This paper investigates the conformal isometry hypothesis as a potential explanation for the hexagonal periodic patterns in grid cell response maps. We posit that grid cell activities form a high-dimensional vector in neural space, encoding the agent's position in 2D physical space. As the agent moves, this vector rotates within a 2D manifold in the neural space, driven by a recurrent neural network. The conformal hypothesis proposes that this neural manifold is a conformal isometric embedding of 2D physical space, where local physical distance is preserved by the embedding up to a scaling factor (or unit of metric). Such distance-preserving position embedding is indispensable for path planning in navigation, especially planning local straight path segments. We conduct numerical experiments to show that this hypothesis leads to the hexagonal grid firing patterns by learning maximally distance-preserving position embedding, agnostic to the choice of the recurrent neural network. Furthermore, we present a theoretical explanation of why hexagon periodic patterns emerge by minimizing our loss function by showing that hexagon flat torus is maximally distance preserving.

Previous studies have compared neural activities in the visual cortex to representations in deep neural networks trained on image classification. Interestingly, while some suggest that their representations are highly similar, others argued the opposite. Here, we propose a new approach to characterize the similarity of the decision strategies of two observers (models or brains) using decision variable correlation (DVC). DVC quantifies the image-by-image correlation between the decoded decisions based on the internal neural representations in a classification task. Thus, it can capture task-relevant information rather than general representational alignment. We evaluate DVC using monkey V4/IT recordings and network models trained on image classification tasks. We find that model–model similarity is comparable to monkey-monkey similarity, whereas model–monkey similarity is consistently lower. Strikingly, DVC decreases with increasing network performance on ImageNet-1k. Adversarial training does not improve model–monkey similarity in task-relevant dimensions assessed using DVC, although it markedly increases the model–model similarity. Similarly, pre-training on larger datasets does not improve model–monkey similarity. These results suggest a divergence between the task-relevant representations in monkey V4/IT and those learned by models trained on image classification tasks.

Relational cognition-the ability to infer relationships that generalize to novel combinations of objects-is fundamental to human and animal intelligence. Despite this importance, it remains unclear how relational cognition is implemented in the brain due in part to a lack of hypotheses and predictions at the levels of collective neural activity and behavior. Here we discovered, analyzed, and experimentally tested neural networks (NNs) that perform transitive inference (TI), a classic relational task (if A > B and B > C, then A > C). We found NNs that (i) generalized perfectly, despite lacking overt transitive structure prior to training, (ii) generalized when the task required working memory (WM), a capacity thought to be essential to inference in the brain, (iii) emergently expressed behaviors long observed in living subjects, in addition to a novel order-dependent behavior, and (iv) expressed different task solutions yielding alternative behavioral and neural predictions. Further, in a large-scale experiment, we found that human subjects performing WM-based TI showed behavior inconsistent with a class of NNs that characteristically expressed an intuitive task solution. These findings provide neural insights into a classical relational ability, with wider implications for how the brain realizes relational cognition.

MOTIVATION: Deep learning-based molecule generation becomes a new paradigm of de novo molecule design since it enables fast and directional exploration in the vast chemical space. However, it is still an open issue to generate molecules, which bind to specific proteins with high-binding affinities while owning desired drug-like physicochemical properties. RESULTS: To address these issues, we elaborate a novel framework for controllable protein-oriented molecule generation, named CProMG, which contains a 3D protein embedding module, a dual-view protein encoder, a molecule embedding module, and a novel drug-like molecule decoder. Based on fusing the hierarchical views of proteins, it enhances the representation of protein binding pockets significantly by associating amino acid residues with their comprising atoms. Through jointly embedding molecule sequences, their drug-like properties, and binding affinities w.r.t. proteins, it autoregressively generates novel molecules having specific properties in a controllable manner by measuring the proximity of molecule tokens to protein residues and atoms. The comparison with state-of-the-art deep generative methods demonstrates the superiority of our CProMG. Furthermore, the progressive control of properties demonstrates the effectiveness of CProMG when controlling binding affinity and drug-like properties. After that, the ablation studies reveal how its crucial components contribute to the model respectively, including hierarchical protein views, Laplacian position encoding as well as property control. Last, a case study w.r.t. protein illustrates the novelty of CProMG and the ability to capture crucial interactions between protein pockets and molecules. It's anticipated that this work can boost de novo molecule design. AVAILABILITY AND IMPLEMENTATION: The code and data underlying this article are freely available at https://github.com/lijianing0902/CProMG.

Understanding how grid cells perform path integration calculations remains a fundamental problem. In this paper, we conduct theoretical analysis of a general representation model of path integration by grid cells, where the 2D self-position is encoded as a higher dimensional vector, and the 2D self-motion is represented by a general transformation of the vector. We identify two conditions on the transformation. One is a group representation condition that is necessary for path integration. The other is an isotropic scaling condition that ensures locally conformal embedding, so that the error in the vector representation translates conformally to the error in the 2D self-position. Then we investigate the simplest transformation, i.e., the linear transformation, uncover its explicit algebraic and geometric structure as matrix Lie group of rotation, and explore the connection between the isotropic scaling condition and a special class of hexagon grid patterns. Finally, with our optimization-based approach, we manage to learn hexagon grid patterns that share similar properties of the grid cells in the rodent brain. The learned model is capable of accurate long distance path integration. Code is available at https://github.com/ruiqigao/grid-cell-path.

The ability to record activities from hundreds of neurons simultaneously in the brain has placed an increasing demand for developing appropriate statistical techniques to analyze such data. Recently, deep generative models have been proposed to fit neural population responses. While these methods are flexible and expressive, the downside is that they can be difficult to interpret and identify. To address this problem, we propose a method that integrates key ingredients from latent models and traditional neural encoding models. Our method, pi-VAE, is inspired by recent progress on identifiable variational auto-encoder, which we adapt to make appropriate for neuroscience applications. Specifically, we propose to construct latent variable models of neural activity while simultaneously modeling the relation between the latent and task variables (non-neural variables, e.g. sensory, motor, and other externally observable states). The incorporation of task variables results in models that are not only more constrained, but also show qualitative improvements in interpretability and identifiability. We validate pi-VAE using synthetic data, and apply it to analyze neurophysiological datasets from rat hippocampus and macaque motor cortex. We demonstrate that pi-VAE not only fits the data better, but also provides unexpected novel insights into the structure of the neural codes.

Previous studies in neurophysiology have shown that neurons exhibit trial-by-trial correlated activity and that such noise correlations (NCs) greatly impact the accuracy of population codes. Meanwhile, multivariate pattern analysis (MVPA) has become a mainstream approach in functional magnetic resonance imaging (fMRI), but it remains unclear how NCs between voxels influence MVPA performance. Here, we tackle this issue by combining voxel-encoding modeling and MVPA. We focus on a well-established form of NC, tuning-compatible noise correlation (TCNC), whose sign and magnitude are systematically related to the tuning similarity between two units. We show that this form of voxelwise NCs can improve MVPA performance if NCs are sufficiently strong. We also confirm these results using standard information-theoretic analyses in computational neuroscience. In the same theoretical framework, we further demonstrate that the effects of noise correlations at both the neuronal level and the voxel level may manifest differently in typical fMRI data, and their effects are modulated by tuning heterogeneity. Our results provide a theoretical foundation to understand the effect of correlated activity on population codes in macroscopic fMRI data. Our results also suggest that future fMRI research could benefit from a closer examination of the correlational structure of multivariate responses, which is not directly revealed by conventional MVPA approaches.

Neural codes are inevitably shaped by various kinds of biological constraints, \emph{e.g.} noise and metabolic cost. Here we formulate a coding framework which explicitly deals with noise and the metabolic costs associated with the neural representation of information, and analytically derive the optimal neural code for monotonic response functions and arbitrary stimulus distributions. For a single neuron, the theory predicts a family of optimal response functions depending on the metabolic budget and noise characteristics. Interestingly, the well-known histogram equalization solution can be viewed as a special case when metabolic resources are unlimited. For a pair of neurons, our theory suggests that under more severe metabolic constraints, ON-OFF coding is an increasingly more efficient coding scheme compared to ON-ON or OFF-OFF. The advantage could be as large as one-fold, substantially larger than the previous estimation. Some of these predictions could be generalized to the case of large neural populations. In particular, these analytical results may provide a theoretical basis for the predominant segregation into ON- and OFF-cells in early visual processing areas. Overall, we provide a unified framework for optimal neural codes with monotonic tuning curves in the brain, and makes predictions that can be directly tested with physiology experiments.

Fisher information is generally believed to represent a lower bound on mutual information (Brunel & Nadal, 1998), a result that is frequently used in the assessment of neural coding efficiency. However, we demonstrate that the relation between these two quantities is more nuanced than previously thought. For example, we find that in the small noise regime, Fisher information actually provides an upper bound on mutual information. Generally our results show that it is more appropriate to consider Fisher information as an approximation rather than a bound on mutual information. We analytically derive the correspondence between the two quantities and the conditions under which the approximation is good. Our results have implications for neural coding theories and the link between neural population coding and psychophysically measurable behavior. Specifically, they allow us to formulate the efficient coding problem of maximizing mutual information between a stimulus variable and the response of a neural population in terms of Fisher information. We derive a signature of efficient coding expressed as the correspondence between the population Fisher information and the distribution of the stimulus variable. The signature is more general than previously proposed solutions that rely on specific assumptions about the neural tuning characteristics. We demonstrate that it can explain measured tuning characteristics of cortical neural populations that do not agree with previous models of efficient coding.

A common challenge for Bayesian models of perception is the fact that the two fundamental Bayesian components, the prior distribution and the likelihood func- tion, are formally unconstrained. Here we argue that a neural system that emulates Bayesian inference is naturally constrained by the way it represents sensory infor- mation in populations of neurons. More specifically, we show that an efficient coding principle creates a direct link between prior and likelihood based on the underlying stimulus distribution. The resulting Bayesian estimates can show bi- ases away from the peaks of the prior distribution, a behavior seemingly at odds with the traditional view of Bayesian estimation, yet one that has been reported in human perception. We demonstrate that our framework correctly accounts for the repulsive biases previously reported for the perception of visual orientation, and show that the predicted tuning characteristics of the model neurons match the reported orientation tuning properties of neurons in primary visual cortex. Our results suggest that efficient coding is a promising hypothesis in constrain- ing Bayesian models of perceptual inference.