Search

Cognitive functions arise from the coordinated activity of neural populations distributed over large-scale brain networks. However, it is challenging to understand and measure how specific aspects of neural dynamics translate into operations of information processing, and, ultimately, cognitive functions. An obstacle is that simple circuit mechanisms–such as self-sustained or propagating activity and nonlinear summation of inputs–do not directly give rise to high-level functions. Nevertheless, they already implement simple transformations of the information carried by neural activity.Here, we propose that distinct neural circuit functions, such as stimulus representation, working memory, or selective attention stem from different combinations and types of low-level manipulations of information, or information processing primitives. To test this hypothesis, we combine approaches from information theory with computational simulations of canonical neural circuits involving one or more interacting brain regions that emulate well-defined cognitive functions. More specifically, we track the dynamics of information emergent from dynamic patterns of neural activity, using suitable quantitative metrics to detect where and when information is actively buffered (“active information storage”), transferred (“information transfer”) or non-linearly merged (“information modification”), as possible modes of low-level processing. We find that neuronal subsets maintaining representations in working memory or performing attention-related gain modulation are signaled by their boosted involvement in operations of active information storage or information modification, respectively.Thus, information dynamics metrics, beyond detectingwhichnetwork units participate in cognitive processing, also promise to specifyhow and whenthey do it, i.e., through which type of primitive computation, a capability that may be exploited for the parsing of actual experimental recordings.

Idea of the model, specification of the model and tests, implementation of the model, tests, data analysis, analysis of results, writing-up. ‡Idea of the model, specification of the model and tests, analysis of results, writing-up. ¤Specification of the model and tests, analysis of results, writing-up. * Abstract In mammals, goal-directed and planning processes support flexible behaviour usable to face new situations or changed conditions that cannot be tackled through more efficient but rigid habitual behaviours. Within the Bayesian modelling approach of brain and behaviour, probabilistic models have been proposed to perform planning as a probabilistic inference. Recently, some models have started to face the important challenge met by this approach: grounding such processes on the computations implemented by brain spiking networks. Here we propose a model of goal-directed behaviour that has a probabilistic interpretation and is centred on a recurrent spiking neural network representing the world model. The model, building on previous proposals on spiking neurons and plasticity rules having a probabilistic interpretation, presents these novelties at the system level: (a) the world model is learnt in parallel with its use for planning, and an arbitration mechanism decides when to exploit the world-model knowledge with planning, or to explore, on the basis of an entropy-based confidence on the world model knowledge; (b) the world model is a hidden Markov model (HMM) able to simulate sequences of states and actions, thus planning selects actions through the same neural generative process used to predict states; (c) the world model learns the hidden causes of observations, and their temporal dependencies, through a biologically plausible unsupervised learning mechanism. The model is tested with a visuomotor learning task and validated by comparing its behaviour with the performance and reaction times of human participants solving the same task. The model represents a further step towards the construction of an autonomous architecture bridging goal-directed behaviour as probabilistic inference to brain-like computations. Author summary Goal-directed behaviour relies on brain processes supporting planning of actions based on the prediction of their consequences before performing them in the environment. An important computational modelling approach of these processes sees the brain as a probabilistic machine implementing goal-directed processes relying on probability distributions and operations on them. An important challenge for this approach is to explain how these distributions and operations might be grounded on the brain spiking doi: bioRxiv preprint neurons and learning processes. Here we propose a hypothesis of how this might happen by presenting a computational model of goal-directed processes based on artificial spiking neural networks. The model presents three main novelties. First, it can plan even while it is still learning the consequences of actions by deciding if planning or exploring the environment based on how confident it is on its predictions. Second, it is able to 'think' alternative possible actions, and their consequences, by relying on the low-level stochasticity of neurons. Third, it can learn to anticipate the consequences of actions in an autonomous fashion based on experience. Overall, the model represents a novel hypothesis on how goal-directed behaviour might rely on the stochastic spiking processes and plasticity mechanisms of the brain neurons.

Quantifying both the amount and content of the information transferred between neuronal populations is crucial to understand brain functions. Traditional data-driven methods based on Wiener-Granger causality quantify information transferred between neuronal signals, but do not reveal whether transmission of information refers to one specific feature of external stimuli or another. Here, we developed a new measure called Feature-specific Information Transfer (FIT), that quantifies the amount of information transferred between neuronal signals about specific stimulus features. The FIT quantifies the feature-related information carried by a receiver that was previously carried by a sender, but that was never carried by the receiver earlier. We tested the FIT on simulated data in various scenarios. We found that, unlike previous measures, FIT successfully disambiguated genuine feature-specific communication from non-feature specific communication, from external confounding inputs and synergistic interactions. Moreover, the FIT had enhanced temporal sensitivity that facilitates the estimation of the directionality of transfer and the communication delay between neuronal signals. We validated the FIT’s ability to track feature-specific information flow using neurophysiological data. In human electroencephalographic data acquired during a face detection task, the FIT demonstrated that information about the eye in face pictures flowed from the hemisphere contralateral to the eye to the ipsilateral one. In multi-unit activity recorded from thalamic nuclei and primary sensory cortices of rats during multimodal stimulation, FIT, unlike Wiener-Granger methods, credibly detected both the direction of information flow and the sensory features about which information was transmitted. In human cortical high-gamma activity recorded with magnetoencephalography during visuomotor mapping, FIT showed that visuomotor-related information flowed from superior parietal to premotor areas. Our work suggests that the FIT measure has the potential to uncover previously hidden feature-specific information transfer in neuronal recordings and to provide a better understanding of brain communication.Author summaryThe emergence of coherent percepts and behavior relies on the processing and flow of information about sensory features, such as the color or shape of an object, across different areas of the brain. To understand how computations within the brain lead to the emergence of these functions, we need to map the flow of information about each specific feature. Traditional methods, such as those based on Wiener-Granger causality, quantify whether information is transmitted from one brain area to another, but do not reveal if the information being transmitted is about a certain feature or another feature. Here, we develop a new mathematical technique for the analysis of brain activity recordings, called Feature-specific Information Transfer (FIT), that can reveal not only if any information is being transmitted across areas, but whether or not such transmitted information is about a certain sensory feature. We validate the method with both simulated and real neuronal data, showing its power in detecting the presence of feature-specific information transmission, as well as the timing and directionality of this transfer. This work provides a tool of high potential significance to map sensory information processing in the brain.