To explore whether the eFFR and eACC can be used as an objective measure of rate processing in CI users we assessed the phase-locking ability of the brainstem by means of the eFFR to four different pulse rates, namely 94, 128, 164, 196 pps. Furthermore, the ability of the auditory cortex to detect changes in rate was assessed with the eACC to changes from 94 pps (base rate) to 128, 164, and 196 pps (deviant rates). We assessed both the eFFR and eACC in a single experiment, by constructing a stimulation sequence (i.e. a single trial) that consisted out of a base rate (94 pps) and a deviant rate that was either 128, 164, or 196 pps. Each part of a trial had a duration of 2.048 s, and pulse rates were adjusted so that an integer number of pulse rates fitted in a single part of a trial (only rounded numbers are reported). Pulse rates below 200 pps were chosen as they are well within the range of rates that CI users perceive (Zeng, 2002). In addition, the changes in rate, which were 36, 72, and 108% were multiples of the ~35% rate change that most CI users are able to detect (Carlyon et al., 2008). Furthermore, these approximately-equally-spaced pulse rates enable a correct distinction between a neural response and a stimulation artifact. One efficient way to determine correct artifact removal and the presence of a phase-locked neural response is to assess the phase lag of a response at different pulse rates (Gransier et al., 2016; Hofmann and Wouters, 2012). The obtained group delay will correspond to the latency of the generator if the eFFR across pulse rates originates from the same generator(s) (i.e. there is a linear relationship between the pulse rates and the response phase across pulse rates). In contrast, an artifact-dominated response will yield a group delay of 0 ms (Gransier et al., 2016; Hofmann and Wouters, 2012). In order to estimate the group delay accurately, two criteria need to be met: first the responses need to be significant and second the difference in phase lag between two neighboring rates used for testing should not be more than 360°. If the latter requirement is not met, then erroneous unwrapping will occur. Based on the literature on FFRs obtained from normal-hearing listeners we expect FFR latencies for this range of pulse rates between 5 and 10 ms (King et al., 2016), which corresponds, respectively to a phase difference of 180 and 360° over a range of 100 pps. Therefore, the four pulse rates that are used here are sufficient for a correct estimation of the latency of a brainstem source.

All stimuli were generated in Matlab (version R2016b) and were delivered directly to the implant by means of a research interface, which consisted of a laptop with custom-written software interfacing with the Nucleus Implant Communicator (version 3) and connected to the implant through a programming device and an L34 research processor. The hardware and the Nucleus Implant Communicator were provided by Cochlear Ltd. The pulse trains consisted of symmetric biphasic cathodic-first pulses, with a phase width of 25 µs and an interphase gap of 8 µs. The most apical electrode (e22) was used for stimulation in monopolar mode (i.e. both the extracochlear electrode on the casing and the extracochlear ball electrode were used as the return electrode). The most apical electrode was used since this electrode was assumed to have the highest chance of activating neurons of the auditory nerve that innervate apical cochlear regions and that are important for rate encoding (Macherey et al., 2011; Middlebrooks and Snyder, 2010; Stahl et al., 2016).

Stimulation levels were determined behaviorally at the beginning of each experiment by using a 7-point categorical loudness scale (i.e. “inaudible”, “very soft”, “comfortably loud”, “loud”, “very loud”, and “unbearable”). First, the maximum comfort level for the base rate (i.e. 94 pps) was assessed by starting at 10 current levels1 and the current was then gradually increased to obtain the most comfortable loudness level which is defined as the current level prior to the transition between “comfortably loud” and “loud” on the categorical loudness scale. The deviant rates (i.e. 128, 164, and 196 pps) were then loudness-balanced to the base rate. This was done to minimize the effect of overall loudness differences between the base and deviant rate on the ACC. During the loudness-balancing task, a two-down, one-up procedure was used and the participant had to indicate which presentation (either the base-rate or the deviant-rate pulse train) was louder, using a two-alternative forced choice paradigm. Step sizes of two or four levels were used depending on the discrimination ability of the participant. The loudness-balancing procedure was stopped after eight reversals and the mean of the last six reversals was used as the loudness-balanced level. Differences between the base rate and the deviant rate were on average 0.7 CU and ranged from 0 to 3 CU with the largest difference always for the largest difference in pulse rates.

For the CIC4 implants used by some participants an additional RF-power pulse was placed before each stimulation pulse. It had the same duration as the stimulation pulse, and the inter-pulse interval between the RF-power pulse and the stimulation pulse was 8 µs. This was necessitated by the greater power requirements of the CIC4 implants, compared to the other implants used by the participants (Table 1). Unfortunately, we found during our experiments that this RF-power pulse was insufficient for Subject S1 and S5 when stimulating at the 94-pps pulse rate (see Fig. 1), leading to a compliance issue that caused the pulse amplitude (and therefore the loudness) to decrease towards the end of each 94-pps 2.048-s epoch. When the pulse rate increased at the start of the deviant-rate epoch the “silent” intervals between the pulses decreased, thereby removing the compliance issue and causing the pulse amplitude to increase. This in turn could have led to a perceived loudness cue that could have biased the ACC for the base-to-deviant change. The ACC to the base-to-deviant change was therefore excluded from the analysis for S1 and S5. Although not included here, we assessed, after observing these issues, what the minimum length needed to be for the CIC4 implants to have no compliance issues at the 94-pps pulse rate by means of an implant-in-the-box and an oscilloscope. No compliance issues were observed for a wide range of loads when the RF-power pulse length was 3 times the stimulation pulse duration.

Information about the participants.

The EEG of two trials of the 94 – 196 pps condition. The first 2.048 s were stimulated at 94 pps and the second 2.048 were stimulated at 196 pps. The red dashed line indicates the deviant-to-base change, the orange dashed line the base-to-deviant change, and the blue dashed line indicates the timepoint where the reduction in amplitude due the compliance issue starts . (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Note: The content above has been extracted from a research article, so it may not display correctly.



Q&A
Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.



We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.