Research shows that our human brains are shrinking rapidly. 

Before we all panic and start trying drastic and somewhat incomprehensible measures like “brain push ups” or “doing all our thinking now rather than later” this brain shrinkage is rapid in evolutionary time – about 10% in 20,000 years (Bailey & Geary, 2009).  If the shrinkage rate is constant, our brains today are similar in size to brains of the year 0 AD and will still be similar in size to brains of the year 4000 AD.  To say our brains are shrinking is akin to saying our sun is expanding and burning out.  Both are true, but of very little immediate, near term, or generational concern.

Nevertheless, we live in an age of anxiety.  Perhaps we constantly feel at a race with competition – in this case, our machines.  Moore’s Law shows computing power expanding exponentially.  Computational power grows roughly 100% every 2 years.  Meanwhile, our brains are shrinking.  Any way we extrapolate this leaves us one day, as Ken Jennings once famously joked on Jeopardy, “welcom[ing] our new computer overlords.”     

A technologist might start building machines to curry favor.  A futurist might start popping vitamins to survive and bear witness.  A business might invent and market the concept of ripped Brain Exercise Centers to capitalize on the anxiety.  A scientist would ask what these empirical trends really mean.

Some current theories on why brains are shrinking include: brains are shrinking because humans are getting smarter; brains are shrinking because humans are getting dumber and mellower; and climate change is causing shrinkage with no discernible effect on intelligence.  That researchers can propose these and other theories seriously shows that all have some merit.  That all have merit shows that we may lack a comprehensive concept of what brain size means.  Does a big brain cause problems like an enlarged heart?  Does a small brain reduce functionality like in hypogonadism?

Drilling down in the details shows that brains are not homogenous; therefore, “more brain” can mean many different things.  While there are many ways to subdivide the brain, one simple way of viewing it is in three parts: the core R-complex, the middle layer limbic system, and the outermost neocortex.  The R-complex appears in all vertebrates and affects wakefulness, hunger, desires, and motor movement.  The limbic system appears in mammals and affects emotions and feelings.  The neocortex appears in primates – mostly humans – and affects conscious thought.  This division is a gross simplification, but it can show that not all brain is equal.  A large brain full of R-complex might serve to make a powerful cold-blooded lizard.  A small brain consisting mainly of neocortex may be useful if the organism does not have many tails or legs and none need to be accelerated to very high velocities to knock down a meal.  It is clear then that there must be a balance of components and size in the brain for it to excel at its goal.  What is the brain’s goal?

If the brain’s goal is to be fast, as in GigaHertz fast like modern computers, then it could do this with a larger population of cells sequentially cycling.  If the brain’s goal is to store many memories, as in TerraByte size in a bit-wise system, then it could do this with a larger population of cells storing a pattern of active and inactive states.  These goals are the computer’s goals, in order to calculate instructions faster and to store more results.  If these truly are also the brain’s goals, then there may not be any need for extrapolating with Moore’s Law.   We should welcome our computer overlords today.  But if these are not the brain’s goals, then comparing computing power growth under Moore’s Law with human brain size over the past 20,000 years is akin to comparing a hammer and a screwdriver.  No matter how big or dense or fast or effective new generations of hammers become, they can never be anything more than a very, very poor screwdriver and vice versa. 

The simplistic evolutionary goal of the brain is to guide the organism to effectively maintain and disseminate its genetic package.  The brain is inseparably linked to the peripheral nerves, muscles, bones, organs, and by extension, to the environment.  This is an interpretation of the currently favored socio-cultural learning theory (Siegler & Alibali, 2005).  Applied to brain function and perceived intelligence, a quick-witted person cleverly and aptly responding in a conversation does not exhibit high-speed brain calculations but rather shows a wide variety of past, lively, and exciting life experiences to draw upon.  This signals a healthy, wealthy body able to travel and enjoy some comfort of leisure – always attractive in a mate.  A person who talks for hours constructing a high fidelity image of a brick wall 3 feet from their window is universally less attractive than a person who lightly speaks of past visits to London, Barcelona, Paris, or Prague all the while subtly tailoring their conversation based on listener reactions for the same reasons.  Using an accent slightly foreign to the local listeners and passably consistent with such places would be an added benefit. 

Brains are getting smaller and computers are getting more powerful.  This means very little.  We are still alive, eating, breathing, reading, and growing today because it means very little.  It provides much needed evidence that needs to fit a model of the brain.  It means that one must not think of brains as being natural counterparts to “electric brains,” as rough translations of some languages’ word for computer go.  It means that we may need to expand our modeling to encompass the social learning environment in order to better understand the individual brain and its role.  But in terms of the arms race between computers and humans, it means very little.  Unless, of course society changes so dramatically and universally to value high fidelity images of brick walls and we should learn the phrase, “1001000 1101001.”