There's a difference between sampling rates for reception and perception. The 4 framers per second you refer to is perception. At the level of input and receptor cell signal capture, it may be sampling at 100 cps or hz. You have to keep in mind that there's a difference between wetware and hardware. Hardware does have the ecological and physiological constraints that impose operational limitation on a biological neuron.
1. Neurotransmitters are required to open ion channels and the post synaptic neuron has to replenish them and the vesicles. When stores are exhausted, neuron fatigue sets in and a recovery period ensues.
2. Sodium and potassium ion concentrations in the immediate environment available to can very and the diffusion rate through the channels only allows one ion at a time to pass in and out of the cell.
3. The so called sodium pump that removes the excess sodium from the synaptic compartment takes time to re-establish the membrane potential.
4. Metabolic waste products like CO2 and lactic acid can attenuate catabolism and the production of ATP.
5. High rates of metabolism can consume catabolic dietary metabolites and cofactors disrupting homeostasis and the cells duty cycle.
As it is, the brain consumes 25% of the caloric intake for the entire body. Mother Nature has kept the sampling rate low yet at rates high enough to detect, capture, and deal with transitions taking place in the environment. The slow sampling rate and serial nature of perception has been offset by broadcasting the captured frames to banks of feature collectors in a process called priming. Priming is essentially a hybridization of an analog process. One cycle and every feature cell gets a sub-sample of the present frame.
As for ever understanding how the Cerebellum functions, I think it is further off than you might expect. The pyramidal neuron with its 5,000-20,000 connections is bad enough. The cerebellum's purkinji neurons with their 100,000-300,000 connections is just nuts! I suppose it's safe to say that it has something to do with perception, motor coordination, and TIMING. I suppose as robotics progresses to replace motors with artificial muscles, the need for such complex neurons will become apparent and some real insights will ensue.
