Reading comprehension in L1 and L2 readers: neurocomputational mechanisms revealed through large language models.

While evidence has accumulated to support the argument of shared computational mechanisms underlying language comprehension between humans and large language models (LLMs), few studies have examined this argument beyond native-speaker populations. This study examines whether and how alignment between LLMs and human brains captures the homogeneity and heterogeneity in both first-language (L1) and second-language (L2) readers. We recorded brain responses of L1 and L2 English readers of texts and assessed reading performance against individual difference factors.

Emergence of a temporal processing gradient from naturalistic inputs and network connectivity.

Natural language unfolds over multiple nested timescales: Words form sentences, sentences form paragraphs, and paragraphs build into full narratives. Correspondingly, the brain exhibits a hierarchy of processing timescales, spanning from lower- to higher-order regions. During narrative comprehension, neural activation patterns have been shown to propagate along this cortical hierarchy with increasing temporal delays (lags).

The "Podcast" ECoG dataset for modeling neural activity during natural language comprehension.

Naturalistic electrocorticography (ECoG) data are a rare but essential resource for studying the brain's linguistic capabilities. ECoG offers high temporal resolution suitable for investigating processes at multiple temporal timescales and frequency bands. It also provides broad spatial coverage, often along critical language areas.

Cortical language areas are coupled via a soft hierarchy of model-based linguistic features.

Natural language comprehension is a complex task that relies on coordinated activity across a network of cortical regions. In this study, we propose that regions of the language network are coupled to one another through subspaces of shared linguistic features. To test this idea, we developed a model-based connectivity framework to quantify stimulus-driven, feature-specific functional connectivity language areas during natural language comprehension.

Large language models without grounding recover non-sensorimotor but not sensorimotor features of human concepts.

To what extent can language give rise to complex conceptual representation? Is multisensory experience essential? Recent large language models (LLMs) challenge the necessity of grounding for concept formation: whether LLMs without grounding nevertheless exhibit human-like representations. Here we compare multidimensional representations of ~4,442 lexical concepts between humans (the Glasgow Norms, N = 829; and the Lancaster Norms, N = 3,500) and state-of-the-art LLMs with and without visual learning, across non-sensorimotor, sensory and motor domains. We found that (1) the similarity between model and human representations decreases from non-sensorimotor to sensory domains and is minimal in motor domains, indicating a systematic divergence, and (2) models with visual learning exhibit enhanced similarity with human representations in visual-related dimensions.

Monitoring Attention in Self and Others.

The ability of the brain to monitor its own attention is important for controlling attention. The ability to reconstruct and monitor the attention of others is important for behavioral prediction and therefore interaction with others. Do the same cortical networks participate in constructing a metacognitive representation of attention, whether one's own or someone else's attention? We studied the brain activity of human participants in an fMRI scanner.

Context modulates brain state dynamics and behavioral responses during narrative comprehension.

Narrative comprehension is inherently context-sensitive, yet the brain and cognitive mechanisms by which brief contextual priming shapes story interpretation remain unclear. Using hidden Markov modeling (HMM) of fMRI data, we identified dynamic brain states as participants listened to an ambiguous spoken story under two distinct narrative contexts (affair vs. paranoia).

A unified acoustic-to-speech-to-language embedding space captures the neural basis of natural language processing in everyday conversations.

This study introduces a unified computational framework connecting acoustic, speech and word-level linguistic structures to study the neural basis of everyday conversations in the human brain. We used electrocorticography to record neural signals across 100 h of speech production and comprehension as participants engaged in open-ended real-life conversations. We extracted low-level acoustic, mid-level speech and contextual word embeddings from a multimodal speech-to-text model (Whisper).

Linguistic coupling between neural systems for speech production and comprehension during real-time dyadic conversations.

The core use of human language is communicating complex ideas from one mind to another in everyday conversations. In conversations, comprehension and production processes are intertwined, as speakers soon become listeners, and listeners become speakers. Nonetheless, the neural systems underlying these faculties are typically studied in isolation using paradigms that cannot fully engage our capacity for interactive communication.

The "Podcast" ECoG dataset for modeling neural activity during natural language comprehension.

Naturalistic electrocorticography (ECoG) data are a rare but essential resource for studying the brain's linguistic capabilities. ECoG offers a high temporal resolution suitable for investigating processes at multiple temporal timescales and frequency bands. It also provides broad spatial coverage, often along critical language areas.

Behaviorally-relevant features of observed actions dominate cortical representational geometry in natural vision.

We effortlessly extract behaviorally relevant information from dynamic visual input in order to understand the actions of others. In the current study, we develop and test a number of models to better understand the neural representational geometries supporting action understanding. Using fMRI, we measured brain activity as participants viewed a diverse set of 90 different video clips depicting social and nonsocial actions in real-world contexts.

Behaviorally-relevant features of observed actions dominate cortical representational geometry in natural vision.

We effortlessly extract behaviorally relevant information from dynamic visual input in order to understand the actions of others. In the current study, we develop and test a number of models to better understand the neural representational geometries supporting action understanding. Using fMRI, we measured brain activity as participants viewed a diverse set of 90 different video clips depicting social and nonsocial actions in real-world contexts.

How a speaker herds the audience: multibrain neural convergence over time during naturalistic storytelling.

Storytelling-an ancient way for humans to share individual experiences with others-has been found to induce neural alignment among listeners. In exploring the dynamic fluctuations in listener-listener (LL) coupling throughout stories, we uncover a significant correlation between LL coupling and lagged speaker-listener (lag-SL) coupling over time. Using the analogy of neural pattern (dis)similarity as distances between participants, we term this phenomenon the "herding effect.

A shared model-based linguistic space for transmitting our thoughts from brain to brain in natural conversations.

Effective communication hinges on a mutual understanding of word meaning in different contexts. We recorded brain activity using electrocorticography during spontaneous, face-to-face conversations in five pairs of epilepsy patients. We developed a model-based coupling framework that aligns brain activity in both speaker and listener to a shared embedding space from a large language model (LLM).

Scale matters: Large language models with billions (rather than millions) of parameters better match neural representations of natural language.

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.

Shared functional specialization in transformer-based language models and the human brain.

When processing language, the brain is thought to deploy specialized computations to construct meaning from complex linguistic structures. Recently, artificial neural networks based on the Transformer architecture have revolutionized the field of natural language processing. Transformers integrate contextual information across words via structured circuit computations.

Alignment of brain embeddings and artificial contextual embeddings in natural language points to common geometric patterns.

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.

Onscreen presence of instructors in video lectures affects learners' neural synchrony and visual attention during multimedia learning.

COVID-19 forced students to rely on online learning using multimedia tools, and multimedia learning continues to impact education beyond the pandemic. In this study, we combined behavioral, eye-tracking, and neuroimaging paradigms to identify multimedia learning processes and outcomes. College students viewed four video lectures including slides with either an onscreen human instructor, an animated instructor, or no onscreen instructor.

Cross-movie prediction of individualized functional topography.

Participant-specific, functionally defined brain areas are usually mapped with functional localizers and estimated by making contrasts between responses to single categories of input. Naturalistic stimuli engage multiple brain systems in parallel, provide more ecologically plausible estimates of real-world statistics, and are friendly to special populations. The current study shows that cortical functional topographies in individual participants can be estimated with high fidelity from naturalistic stimuli.

How a speaker herds the audience: Multi-brain neural convergence over time during naturalistic storytelling.

Storytelling-an ancient way for humans to share individual experiences with others-has been found to induce neural synchronization among listeners. In our exploration of the dynamic fluctuations in listener-listener (LL) coupling throughout stories, we uncover a significant correlation between LL and lag-speaker-listener (lag-SL) couplings over time. Using the analogy of neural pattern (dis)similarity as distances between participants, we term this phenomenon the "herding effect": like a shepherd guiding a group of sheep, the more closely listeners follow the speaker's prior brain activity patterns (higher lag-SL similarity), the more tightly they cluster together (higher LL similarity).

Modeling naturalistic face processing in humans with deep convolutional neural networks.

Deep convolutional neural networks (DCNNs) trained for face identification can rival and even exceed human-level performance. The ways in which the internal face representations in DCNNs relate to human cognitive representations and brain activity are not well understood. Nearly all previous studies focused on static face image processing with rapid display times and ignored the processing of naturalistic, dynamic information.

A shared linguistic space for transmitting our thoughts from brain to brain in natural conversations.

Effective communication hinges on a mutual understanding of word meaning in different contexts. The embedding space learned by large language models can serve as an explicit model of the shared, context-rich meaning space humans use to communicate their thoughts. We recorded brain activity using electrocorticography during spontaneous, face-to-face conversations in five pairs of epilepsy patients.

The single- and dual-brain mechanisms underlying the adviser's confidence expression strategy switching during influence management.

Effective influence management during advice-giving requires individuals to express confidence in the advice properly and switch timely between the 'competitive' strategy and the 'defensive' strategy. However, how advisers switch between these two strategies, and whether and why there exist individual differences during this process remain elusive. We used an advice-giving game that manipulated incentive contexts (Incentivized/Non-Incentivized) to induce the adviser's confidence expression strategy switching and measured the brain activities of adviser and advisee concurrently using functional near-infrared spectroscopy (fNIRS).