Abstract Brain activity unfolds across space and time, giving rise to spatiotemporal dynamics that can manifest as cortical traveling waves –smooth phase shifts propagating across the cortex. Although traveling waves have been observed across species and measurement scales, their functional role in human cognition remains largely inferred from correlational evidence. In healthy humans, large-scale neural dynamics must be studied non-invasively, which provides limited access to their neural origins. Here, we tested whether global traveling waves causally contribute to inter-areal communication during attention. Using transcranial magnetic stimulation (TMS), we induced long-range traveling waves detectable with electroencephalography and assessed their neural and behavioral consequences during an attentional search task. Double-pulse TMS applied over the right frontal eye field transiently disrupted ongoing global waves and selectively induced direction- and frequency-specific traveling waves propagating toward the occipital visual cortex. These TMS-induced anterior-to-posterior theta traveling waves were rhythmically modulated by stimulation latency relative to search-trial onset. Critically, behavioral performance exhibited the same rhythmic pattern, with improved attentional search performance when TMS-induced traveling waves were more prominent. Together, these findings provide causal evidence that large-scale traveling waves support inter-areal communication during rhythmic visual attention.

Humans perform tasks involving the manipulation of inputs regardless of how these signals are perceived by the brain, thanks to representations that are invariant to the stimulus modality. In this paper, we present modality-agnostic decoders that leverage such modality-invariant representations to predict which stimulus a subject is seeing, irrespective of the modality in which the stimulus is presented. Training these modality-agnostic decoders is made possible thanks to our new large-scale fMRI dataset SemReps-8K, released publicly along with this paper. It comprises six subjects watching both images and short text descriptions of such images, as well as the conditions during which the subjects were imagining visual scenes. We find that modality-agnostic decoders can perform as well as modality-specific decoders and even outperform them when decoding captions and mental imagery. Furthermore, a searchlight analysis revealed that large areas of the brain contain modality-invariant representations. Such areas are also particularly suitable for decoding visual scenes from the mental imagery condition.

Global Workspace Theory (GWT), inspired by cognitive neuroscience, posits that flexible cognition could arise via the attentional selection of a relevant subset of modalities within a multimodal integration system. This cognitive framework can inspire novel computational architectures for multimodal integration. Indeed, recent implementations of GWT have explored its multimodal representation capabilities, but the related attention mechanisms remain understudied. Here, we propose and evaluate a top-down attention mechanism to select modalities inside a global workspace. First, we demonstrate that our attention mechanism improves noise robustness of a global workspace system on two multimodal datasets of increasing complexity: Simple Shapes and MM-IMDb 1.0. Second, we highlight various cross-task and cross-modality generalization capabilities that are not shared by multimodal attention models from the literature. Comparing against existing baselines on the MM-IMDb 1.0 benchmark, we find our attention mechanism makes the global workspace competitive with the state of the art.

We present a model inspired by the Global Workspace Theory that integrates specialized modules to perform a sequential reasoning task. A controller selectively routes information between modules through the workspace using a gating mechanism. This approach allows the model to chain operations by iteratively broadcasting information between specialized domains, mimicking System-2 reasoning. We evaluate the model's performance on a simple addition task, where two addends must be summed. The task can be solved by routing information sequentially through an Input module, an Increment module (multiple times), and finally an Output module. We consider two implementations of this system with increasing complexity. First, using hand-designed modules operating on one-hot digit representations, the controller (a LSTM recurrent network) learns to select the appropriate modules (input, increment, output) in the appropriate sequence. Second, we replace the hand-designed modules with learned representation modules for MNIST images and an increment module trained on the task objectives; here again, the controller learns the appropriate sequential module selection to solve the task. Finally, we show that the Global Workspace model, while having fewer parameters, outperforms LSTMs and Transformers when tested on unseen addition operations (both interpolations and extrapolations of addition operations seen during training). Our results highlight the potential of architectures inspired by the Global Workspace Theory to enhance deep learning's reasoning capabilities.

What is the shortest path between two data points lying in a high-dimensional space? While the answer is trivial in Euclidean geometry, it becomes significantly more complex when the data lies on a curved manifold -- requiring a Riemannian metric to describe the space's local curvature. Estimating such a metric, however, remains a major challenge in high dimensions. In this work, we propose a method for deriving Riemannian metrics directly from pretrained Energy-Based Models (EBMs) -- a class of generative models that assign low energy to high-density regions. These metrics define spatially varying distances, enabling the computation of geodesics -- shortest paths that follow the data manifold's intrinsic geometry. We introduce two novel metrics derived from EBMs and show that they produce geodesics that remain closer to the data manifold and exhibit lower curvature distortion, as measured by alignment with ground-truth trajectories. We evaluate our approach on increasingly complex datasets: synthetic datasets with known data density, rotated character images with interpretable geometry, and high-resolution natural images embedded in a pretrained VAE latent space. Our results show that EBM-derived metrics consistently outperform established baselines, especially in high-dimensional settings. Our work is the first to derive Riemannian metrics from EBMs, enabling data-aware geodesics and unlocking scalable, geometry-driven learning for generative modeling and simulation.

Objects we encounter often change appearance as we interact with them. Changes in illumination (shadows), object pose, or the movement of non-rigid objects can drastically alter available image features. How do biological visual systems track objects as they change? One plausible mechanism involves attentional mechanisms for reasoning about the locations of objects independently of their appearances --- a capability that prominent neuroscience theories have associated with computing through neural synchrony. Here, we describe a novel deep learning circuit that can learn to precisely control attention to features separately from their location in the world through neural synchrony: the complex-valued recurrent neural network (CV-RNN). Next, we compare object tracking in humans, the CV-RNN, and other deep neural networks (DNNs), using FeatureTracker: a large-scale challenge that asks observers to track objects as their locations and appearances change in precisely controlled ways. While humans effortlessly solved FeatureTracker, state-of-the-art DNNs did not. In contrast, our CV-RNN behaved similarly to humans on the challenge, providing a computational proof-of-concept for the role of phase synchronization as a neural substrate for tracking appearance-morphing objects as they move about.

In AI, the existential risk denotes the hypothetical threat posed by an artificial system that would possess both the capability and the objective, either directly or indirectly, to eradicate humanity. This issue is gaining prominence in scientific debate due to recent technical advancements and increased media coverage. In parallel, AI progress has sparked speculation and studies about the potential emergence of artificial consciousness. The two questions, AI consciousness and existential risk, are sometimes conflated, as if the former entailed the latter. Here, I explain that this view stems from a common confusion between consciousness and intelligence. Yet these two properties are empirically and theoretically distinct. Arguably, while intelligence is a direct predictor of an AI system's existential threat, consciousness is not. There are, however, certain incidental scenarios in which consciousness could influence existential risk, in either direction. Consciousness could be viewed as a means towards AI alignment, thereby lowering existential risk; or, it could be a precondition for reaching certain capabilities or levels of intelligence, and thus positively related to existential risk. Recognizing these distinctions can help AI safety researchers and public policymakers focus on the most pressing issues.

We propose a mathematical framework to systematically explore the propagation properties of a class of continuous in time nonlinear neural network models comprising a hierarchy of processing areas, mutually connected according to the principles of predictive coding. We precisely determine the conditions under which upward propagation, downward propagation or even propagation failure can occur in both bi-infinite and semi-infinite idealizations of the model. We also study the long-time behavior of the system when either a fixed external input is constantly presented at the first layer of the network or when this external input consists in the presentation of constant input with large amplitude for a fixed time window followed by a reset to a down state of the network for all later times. In both cases, we numerically demonstrate the existence of threshold behavior for the amplitude of the external input characterizing whether or not a full propagation within the network can occur. Our theoretical results are consistent with predictive coding theories and allow us to identify regions of parameters that could be associated with dysfunctional perceptions.

Humans leverage rich internal models of the world to reason about the future, imagine counterfactuals, and adapt flexibly to new situations. In Reinforcement Learning (RL), world models aim to capture how the environment evolves in response to the agent's actions, facilitating planning and generalization. However, typical world models directly operate on the environment variables (e.g. pixels, physical attributes), which can make their training slow and cumbersome; instead, it may be advantageous to rely on high-level latent dimensions that capture relevant multimodal variables. Global Workspace (GW) Theory offers a cognitive framework for multimodal integration and information broadcasting in the brain, and recent studies have begun to introduce efficient deep learning implementations of GW. Here, we evaluate the capabilities of an RL system combining GW with a world model. We compare our GW-Dreamer with various versions of the standard PPO and the original Dreamer algorithms. We show that performing the dreaming process (i.e., mental simulation) inside the GW latent space allows for training with fewer environment steps. As an additional emergent property, the resulting model (but not its comparison baselines) displays strong robustness to the absence of one of its observation modalities (images or simulation attributes). We conclude that the combination of GW with World Models holds great potential for improving decision-making in RL agents.

Recent deep learning models can efficiently combine inputs from different modalities (e.g., images and text) and learn to align their latent representations, or to translate signals from one domain to another (as in image captioning, or text-to-image generation). However, current approaches mainly rely on brute-force supervised training over large multimodal datasets. In contrast, humans (and other animals) can learn useful multimodal representations from only sparse experience with matched cross-modal data. Here we evaluate the capabilities of a neural network architecture inspired by the cognitive notion of a "Global Workspace": a shared representation for two (or more) input modalities. Each modality is processed by a specialized system (pretrained on unimodal data, and subsequently frozen). The corresponding latent representations are then encoded to and decoded from a single shared workspace. Importantly, this architecture is amenable to self-supervised training via cycle-consistency: encoding-decoding sequences should approximate the identity function. For various pairings of vision-language modalities and across two datasets of varying complexity, we show that such an architecture can be trained to align and translate between two modalities with very little need for matched data (from 4 to 7 times less than a fully supervised approach). The global workspace representation can be used advantageously for downstream classification tasks and for robust transfer learning. Ablation studies reveal that both the shared workspace and the self-supervised cycle-consistency training are critical to the system's performance.

In recent years, Transformer architectures have revolutionized most fields of artificial intelligence, relying on an attentional mechanism based on the agreement between keys and queries to select and route information in the network. In previous work, we introduced a novel, brain-inspired architecture that leverages a similar implementation to achieve a global 'routing by agreement' mechanism. Such a system modulates the network's activity by matching each neuron's key with a single global query, pooled across the entire network. Acting as a global attentional system, this mechanism improves noise robustness over baseline levels but is insufficient for multi-classification tasks. Here, we improve on this work by proposing a novel mechanism that combines aspects of the Transformer attentional operations with a compelling neuroscience theory, namely, binding by synchrony. This theory proposes that the brain binds together features by synchronizing the temporal activity of neurons encoding those features. This allows the binding of features from the same object while efficiently disentangling those from distinct objects. We drew inspiration from this theory and incorporated angular phases into all layers of a convolutional network. After achieving phase alignment via Kuramoto dynamics, we use this approach to enhance operations between neurons with similar phases and suppresses those with opposite phases. We test the benefits of this mechanism on two datasets: one composed of pairs of digits and one composed of a combination of an MNIST item superimposed on a CIFAR-10 image. Our results reveal better accuracy than CNN networks, proving more robust to noise and with better generalization abilities. Overall, we propose a novel mechanism that addresses the visual binding problem in neural networks by leveraging the synergy between neuroscience and machine learning.

Neural synchrony is hypothesized to play a crucial role in how the brain organizes visual scenes into structured representations, enabling the robust encoding of multiple objects within a scene. However, current deep learning models often struggle with object binding, limiting their ability to represent multiple objects effectively. Inspired by neuroscience, we investigate whether synchrony-based mechanisms can enhance object encoding in artificial models trained for visual categorization. Specifically, we combine complex-valued representations with Kuramoto dynamics to promote phase alignment, facilitating the grouping of features belonging to the same object. We evaluate two architectures employing synchrony: a feedforward model and a recurrent model with feedback connections to refine phase synchronization using top-down information. Both models outperform their real-valued counterparts and complex-valued models without Kuramoto synchronization on tasks involving multi-object images, such as overlapping handwritten digits, noisy inputs, and out-of-distribution transformations. Our findings highlight the potential of synchrony-driven mechanisms to enhance deep learning models, improving their performance, robustness, and generalization in complex visual categorization tasks.

Rapid progress in artificial intelligence (AI) capabilities has drawn fresh attention to the prospect of consciousness in AI. There is an urgent need for rigorous methods to assess AI systems for consciousness, but significant uncertainty about relevant issues in consciousness science. We present a method for assessing AI systems for consciousness that involves exploring what follows from existing or future neuroscientific theories of consciousness. Indicators derived from such theories can be used to inform credences about whether particular AI systems are conscious. This method allows us to make meaningful progress because some influential theories of consciousness, notably including computational functionalist theories, have implications for AI that can be investigated empirically.

Electrophysiological and neuroimaging studies have revealed how the brain encodes various visual categories and concepts. An open question is how combinations of multiple visual concepts are represented in terms of the component brain patterns: are brain responses to individual concepts composed according to algebraic rules? To explore this, we generated "conceptual perturbations" in neural space by averaging fMRI responses to images with a shared concept (e.g., "winter" or "summer"). After thresholding to ensure specificity, we applied these perturbations to the neural pattern associated with a base image, forming new brain patterns that incorporate the added concept. These modified brain patterns were then decoded into images using a pretrained fMRI-to-image decoding model. Qualitative and quantitative inspection of the resulting images provides insight into how the brain might combine visual concepts. For example, adding a "winter" perturbation to the brain pattern of a man on a skateboard yields a new pattern representing a man on a snowboard in a winter scene-even when the perturbation modifies only a small subset of voxels. Our findings reveal that compositional processes in neural representations may lead to predictable perceptual outcomes, as interpreted by our decoding model. This suggests that the brain's combinatory encoding of concepts may follow a systematic, algebraic-like process-what we term "brain algebra." Although our study is model-driven, it opens avenues for future empirical work into the mechanisms of compositionality in the brain.

Automotive radar sensors provide valuable information for advanced driving assistance systems (ADAS). Radars can reliably estimate the distance to an object and the relative velocity, regardless of weather and light conditions. However, radar sensors suffer from low resolution and huge intra-class variations in the shape of objects. Exploiting the time information (e.g., multiple frames) has been shown to help to capture better the dynamics of objects and, therefore, the variation in the shape of objects. Most temporal radar object detectors use 3D convolutions to learn spatial and temporal information. However, these methods are often non-causal and unsuitable for real-time applications. This work presents RECORD, a new recurrent CNN architecture for online radar object detection. We propose an end-to-end trainable architecture mixing convolutions and ConvLSTMs to learn spatio-temporal dependencies between successive frames. Our model is causal and requires only the past information encoded in the memory of the ConvLSTMs to detect objects. Our experiments show such a method's relevance for detecting objects in different radar representations (range-Doppler, range-angle) and outperform state-of-the-art models on the ROD2021 and CARRADA datasets while being less computationally expensive.

Previous brain decoding research primarily involves single-subject studies, reconstructing stimuli via fMRI activity from the same subject. Our study aims to introduce a generalization technique for cross-subject brain decoding, facilitated by exploring data alignment methods. We utilized the NSD dataset, a comprehensive 7T fMRI vision experiment involving multiple subjects exposed to 9841 images, 982 of which were viewed by all. Our approach involved training a decoding model on one subject, aligning others' data to this space, and testing the decoding on the second subject. We compared ridge regression, hyper alignment, and anatomical alignment techniques for fMRI data alignment. We established that cross-subject brain decoding is feasible, even using around 10% of the total data, or 982 common images, with comparable performance to single-subject decoding. Ridge regression was the best method for functional alignment. Through subject alignment, we achieved superior brain decoding and a potential 90% reduction in scan time. This could pave the way for more efficient experiments and further advancements in the field, typically requiring an exorbitant 20-hour scan time per subject.

Humans perceive the world through multiple senses, enabling them to create a comprehensive representation of their surroundings and to generalize information across domains. For instance, when a textual description of a scene is given, humans can mentally visualize it. In fields like robotics and Reinforcement Learning (RL), agents can also access information about the environment through multiple sensors; yet redundancy and complementarity between sensors is difficult to exploit as a source of robustness (e.g. against sensor failure) or generalization (e.g. transfer across domains). Prior research demonstrated that a robust and flexible multimodal representation can be efficiently constructed based on the cognitive science notion of a 'Global Workspace': a unique representation trained to combine information across modalities, and to broadcast its signal back to each modality. Here, we explore whether such a brain-inspired multimodal representation could be advantageous for RL agents. First, we train a 'Global Workspace' to exploit information collected about the environment via two input modalities (a visual input, or an attribute vector representing the state of the agent and/or its environment). Then, we train a RL agent policy using this frozen Global Workspace. In two distinct environments and tasks, our results reveal the model's ability to perform zero-shot cross-modal transfer between input modalities, i.e. to apply to image inputs a policy previously trained on attribute vectors (and vice-versa), without additional training or fine-tuning. Variants and ablations of the full Global Workspace (including a CLIP-like multimodal representation trained via contrastive learning) did not display the same generalization abilities.