Hybrid predictive coding: Inferring, fast and slow

Adaptation in the visual system: Networked fatigue or suppressed prediction error signalling?

Our brains are constantly adapting to changes in our visual environments. Neural adaptation exerts a persistent influence on the activity of sensory neurons and our perceptual experience, however there is a lack of consensus regarding how adaptation is implemented in the visual system. One account describes fatigue-based mechanisms embedded within local networks of stimulus-selective neurons (networked fatigue models). Another depicts adaptation as a product of stimulus expectations (predictive coding models). In this review, I evaluate neuroimaging and psychophysical evidence that poses fundamental problems for predictive coding models of neural adaptation. Specifically, I discuss observations of distinct repetition and expectation effects, as well as incorrect predictions of repulsive adaptation aftereffects made by predictive coding accounts. Based on this evidence, I argue that networked fatigue models provide a more parsimonious account of adaptation effects in the visual system. Although stimulus expectations can be formed based on recent stimulation history, any consequences of these expectations are likely to co-occur (or interact) with effects of fatigue-based adaptation. I conclude by proposing novel, testable hypotheses relating to interactions between fatigue-based adaptation and other predictive processes, focusing on stimulus feature extrapolation phenomena.

The subcortical brain regions influence the cortical areas during resting-state: an fMRI study

Introduction

Numerous modes or patterns of neural activity can be seen in the brain of individuals during the resting state. However, those functions do not persist long, and they are continuously altering in the brain. We have hypothesized that the brain activations during the resting state should themselves be responsible for this alteration of the activities.

Methods

Using the resting-state fMRI data of 63 healthy young individuals, we estimated the causality effects of each resting-state activation map on all other networks. The resting-state networks were identified, their causality effects on the other components were extracted, the networks with the top 20% of the causality were chosen, and the networks which were under the influence of those causal networks were also identified.

Results

Our results showed that the influence of each activation component over other components is different. The brain areas which showed the highest causality coefficients were subcortical regions, such as the brain stem, thalamus, and amygdala. On the other hand, nearly all the areas which were mostly under the causal effects were cortical regions.

Discussion

In summary, our results suggest that subcortical brain areas exert a higher influence on cortical regions during the resting state, which could help in a better understanding the dynamic nature of brain functions.

Active sensing with predictive coding and uncertainty minimization

We present an end-to-end architecture for embodied exploration inspired by two biological computations: predictive coding and uncertainty minimization. The architecture can be applied to any exploration setting in a task-independent and intrinsically driven manner. We first demonstrate our approach in a maze navigation task and show that it can discover the underlying transition distributions and spatial features of the environment. Second, we apply our model to a more complex active vision task, whereby an agent actively samples its visual environment to gather information. We show that our model builds unsupervised representations through exploration that allow it to efficiently categorize visual scenes. We further show that using these representations for downstream classification leads to superior data efficiency and learning speed compared to other baselines while maintaining lower parameter complexity. Finally, the modular structure of our model facilitates interpretability, allowing us to probe its internal mechanisms and representations during exploration.

Predictive coding with spiking neurons and feedforward gist signaling

Predictive coding (PC) is an influential theory in neuroscience, which suggests the existence of a cortical architecture that is constantly generating and updating predictive representations of sensory inputs. Owing to its hierarchical and generative nature, PC has inspired many computational models of perception in the literature. However, the biological plausibility of existing models has not been sufficiently explored due to their use of artificial neurons that approximate neural activity with firing rates in the continuous time domain and propagate signals synchronously. Therefore, we developed a spiking neural network for predictive coding (SNN-PC), in which neurons communicate using event-driven and asynchronous spikes. Adopting the hierarchical structure and Hebbian learning algorithms from previous PC neural network models, SNN-PC introduces two novel features: (1) a fast feedforward sweep from the input to higher areas, which generates a spatially reduced and abstract representation of input (i.e., a neural code for the gist of a scene) and provides a neurobiological alternative to an arbitrary choice of priors; and (2) a separation of positive and negative error-computing neurons, which counters the biological implausibility of a bi-directional error neuron with a very high baseline firing rate. After training with the MNIST handwritten digit dataset, SNN-PC developed hierarchical internal representations and was able to reconstruct samples it had not seen during training. SNN-PC suggests biologically plausible mechanisms by which the brain may perform perceptual inference and learning in an unsupervised manner. In addition, it may be used in neuromorphic applications that can utilize its energy-efficient, event-driven, local learning, and parallel information processing nature.

A new predictive coding model for a more comprehensive account of delusions

Attempts to understand psychosis-the experience of profoundly altered perceptions and beliefs-raise questions about how the brain models the world. Standard predictive coding approaches suggest that it does so by minimising mismatches between incoming sensory evidence and predictions. By adjusting predictions, we converge iteratively on a best guess of the nature of the reality. Recent arguments have shown that a modified version of this framework-hybrid predictive coding-provides a better model of how healthy agents make inferences about external reality. We suggest that this more comprehensive model gives us a richer understanding of psychosis compared with standard predictive coding accounts. In this Personal View, we briefly describe the hybrid predictive coding model and show how it offers a more comprehensive account of the phenomenology of delusions, thereby providing a potentially powerful new framework for computational psychiatric approaches to psychosis. We also make suggestions for future work that could be important in formalising this novel perspective.

Predictive coding networks for temporal prediction

One of the key problems the brain faces is inferring the state of the world from a sequence of dynamically changing stimuli, and it is not yet clear how the sensory system achieves this task. A well-established computational framework for describing perceptual processes in the brain is provided by the theory of predictive coding. Although the original proposals of predictive coding have discussed temporal prediction, later work developing this theory mostly focused on static stimuli, and key questions on neural implementation and computational properties of temporal predictive coding networks remain open. Here, we address these questions and present a formulation of the temporal predictive coding model that can be naturally implemented in recurrent networks, in which activity dynamics rely only on local inputs to the neurons, and learning only utilises local Hebbian plasticity. Additionally, we show that temporal predictive coding networks can approximate the performance of the Kalman filter in predicting behaviour of linear systems, and behave as a variant of a Kalman filter which does not track its own subjective posterior variance. Importantly, temporal predictive coding networks can achieve similar accuracy as the Kalman filter without performing complex mathematical operations, but just employing simple computations that can be implemented by biological networks. Moreover, when trained with natural dynamic inputs, we found that temporal predictive coding can produce Gabor-like, motion-sensitive receptive fields resembling those observed in real neurons in visual areas. In addition, we demonstrate how the model can be effectively generalized to nonlinear systems. Overall, models presented in this paper show how biologically plausible circuits can predict future stimuli and may guide research on understanding specific neural circuits in brain areas involved in temporal prediction.

Brain-like Flexible Visual Inference by Harnessing Feedback-Feedforward Alignment

In natural vision, feedback connections support versatile visual inference capabilities such as making sense of the occluded or noisy bottom-up sensory information or mediating pure top-down processes such as imagination. However, the mechanisms by which the feedback pathway learns to give rise to these capabilities flexibly are not clear. We propose that top-down effects emerge through alignment between feedforward and feedback pathways, each optimizing its own objectives. To achieve this co-optimization, we introduce Feedback-Feedforward Alignment (FFA), a learning algorithm that leverages feedback and feedforward pathways as mutual credit assignment computational graphs, enabling alignment. In our study, we demonstrate the effectiveness of FFA in co-optimizing classification and reconstruction tasks on widely used MNIST and CIFAR10 datasets. Notably, the alignment mechanism in FFA endows feedback connections with emergent visual inference functions, including denoising, resolving occlusions, hallucination, and imagination. Moreover, FFA offers bio-plausibility compared to traditional back-propagation (BP) methods in implementation. By repurposing the computational graph of credit assignment into a goal-driven feedback pathway, FFA alleviates weight transport problems encountered in BP, enhancing the bio-plausibility of the learning algorithm. Our study presents FFA as a promising proof-of-concept for the mechanisms underlying how feedback connections in the visual cortex support flexible visual functions. This work also contributes to the broader field of visual inference underlying perceptual phenomena and has implications for developing more biologically inspired learning algorithms.

Correction: Hybrid predictive coding: Inferring, fast and slow

[This corrects the article DOI: 10.1371/journal.pcbi.1011280.].