By Gary Marcus
May 4, 2008
How many times has this happened to you? You leave work, decide that you need
to get groceries on the way home, take a cellphone call and forget all about
your plan. Next thing you know, you've driven home and forgotten all about the
groceries.
Or this. You decide, perhaps circa Jan. 1, that it's time to lose weight; you
need to eat less, eat better and exercise more. But by the first of May, your
New Year's resolutions are a distant memory.
Human beings are, to put it gently, in a unique position in the animal world.
We're the only species smart enough to plan systematically for the future --
yet we remain dumb enough to ditch even our most carefully made plans in favor
of short-term gratification. ("Did I say I was on a diet? Mmm, but three-layer
chocolate mousse is my favorite. Maybe I'll start my diet tomorrow.")
In a wonderful study conducted at Stanford University in the late 1960s,
psychologist Walter Mischel offered preschoolers a choice: a marshmallow now,
or two marshmallows if they could wait until he returned. And then, cruelly,
he left them alone with nothing more than themselves, the single marshmallow,
a hidden camera and no indication of when he would return.
A few of the kids ate the oh-so-tempting marshmallow the minute he left the
room. But most kids wanted the bigger bonus and endeavored to wait. So they
tried. Hard. But with nothing else to do in the room, the torture was visible.
The kids did just about anything they could to distract themselves from the
tempting marshmallow that stood before them. They talked to themselves,
bounced up and down, covered their eyes, sat on their hands -- strategies that
more than a few adults might on occasion profitably adopt. Even so, for about
half the kids, the 15 to 20 minutes until Mischel returned was just too long
to wait.
Toddlers, of course, aren't the only humans who melt in the face of
temptation. Teenagers often drive at speeds that would be unsafe even on an
autobahn, and people of all ages have been known to engage in unprotected sex
with strangers, even when they are perfectly aware of the risks. (To say
nothing of the daily uncontrollable choices of alcoholics, drug addicts and
compulsive gamblers.)
What gives? Why are we as a species so often so desperately poor at achieving
our goals? If we are, as the selfish-gene theory would have it, organisms that
exist only to serve the interests of our genes, why do we waste so much of our
time doing things that are not, in any obvious way, remotely in the interest
of our genes? How can one explain, for example, why a busy undergraduate would
spend four weeks playing "Halo 3" rather than studying for his exams?
The selfish-gene theory doesn't, in itself, answer these questions, but there
is another facet of evolution that can: The fact that evolution is entirely
blind, unable to look forward, backward or to the side. As Charles Darwin
observed, evolution invariably proceeds through a process called "descent with
modification." In lay language, this means that Mother Nature never starts
from scratch, no matter how useful an overhaul might be. Everything that
evolves necessarily builds on that which came before. Our arms, to take one
simple example, are adaptations of the front legs of our primate ancestors.
In practical terms, that means that evolution's products aren't always
particularly sound. Truly dismal solutions are quickly weeded out; if someone
has a genetic condition that brings them into the world without a functioning
heart, they don't live long enough to reproduce. But merely adequate solutions
(what engineers call "kluges") -- like the awkward, injury-prone human spine,
good enough but far from perfect -- can stick around indefinitely if better
solutions are too far away on the evolutionary landscape.
In the mental machinery that governs our everyday decisions, kluges abound.
Take, for example, the scenario described in the beginning of the essay -- the
fellow who forgets his errand on the way home. His problem is clearly not in
finding his way to the grocery store -- it's in remembering to go in the first
place.
The problem is that evolution failed to realize that remembering goals is not
like recognizing objects. When your brain sees a lion, the thing to do is to
decide, lickety-split, to get out of the way. Run first; ask questions later.
We're programmed for just that kind of split-second decision; just about every
creature on the planet is built such that it can identify things like
predators and prey very rapidly. We're not programmed to remember precise
episodes from the past. Why not? Because remembering the exact date on which
you last saw a lion is not particularly helpful when you're trying to get out
of the way.
Alas, evolution didn't have the foresight to realize that different kinds of
tasks require different kinds of memory, and it used the same basic sort of
memory for everything, not just for remembering what lions and tigers look
like (in which general tendencies suffice) but also for cases -- like tracking
our goals -- where a bit more precision would have been helpful. As a result,
trying to remember what to do next can be a little like trying to remember
what you had for breakfast yesterday: There are too many breakfasts and too
many yesterdays for our biological memories to keep track of.
The same thing can happen with our goals. When you sit in your car late in the
day and ask yourself, "What am I supposed to do next?" and all of a sudden the
cellphone rings, your brain can easily lose track of which "next step" is the
right one. Instead of zeroing in on the specific memory it needs, it may well
settle for remembering whatever you've done in the car most often -- and
that's drive home. Voila,
autopilot.
Our attempts to pursue our goals are often thwarted by the fact that evolution
has built our most sophisticated technologies on top of older technologies --
without working out how to integrate the two. We can plan in advance, using
our modern deliberative reasoning systems, but our ancestral reflexive
mechanisms, which evolved first, still basically control the steering wheel.
When the chips are down, it's those mechanisms that our brains turn to, and
that means that our brains frequently wind up relying on machinery that is all
about acting first and asking questions later, squandering some of the efforts
of our deliberative system.
No sensible engineer would have designed things this way. Why design fancy
machinery for making long-term goals if you're not going to use it? Yet the
brain is structured such that the more tired, stressed or distracted we are,
the less likely we are to use our forebrains and the more likely to lean back
on the time-tested but shortsighted machinery we've inherited from our
ancestors.
Still, all is not lost. Even though our short-term desires are pretty good at
grabbing the steering wheel of our consciousness, our more recently evolved
deliberate minds are powerful enough to regain at least some measure of
control.
Consider, for example, the difficulty that most people having in sticking to
abstract goals like "I intend to lose weight" or "I plan to finish this
article before the deadline." Nice thoughts, but not formulated in terms that
your ancestral, reflexive brain might understand. The work-around? Translate
those abstract goals into a form your ancestral systems -- which traffic
largely in dumb reflexes -- can understand: if-then. If you
find yourself in a particular situation, then take
a specific action: "If I see French fries, then I will avoid them." As Peter
Gollwitzer, my colleague in New York University's department of psychology,
has shown, even simple changes like these can markedly increase the chances of
success.
Our conscious, deliberate systems will never have total control, and our
memories will never be perfect, but as they say in Alcoholics Anonymous,
recognition is the first step. If we come to recognize our limitations, and
how they evolved, we just might be able to outwit our inner kluge.
Gary Marcus, a professor of psychology at New York University, is the
author, most recently, of "Kluge: The Haphazard Construction of the Human
Mind."
