Publications

Integrating neurons into digital systems may enable performance infeasible with silicon alone. Here, we develop DishBrain, a system that harnesses the inherent adaptive computation of neurons in a structured environment. In vitro neural networks from human or rodent origins are integrated with in silico computing via a high-density multielectrode array. Through electrophysiological stimulation and recording, cultures are embedded in a simulated game-world, mimicking the arcade game "Pong." Applying implications from the theory of active inference via the free energy principle, we find apparent learning within five minutes of real-time gameplay not observed in control conditions. Further experiments demonstrate the importance of closed-loop structured feedback in eliciting learning over time. Cultures display the ability to self-organize activity in a goal-directed manner in response to sparse sensory information about the consequences of their actions, which we term synthetic biological intelligence. Future applications may provide further insights into the cellular correlates of intelligence.

Procedures that can predict cognitive abilities from brain imaging data are potentially relevant to educational assessments and studies of functional anatomy in the developing brain. Our aim in this work was to quantify the degree to which IQ change in the teenage years could be predicted from structural brain changes. Two well-known k-fold cross-validation analyses were applied to data acquired from 33 healthy teenagers - each tested at Time 1 and Time 2 with a 3.5 year interval. One approach, a Leave-One-Out procedure, predicted IQ change for each subject on the basis of structural change in a brain region that was identified from all other subjects (i.e., independent data). This approach predicted 53% of verbal IQ change and 14% of performance IQ change. The other approach used half the sample, to identify regions for predicting IQ change in the other half (i.e., a Split half approach); however--unlike the Leave-One-Out procedure--regions identified using half the sample were not significant. We discuss how these out-of-sample estimates compare to in-sample estimates; and draw some recommendations for k-fold cross-validation procedures when dealing with small datasets that are typical in the neuroimaging literature.

Abstract Functional recovery after brain damage varies widely and depends on many factors, including lesion site and extent. When a neuronal system is damaged, recovery may occur by engaging residual (e.g., perilesional) components. When damage is extensive, recovery depends on the availability of other intact neural structures that can reproduce the same functional output (i.e., degeneracy). A system’s response to damage may occur rapidly, require learning or both. Here, we simulate functional recovery from four different types of lesions, using a generative model of word repetition that comprised a default premorbid system and a less used alternative system. The synthetic lesions (i) completely disengaged the premorbid system, leaving the alternative system intact, (ii) partially damaged both premorbid and alternative systems, and (iii) limited the experience-dependent plasticity of both. The results, across 1000 trials, demonstrate that (i) a complete disconnection of the premorbid system naturally invoked the engagement of the other, (ii) incomplete damage to both systems had a much more devastating long-term effect on model performance and (iii) the effect of reducing learning capacity within each system. These findings contribute to formal frameworks for interpreting the effect of different types of lesions.

Oscillatory activity in the beta range, in human primary motor cortex (M1), shows interesting dynamics that are tied to behaviour and change systematically in disease. To investigate the pathophysiology underlying these changes, we must first understand how changes in beta activity are caused in healthy subjects. We therefore adapted a canonical (repeatable) microcircuit model used in dynamic causal modelling (DCM) previously used to model induced responses in visual cortex. We adapted this model to accommodate cytoarchitectural differences between visual and motor cortex. Using biologically plausible connections, we used Bayesian model selection to identify the best model of measured MEG data from 11 young healthy participants, performing a simple handgrip task. We found that the canonical M1 model had substantially more model evidence than the generic canonical microcircuit model when explaining measured MEG data. The canonical M1 model reproduced measured dynamics in humans at rest, in a manner consistent with equivalent studies performed in mice. Furthermore, the changes in excitability (self-inhibition) necessary to explain beta suppression during handgrip were consistent with the attenuation of sensory precision implied by predictive coding. These results establish the face validity of a model that can be used to explore the laminar interactions that underlie beta-oscillatory dynamics in humans in vivo. Our canonical M1 model may be useful for characterising the synaptic mechanisms that mediate pathophysiological beta dynamics associated with movement disorders, such as stroke or Parkinson's disease.

No abstract is provided for this article.

Artificial intelligence (AI) is rapidly becoming one of the key technologies of this century. The majority of results in AI thus far have been achieved using deep neural networks trained with a learning algorithm called error backpropagation, always considered biologically implausible. To this end, recent works have studied learning algorithms for deep neural networks inspired by the neurosciences. One such theory, called predictive coding (PC), has shown promising properties that make it potentially valuable for the machine learning community: it can model information processing in different areas of the brain, can be used in control and robotics, has a solid mathematical foundation in variational inference, and performs its computations asynchronously. Inspired by such properties, works that propose novel PC-like algorithms are starting to be present in multiple sub-fields of machine learning and AI at large. Here, we survey such efforts by first providing a broad overview of the history of PC to provide common ground for the understanding of the recent developments, then by describing current efforts and results, and concluding with a large discussion of possible implications and ways forward.

This positron emission tomography study dissociates the neural correlates of object recognition and naming. Stimuli comprised coloured outline drawings of objects and coloured nonsense shapes. Subjects either viewed or explicitly named objects, similarly they viewed or named the colour of the shapes. Activations common to object and colour naming were identified by contrasting the explicit naming conditions (objects and colours) with the control (viewing) conditions. Activations associated with object recognition were identified by contrasting both object conditions (naming and viewing) with both shape conditions and activations specific to object or colour naming were identified by contrasting object naming (relative to object viewing) with colour naming (relative to shape viewing). The results associate: (i) object recognition with left middle occipital and bilateral anterior temporal cortices; (ii) modality independent naming with left posterior basal temporal lobe and the left prefrontal cortex; (iii) areas specific to object naming with left temporal extrasylvian regions, left anterior insula and right cerebellum; and (iv) areas specific to colour naming with left posterior lingual and fusiform gyri and midline cerebellum. These results are discussed in relation to previous neuroimaging and neuropsychological findings.