Volume conduction models were used qualitatively to model surface potential
s from cochlear implant patients recorded earlier by the authors. These rec
orded potentials reflected the equivalent dipole orientation in the head in
patients who are deaf due to otosclerosis, but increased uniformly with th
e distance between the stimulating electrodes along the basilar membrane in
other patients, which suggested a low and high resistivity of the cochlear
bone, respectively. Several models of the head were constructed, with comp
artments representing the skin, skull, brain, cochlea, internal and externa
l ear canal. In the "petrous bone" model, the cochlea was modelled as a cav
ity in a bony layer surrounded by the brain compartment. Of all models, the
petrous bone model using a high resistivity ratio (1:100) between the bony
and the other compartments was the only one that produced outcomes similar
to the potentials observed in non-otosclerosis patients. In conclusion, th
e results suggested that the surface potentials observed in nonotosclerosis
patients are sufficiently explained by a high impedance between cochlear t
urns and a non-specific return of current via the wall of the petrous bone
into the larger brain compartment.