86 / 162
84 | Chapter 4
Recorded responses are larger when the recording contact is close to the responding neural fibers. The
presence of a significant shift in the SOE illustrates that, for both scanning and selectivity measures, the
outcome is to some extent influenced by current spread from the fibres towards the recording contact,
leading to a skewing of the SOE curve towards the recording electrode. Consequently, it was evident that
apical recordings (i.e. with the recording electrode apical to the fixed probe or masker) shift the recorded
flank of the SOE curve apical-wards, and basal recordings basal-wards. As a consequence, the data presented
here (which have mainly been recorded from an apical position) show a shift of the SOE in the apical
direction, resulting in a smaller SOE width for EA-B (Figure 2) and an increase in SOE width for the EB-A
condition (Table 4). In contrast, the middle electrode (EM) did not show such a clear shift. However, this
may be because most of the EM-A curves did not reach the 60% criterion resulting in arbitrary, fixed values,
reducing the possibility of showing the effect of the recording locations.
Our findings are in contrast to data presented by Cohen et al, who reported little overall difference between
selectivity recordings made apically or basally (Cohen et al, 2003, 2004). And although Hughes and Stille
(2010) describe that for recording positions equidistant from the probe, amplitudes were generally larger
when recorded from the apical side, only in a small minority of cases different recording positions resulted
in a significant shift in selectivity measures. There is also a lack of consistency in other previous studies.
eCAP measurements have sometimes been obtained from apical recordings (Busby et al, 2008), sometimes
from basal recordings (Lai et al, 2009), and sometimes not clearly specified (Cohen, 2009). This lack in
consistency in the use of recording contacts makes interpretation of conclusions about SOE measurements
difficult.
For the selectivity measures, asymmetry along the array was seen, with wider SOE functions at the apical
contact than at the basal contact (Table 2). This is generally in agreement with other studies (Eisen &
Franck, 2005). Moreover, in line with previous research, an asymmetry in the middle of the array was
evident (Cohen et al, 2003; Cohen, 2009). However, asymmetry in the middle part does not imply an
asymmetry in the neural excitation (Cohen, 2009). This can be explained as follows: the forward masking
paradigm measures overlap of excitation produced by different contacts. This overlap consists of neurons
excited by the contact of interest (the probe) and neurons excited by the masker. If the width of the pattern
of excited nerve fibres is not constant along the array, but wider in the apex than in the base, then the
overlap of a probe in the middle with an apical masking contact would be larger than with a basal masking
contact. The resulting SOE curve would thus become asymmetric towards the apex, as is seen in the data
of the present study. Psychophysical experiments of SOE determine thresholds of masking and are thus
theoretically less influenced by the width of excitation of the masking contact. Accordingly, psychophysically
obtained forward masking curves have shown no significant asymmetry (Cohen et al, 2004; Nelson et al,
2008). The differences between the SOE seen in the apical and basal parts of the cochlea may relate to
the smaller distance to the modiolus or a smaller volume of the apical cochlea. Alternatively, a larger SOE
apically could be due to crossturn stimulation, which is known to be more likely apically where the cochlea
is more tightly coiled (Frijns et al, 2001). These factors caused by the tapered morphology of the cochlea