Thermodynamics of Intelligence and Cognitive Enhancement


Brain en­ergy is of­ten con­fused with mo­ti­va­tion, but these are two dis­tinct phe­nom­ena. Brain en­ergy is the ac­tual metabolic en­ergy available to the neu­rons, in the form of adeno­sine triphos­phate (ATP) molecules. ATP is the “en­ergy cur­rency” of the cell, and is pro­duced pri­mar­ily by ox­ida­tive metabolism of en­ergy from food. High mo­ti­va­tion in­creases the use of this en­ergy, but in the ab­sence of suffi­cient metabolic ca­pac­ity it even­tu­ally re­sults in stress, de­pres­sion, and burnout as seen in manic de­pres­sion. Most at­tempts at cog­ni­tive en­hance­ment only ad­dress the mo­ti­va­tion side of the equa­tion.

The “smart drug” cul­ture has gen­er­ally been think­ing phar­ma­ceu­ti­cally rather than biolog­i­cally. Be­hind that phar­ma­ceu­ti­cal ori­en­ta­tion there is some­times the idea that the in­di­vi­d­ual just isn’t try­ing hard enough, or doesn’t have quite the right genes to ex­cel men­tally.

-Ray Peat, PhD

Cel­lu­lar Thermodynamics

Any sim­ple ma­jor en­hance­ment to hu­man in­tel­li­gence is a net evolu­tion­ary dis­ad­van­tage.

-Eliezer Yud­kowsky (Alger­non’s Law)

I pro­pose that this con­strain is im­posed by the en­ergy cost of in­tel­li­gence. The con­ven­tional text­book view of neu­rol­ogy sug­gests that much of the brain’s en­ergy is “wasted” in over­com­ing the con­stant diffu­sion of ions across the mem­branes of neu­rons that aren’t ac­tively in use. This is nec­es­sary to keep the neu­rons in a ‘ready state’ to fire when called upon.

Why haven’t we evolved some mechanism to con­trol this mas­sive waste of en­ergy?

The As­so­ci­a­tion-In­duc­tion hy­poth­e­sis for­mu­lated by Gilbert Ling is an al­ter­nate view of cell func­tion, which sug­gests a dis­tinct func­tional role of en­ergy within the cell. I won’t re­view it in de­tail here, but you can find an easy to un­der­stand and com­pre­hen­sive in­tro­duc­tion to this hy­poth­e­sis in the book “Cells, Gels and the Eng­ines of Life” by Ger­ald H. Pol­lack (ama­zon link). This idea has a long his­tory with con­sid­er­able ex­per­i­men­tal ev­i­dence, which is too ex­ten­sive to re­view in this ar­ti­cle.

The As­so­ci­a­tion-In­duc­tion hy­poth­e­sis states that ion ex­clu­sion in the cell is main­tained by the struc­tural or­der­ing of wa­ter within the cy­to­plasm, by an in­ter­ac­tion be­tween the cy­toskele­tal pro­teins, wa­ter molecules, and ATP. En­ergy (in the form of ATP) is used to un­fold pro­teins, pre­sent­ing a reg­u­lar pat­tern of sur­face charges to cell wa­ter. This or­ders the cell wa­ter into a ‘gel like’ phase which ex­cludes spe­cific ions, be­cause their pres­ence within the struc­ture is en­er­get­i­cally un­fa­vor­able. Other ions are se­lec­tively re­tained, be­cause they are ad­sorbed to charged sites on pro­tein sur­faces. This struc­tured state can be main­tained with no ad­di­tional en­ergy. When a neu­ron fires, this or­ga­ni­za­tion col­lapses, which re­leases en­ergy and performs work. The neu­ron uses sig­nifi­cant en­ergy only to re­store this struc­tured low en­tropy state, af­ter the neu­ron fires.

This figure (bor­rowed from Gilbert Ling) sum­ma­rizes this phe­nom­ena, show­ing a folded pro­tein (on the left) and an un­folded pro­tein cre­at­ing a low en­tropy gel (on the right).

To sum­ma­rize, main­tain­ing the low en­tropy liv­ing state in a non-firing neu­ron re­quires lit­tle en­ergy. This im­plies that the brain may already be very effi­cient, where nearly all en­ergy is used to func­tion, grow, and adapt rather than pump the same ions ‘up­hill’ over and over.

Cost of Intelligence

To quote Eliezer Yud­kowsky again, “the evolu­tion­ary rea­sons for this are so ob­vi­ous as to be worth be­la­bor­ing.” Mam­malian brains may already be nearly as effi­cient as their physics and struc­ture al­lows, and any in­crease in in­tel­li­gence comes with a cor­re­spond­ing in­crease in en­ergy de­mand. Brain en­ergy con­sump­tion ap­pears cor­re­lated with in­tel­li­gence across differ­ent mam­mals, and hu­mans have un­usu­ally high en­ergy re­quire­ments due to our in­tel­li­gence and brain size.

There­fore if an or­ganism is go­ing to com­pete while hav­ing a greater in­tel­li­gence, it must be in a situ­a­tion where this ex­tra in­tel­li­gence offers a com­pet­i­tive ad­van­tage. Once in­tel­li­gence is ad­e­quate to meet the de­mands of sur­vival in a given en­vi­ron­ment, ex­tra in­tel­li­gence merely im­poses un­nec­es­sary nu­tri­tional re­quire­ments.

Th­ese ther­mo­dy­namic re­al­ities of in­tel­li­gence lead to the fol­low­ing corol­lary to Alger­non’s Law:

Any in­crease in in­tel­li­gence im­plies a cor­re­spond­ing in­crease in brain en­ergy con­sump­tion.

Po­ten­tial Implications

What is called ge­nius is the abun­dance of life and health.

-Henry David Thoreau

This idea can be ap­plied to both eval­u­ate nootrop­ics, and to un­der­stand and treat cog­ni­tive prob­lems. It’s un­likely that any drug will in­crease in­tel­li­gence with­out ad­verse effects, un­less it also acts to in­crease en­ergy availa­bil­ity in the brain. From this per­spec­tive, we can cat­e­gor­i­cally ex­clude any nootropic ap­proaches which fail to in­crease ox­ida­tive metabolism in the brain.

This idea shifts the search for nootrop­ics from neu­ro­trans­mit­ter like drugs that im­prove fo­cus and mo­ti­va­tion, to those com­pounds which reg­u­late and sup­port ox­ida­tive metabolism such as glu­cose, thy­roid hor­mones, some steroid hor­mones, choles­terol, oxy­gen, car­bon diox­ide, and en­zyme co­fac­tors.

Why haven’t we already found that these sub­stances in­crease in­tel­li­gence?

Defi­cien­cies in all of these sub­stances do re­duce in­tel­li­gence. Fur­ther rais­ing brain metabolism above nor­mal healthy lev­els should be ex­pected to be a com­plex prob­lem be­cause of the in­ter­re­la­tion be­tween the molecules re­quired to sup­port metabolism:

If you in­crease ox­ida­tive metabolism, the de­mand for all raw ma­te­ri­als of metabolism is cor­re­spond­ingly in­creased. Any sin­gle defi­ciency poses a bot­tle­neck, and may re­sult in the op­po­site of the in­tended re­sult.

So this sug­gests a ‘sys­tems biol­ogy’ ap­proach to cog­ni­tive en­hance­ment. It’s nec­es­sary to con­sider how metabolism is reg­u­lated, and what sub­strates it re­quires. To raise in­tel­li­gence in a safe and effec­tive way, all of these sub­strates must have in­creased availa­bil­ity to the neu­ron, in ap­pro­pri­ate ra­tios.

I am always leery of draw­ing analo­gies be­tween brains and com­put­ers but this ap­proach to cog­ni­tive en­hance­ment is very loosely analo­gous to over-clock­ing a CPU. Over-clock­ing re­quires rais­ing both the clock rate, and the en­ergy availa­bil­ity (voltage). In the case of the brain, the effec­tive ‘clock rate’ is con­trol­led by hor­mones (pri­mar­ily triiodothy­ro­nine aka T3), and en­ergy availa­bil­ity is pro­vided by glu­cose and other nu­tri­ents.

It’s not clear if merely rais­ing brain metabolism in this way will ac­tu­ally re­sult in a cor­re­spond­ing in­crease in in­tel­li­gence, how­ever I think it’s un­likely that the op­po­site is pos­si­ble (in­creas­ing in­tel­li­gence with­out rais­ing brain metabolism).