Long-Term Memory Neural Circuits, Fast and Precise - Brain Theory from a Circuits and Systems Perspective - page 144

learning, they develop a state machine held by interneurons that encourage synaptic
growth. Once in place, subconscious long-term memory jumps from one word of
information to another as fast as the information can be expressed, without having
to run each step through conscious STM. In this way, without knowing how, a
savant remembers amazing sequences.

References

1. Pagiamtzis K, Sheikholeslami A (2006) Content-addressable memory (CAM) circuits and
architectures: a tutorial and survey, IEEE J Solid State Circuits 41(3):712-727
2. Burger JR (2009) Human memory modeled with standard analog and digital circuits: inspiration
for man-made computers. Wiley, Hoboken, NJ
3. Huettel SA, Song AW, McCarthy G (2004) Functional magnetic resonance imaging. Siniauer
Associates, Inc, Sunderland, MA, pp 162-170
4. Ogawa S, Lee TM, Nayuak AS, Glynn P (1990) Oxygenation-sensitive contrast in magnetic
resonance image of rodent brain at high magnetic fields. Magn Reson Med 14(1):68-78
5. Burger JR (2011) Qubits underlie gifted savants. NeuroQuantology 9:351-360

Self-Study Exercises

(If you desire to keep a record of your efforts, please explain your answers in a clear
way)

Memory Circuits

1. A neuron may accomplish Boolean logic:

(a) List major differences between neurons and CMOS (HINT: Pulses vs.
voltage levels; dendritic logic with timing vs. enabled logic)
(b) The XOR is generated by a standard Boolean equation. Show a logic circuit
for z that corresponds to this Boolean equation (HINT: You may use five
basic gates)

xy 0 þ x 0 y z ¼ :

(c) Modify the XOR circuit to give a non-zero output only if an enable signal E
is true (HINT: Use three-input AND gates with one input an enable signal)
(d) Demonstrate with reference to the XOR equation that it can be used to
generate a NOT function of x. Illustrate the NOT using an XOR circuit
(HINT: Hold one input to true)