Publications

We characterised the pathophysiology of seizure onset in terms of slow fluctuations in synaptic efficacy using EEG in patients with anti-N-methyl-d-aspartate receptor (NMDA-R) encephalitis. EEG recordings were obtained from two female patients with anti-NMDA-R encephalitis with recurrent partial seizures (ages 19 and 31). Focal electrographic seizure activity was localised using an empirical Bayes beamformer. The spectral density of reconstructed source activity was then characterised with dynamic causal modelling (DCM). Eight models were compared for each patient, to evaluate the relative contribution of changes in intrinsic (excitatory and inhibitory) connectivity and endogenous afferent input. Bayesian model comparison established a role for changes in both excitatory and inhibitory connectivity during seizure activity (in addition to changes in the exogenous input). Seizures in both patients were associated with a sequence of changes in inhibitory and excitatory connectivity; a transient increase in inhibitory connectivity followed by a transient increase in excitatory connectivity and a final peak of excitatory-inhibitory balance at seizure offset. These systematic fluctuations in excitatory and inhibitory gain may be characteristic of (anti NMDA-R encephalitis) seizures. We present these results as a case study and replication to motivate analyses of larger patient cohorts, to see whether our findings generalise and further characterise the mechanisms of seizure activity in anti-NMDA-R encephalitis.

Selective attention implements preferential routing of attended stimuli, likely through increasing the influence of the respective synaptic inputs on higher-area neurons. As the inputs of competing stimuli converge onto postsynaptic neurons, presynaptic circuits might offer the best target for attentional topdown influences. If those influences enabled presynaptic circuits to selectively entrain postsynaptic neurons, this might lead to selective routing. Indeed, when two visual stimuli induce two gamma rhythms in V1, only the gamma induced by the attended stimulus entrains gamma in V4. Here, we modeled this selective entrainment with a Dynamic Causal Model for Cross-Spectral Densities and found that it can be explained by attentional modulation of intrinsic V1 connections. Specifically, local inhibition was decreased in the granular input layer and increased in the supragranular output layer of the V1 circuit that processed the attended stimulus. Thus, presynaptic attentional influences and ensuing entrainment were sufficient to mediate selective routing.

There is a steadily growing literature on the role of the immune system in psychiatric disorders. So far, these advances have largely taken the form of correlations between specific aspects of inflammation (e.g. blood plasma levels of inflammatory markers, genetic mutations in immune pathways, viral or bacterial infection) with the development of neuropsychiatric conditions such as autism, bipolar disorder, schizophrenia and depression. A fundamental question remains open: why are psychiatric disorders and immune responses intertwined? To address this would require a step back from a historical mind-body dualism that has created such a dichotomy. We propose three contributions of active inference when addressing this question: translation, unification, and simulation. To illustrate these contributions, we consider the following questions. Is there an immunological analogue of sensory attenuation? Is there a common generative model that the brain and immune system jointly optimise? Can the immune response and psychiatric illness both be explained in terms of self-organising systems responding to threatening stimuli in their external environment, whether those stimuli happen to be pathogens, predators, or people? Does false inference at an immunological level alter the message passing at a psychological level (or vice versa) through a principled exchange between the two systems?

<title>Abstract</title> Autonomous driving and robotics are increasingly focusing on improving safety, particularly in collision avoidance systems. Detecting unexpected obstacles quickly is essential for preventing accidents, especially in high-speed driving scenarios where the sudden appearance of pedestrians presents a significant challenge. Conventional systems often rely on prediction, but they struggle with obstacles emerging suddenly from blind spots, introducing delays in detection. Here, we show that spike cameras, can significantly reduce these delays by enabling rapid object detection. Our system integrates perception, reasoning, and action into an end-to-end, ultra-low-latency closed-loop that allows for faster reactions. Our system reduces the overall latency of detecting sudden appearance of objects to under 5 ms: in real-world ghosting pedestrian scenarios, it achieves a 97% success rate in obstacle avoidance tests. Additionally, even at an equivalent real vehicle speed of 118 km/h, the system maintains a success rate of 92.5%. These results represent a meaningful improvement in autonomous driving safety, providing faster, more accurate obstacle detection in high-speed environments, where quick reactions are crucial for avoiding accidents and ensuring the safety of all road users.

In its full sense, perception rests on an agent's model of how its sensory input comes about and the inferences it draws based on this model. These inferences are necessarily uncertain. Here, we illustrate how the Hierarchical Gaussian Filter (HGF) offers a principled and generic way to deal with the several forms that uncertainty in perception takes. The HGF is a recent derivation of one-step update equations from Bayesian principles that rests on a hierarchical generative model of the environment and its (in)stability. It is computationally highly efficient, allows for online estimates of hidden states, and has found numerous applications to experimental data from human subjects. In this paper, we generalize previous descriptions of the HGF and its account of perceptual uncertainty. First, we explicitly formulate the extension of the HGF's hierarchy to any number of levels; second, we discuss how various forms of uncertainty are accommodated by the minimization of variational free energy as encoded in the update equations; third, we combine the HGF with decision models and demonstrate the inversion of this combination; finally, we report a simulation study that compared four optimization methods for inverting the HGF/decision model combination at different noise levels. These four methods (Nelder-Mead simplex algorithm, Gaussian process-based global optimization, variational Bayes and Markov chain Monte Carlo sampling) all performed well even under considerable noise, with variational Bayes offering the best combination of efficiency and informativeness of inference. Our results demonstrate that the HGF provides a principled, flexible, and efficient-but at the same time intuitive-framework for the resolution of perceptual uncertainty in behaving agents.

Depression with cognitive impairment, so called depressive pseudodementia, is commonly mistaken for a neurodegenerative dementia. Using positron emission tomography (PET) derived measures of regional cerebral blood flow (rCBF) a cohort of 33 patients with major depression was studied. Ten patients displayed significant and reversible cognitive impairment. The patterns of rCBF of these patients were compared with a cohort of equally depressed non-cognitively impaired depressed patients. In the depressed cognitively impaired patients a profile of rCBF abnormalities was identified consisting of decreases in the left anterior medial prefrontal cortex and increases in the cerebellar vermis. These changes were additional to those seen in depression alone and are distinct from those described in neurodegenerative dementia. The cognitive impairment seen in a proportion of depressed patients would seem to be associated with dysfunction of neural systems distinct from those implicated in depression alone or the neurodegenerative dementias.

The neural processes underlying tactile decisions in the human brain remain elusive. We addressed this question in a functional magnetic resonance imaging study using a somatosensory discrimination task, requiring participants to compare the frequency of two successive tactile stimuli. Tactile stimuli per se engaged somatosensory, parietal, and frontal cortical regions. Using a statistical model that accounted for the relative difference in frequencies (i.e., Weber fraction) and discrimination accuracy (i.e., correct or incorrect), we show that trial-by-trial relative frequency difference is represented linearly by activity changes in the left dorsolateral prefrontal cortex (DLPFC), the dorsal anterior cingulate cortex, and bilateral anterior insular cortices. However, a circumscribed region within the left DLPFC showed a different response pattern expressed as activity changes that were monotonically related to relative stimulation difference only for correct but not for incorrect trials. Our findings suggest that activity in the left DLPFC encodes stimulus representations that underlie veridical tactile decisions in humans.

Action observation (AO) is crucial for motor planning, imitation learning, and social interaction, but it is not clear whether and how an action execution-observation network (AEON) processes the effort of others engaged in performing actions. In this functional magnetic resonance imaging (fMRI) study, we used a "squeeze ball" task involving different grip forces to investigate whether AEON activation showed similar patterns when executing the task or observing others performing it. Both in action execution, AE (subjects performed the visuomotor task) and action observation, AO (subjects watched a video of the task being performed by someone else), the fMRI signal was detected in cerebral and cerebellar regions. These responses showed various relationships with force mapping onto specific areas of the sensorimotor and cognitive systems. Conjunction analysis of AE and AO was repeated for the "0th" order and linear and nonlinear responses, and revealed multiple AEON nodes remapping the detection of actions, and also effort, of another person onto the observer's own cerebrocerebellar system. This result implies that the AEON exploits the cerebellum, which is known to process sensorimotor predictions and simulations, performing an internal assessment of forces and integrating information into high-level schemes, providing a crucial substrate for action imitation.