CIAP 2023 CI Pulse Timing Intervals

Pulse timing interval sensitivity in the inferior colliculus of cochlear implanted rats

Fei Peng; Shiyi Fang; Muhammad Zeeshan; Bruno Castellaro; Qinjie Zhang; Jan Schnupp

In normal hearing, the neural processing of temporal fine structure information provides valuable pitch cues which help the perception of speech prosody, tonal language reception, musical melody appreciation and acoustic scene analysis. However, cochlear implant (CI) processors in current clinical use provide only limited amounts of fine structure in their electrical pulse trains, and the perceptual sensitivity of CI users is usually very poor compared to that of normal hearing individuals. To understand this deficit, several researchers have probed the sensitivity of CI patients to changes in the timing of pulses delivered via experimental processors. When investigating this phenomenon, one needs to control for confounds that can arise when changes in stimulus pulse rate also imply a change in stimulus intensity and therefore perceived loudness. One recent study controlled for this by delivering CI pulses in pairs and changing the interval in each pair while keeping the rate at which pairs were delivered constant. They observed a psychoacoustic interpulse interval discrimination threshold of 3 ms for pulse pairs delivered at a rate of 100 pairs per second in CI users. We suspect that CI patients who spend many months listening to stimuli with entirely fixed pulse rates over their clinical processors may become desensitized to small pulse interval changes, and that CI supplied, neonatally deafened rats which have not been similarly desensitized might exhibit lower inter-pulse interval detection thresholds. We therefore used a multichannel probe to record neural responses in the inferior colliculus (IC) of anesthetized, deafened rats to continuous trains of electrical pulse pulse pairs with inter-pair intervals that changed once per second from 1 ms to 1.04, 1.08, 1.12, 1.16, 1.2, 2, 3, 4, or 5 ms. We analyzed the recorded multiunit responses by computing their spectral power at 100 Hz. Using paired-sign rank tests we compared the power of the neural response during the stimulus periods where the interval in each pair was 1 ms against each of the other intervals. We found significant differences in the neural responses for interpulse interval changes as small as 1 ms to 1.04 ms in as many as 48 out of 606 IC multiunits. Our findings suggest that the IC of neonatally deafened rats is sensitive to much smaller changes in inter-pulse intervals than those previously reported in a psychoacoustic study of human CI users, which suggests the intriguing possibility that much better temporal fine structure sensitivity might be achievable in CI patients with better signal processing and treatment strategies.

Click here to download a pdf of the poster.