The 3rd wave won’t be synaptic

The neuron is a highly specialized cell type, strongly compartmentalized with extensive axons and dendrites. Each compartment, dendritic or axonal, connects directly and specifically to other neurons. For this reason, there has always been a strong focus within neuroscience on the synapse as a specialized connective element between neurons, with its postsynaptic density protein composition, and its presynaptic vesicle release mechanism. This has led to theoretical neuroscience focusing on adjusting synaptic weights as the main memory mechanism. But by now the experimental literature on neuronal plasticity is vast, and synapses make up only a small part of it. Intrinsic forms of plasticity via ion channels and NM receptors are well-known and have been researched in great detail. The vertical, internal dimension of the neuron with its intracellular signaling network and the nuclear processes, genetic and epigenetic, has particular significance in the fields of psychopharmacology and disease modeling. But the development of a theoretical framework, linking those aspects of neural plasticity, has been missing. We have begun to fill the void by re-thinking the concept of neural plasticity from the perspective of the individual neuron. We argue thateven synaptic plasticity is incompletely understood when placed outside of the context of cellular functioning. Purely horizontal models with weight adaptation, like current neural network models, can only be programmed by input. That is very restrictive. Huge networks with billions of parameters are needed to solve simple problems by brute-force massive storage. Those restrictions are systematic and mathematically explainable, since the expressive power of networks with adjustable weights is weak. A programmable memory at each neuronal site allows for more complex and interesting operations. For instance, patterns from the horizontal plane can get stored into a small set of dedicated neurons for a particular problem. Such models may then be used to build complex knowledge structures. We believe that the ”third wave of AI” will have to employ some kind of horizontal-vertical brain model to make use of the opportunities of linking intracellular intelligence with large-scale neuron modeling to achieve a truly intelligent,self-programmable system. An immediate challenge will be to create models which run in a self-contained way, and build up internal structures from the initial set-up and the processing of input patterns.