Distinct cortical encoding of acoustic and electrical cochlear stimulation

Reviewer #1 (Public review):

Summary:

This manuscript addresses an important question: whether cortical population codes for cochlear-implant (CI) stimulation resemble those for natural acoustic input or constitute a qualitatively different representation. The authors record intracranial EEG (µECoG) responses to pure tones in normal-hearing rats and to single-channel CI pulses in bilaterally deafened, acutely implanted rats, analysing the data with ERP/high-gamma measures, tensor component analysis (TCA), and information-theoretic decoding. Across several readouts, the acoustic condition supports better single-trial stimulus classification than the CI condition. However, stronger decoding does not, on its own, establish that the acoustic responses instantiate a "richer" cortical code, and the evidence for orderly spatial organisation is not compelling for CI, and is also less evident than expected for normal-hearing, given prior knowledge. The overall narrative is interesting, but at present, the conclusions outpace the data because of statistical, methodological, and presentation issues.

Strengths:

The study poses a timely, clinically relevant question with clear implications for CI strategy. The analytical toolkit is appropriate: µECoG captures mesoscale patterns; TCA offers a transparent separation of spatial and temporal structure; and mutual-information decoding provides an interpretable measure of single-trial discriminability. Within-subject recordings in a subset of animals, in principle, help isolate modality effects from inter-animal variability. Where analyses are most direct, the acoustic condition yields higher single-trial decoding accuracy, which is a meaningful and clearly presented result.

Weaknesses:

Several limitations constrain how far the conclusions can be taken. Parts of the statistical treatment do not match the data structure: some comparisons mix paired and unpaired animals but are analysed as fully paired, raising concerns about misestimated uncertainty. Methodological reporting is incomplete in places; essential parameters for both acoustic and electrical stimulation, as well as objective verification of implantation and deafening, are not described with sufficient detail to support confident interpretation or replication. Figure-level clarity also undermines the message. In Figure 2, non-significant slopes for CI, repeated identification of a single "best channel," mismatched axes, and unclear distinctions between example and averaged panels make the assertion of spatial organisation unconvincing; importantly, the normal-hearing panels also do not display tonotopy as clearly as expected, which weakens the key contrast the paper seeks to establish. Finally, the decoding claims would be strengthened by simple internal controls, such as within-modality train/test splits and decoding on raw ERP/high-gamma features to demonstrate that poor cross-modal transfer reflects genuine differences in the underlying responses rather than limitations of the modelling pipeline.

Reviewer #2 (Public review):

Summary:

This article reports measurements of iEEG signals on the rat auditory cortex during cochlear implant or sound stimulation in separate groups of rats. The observations indicate some spatial organization of cochlear implant stimuli, but that is very different from cochlear implants.

Strengths:

The study includes interesting analyses of the sound and cochlear implant representation structure based on decoders.

Weaknesses:

The observation that responses to cochlear implant stimulation (stimulation) are spatially organized is not new (e.g., Adenis et al. 2024).

The claim that spatial and temporal dimensions contribute information about the sound is also not new; there is a large literature on this topic. Moreover, the results shown here are extremely weak. They show similar levels of information in the spatial and temporal dimensions, and no synergy between the two dimensions. This is however, likely the consequence of high measurement noise leading to poor accuracy in the information estimates, as the authors state.

The main claim of the study - the mismatch between cochlear implant and sound representation - is not supported. The responses to each modality are measured in different animals. The authors do not show that they actually can compare representations across animals (e.g., for the same sounds). Without this positive control, there is no reason to think that it is possible to decode from one animal with a decoder trained on another, and the negative result shown by the authors is therefore not surprising.

Reviewer #3 (Public review):

Summary:

Through micro-electroencephalography, Hight and colleagues studied how the auditory cortex in its ensemble responds to cochlear implant stimulation compared to the classic pure tones. Taking advantage of a double-implanted rat model (Micro-ECoG and Cochlear Implant), they tracked and analyzed changes happening in the temporal and spatial aspects of the cortical evoked responses in both normal hearing and cochlear-implanted animals. After establishing that single-trial responses were sufficient to encode the stimuli's properties, the authors then explored several decoder architectures to study the cortex's ability to encode each stimulus modality in a similar or different manner. They conclude that a) intracranial EEG evoked responses can be accurately recorded and did not differed between normal hearing and cochlear-implanted rats; b) Although coarsely spatially organized, CI-evoked responses had higher trial-by-trial variability than pure tones; c) Stimulus identity is independently represented by temporal and spatial aspect of cortical representations and can be accurately decoded by various means from single trials; d) and that Pure tones trained decoder can't decode CI-stimulus identity accurately.

Strength:

The model combining micro-eCoG and cochlear implantation and the methodology to extract both the Event Related Potentials (ERPs) and High-Gammas (HGs) is very well designed and appropriately analyzed. Likewise, the PCA-LDA and TCA-LDA are powerful tools that take full advantage of the information provided by the cortical ensembles.

The overall structure of the paper, with a paced and exhaustive progress through each step and evolution of the decoder, is very appreciable and easy to follow. The exploration of single-trial encoding and stimulus identity through temporal and spatial domains is providing new avenues to characterize the cortical responses to CI stimulations and their central representation. The fact that single trials suffice to decode the stimulus identity regardless of their modality is of great interest and noteworthy. Although the authors confirm that iEEG remains difficult to transpose in the clinic, the insights provided by the study confirm the potential benefit of using central decoders to help in clinic settings.

Weaknesses:

The conclusion of the paper, especially the concept of distinct cortical encoding for each modality, is unfortunately partially supported by the results, as the authors did not adequately consider fundamental limitations of CI-related stimulation.

First, the reviewer assumed that the authors stimulated in a Monopolar mode, which, albeit being clinically relevant, notoriously generates a high current spread in rodent models. Second, comparing the averaged BF maps for iEEG (Figure 2A, C), BFs ranged from 4 to 16kHz with a predominance of 4kHz BFs. The lack of BFs at higher frequencies hints at a potential location mismatch between the frequency range sampled at the level of the cortex (low to medium frequencies) and the frequency range covered by the CI inserted mostly in the first turn-and-a-half of the cochlea (high to medium frequencies). Looking at Figure 2F (and to some extent 2A), most of the CI electrodes elicited responses around the 4kHz regions, and averaged maps show a predominance of CI-3-4 across the cortex (Figure 2C, H) from areas with 4kHz BF to areas with 16kHz BF. It is doubtful that CI-3-4 are located near the 4kHz region based on Müller's work (1991) on the frequency representation in the rat cochlea.

Taken together with the Pearsons correlations being flat, the decoder examples showing a strong ability to identify CI-4 and 3 and the Fig-8D, E presenting a strong prediction of 4kHz and 8kHz for all the CI electrodes when using a pure tone trained decoder, it is possible that current spread ended stimulating indistinctly higher turns of the cochlea or even the modiolus in a non-specific manner, greatly reducing (or smearing) the place-coding/frequency resolution of each electrode, which in turn could explain the coarse topographic (or coarsely tonotopic according to the manuscript) organization of the cortical responses. Thus, the conclusion that there are distinct encodings for each modality is biased, as it might not account for monopolar smearing. To that end, and since it is the study's main message and title, it would have benefited from having a subgroup of animals using bipolar stimulations (or any focused strategy since they provide reduced current spread) to compare the spatial organization of iEEG responses and the performances of the different decoders to dismiss current spread and strengthen their conclusion.

Nevertheless, the reviewer wants to reiterate that the study proposed by Hight et al. is well constructed, relevant to the field, and that the overall proposal of improving patient performances and helping their adaptation in the first months of CI use by studying central responses should be pursued as it might help establish new guidelines or create new clinical tools.