Faculty Bibliography

OBJECTIVE/UNASSIGNED:This study investigates speech decoding from neural signals captured by intracranial electrodes. Most prior works can only work with electrodes on a 2D grid (i.e., Electrocorticographic or ECoG array) and data from a single patient. We aim to design a deep-learning model architecture that can accommodate both surface (ECoG) and depth (stereotactic EEG or sEEG) electrodes. The architecture should allow training on data from multiple participants with large variability in electrode placements and the trained model should perform well on participants unseen during training. APPROACH/UNASSIGNED:We propose a novel transformer-based model architecture named SwinTW that can work with arbitrarily positioned electrodes, by leveraging their 3D locations on the cortex rather than their positions on a 2D grid. We train both subject-specific models using data from a single participant as well as multi-patient models exploiting data from multiple participants. MAIN RESULTS/UNASSIGNED:The subject-specific models using only low-density 8x8 ECoG data achieved high decoding Pearson Correlation Coefficient with ground truth spectrogram (PCC=0.817), over N=43 participants, outperforming our prior convolutional ResNet model and the 3D Swin transformer model. Incorporating additional strip, depth, and grid electrodes available in each participant (N=39) led to further improvement (PCC=0.838). For participants with only sEEG electrodes (N=9), subject-specific models still enjoy comparable performance with an average PCC=0.798. The multi-subject models achieved high performance on unseen participants, with an average PCC=0.765 in leave-one-out cross-validation. SIGNIFICANCE/UNASSIGNED:The proposed SwinTW decoder enables future speech neuroprostheses to utilize any electrode placement that is clinically optimal or feasible for a particular participant, including using only depth electrodes, which are more routinely implanted in chronic neurosurgical procedures. Importantly, the generalizability of the multi-patient models suggests the exciting possibility of developing speech neuroprostheses for people with speech disability without relying on their own neural data for training, which is not always feasible.

Recent research has used large language models (LLMs) to study the neural basis of naturalistic language processing in the human brain. LLMs have rapidly grown in complexity, leading to improved language processing capabilities. However, neuroscience researchers haven't kept up with the quick progress in LLM development. Here, we utilized several families of transformer-based LLMs to investigate the relationship between model size and their ability to capture linguistic information in the human brain. Crucially, a subset of LLMs were trained on a fixed training set, enabling us to dissociate model size from architecture and training set size. We used electrocorticography (ECoG) to measure neural activity in epilepsy patients while they listened to a 30-minute naturalistic audio story. We fit electrode-wise encoding models using contextual embeddings extracted from each hidden layer of the LLMs to predict word-level neural signals. In line with prior work, we found that larger LLMs better capture the structure of natural language and better predict neural activity. We also found a log-linear relationship where the encoding performance peaks in relatively earlier layers as model size increases. We also observed variations in the best-performing layer across different brain regions, corresponding to an organized language processing hierarchy.

BACKGROUND AND OBJECTIVES/OBJECTIVE:Generalized convulsive seizures (GCSs) are the main risk factor of sudden unexpected death in epilepsy (SUDEP), which is likely due to peri-ictal cardiorespiratory dysfunction. The incidence of GCS-induced cardiac arrhythmias, their relationship to seizure severity markers, and their role in SUDEP physiopathology are unknown. The aim of this study was to analyze the incidence of seizure-induced cardiac arrhythmias, their association with electroclinical features and seizure severity biomarkers, as well as their specific occurrences in SUDEP cases. METHODS:-score test for 2 population proportions was used to test whether the proportion of seizures and patients with postconvulsive ESAWB or bradycardia differed between SUDEP cases and survivors. RESULTS:> 0.05). DISCUSSION/CONCLUSIONS:Markers of seizure severity are not related to seizure-induced arrhythmias of interest, suggesting that other factors such as occult cardiac abnormalities may be relevant for their occurrence. Seizure-induced ESAWB and bradycardia were more frequent in SUDEP cases, although this observation was based on a very limited number of SUDEP patients. Further case-control studies are needed to evaluate the yield of arrhythmias of interest along with respiratory changes as potential SUDEP biomarkers.

OBJECTIVE:Dravet syndrome is a rare, early childhood-onset epileptic and developmental encephalopathy. Responses to placebo in clinical trials for epilepsy therapies range widely, but factors influencing placebo response remain poorly understood. This study explored placebo response and its effects on safety, efficacy, and quality of life outcomes in patients with Dravet syndrome. METHODS:We performed exploratory post-hoc analyses of pooled data from placebo-treated patients from the GWPCARE 1B and GWPCARE 2 randomized controlled phase III trials, comparing cannabidiol and matched placebo in 2-18 year old Dravet syndrome patients. All patients had ≥4 convulsive seizures during a baseline period of 4 weeks. RESULTS:124 Dravet syndrome-treated patients were included in the analysis (2-5 years: n = 35; 6-12 years: n = 52; 13-18 years: n = 37). Convulsive seizures were experienced by all placebo group patients at all timepoints, with decreased median convulsive seizure frequency during the treatment period versus baseline; the number of convulsive seizure-free days was similar to baseline. Convulsive seizure frequency had a nominally significant positive correlation with age and a nominally significant negative correlation with body mass index. Most placebo-treated patients experienced a treatment-emergent adverse event; however, most resolved quickly, and serious adverse events were infrequent. Placebo treatment had very little effect on reported Caregiver Global Impression of Change outcomes versus baseline. INTERPRETATION/CONCLUSIONS:Placebo had little impact on convulsive seizure-free days and Caregiver Global Impression of Change versus baseline, suggesting that these metrics may help differentiate placebo and active treatment effects in future studies. However, future research should further assess placebo responses to confirm these results.

OBJECTIVE:Dravet syndrome is a rare developmental epilepsy syndrome associated with severe, treatment-resistant seizures. Since seizures and seizure clusters are linked to morbidity, reduced quality of life, and premature mortality, a greater understanding of these outcomes could improve trial designs. This analysis explored seizure types, seizure clusters, and factors affecting seizure cluster variability in Dravet syndrome patients. METHODS:Pooled post-hoc analyses were performed on data from placebo-treated patients in GWPCARE 1B and GWPCARE 2 randomized controlled phase III trials comparing cannabidiol and placebo in Dravet syndrome patients aged 2-18 years. Multivariate stepwise analysis of covariance of log-transformed convulsive seizure cluster frequency was performed, body weight and body mass index z-scores were calculated, and incidence of adverse events was assessed. Data were summarized in three age groups. RESULTS:We analyzed 124 placebo-treated patients across both studies (2-5 years: n = 35; 6-12 years: n = 52; 13-18 years: n = 37). Generalized tonic-clonic seizures followed by myoclonic seizures were the most frequent seizure types. Mean and median convulsive seizure cluster frequency overall decreased between baseline and maintenance period but did not change significantly during the latter; variation in convulsive seizure cluster frequency was observed across age groups. Multivariate analysis suggested correlations between convulsive seizure cluster frequency and age (positive), and body mass index (BMI) (negative). INTERPRETATION/CONCLUSIONS:Post-hoc analyses suggested that potential relationships could exist between BMI, age and convulsive seizure cluster variation. Results suggested that seizure cluster frequency may be a valuable outcome in future trials. Further research is needed to confirm our findings.

Neural responses in visual cortex adapt to prolonged and repeated stimuli. While adaptation occurs across the visual cortex, it is unclear how adaptation patterns and computational mechanisms differ across the visual hierarchy. Here we characterize two signatures of short-term neural adaptation in time-varying intracranial electroencephalography (iEEG) data collected while participants viewed naturalistic image categories varying in duration and repetition interval. Ventral- and lateral-occipitotemporal cortex exhibit slower and prolonged adaptation to single stimuli and slower recovery from adaptation to repeated stimuli compared to V1-V3. For category-selective electrodes, recovery from adaptation is slower for preferred than non-preferred stimuli. To model neural adaptation we augment our delayed divisive normalization (DN) model by scaling the input strength as a function of stimulus category, enabling the model to accurately predict neural responses across multiple image categories. The model fits suggest that differences in adaptation patterns arise from slower normalization dynamics in higher visual areas interacting with differences in input strength resulting from category selectivity. Our results reveal systematic differences in temporal adaptation of neural population responses between lower and higher visual brain areas and show that a single computational model of history-dependent normalization dynamics, fit with area-specific parameters, accounts for these differences.

Although sudden unexpected death in epilepsy (SUDEP) is the most feared epilepsy outcome, there is a dearth of SUDEP counseling provided by neurologists. This may reflect limited time, as well as the lack of guidance on the timing and structure for counseling. We evaluated records from SUDEP cases to examine frequency of inpatient and outpatient SUDEP counseling, and whether counseling practices were influenced by risk factors and biomarkers, such as post-ictal generalized EEG suppression (PGES). We found a striking lack of SUDEP counseling despite modifiable SUDEP risk factors; counseling was limited to outpatients despite many patients having inpatient visits within a year of SUDEP. PGES was inconsistently documented and was never included in counseling. There is an opportunity to greatly improve SUDEP counseling by utilizing inpatient settings and prompting algorithms incorporating risk factors and biomarkers.

OBJECTIVE:Epilepsy patients often report memory deficits despite normal objective testing, suggesting that available measures are insensitive or that non-mnemonic factors are involved. The Visual Paired Comparison Task (VPCT) assesses novelty preference, the tendency to fixate on novel images rather than previously viewed items, requiring recognition memory for the "old" images. As novelty preference is a sensitive measure of hippocampal-dependent memory function, we predicted impaired VPCT performance in epilepsy patients compared to healthy controls. METHODS:We assessed 26 healthy adult controls and 31 epilepsy patients (16 focal-onset, 13 generalized-onset, 2 unknown-onset) with the VPCT using delays of 2 or 30 s between encoding and recognition. Fifteen healthy controls and 17 epilepsy patients (10 focal-onset, 5 generalized-onset, 2 unknown-onset) completed the task at 2-, 5-, and 30-minute delays. Subjects also performed standard memory measures, including the Medical College of Georgia (MCG) Paragraph Test, California Verbal Learning Test-Second Edition (CVLT-II), and Brief Visual Memory Test-Revised (BVMT-R). RESULTS:The epilepsy group was high functioning, with greater estimated IQ (p = 0.041), greater years of education (p = 0.034), and higher BVMT-R scores (p = 0.024) compared to controls. Both the control group and epilepsy cohort, as well as focal- and generalized-onset subgroups, had intact novelty preference at the 2- and 30-second delays (p-values ≤ 0.001) and declined at 30 min (p-values > 0.05). Only the epilepsy patients had early declines at 2- and 5-minute delays (controls with intact novelty preference at p = 0.003 and p ≤ 0.001, respectively; epilepsy groups' p-values > 0.05). CONCLUSIONS:Memory for the "old" items decayed more rapidly in overall, focal-onset, and generalized-onset epilepsy groups. The VPCT detected deficits while standard memory measures were largely intact, suggesting that the VPCT may be a more sensitive measure of temporal lobe memory function than standard neuropsychological batteries.

Many developmental and epileptic encephalopathies (DEEs) result from variants in cation channel genes. Using mRNA transfection, we generated and characterised an induced pluripotent stem cell (iPSC) line from the fibroblasts of a male late-onset DEE patient carrying a heterozygous missense variant (E1211K) in Na_v1.2(SCN2A) protein. The iPSC line displays features characteristic of the human iPSCs, colony morphology and expression of pluripotency-associated marker genes, ability to produce derivatives of all three embryonic germ layers, and normal karyotype without SNP array-detectable abnormalities. We anticipate that this iPSC line will aid in the modelling and development of precision therapies for this debilitating condition.

Contextual embeddings, derived from deep language models (DLMs), provide a continuous vectorial representation of language. This embedding space differs fundamentally from the symbolic representations posited by traditional psycholinguistics. We hypothesize that language areas in the human brain, similar to DLMs, rely on a continuous embedding space to represent language. To test this hypothesis, we densely record the neural activity patterns in the inferior frontal gyrus (IFG) of three participants using dense intracranial arrays while they listened to a 30-minute podcast. From these fine-grained spatiotemporal neural recordings, we derive a continuous vectorial representation for each word (i.e., a brain embedding) in each patient. Using stringent zero-shot mapping we demonstrate that brain embeddings in the IFG and the DLM contextual embedding space have common geometric patterns. The common geometric patterns allow us to predict the brain embedding in IFG of a given left-out word based solely on its geometrical relationship to other non-overlapping words in the podcast. Furthermore, we show that contextual embeddings capture the geometry of IFG embeddings better than static word embeddings. The continuous brain embedding space exposes a vector-based neural code for natural language processing in the human brain.