Drivers and modulators
Driver inputs to the thalamus, whether cortical or subcortical, tend to be "efference copies" -- the axons providing the input branch off and the other brnach also goes to some subcortical motor or premotor structure.
signal propagation
prediction
function approximation
anti-correlation
clustering
dimensionality reduction
universal turing machine
regexp
FSM
routing
memory (map)
memory (autoassociative)
pointwise arithmetic: +, -, *, /
convolutions
reinforcement learning
planning
NP-complete solving
optimization
CPG (rhythmic output)
fourier/wavelet transform
multiplexing
selective signal amplification tune into one frequency
recirculate information for awhile (circuit-level short term memory)
locks/semaphores
boolean logic gates
(database) transactions
addressing models: linear none tree
some peculiarly cognitive ones:
imitation
analogical reasoning
modeling self (i.e. introspection)
object recognition
"workspace model of consciousness" (i.e. broadcast)
coding regimes (single neurons):
rate coding "spike" coding phase coding AM FM
coding regimes (ensemble):
linear "sparse" hash
possible objectives being optimized:
minimize: wiring length energy MSE transmission time or computation time
maximize: information ML predictions/reward (?) MAP predictions/reward (?) precision/resolution number of patterns stored in memory S/R ratio