Search

The neuronal substrates that implement the free-energy principle and ensuing active inference at the neuron and synapse level have not been fully elucidated. This Review considers possible neuronal substrates underlying the principle. First, the foundations of the free-energy principle are introduced, and then its ability to empirically explain various brain functions and psychological and biological phenomena in terms of Bayesian inference is described. Mathematically, the dynamics of neural activity and plasticity that minimise a cost function can be cast as performing Bayesian inference that minimises variational free energy. This equivalence licenses the adoption of the free-energy principle as a universal characterisation of neural networks. Further, the neural network structure itself represents a generative model under which an agent operates. A virtue of this perspective is that it enables the formal association of neural network properties with prior beliefs that regulate inference and learning. The possible neuronal substrates that implement prior and posterior beliefs and how to empirically examine the theory are discussed. This perspective renders brain activity explainable, leading to a deeper understanding of the neuronal mechanisms underlying basic psychology and psychiatric disorders in terms of an implicit generative model.

Theoretical simulations suggest that spike rate is regulated by varying both membrane potential and its fluctuation. We investigated whether membrane potential fluctuation functionally changes in motor cortex and striatum neurons during discrete forelimb movements and pauses, or at rest, using whole-cell recording in task-performing rats. Membrane potential fluctuation was diminished by task performance, but maintained overall in the alpha/beta and gamma bands during forelimb movements and pauses. By contrast, membrane potential itself was correlated with spike rate in task-related neurons. Thus, membrane potential, but not its fluctuation, is a critical determinant of execution and pausing of discrete movements.

The hippocampus is a remarkable neural structure that displays a variety of synchronous oscillations that may be physiological or pathophysiological, such as theta rhythms and epileptic seizures. Electrically induced seizure-like afterdischarges are an excellent system for elucidating the network mechanisms underlying neuronal synchronization and rhythm generation of epileptic synchronous oscillations in extremely hyperactive hippocampal networks. In this Update Article, we review key findings of studies on these electrically induced seizure-like afterdischarges in vitro. During these afterdischarges, GABAergic responses become transiently depolarizing and even excitatory as chloride rapidly accumulates postsynaptically in pyramidal cells. Glutamate and potassium enhance this transient GABAergic excitation. Neuronal synchronization of afterdischarge is achieved by GABAergic and glutamatergic excitation of pyramidal cells and interneurons localized in the stratum pyramidale and stratum oriens. Rhythm generation in seizure-like synchronous oscillations is not yet understood but is the subject of intensive study.

Strong tetanization induces synchronous membrane potential oscillations (seizure-like afterdischarge) in mature pyramidal cells of the hippocampal CA1 region. To investigate whether local networks in other brain regions can generate such an afterdischarge independently, we studied the inducibility of afterdischarge in individual ‘isolated slices’ of the rat hippocampal CA1 and CA3 regions, dentate gyrus (DG), entorhinal cortex (EC), and temporal cortex (TC) using intracellular and extracellular recordings. The strong tetanization constantly induced afterdischarges in the CA1 and CA3 pyramidal cells as well as in the EC and TC superficial principal cells. However, parameters of the afterdischarges, such as the frequency and duration of afterdischarges, varied among the regions. A mixture of N -methyl- d -aspartate (NMDA) and non-NMDA receptor antagonists or a GABA A receptor antagonist completely blocked the afterdischarges. Local GABA application during the afterdischarge elicited depolarization, rather than hyperpolarization. Moreover, reversal potentials of the afterdischarge were around −40 mV. In contrast, the tetanization resulted in occasional afterdischarge-like activities in DG slices, which were blocked by the non-NMDA or GABA A receptor antagonist. These findings suggest that the afterdischarges mediated through the excitatory GABAergic and glutamatergic transmissions might be common to, but be modulated differently by individual local networks in the hippocampus and cortex.

5-Hydroxy- l -tryptophan (5-HTP) and l -3,4-dihydroxyphenylalanine ( l -DOPA) were injected intraperitoneally (i.p.) into the laboratory shrew (Suncus murinus). Immunocytochemical and immunofluorescence studies were carried out on serial or same sections of the brain, which were reacted with specific antisera to dopamine (DA) or serotonin (5-HT) produced in our laboratory. We observed that cell bodies and nerve terminals of many catecholaminergic (CAnergic) neurons exhibited 5-HTP uptake and conversion of the precursor into 5-HT. However, the locus ceruleus showed scarcely any 5-HT immunoreactiviyt. This suggests that the precursor uptake mechanism may be different among various CAnergic groups. In contrast to these findings on CAnergic neurons, all serotoninergic (5-HTnergic) neurons after l -DOPA administration showed DA immunopositive reaction in their cell bodies and nerve terminals, suggesting that 5-HTnergic neurons may have the same capacity for precursor uptake. On the other hand, we observed that all aromatic l -amino acid decarboxylase (AADC)-only-positive neuron groups showed both DA and 5-HT immunoreactions after l -DOPA and 5-HTP administration, respectively, in the double-staining immunofluorescence method. From these results, AADC-only-positive neurons may be considered to belong to the amine precursor uptake and decarboxylation (APUD) system.

By means of immunohistochemistry using anti-γ-aminobutyric acid (GABA) antibodies, characteristic neurons of pyriform, bipolar or pleomorphic shapes, regarded as the Cajal-Retzius (CR) neurons, were clearly demonstrated in layer I of the rat cerebral cortex at the various experimental periods. On embryonic day 15, ovoid neurons only in the marginal zone indicated immunoreactivity for GABA. They gradually extended thick processes often in parallel with the pial surface and formed a dense GABA fiber network in immature layer I during the early postnatal periods. Some GABA neurons seemed to migrate into the underlying layers to settle as nonpyramidal cells. With the expansion of brain volume, GABA neurons relatively diminished and decreased in number. Nevertheless, a small number of GABA neurons did exist as essential CR neurons in layer I even in the adult rats aged 9 months.