
relies on a proportionality between IHCs stereocilia deflection
amplitude and BM velocity up to the units’ best frequency.
By assuming well-established features of IHC processing, i.e.
a Boltzmann type transduction nonlinearity followed by the
low-pass filtering action of the IHC basolateral membrane,
we derive the theoretical background for this work. We
then validate our hypothesis using numerical simulations in
a computational model of the inner hair cell/auditory nerve
complex.
Unfortunately, the demonstrated proportionality between
BM velocity and stereocilia deflection seems in contrast
with recent in-vivo recordings from the ~18 kHz region of
the guinea-pig cochlea that show a level-dependent sharper
tuning of the reticular lamina (RL) compared to BM tuning.
To resolve this discrepancy, we propose the existence of a
mode of RL vibration that reflects the motion of the BM in a
more basal cochlear region at a fixed distance. This mode
of vibration can account for sharper RL than BM tuning in
the 18 kHz region through a phase cancellation mechanism
and we show this numerically using a linear nonlocal model
that relates RL and BM vibrations. The suggested phase
cancellation mechanism results in a mode that enhances RL
vibrations at high-frequencies (above 10 kHz), but that leaves
lower-frequency responses unaffected in agreement with
the differences observed in the high (~18 kHz; guinea pig)
vs lower (~8 kHz; chinchilla) frequency cochlear mechanics.
Additionally, the proposed linear model is capable of capturing
the apparent nonlinear relationship between RL and BM
motion.
We propose a description of BM-RL-IHC transduction that
is based on previous BM and IHC models, combined with
(i) frequency dependent stereocilia low-pass filtering and (ii)
a feedforward model of RL motion, and demonstrate that it
can account for well-documented relationships between
mechanical and neural tuning. We lastly point out that
established functional models of the IHC/auditory-nerve
complex tend to overestimate the neural response to lowfrequency
stimuli, resulting from an excessive low-pass
filtering in the IHC stage.
Funding
Work supported by Aalto ELEC doctoral school and DFG
Cluster of Excellence EXC 1077/1 "Hearing4all".}
}