Frontiers in Neuroscience

Despite its high prevalence and socioeconomic costs, the condition of tinnitus shares with related neuropsychiatric disorders the characteristic that, to this day, it cannot be cured. Two contradictory views of the origin of tinnitus (peripheral hyperexcitability and central brain oscillation changes linked to prediction error) are currently discussed without any regard for one another. We now firstly used a compact 64 sensor optically pumped magnetometer (OPM)-MEG system to study a group of tinnitus subjects without co-morbidity of hyperacusis. This new technology provided an unprecedented opportunity for analyzing hemisphere-specific brain activity changes with high spatial resolution in response to pure-tones with a pitch within or outside the tinnitus frequency. We observed in tinnitus smaller ABR amplitudes (reflecting reduced cochlear output synchrony) linked with reduced alpha and enhanced gamma activity at rest (reflecting elevated excitement of intracortical circuits). In tinnitus, reduced alpha and enhanced gamma brain activity at rest were associated with reduced evoked alpha, beta, and gamma in response to pure-tones within tinnitus frequencies (reflecting low signal-to-noise ratios in auditory target regions). Furthermore, elevated gamma activity in key regions in the brain involved in attention control was observed in tinnitus subjects: hypergamma activity was seen in posterior/frontal regions, that -when hyperactive - are predicted to trigger excessive attention to irrelevant stimuli. Weakened cochlear output synchrony, possibly through lowering tonic inhibitory strength in the ascending auditory pathway, can thus reduce alpha activity (default-mode network) and unleash cortical regions that control attention to irrelevant stimuli -- tracing tinnitus to perception.

High performance intracortical brain-computer interface (iBCI) control has been demonstrated in research settings, but performance can still vary within and between sessions. One potential source of this variability is the change in attentional load that comes from processing naturally occurring distractors such as thoughts, sounds, fatigue, or pain. To improve the consistency of iBCI performance in real-world environments where this sort of multi-tasking is inevitable, we must understand how shifts in attention can impact performance. Here we examined the effect of attentional load on iBCI performance and movement-related neural activity using a 2D cursor translation + click iBCI task paired with an N-Back working memory task to increase attentional load during dual-task performance. Two participants (P2 and P4) with tetraplegia completed the study while enrolled in a long-term clinical trial of an iBCI device (NCT1894802). Common neural correlates of attention (theta and alpha band power) were measured with simultaneously recorded scalp electroencephalography (EEG). While the EEG recordings and difficulty ratings suggested increased attentional load during dual tasking, iBCI performance was quite robust across the various dual tasking conditions. One participant, P2, experienced a small but significant increase in trial completion time and normalized path length during the mild attentional load condition. Signal quality differences between the two participants may have impacted the results, as P2 had lower signal quality and was therefore likely more vulnerable to attentional load. P4s higher signal quality likely allowed him to accommodate increased attentional load without a drop in performance. Overall, iBCI performance appears to be robust to attentional load, but the complex trends observed here reflect a need for continued investigation of BCI use under different cognitive states to elucidate potential challenges and compensatory mechanisms across participants.

Non-invasive brain-computer interfaces (BCIs) have substantially advanced in the field of continuous cursor control over the past decade. Yet, current methods lack key control aspects such as initiation and termination of cursor movements as well as evaluation in real-world applications. In this study, we introduce a framework for continuous, electroencephalography-based cursor control that supports both active movement and no-movement states, thereby allowing for inactive periods of the user when no control input is desired. We demonstrate its applicability in healthy participants and show its performance in real-world application through the selection of targets on a screen. This demonstrates that participants can leverage the continuous control cursor control and the intentional starting and stopping of motions to effectively select targets on a screen through dwell-time selection. On average, 7.1 out of 40 targets were correctly selected (level of significant performance: 4.5 targets), while experienced BCI users achieved an average of 12.8 targets. The proposed framework additionally demonstrates compatibility with motor-impaired people without residual hand motions since it does not rely on observable movements for model training.

The Empatica E4 wristband provides continuous multi-modal physiological monitoring including blood volume pulse (BVP), electrodermal activity (EDA) and skin temperature (TEMP) but its validity for sleep-stage-specific autonomic and thermoregulatory monitoring has not been systematically evaluated against concurrent polysomnography (PSG). Using the Wearanize+ dataset which provides synchronised PSG, Empatica E4, and Zmax EEG recordings from 100 home-recorded participants; a systematic validation of Empatica E4 physiological signals against PSG ground truth across five sleep stages was conducted. Of 100 participants, 92 had Empatica data; 69 met Zmax EEG signal quality criteria and formed the analysis sample. Heart rate (HR) from the pre-computed Empatica HR channel showed valid stage-specific patterns (Wake: 70.9 bpm, N3: 61.2 bpm) and moderate inter-device MeanNN correspondence with PSG ECG (Spearman r=0.35-0.42 across stages). Skin temperature showed the expected thermoregulatory pattern (Wake: 33.92C, N3: 35.48C) and is recommended for downstream analyses. Tonic EDA showed an inverted stage pattern attributable to wrist sweat accumulation during deep sleep, representing a known confound for wrist-worn EDA during sleep. Phasic EDA showed plausible patterns and may be used with caution. These findings establish a validated feature set for Empatica E4 sleep research and directly inform multimodal psychiatric biomarker studies using the Wearanize+ dataset.

Parkinsons disease is predominantly characterized by dopaminergic neurodegeneration linked to toxic aggregation of -synuclein. Genipin, a bioactive iridoid, was previously shown to improve the motility and survival deficits caused by pan-neuronal expression of native -synuclein in a transgenic Drosophila melanogaster model system. We show that expression of -synuclein causes sleep deficits and that genipin treatment rescued these sleep deficits, increasing total sleep and consolidating nighttime sleep relative to untreated -synuclein-expressing fruit flies. Our findings extend genipins protective profile in Drosophila melanogaster and highlight sleep regulation as an additional phenotype responsive to -synuclein-targeted interventions.

Rodent brain matrices that produce coronal or sagittal brain sections for histology confer reproducibility and enable high throughput processing of tissues. However, a stainless steel or acrylic brain matrix that produces tissue sections in a transversal (or horizontal) orientation is currently unavailable as a standard tool. This limits the direct comparison of bilateral brain hemispheres within a single histological section, as freehand trimming to obtain horizontal planes is not easily replicable across samples. To mitigate this challenge, we designed a low-cost (USD 7 per unit), 3D-printed resin-based transverse brain matrix that accommodates mouse brains ranging from 12 to 16 mm in length from the olfactory bulb to the brainstem. Our matrix reproducibly generates horizontal tissue sections with a minimum of 1-mm-thickness without causing visible tissue deformation, which is comparable to the performance of commercial rodent brain matrices. Users may adapt the accompanying CAD code using our video tutorials to customize the transverse brain matrix for their specific needs, including alternative brain size, shape, and tissue thickness.

Purpose. To evaluate whether a Geographic Atrophy Morphology-based Mapping Algorithm (GAMMA) grid, informed by geographic atrophy (GA) lesion morphology, can accelerate functional progression detection compared with a conventional 10-2 grid and a dense grid (129 locations). This work is motivated by emerging regulatory expectations requiring at least five locations to worsen by [&ge;]7 dB from baseline. Methods. Binary atrophy masks from six autofluorescence images were used to simulate GA expansion over 3 years at 3-month intervals using a stochastic perimeter-growth model with a fixed preferential expansion direction (Pdir). For each image, 32 independent growth histories and 32 microperimetric test realisations per history were generated. For each grid (10-2, Dense, and GAMMA), 5-point clusters were selected outside the baseline GA lesion along three directions (0{degrees}, 30{degrees}, 120{degrees}) away from Pdir, simulating full, partial, and no prior knowledge of Pdir. Ground-truth sensitivities were <0 dB inside the GA lesion and normal outside, calculated using a published normative equation. Response variability was simulated following Henson et al. with baseline averaging. Detection time was the first visit at which all five selected locations showed [&ge;]7 dB loss from baseline. Survival curves and median detection times (T50) were used to compare grid performance. Results. The GAMMA grid achieved the earliest progression detection across all scenarios. Under full knowledge of the expansion direction, T50 was 1.0 year for GAMMA versus 1.25 and 1.5 years for Dense and 10-2, respectively. With partial knowledge, GAMMA's T50 was 1.25 years versus 1.5 and 2.0 years for Dense and 10-2. Even under no knowledge, GAMMA detected progression earliest (T50 = 1.5 years), while Dense required 6 months longer and 10-2 nearly double the time (2.75 years). Conclusions. The automatic GAMMA grid accelerates detection of localised functional progression compared with conventional and dense grids. Structure-informed grid optimisation may better align testing with likely expansion paths, potentially reducing follow-up duration and sample sizes in perimetry-based interventional trials.

Individuals with inherited retinal dystrophies (IRDs) undergo a slow, genetically heterogeneous loss of vision, yet how the visual cortex and non-visual sensory, motor, and psychological systems adapt to this deprivation remains poorly characterized. Existing evidence comes mainly from single-modality, cross-sectional studies that rarely account for genetic heterogeneity, making it hard to distinguish adaptive change from a direct, non-retinal mutation effect, since several IRD genes are not retina-specific. To address this gap, we designed an observational, longitudinal, multimodal protocol that combines ophthalmological, genetic, and in silico characterization with electrophysiological (steady-state visual evoked potentials and TMS-EEG), chemosensory, sensorimotor, and psycho-personological assessments. Patients aged 18 to 75 years with rod-cone (retinitis pigmentosa, Usher syndrome) or cone and cone-rod dystrophies will be assessed at baseline (T0) and at an 18-month follow-up (T1); sighted controls, matched for age, sex, and handedness, will complete the same battery once. Importantly, pairing genotypic with phenotypic data allows changes in non-visual domains to be interpreted against, rather than independently of, each patient's molecular background. We expect individuals with IRDs to differ from controls in visual cortical responsiveness and in selected non-visual sensory and sensorimotor measures, with genotype-related differences explored where sample size permits. Given the rarity of IRDs, the design is exploratory and emphasizes effect sizes and individual variability over large-sample inference. The protocol was approved by the Ethics Committee of the University of Verona (CARP 08.R1/2024) and follows the Declaration of Helsinki and the GDPR; findings will be disseminated through peer-reviewed publications and shared with patients and IRD patient associations.

In this study, we investigated the neural and behavioral basis of motivationally objective versus subjective value-based decisions. Using a within-subject fMRI design, healthy participants performed a risky decision-making task that elicited different levels of subjectivity in decision-making in two task conditions. In the Best or objective condition, choices were rewarded only when they were objectively best on a given trial, incentivizing decisions based on externally specified per-trial point maximization. In the Choice or subjective condition, participants received the reward associated with the chosen option, irrespective of how it compared to the unchosen option, allowing greater freedom to exercise subjective preferences in decision policy. Behaviorally, participants relied more on objectively optimal policy in the Best than the Choice condition. There was also a greater consensus across participants in behaviorally displayed and self-reported policies in the Best condition as well as a greater commitment to a single policy by individual participants in this condition, further confirming more objective behavior in the Best condition, compared to Choice. Moreover, behavioral inferences showed a greater agreement with self-report in the Best condition. Our fMRI results showed that the decision-making in Choice, relative to the Best condition, was associated with greater BOLD response in mid-cingulum/posterior cingulate cortex and dorsal anterior cingulate cortex, suggesting their involvement in less externally constrained, or motivationally subjective, decision-making.

Low-cost and open-source neural recording systems are increasingly important for expanding access to electrophysiological research. However, many existing platforms still rely on specialized hardware or limited modularity, restricting flexibility for laboratories seeking customizable solutions. Here, we developed and evaluated a modular neural recording platform constructed entirely from commercially available components. Recordings were compared against the ground truth. The platform successfully recovered local field potential (LFP)-like waveforms in most conditions and detected spike-like activity during direct connection recordings. Principal component analysis and k-means clustering further demonstrated the ability to distinguish multiple simulated spike waveforms. Signal quality varied across configurations, with saline recordings and preamplifier integration introducing increased noise and reduced detectability. These findings demonstrate the feasibility of building affordable and modular electrophysiology systems using widely accessible hardware. Although the current implementation has limitations in sampling rate, noise performance, and in vivo validation, the presented framework provides a practical foundation for future customizable open-source neural recording.

Cochlear implants (CIs) restore hearing by stimulating auditory neurons to encode amplitude envelopes across frequency bands, providing essential cues for speech recognition. This study investigated how stimulation pulse rate constrains temporal envelope processing and speech cue perception in ten post-lingually deaf CI users by evaluating amplitude modulation (AM) detection thresholds and consonant identification performance across pulse rates. The effects of pulse rate on temporal processing and speech perception were examined using both standard clinical multi-channel strategies and single-channel strategies designed to isolate within-channel envelope representations. Results revealed a significant decline in AM detection and consonant recognition performance at the lowest tested pulse rate of 125 pulses per second (pps), consistent with perceptual constraints on temporal processing at low carrier rates, rather than inadequate envelope sampling. At the highest pulse rate of 4000pps, a non-significant reduction in AM detection was observed which may be consistent with previously reported reductions in amplitude discrimination at high pulse rates. Consonant recognition performance remained stable across clinically relevant pulse rates (250-2000pps), though listener-specific pulse rate effects were observed. Notably, significant correlations were found between single-channel and multi-channel performance in AM detection and consonant recognition tasks. These findings support an important contribution of within-electrode temporal envelope processing to multi-channel speech perception and highlight the clinical relevance of individual variability in pulse rate effects.

Accurate determination of surgical margins is critical in tumor resection to ensure complete removal of tumor-infiltrated tissue while preserving healthy tissue. Pathological assessment provides reliable information but is time-consuming. This study investigates the feasibility of using impedance spectroscopy to detect tissue transitions at a macroscopic level. Two electrode arrays--one-dimensional and two-dimensional--were applied to ex vivo porcine brain tissue. Measurements were performed using both two- and four-electrode configurations, and data were corrected using the multiple-load compensation method. Results demonstrate that the one-dimensional array provides continuous conductivity profiles corresponding to tissue transitions, while the two-dimensional array showed less consistent results. These findings suggest that impedance spectroscopy is a promising tool for intraoperative margin detection, but further optimization of electrode geometry and measurement data processing is required.

Surgical planning for drug-resistant epilepsy often relies on stereo-EEG (sEEG) recordings obtained with cylindrical ring electrodes. Prior modeling studies, including lead-field analysis, suggest that microelectrodes distributed around an sEEG-sized insulating body offer superior source amplification and directional sensitivity that are not available with either a ring design or microelectrodes on micro-structures, e.g. Neuropixel. However, these advantages have not been demonstrated in seizure models. This study evaluated high-density sEEG recordings using directional microelectrode arrays in a kainate-mediated rat model (n=6). Two 64-channel microelectrode arrays were implanted near the hippocampus, and the signals were spatially averaged to emulate virtual ring electrodes for comparison. Device locations were reconstructed and placed in copies of the Waxholm Space Rat Brain Atlas registered to subject-specific MRI scans. In subjects exhibiting seizures (n=4), automated line length detection showed that microelectrode signals identified epileptiform activity sooner and with greater specificity than ring electrodes. In subjects that only seized post-kainate injection (n=3), manual review by a board-certified epileptologist confirmed that microelectrodes provided the earliest onset detection times. Furthermore, the microelectrode arrays high resolution revealed distinct instances of hyperactivity occurring at similar depths but from different directions -- a feature indistinguishable to standard ring electrodes. These results demonstrate that microelectrodes on a large insulating body significantly enhance signal quality and spatial localization. This technology offers a potential advancement over current clinical standards for identifying seizure foci during surgical planning. SIGNIFICANCESurgical treatment for drug-resistant epilepsy depends on accurately identifying the brain regions where seizures begin. Current stereo-EEG electrodes sample activity with ring contacts that average signals around the probe shaft, potentially obscuring directional differences in nearby neural activity. This study shows that microelectrodes distributed around an sEEG-sized insulating body can improve seizure-related signal detection and spatial localization compared with ring-like recordings from the same implant locations. By resolving activity that conventional ring electrodes cannot distinguish, high-density directional sEEG may provide more informative recordings for seizure mapping. These findings support the development of next-generation intracranial electrodes for improving epilepsy surgical planning.

ObjectiveElectroencephalography (EEG) is a key tool for studying auditory processing in cochlear implant (CI) users. In particular, EEG recordings obtained during continuous speech are becoming increasingly important for assessing speech and language processing in CI users, and may be utilized for neurofeedback. However, CIs also induce strong stimulation artifacts that are time-locked to the stimulus and mask the neural responses that have smaller magnitudes. Existing artifact reduction methods are typically based on event-related potentials (ERPs) or require manual component selection, making them unsuitable for naturalistic listening conditions or large datasets. ApproachWe develop CORSICA (CORrelation-baSed ICA artifact rejection), a reproducible, parameter-efficient method for CI artifact reduction in EEG responses to continuous speech. CORSICA operates on independent components (ICs) obtained through Infomax ICA and requires no manual component labelling, with performance governed by a single tunable threshold. It exploits the observation that CI artifacts temporally follow the audio signal without delay, whereas neural responses have an inherent lag due to auditory pathway latencies. For each IC, CORSICA computes the cross-correlation with the speech stimulus. Artifacts are identified by a high signal-to-noise ratio (SNR) of the correlation peak near zero lag, and the component is rejected if this SNR exceeds a threshold. To benchmark CORSICA, we evaluate two alternatives: a TRF-based SNR method, in which temporal response functions are fitted to each IC and artifact-driven peaks near zero lag are used for rejection, and a variant replacing ICA with second-order blind identification (SOBI) as the source separation step. Main resultsCORSICA effectively suppressed CI artifacts while preserving neural activity, enabling recovery of physiologically plausible TRFs with only 2% of ICs rejected. Both benchmark methods confirmed the validity of the SNR-based rejection framework, but CORSICA outperformed the TRF-based alternative in artifact suppression quality. Replacing ICA with SOBI as the source separation step required more ICs to be rejected, further supporting ICA as the preferred backbone for CORSICA. SignificanceCORSICA provides a fully objective, label-free approach to identifying CI artifacts in speech-evoked EEG data, with no manual intervention required. By centering artifact rejection on a single interpretable threshold, it offers a reproducible preprocessing standard for future EEG studies on speech processing in CI users. ConclusionOur findings demonstrate that objective CI artifact suppression in speech-evoked EEG data is feasible on the basis of the ICs temporal response patterns.

Electroretinography (ERG) measures the functional response of distinct retinal cells to light, but was largely displaced by structural imaging in the 2000s. Standardization efforts by the International Society for Clinical Electrophysiology of Vision (ISCEV) began in the late 1980s, and collapsed the rich time-series traces into reproducible components and implicit times. Recent improvements in hardware (RETeval) and software (artificial intelligence) may increase the utility of ERG data. However, no ERG-specific foundation models exist, and there are not enough public datasets to train one. We asked whether time-series foundation models (FMs) trained without ERG-specific pre-training could be adapted through transfer learning. Using two public datasets, PERG-IOBA (pattern ERG with ocular diagnoses), and LEOPs (full-field ERG focusing on Autism Spectrum Disorder, ASD), we interrogated how FMs could improve over smaller within-domain models. We measured the binary (healthy/typically developing vs any annotation) and multiclass (specific family/diagnosis) classification performance of both frozen and fine-tuned FMs, alongside custom autoencoder and multiscale models, using patient-aware splits for cross validation. We benchmark the same architectures against PTB-XL, a large 12-lead ECG corpus, as both a control for each approach and to explore scaling behavior. We show that 1) pre-trained FMs can reconstruct masked traces from all three datasets, 2) frozen and fine-tuned embeddings, especially combined with multimodal metadata through masked autoencoders, performed best on classification tasks. Performance on PTB-XL was maintained down to 300 records, comparable in size to the ERG datasets. We could not reproduce published classification performance on the ASD task. Taken together, these results support general purpose foundation models as a practical approach to ERG analysis.

Substantial progress has been made in recent years on understanding how the human brain parses and processes natural speech. Much of this progress has been based on modeling how brain activity relates to the different acoustic and linguistic features of speech. By fitting and testing models based on those features, one can test hypotheses about the kinds of computations and representations the brain uses to convert speech sounds into understanding. While much of this work has focused on modeling BOLD activity using functional neuroimaging or intracranially recorded electrophysiological signals, the approach has also proven useful with MEG and EEG. Indeed, noninvasive EEG has certain advantages for studying speech processing in terms of translational research and application. Research over the last decade or so has shown that EEG can be successfully modeled based on numerous acoustic, linguistic, and paralinguistic speech features. However, an important unanswered question hangs over all of this work: namely, what constitutes a good model of EEG responses to natural speech? Or, to put it another way, how much variance in EEG recorded during natural speech listening is explainable as having derived from that speech input? The present study aims to tackle this issue. We do so under the assumption that the best model for a person's EEG response to natural speech is a set of EEG responses from other people listening to the same speech. Using this assumption, we construct inter-subject models using EEG from 19 healthy adult native speakers of English who all listened to the same audiobook. The model for each subject involves predicting their EEG data using (dimensionality-reduced) EEG from different numbers of other subjects and then extrapolating to estimate the total explainable variance in the target individual's response to speech. Following this, we show that linear models (temporal response functions) based on several commonly used acoustic and linguistic speech features can predict most - but importantly not all - of the estimated total explainable variance in EEG responses across subjects.

Crossmodal plasticity refers to the phenomenon by which typical sensory areas respond to stimulation in other sensory modalities when their main input is absent or reduced. In this study, we examined the effect of age-related hearing loss (ARHL) on crossmodal plasticity in auditory brain regions. ARHL is prevalent in a large proportion of older adults, and it is associated with an increased risk of cognitive decline and dementia. Understanding how the brain adapts to ARHL is essential in aiding the development of adequate therapies and interventions for a phenomenon experienced by most adults aged 65 and over. Previous research in congenitally deaf adults has shown that crossmodal plasticity effects are stronger for conditions that require higher executive demands. We conducted an fMRI experiment in older adults to determine whether crossmodal recruitment of auditory cortical regions during higher executive demands also occurs following ARHL. In the MRI scanner, participants with and without ARHL completed a visual task-switching paradigm and a visual working memory task, each comprising high and low executive-demand conditions. Results from both groups of participants showed that the high-demand conditions reliably activated canonical frontoparietal regions involved in executive function and cognitive control. Crucially, we also observed significant recruitment of auditory regions during these visual tasks, particularly under the higher executive demands condition of the task-switching paradigm. Crossmodal activations in auditory areas occurred in both groups, with no significant effects of hearing level or age. These findings indicate that auditory regions are involved in high executive demand visual processing, and that crossmodal plasticity effects are not restricted to early sensitive periods. We propose that reduced auditory input and age-related changes in cortical processing jointly contribute to the reorganisation of the auditory cortex in later life. Understanding these mechanisms is critical for clarifying the neural consequences of ARHL and their impact on cognitive decline.

Deep brain stimulation (DBS) is transforming from a static therapy toward adaptive systems that adjust stimulation based on neural biomarkers. However, the detection of reliable biomarkers that capture the multi-dimensional nature of complex symptoms is often challenging. Here we demonstrate volitional DBS (vDBS)--a paradigm in which patients use brain-computer interface (BCI) training to learn self-regulation of a neural signal that then controls closed-loop DBS. Two patients with Parkinson's disease implanted with sensing-enabled neurostimulators completed chronic, at-home BCI training by playing an airplane simulation game. Through training, they were able to effectively down-regulate their cortical beta signal (p's < 1e-10), represented as the real-time position of a plane in the BCI game. Following training, this cortical beta signal served as the input to a closed-loop DBS algorithm. By modulating their beta signal to cross personalized thresholds, patients voluntarily increased or decreased neurostimulation amplitude at will, in the absence of physical movement (p's < 1e-10). This proof-of-principle demonstration establishes that volitional control of intracranial neurostimulation is achievable without the need of an externalized manual controller. BCI-vDBS could potentially be used for a range of neuropsychiatric conditions and brain rehabilitation to support personalized control of neurostimulation.

This study examined the extent to which goodness of pronunciation (GoP) scores and phonological posterior probabilities capture perceptual ratings of speech severity in individuals with motor speech disorders (MSD). Speech recordings of the word catastrophe were obtained from 489 participants, including 333 neurologically typical controls and 156 individuals with MSD. GoP scores were derived using traditional acoustic features and self-supervised speech representations, including WavLM and XLS-R, across multiple modeling approaches, while phonological posterior probabilities were extracted using Phonet. Model performance was evaluated using Kendall's rank correlations, regression, and receiver operating characteristic analyses against speech-language pathologists' perceptual ratings of sound distortion and intelligibility. Both GoP and phonological posterior probabilities were significantly associated with perceptual ratings. Self-supervised speech representations substantially outperformed traditional acoustic features, with WavLM-based GoP using k-nearest neighbors achieving the strongest performance. Across correlation, regression, and classification analyses, GoP consistently outperformed phonological posterior probabilities for both sound distortion and intelligibility. Age and gender had minimal influence on model-derived measures or their relationships with perceptual ratings. These findings demonstrate the value of self-supervised GoP as an objective measure of speech impairment while highlighting the complementary role of phonological posterior probabilities in characterizing articulatory aspects of motor speech disorders.

Carotid endarterectomy carries the risk of intraoperative cerebral ischemia, which is monitored by expert neurophysiologists through continuous electroencephalography (cEEG). Because expert availability is limited, we developed a hybrid novice-artificial intelligence (AI) system that detects ischemia using novice monitors with limited cEEG training. The hybrid system dynamically weights novice and AI inputs to arrive at a final output. Using four novices, we compared hybrid systems against experts alone, novices alone, and AI alone. Hybrid systems were statistically non-inferior to experts in sensitivity and false-positive rate (FPR), whereas novices alone were not. At 80% sensitivity, hybrid systems reduced FPR by half compared with the AI-only system, with similar benefits at 90% sensitivity. Further, the area under the precision-recall curve improved from 0.546 to 0.610-0.726, the area under the receiver operating characteristic curve improved from 0.957 to 0.967-0.971, and calibration improved compared with AI alone. These results highlight the potential of a hybrid system to monitor intraoperative cerebral ischemia.