Biology subjects

Paulo, D.

Publications and source records attributed to Paulo, D..

12 recordsLinked to original sources

bioRxiv2026

Electrophysiological features of signals recorded from white matter

Intracranial neurophysiology studies have typically ignored signals from electrodes located in white matter (WM), assuming that their information content is artifactual or related to nearby gray matter (GM). Here, we tested the electrophysiological and functional features of signals recorded from different WM locations. Signals were recorded from 19 patients undergoing intracranial monitoring for drug-resistant epilepsy by means of stereo-electroencephalography (sEEG). Each sEEG electrode was classified into WM or GM based on the surrounding tissue. We obtained recordings from a total of 1,717 sEEG electrode contacts, 36% in WM, while the patients were in awake resting state (5 minutes). For each sEEG electrode, we employed a model-based spectral decomposition to separate periodic and aperiodic components, and we computed signal complexity metrics. For a subset of participants, we computed WM structural information from diffusion-weighted magnetic resonance imaging and we evaluated functional signals during a cognitive control task. Our results show that signals recorded from WM have different spectral features and higher complexity than GM. Complexity correlates positively with fractional anisotropy, and modulations related to behavior during the task were detected in WM. Overall, this indicates that WM signals carry information that may reflect signal propagation across WM fiber tracts.

neuroscience↗

bioRxiv2026

A number simplex in the human medial temporal lobe

Humans handle numbers nimbly, suggesting a richer neural manifold structure than the prevalent mental number line model. In populations of medial temporal lobe (MTL) neurons in humans performing two simple tasks (dot counting and arithmetic), we find robust neural coding of numerosity that results in high dimensional, simplex-shaped manifolds. This shape affords more flexibility than a linear manifold due to its high shattering dimensionality and expressibility. Dot arrays and Arabic numerals evoked distinct simplicial population codes, yet they were linked by a linearly transferable latent structure within the same task. We find similar simplicial geometry of number representations in large language models (LLMs). Moreover, subjects internally computed arithmetic results were decodable during the calculation period, with decoding accuracy correlating with individual mathematical capacity. Finally, linear transformations of simplicial operand representations modeled the brains conversion of operands into decodable results, suggesting that the brains arithmetic procedures have some resemblance to the attention architecture of LLMs. Together, these findings establish a high dimensional representational foundation for numerical cognition in the brain.

neuroscience↗

bioRxiv2026

Polysemanticity in human hippocampal neurons

To comprehend language, the brain must navigate a high-dimensional semantic landscape while seamlessly contextualizing meaning. Inspired by recent advances in the mechanistic interpretability of large language models (LLMs), we hypothesized that the brain utilizes polysemanticity, a coding strategy wherein individual neurons represent multiple semantically unrelated features through high-dimensional superposition (Elhage et al., 2022; Olah et al., 2020). We recorded single-unit activity from the human hippocampus during podcast listening. We found that hippocampal neurons exhibit dense semantic codes characterized by multiple tuning peaks with an overdispersed, isotropic geometry. This geometry satisfies the theoretical requirements for interference minimization in superimposed codes. Furthermore, semantic responses are strongly modulated by lexical and speaker-identity context; nonetheless, the underlying population geometry remains stable. This coding strategy permits rapid contextualization without requiring specialized, context-specific neurons. Indeed, we show clear pattern separation of similar terms, along with pattern completion for held-out words. Together, these results demonstrate that the human brain leverages superposition to solve a universal computational problem: maximizing semantic capacity within a constrained representational space.

neuroscience↗

bioRxiv2026

Neural signatures of impaired semantic contextualization in Autism Spectrum Disorder

Social and communicative deficits are defining characteristics of autism spectrum disorder (ASD). Some theories suggest that these challenges, among other autistic traits, may arise from differences in predictive coding, or how the brain uses context to predict and interpret incoming information. This idea has the potential to link symptoms of autism to specific neurocomputational processes, and is especially promising for communication, whose impairment is a hallmark of ASD. Here we leveraged the ability of large language models (LLMs) to quantify semantic contextualization to analyze a unique dataset of responses from hippocampal neurons obtained during language listening in three mild-to-severe autistic individuals with comorbid epilepsy. Key elements of semantic coding were preserved in all three individuals with ASD: single-neuron response dynamics, representation of word-word semantic relationships, and patterns of context-dependent shifts in meaning. However, relative to controls, ASD resulted in reduced neural signatures of contextualization: (1) neuronal responses were aligned with earlier, less contextual layers of GPT-2, (2) ASD patients had lower effective dimensionality of the neural subspace predicting semantics, (3) neural representations of word meaning were less influenced by preceding context, and (4) neural signatures of lexical surprisal were reduced. Together, these results support theories of autism that emphasize impairments in contextualization, and highlight the power of LLMs as a tool for quantifying the computational basis of neurodevelopmental disorders.

neuroscience↗

bioRxiv2026

Semantic axes in the brain support analogical representations

ABSTRACTIn language models of word meaning, directions in the embedding space often correspond to semantic features that can be reused across different words. For example, a single direction corresponding to gender may differentiate word pairs like "boy/girl", "uncle/aunt" and "king/queen". Here we show that the same principle governs semantically driven neural responses in the human brain. We recorded populations of single neurons during podcast listening and identified word sets with consistent meaning differences. Across fifteen sets, including gender, plural, and negation, we observed consistent vectorial directions, resulting in parallelogram structures within the neural manifold. Deviation from parallelism in large language models (LLMs) predicted corresponding deviations in brain-derived parallelism. Among pronouns, vectors corresponding to case, number and person exhibited parallelogram structures individually and, collectively, obeyed the principle of commutativity, resulting in a prismatic structure. Finally, different semantic variables were preferentially associated with discrete groups of neurons, consistent with energy-efficiency theories. Together, these results establish a geometric foundation for the neural encoding of word meaning.

neuroscience↗

bioRxiv2025

Shared neural geometries for bilingual semantic representations

The human brain has the remarkable ability to comprehend and express similar concepts in multiple languages. To understand how it does so, we examined responses of hippocampal neurons during passive listening, directed speaking, and spontaneous conversation, in both English and Spanish, in a small group of balanced bilinguals. We find a small number of putative cross-language neurons, whose responses to equivalent words (e.g., "tierra" and "earth") are correlated. However, neurons semantic tunings differed substantially by language, suggesting language-specific neural implementations. Instead, the crucial driver of translation was a preserved geometric organization of neural responses between the two languages, one that did not depend on neuron level functional overlap. Indeed, that geometry was implemented by a common set of neurons along distinct readout axes; this difference in readout may help prevent cross-language interference. Together, these results suggest that hippocampus encodes a language-independent internal model for meaning.

neuroscience↗

bioRxiv2025

A semantotopic map in human hippocampus

The hippocampus plays a central role in encoding abstract conceptual and semantic information. However, little is known about the topography of that encoding. We leveraged the rare opportunity to examine neural responses densely along a small portion of the human hippocampus, specifically, the mediolateral axis of the anterior body. We collected responses to passive language listening using Neuropixels probes in three anesthetized patients during clinically indicated neurosurgical procedures. We computed semantic tuning functions for each recording site by regressing threshold crossing events and single unit responses against semantic embeddings from GPT-2, Word2Vec, and SBERT. We find that tuning functions of more distant recording sites are more dissimilar, supporting the hypothesis semantotopic organization. Multiple semantic features showed systematic changes along that axis, including animacy, concreteness, and familiarity; notably, effects were individual-specific. Surprisingly, we also found a small but significant increase in semantic similarity as a function of distance between recording sites, on a shorter spatial scale, suggesting a modest periodic organization. Together, these results demonstrate the presence of a multiscale functional organization of semantics in the hippocampus.

neuroscience↗

bioRxiv2025

Mirror manifolds: partially overlapping neural subspaces for speaking and listening

We utilize internal representations of meaning for two purposes: to understand the words we hear and to generate our own speech. This dual requirement necessitates abstract, modality-agnostic representations. Building on work identifying it as a hub for relational mapping, we hypothesized that the hippocampus supports abstract, cross-person representations, and uses shared semantic geometries to do so. We tested this hypothesis by examining hippocampal activity in a remarkable single-neuron dataset derived from conversational speech. Neurons robustly encoded meanings of both spoken and heard words, and used common geometric embeddings for both, leading to abstract meaning performance. Speaker identity was aligned with meaning via partial subspace alignment, which affords speaker-meaning binding by partitioning meaning by speaker while maintaining cross-speaker generalization. Degrees of subspace rotation varied on a single word level and depended systematically on semantic category. Together, these findings indicate how geometric principles allow for abstract cross-personal meanings while preserving binding to speaker identity.

neuroscience↗

bioRxiv2025

Human neuronal firing is modulated by the frequency of local field potential oscillations

Neural oscillations play a critical role in shaping neuronal firing patterns. While phase-locked neuronal firing ("phase tuning") has been extensively studied in animal models and human invasive recordings, much less is known about whether neurons show preferential firing at specific oscillatory frequencies, termed frequency tuning. Here, we employ human intracranial recordings across several brain regions including hippocampus, entorhinal cortex, anterior and posterior cingulate cortex, and orbitofrontal cortex to test the hypothesis that neurons exhibit frequency-specific firing. We analyzed 357 single units recorded simultaneously with local field potentials in 19 neurosurgical patients during awake resting. We estimated the instantaneous frequency of the LFP using adaptive spectral decomposition and assessed frequency tuning of each neuron while controlling for changes in firing rate unrelated to frequency changes. We found 27% neurons exhibited increased or decreased firing within specific frequencies, most commonly within the low-theta range (<10 Hz). Neurons exhibiting frequency tuning were distinct from those displaying phase tuning, and both types of tuning were observed across multiple brain regions with no anatomical preference. Together, our results demonstrate that the instantaneous frequency of neural oscillations modulates neuronal firing which may serve as an additional mechanism for information processing in the human brain, opening new avenues for frequency-targeted neural stimulation.

neuroscience↗

bioRxiv2025

Generalized brain-state modeling with KenazLBM

The large-scale functional state of a human brain remains difficult to characterize, much less predict. Regardless, techniques have been engineered to electrically neuromodulate the brain to treat a subset of neurologic and psychiatric disorders with moderate efficacy. Accurate characterization of a brains instantaneous functional state has stymied the development of more effective neuromodulation paradigms. Advanced computational methods are required to address this gap and enable large-scale neuroscience. Here we define the concept of generalized brain-state modeling across humans as Large Brain-State Modeling (LBM) and present KenazLBM as the worlds first example. KenazLBM can instantaneously characterize the functional state of a persons brain with raw iEEG data, and predict future brain-states. KenazLBM was trained on over 17.9 billion unique multichannel tokens from people undergoing intracranial electroencephalography (iEEG) recordings, and has learned to interrelate brain-states between people into a common interpretable topology. Most importantly, the model generalizes to unseen subject data with significant recording channel heterogeneity from the training set. We offer KenazLBM as a first generalized brain-state model to serve as a new paradigm of basic neuroscience inquiry and potential translation into neuromodulation therapeutics.

neuroscience↗

bioRxiv2025

Attention is all you need (in the brain): semantic contextualization in human hippocampus

Word meanings in language are contextualized by surrounding words. Inspired by the self-attention mechanism in transformer-based large language models (LLMs), we hypothesized that structural composition in the brain arises from combining canonical (non-contextual) word representations with those of nearby words. We analyzed single unit activity in the human hippocampus, a region involved in semantic and contextual processing, while n=10 participants listened to podcasts. We found that hippocampal neurons encoded word position within a clause, using both ordinal and frequency-domain positional encoding. Moreover, neural responses to specific words reflected both the words own lexical semantics and a weighted sum of the embeddings of preceding words. The relative weighting of these contextualizing words correlated with LLM self-attention weights. These findings suggest that contextualization in the brain makes use of vectorial shifts that have a resemblance to attentional reweighting in LLMs, and highlight the role of the mesial temporal lobe within the broader language network.

neuroscience↗

bioRxiv2025

Complementary roles for hippocampus and anterior cingulate in composing continuous choice

Naturalistic, goal directed behavior often requires continuous actions directed at dynamically changing goals. In this context, the closest analogue to choice is a strategic reweighting of multiple goal-specific control policies in response to shifting environmental pressures. To understand the algorithmic and neural bases of choice in continuous contexts, we examined behavior and brain activity in humans performing a continuous prey-pursuit task. Using a newly developed control-theoretic decomposition of behavior, we find pursuit strategies are well described by a meta-controller dictating a mixture of lower-level controllers, each linked to specific pursuit goals. We find that hippocampal neurons encode the policy blending variable in a value-invariant manner and monitor policy switches after they occur. ACC neurons encode policy switches in a value-dependent manner, with value related modulation detectable several hundred ms before the switch, alongside a ramping increase in mean firing rate toward the switch. Meanwhile, OFC activity is consistent with an encoding of the current value structure of the task, rather than policy switching. Together these results are consistent with a tripartite functional division in which hippocampus serves as a controller over behavior, ACC serves as a meta-controller, and OFC provides a value context signal. Overall, our results shed light onto the complex processes associated with choice during naturalistic continuous interactive behavior.

neuroscience↗