Mechanization of Cognition - Biomimetics: Biologically Inspired Technologies - page 113

p(cl) (i.e., roughly, the probability of the two involved symbols being coactive). In addition,
studies of postsynaptic neurotransmitter depolarization transduction response (i.e., within the
neuron receiving the synaptic neurotransmitter output; separate from the transmitting synapse
itself) by Marder and her colleagues (Marder and Prinz, 2002, 2003) and by Turrigiano and her
colleagues (Desai et al., 2002; Turrigiano and Nelson, 2000, 2004; Turrigiano et al., 1998) suggest
that the postsynaptic apparatus of an excitatory cortical synapse (e.g., one landing on a target
symbol neuron) is independently modifiable in efficacy, in multiplicative series with this Hebbian
p(cl) efficacy. This ''post-synaptic signaling efficacy'' is expressed as a neurotransmitter recep-
tivity proportional to a direct function of the reciprocal of that target neuron's average firing rate,
which is essentially 1/p(l). The net result is implementation by this Hebb/Marder/Turrigiano
learning process (as I call it) of an overall link strength directly related to p(cl)/p(l), which by
Bayes law is p(c j l). Thus, it is plausible that biological learning processes at the neuron level can
accumulate the knowledge needed for confabulation.

3.A.5 Implementation of Confabulation

Since only a small subset of the neurons representing target lexicon symbol l are excited by
a knowledge link from source lexicon symbol c, how can confabulation be implemented?
This section, which presents the theory's hypothesized implementation of confabulation,
answers this question and shows that these ''internally sparse'' knowledge links are an essential
element of cortical design. Counterintuitively, if these links were ''fully connected,'' cortex could
not function.
Figure 3.A.6 schematically illustrates how confabulation is implemented in a thalamocortical
(answer lexicon) module. The four boxes on the left are four cortical lexicons, each having exactly
one assumed fact symbol active (symbols a, b, g, and d respectively). Each of these active symbols
is represented by the full complement of the neurons which represent it, which are all active
(illustrated as a complete row of filled circles within that assumed fact symbol's lexicon module,
depicted in the figure in colors green, red, blue, and brown for a, b, g, and d, respectively). As will
be seen below, this is how the symbol(s), which are the conclusions of a confabulation operation are
biologically expressed (namely, all of their representing neurons are active and all other symbol
representing neurons are inactive).
In Figure 3.A.6 the neurons representing each symbol of a module are shown as separated
into their own rows. Of course, in the actual tissue, the neurons of each symbol are scattered
randomly within the relevant layers of the cortical portion of the module implementing the
lexicon. But for clarity, in Figure 3.A.6 each symbol's neurons are shown collected together
into one row. The fact that the same neuron appears in multiple rows (each symbol-representing
neuron typically participates in representing many different symbols) is ignored here, as this small
pairwise overlap between symbol representations causes no significant interference between
symbols.
(Note: This is easy to see: consider the simplified attractor you built and experimented
with above. It always converged to a single pure state x k (at least when the initial state u was
dominated by x k ); meaning that all of the neurons which represent x k are active and all other
neurons are inactive. However, each of the neurons of x k also belongs to many other stable states x i ,
but this does not cause any problems or interference. You may not have seen this aspect of the
system at the time you did your experiments — go check! You will find that even though the
overlap between each pair of x field stable states is relatively small, each individual neuron
participates in many such stable states. The properties of this kind of attractor network are quite
astounding; and they do not even have many of the additional design features that thalamocortical
modules possess.)
The answer lexicon for the elementary confabulation we are going to carry out (based upon
assumed facts a, b, g, and d, just as described in Hecht-Nielsen, 2005) is shown as the box on