Read Mind Hacks™: Tips & Tools for Using Your Brain Online
Authors: Tom Stafford,Matt Webb
Tags: #COMPUTERS / Social Aspects / Human-Computer Interaction
Mind Hacks™: Tips & Tools for Using Your Brain (37 page)
People don’t like change. If you really want people to try something new, you
should just coerce them into giving it a go and chuck the idea of persuading them straight
off.
By default, people side with what already is and what happened last time. We’re curious,
as animals go, but even humans are innately conservative. Like the Dice Man, who delegates
all decisions to chance in Luke Rhinehart’s classic 1970s novel of the same name, was told:
“It’s the way a man chooses to limit himself that determines his character. A man without
habits, consistency, redundancy — and hence boredom — is not human. He’s insane.”
1
In this hack we’re going to look at our preference for the way things are and where this
tendency comes from. I’m not claiming that people don’t change — obviously this happens all
the time and is the most interesting part of life — but, in general, people are consistent and
tend toward consistency. Statistically, if you want to predict what people will do in a
familiar situation, the most useful thing you can measure is what they did last time. Past
action correlates more strongly with their behavior than every other variable psychologists
have tried to measure.
2
If you’re interested in predicting who people will vote for, what they will
buy, what kind of person they will sleep with, anything at all really, finding out what
tendencies they’ve exhibited or what habits they’ve formed before is the most useful
information at your disposal. You’re not after what they
say
they will
do — not what party, brand, or sexual allegiance they tick on a form — nor the choice they think
they’re feeling pressured into making. Check out what they actually did last time and base
your prediction on that. You won’t always be right, but you will be right more often by
basing your guess upon habit than upon any other single variable.
This bias is the result of a number of factors, not least the fact that people’s
previous choice is often the best one or the one that best reflects their character. But
also we have mental biases,
3
like the mental biases we have about numbers
[
Use Numbers Carefully
]
, which produce consistent habits and an
innate conservativism.
Biases in reasoning are tendencies, not absolutes. They make up the mental forces that
push your conclusions one way or the other. No single force ever rules completely, and in
each case, several forces compete. We’re mostly trying to be rational so we keep a look out
for things that might have biased us so we can discount them. Even if we know we can’t be
rational, we mostly try to be at least consistent. This means that often you can’t give the
same person the same problem twice if it’s designed to evoke different biases.
They’ll spot the similarity between the two presentations and know their answers
should be the same.
I’m carelessly using the word “rational” here, in the same way that logicians and
people with a faith in pure reason might. But the study of heuristics and biases should
make us question what a psychological meaning of “rational” could be. In some of the very
arbitrary situations contrived by psychologists, people can appear to be irrational, but
often their behavior would be completely reasonable in most situations, and even rational
considering the kind of uncertainties that normally accompany most choices in the everyday
world.
— T.S.
But some biases are so strong that you can feel them tugging on your reason even when
the rational part of your mind knows they are misleading. These “cognitive illusions” work
even when you present two differently biased versions of the choice side by side. The
example we’re going to see in action is one of these.
I’m going to tell you in advance that the two versions of the problem are logically
identical, but I know — because your brain evolved in the same way mine did — that you’ll feel
as if you want to answer them differently despite knowing this. If your supreme powers of
reason don’t let you feel the tug induced by the superficial features of the problem (the
bit that conveys the bias), take the two versions and present them to two different
friends.
Here we go...
A lethal disease is spreading through the city of which you are mayor. It is
expected to kill 600 people. Your chief medical adviser tells you that there is a choice
between two treatment plans. The first strategy will definitely save 200 people, whereas
the second strategy has a one-third chance of saving 600 people and a two-thirds chance
of saving no one. Which strategy do you choose?
A lethal disease is spreading through the city of which you are mayor. It is
expected to kill 600 people. Your chief medical adviser tells you that there is a choice
between two treatment plans. The first strategy will definitely kill 400 people, whereas
the second strategy has a one-third chance that nobody will die and a two-thirds chance
that 600 people will die.
Do you feel it? The choices feel different, even though you know they are
the same. What’s going on?
At least two things are going on here. The first is the effect of the way the choice
is presented — the
framing effect
. Anyone who has ever tried to
persuade someone of something knows the importance of this. It’s not just what you say,
but how you say it, that is important when presenting people with a choice or argument.
The second thing is a bias we have against risking an already satisfactory situation — we’re
much more willing to take risks when we’re in a losing position to begin with. In the
examples, the first frame makes it look like you stand to gain without having to take a
risk — the choice is between definitely saving 200 people versus an all-or-nothing gamble.
The second frame makes it appear as though you start in a losing position (400 people
down) and you can risk the all-or-nothing gamble to potentially improve your standing. In
experimental studies of this dilemma, around 75% of people favor not gambling in the first
frame, with the situation reversed in the second.
4
So why do we gamble when we think we might lose out, but have a bias to avoid gambling
on gains? I’m going to argue that this is part of a general bias we have toward the way
things are. Let’s call it the “status quo bias.” This is probably built into our minds by
evolution — nature’s way of saying “If it ain’t broke, don’t fix it.”
With habits, it is easy to see why the status quo bias is evolutionary adaptive. If
you did it last time and it didn’t kill you, why do it differently? Sure, you could try
things differently, but why waste the effort, especially if there’s any risk at all of
things getting worse?
There’s a way to hack this habit bias, and it’s well-known to advertisers. If people
generally stick with what they know, the most important thing you can do is get them to
start off with your product in the first place (hence the value of kids as a target
market). But you can make use of the bias: people choose based on what they did before, so
it is more effective to advertise to influence what they choose rather than how they feel
about that choice. Even if there’s no good reason for someone using your product in the
first place, the fact that they did once has established a strong bias for them doing so
again. A computer user may prefer one browser, but if another one comes bundled with her
new operating system, we can bet that’s what she’ll end up relying on. You may have no
rational reason for choosing Brand A
over Brand B when you buy jam, but if the manufacturers of Brand B can get you
to try it (maybe by giving you a free sample or a special offer), they’ve overcome the
major barrier that would have stopped you from buying it next time.
Status quo bias works for beliefs as well as behaviors. In many situations we are
drawn to confirm what we already know, rather than test it in a way that might expose it
to be false
[
Detect Cheaters
]
.
It’s an experience I’ve had a lot when debugging code. I do lots of things that
prove to me that it must be the bug I first think it is, but when I fix that bug, my
code still doesn’t work.
It’s not just me, right?
— T.S.
Another manifestation of our preference for the way things are is the so-called
endowment effect
,
5
whereby once we have something, however we acquired it, we give it more
value than we would give up to obtain it. In one study, students were given a mug with
their university emblem, worth $6. In a trading game they subsequently wanted an average
of around $5 to give up their mug, whereas students without mugs were willing to offer an
average of only around $2 to buy a mug. The mere sense of ownership that came with being
given the mug was enough to create a difference between how the two groups valued the
object. This is just one of the ways in which human behavior violates the rationality
supposed by classical economic theory.
So we can see that if you want people to give something up, you shouldn’t give it to
them in the first place, and if you want to introduce something new, you should make
people try it before trying to persuade them to accept it. If you can’t do this, you
should at least try and introduce the new change elements as part of the familiar
experience.
- Rhinehart, L. (1971).
The Dice Man
. - Ajzen, I. (2002). Residual effects of past on later behavior:
Habituation and reasoned action perspectives.
Personality and Social
Psychology Review, 6
, 107–122. See also: Ouellette, J. A., &
Wood, W. (1998). Habit and intention in everyday life: The multiple processes by which
past behavior predicts future behavior.
Psychological Bulletin,
124
, 54–74. - The Wikipedia has an enjoyable, if unstructured, list of cognitive
biases (
http://en.wikipedia.org/wiki/List_of_cognitive_biases
). A good introduction to cognitive biases and heuristics is Nicholls, N.
(1999). Cognitive illusions, heuristics, and climate prediction.
Bulletin of
the American Meteorological Society, 80
(7), 1385–1397. - Tversky, A., & Kahneman, D. (1981). The framing of decisions
and the psychology of choice.
Science
, 211, 453–458. - Kahneman, D., Knetch, J. L., & Thaler, R. H. (1991).
Anomalies: The endowment effect, loss aversion, and status quo bias.
Journal
of Economic Perspectives, 5
(1), 193–206. A reverse of the endowment
effect is the windfall effect in which people value less highly money they didn’t
expect to come to them (like lottery wins and inheritance).
What makes “this” a word, rather than being simply the adjacently written letters
t
,
h
,
i
,
s
? Or, to ask a similar question, why should we see a single dog
running across a field rather than a collection of legs, ears, hair, and a wet nose flying
over the grass? And why, when the dog knocks us over, do we know to blame the dog?
To put these questions another way: how do we group sensations into whole objects, and how
do we decide that a certain set of perceptions constitutes cause and effect?
It’s not a terribly easy problem to solve. The nature of causality isn’t transmitted in an
easy-to-sense form like color is in light. Rather than sense it directly, we have to guess. We
have built-in heuristics to do just that, and these heuristics are based on various forms of
togetherness. The word “this” hangs together well because the letters are in a straight line,
for example, and they’re closer to one another than the letters in the surrounding words.
Those are both principles by which the brain performs grouping. To take the second question,
we see the parts of the dog as a single animal because they move together. That’s another
heuristic.
This recognition acuity lets us see human forms from the tiniest of clues, but it also — as
we’ll see in
See a Person in Moving Lights
— is not perfect and can be duped.
We’ll see how we can perceive animacy — the aliveness shown by living creatures — where none
exists and how we can ignore the cause in cause and effect. Sometimes that’s the best way to
find out what our assumptions really are, to see when they don’t quite match what’s happening
in the real world.
We group our visual perceptions together according to the gestalt grouping
principles. Knowing these can help your visual information design to sit well with
people’s expectations.
It’s a given that we see the world not as isolated parts, but as groups and single
objects. Instead of seeing fingers and a palm, we see a hand. We see a wall as a unit rather
than seeing the individual bricks. We naturally group things together, trying to make a
coherent picture out of all the individual parts. A few fundamental grouping principles can
be used to do most of the work, and knowing them will help you design well-organized, visual
information yourself.
Automatic grouping is such second nature that we really notice only its absence. When
the arrangement of parts doesn’t sit well with the grouping principles the brain uses,
cracks can be seen.
Figure 8-1
shows some
of these organizational rules coming into play.
1
appear to point either way, depending on which group you see it being part of
2
You don’t see 17 triangles. Instead, you see two groups of eight and one triangle in
the middle. Your similarity drive has formed the arrangement into rows and columns of the
shapes and put them into two groups: one group points to the bottom left, the other points
off to the right.
Each group belongs together partly because the triangles are arranged into a pattern
(two long rows pointing in a direction) and partly because of proximity (shapes that are
closer together are more likely to form a group). The triangle in the middle is a long way
from both groups and doesn’t fall into the same pattern as either. It’s left alone by the
brain’s grouping principles.
You can, however, voluntarily group the lone triangle. By mentally putting it with the
left-hand set, it appears to point down and left along with the other triangles. You can
make it point right by choosing to see it with the other set.
The rules by which the brain groups similar objects together are called
gestalt grouping principles
in psychology. Although there’s no
direct German-to-English translation, “gestalt” means (roughly) “whole.” When we
understand objects and the relationships between them in a single, coherent pattern rather
than as disconnected items, we understand the group as a gestalt. We have a gestalt
comprehension of each of the sets of triangles in
Figure 8-1
, for instance.
Four of the most commonly quoted grouping principles are proximity, similarity,
closure, and continuation. An example of each is shown in
Figure 8-2
.
Proximity
- We preconsciously group items that are close together, so in the picture you see
columns rather than rows or a grid. This principle is the cause of the triangles in
the original diagram coming together into two sets and the reason the lone triangle
didn’t feel part of either of them.
Similarity
- We prefer to group together objects of the same kind. In the example, you see
alternating rows of circles and squares rather than columns of mixed shapes.
Closure
- There’s a tendency to complete patterns. There’s no triangle in the example
pattern, but we see one because the arrangement of the three Pac Man shapes would be
completed if one were there.
Continuation
- Just as we like to see completed patterns, we like seeing shapes that continue
along the same path, smoothly. We see two lines crossing in the example, rather than
two arrowheads touching at their points or four lines meeting together.
When none of these principles apply, it’s still possible to mentally group items
together. When you put the middle triangle in
Figure 8-1
with one group or the other, it
picks up the orientation of the group as a whole. It’s a voluntary grouping that modifies
how you see.
Gestalt principles exist in visual processing not because they are always
right, but because on average, they are useful. They’re good rules of thumb for making
sense of the world. It’s not that similar things can’t be separate; it’s more that most of
the time they aren’t. Although random coincidences can happen, they are vastly outnumbered
by meaningful coincidences.
The world isn’t a mess of disconnected parts, and it’s useful to see the
connections — if you’re hunting an animal, it makes sense to see it as a single gestalt
rather than a paw here and a tail there.
- The gestalt grouping principles are interesting, but are they really
useful? Good print or web page design involves easy comprehension, and knowing how the
principles can conflict or mislead in your layout helps along the way. James Levin has
applied the gestalt grouping principles to web design (
http://tepserver.ucsd.edu:16080/~jlevin/gp
). - Illustration inspired by Fred Attneave’s demonstrations as used in
“How the Mind Works” by Steven Pinker.
- Max Wertheimer was the first to identify the principles in his 1923 paper “Laws of
Organization in Perceptual Forms” (
http://psy.ed.asu.edu/~classics/Wertheimer/Forms/forms.htm
). - As well as the basic grouping principles, which work looking at static objects,
others deduce grouping from behavior over time
[
To Be Noticed, Synchronize in Time
]
.
We tend to group together things that happen at the same time or move in the same way.
It’s poor logic but a great hack for spotting patterns.
It’s a confusing, noisy world out there. It’s easier to understand the world if we
perceive a set of objects rather than just a raw mass of sensations, and one way to do this
is to group together perceptions that appear to have the same cause. The underlying
assumptions involved manifest as the
gestalt grouping principles
, a set
of heuristics used by the brain to lump things together (see
Grasp the Gestalt
for the simplest of these, used for vision).
Perhaps the most powerful of these assumptions is termed
common
fate
. We group together events that occur at the same time, change in the same
way, or move in the same direction. Imagine if you saw, from far off, two points of light
that looked a bit like eyes in the dark. You might think they were
eyes or you could just put it down to a coincidence of two unrelated lights. But
if the points of light moved at the same time, in the same direction, bounced with the
characteristic bounce of a person walking, you’d
know
they were eyes.
Using behavior over time allows you to stringently test spatial data for possible common
cause. If the bouncing lights pass the common fate test, they’re almost certainly a single
object. Visual system tags this certainty by providing you with a correspondingly strong
perceptual experience; if some things move together, it is almost impossible to see them as
separate items instead of a coherent whole.
“Illusion — Motion Capture — Grouping” (
http://psy.ucsd.edu/chip/illu_mot_capt_grpng.html
; a Real video requiring Real Player) demonstrates just how completely your
perception of a single item is altered by global context and common fate. Watch the video
for at least 30 seconds. At first you see just a dot blinking on and off next to a square.
But then other dots are added in the surrounding area, and as the first dot blinks off,
they all shift right. Now your unavoidable impression is of the first dot moving behind
the square. The appearance of the other dots, and their behavior, gives your visual system
correlations that are just too strong to ignore. The single dot is still blinking on and
off — you just can’t see it like that any more.
“A Time Gestalt Principle Example: Common Fate” (
http://tepserver.ucsd.edu/~jlevin/gp/time-example-common-fate
; a Java applet),
1
shown in
Figure 8-3
, is an
interactive demonstration of how your visual system deduces the shape of objects from
movement, without any color or shading clues to help out.
You see a shape with a static-like texture moving across a similarly randomized
background. Click anywhere in the image to start and stop the demo. Frozen, there is no
pattern to see; you see just a random mess. This is the real force of common fate. The
correlations exist only across time, in movement — it’s only when the demo is moving that
you can see an object among the noise.
The gestalt grouping inferences are so preconscious and automatic that it’s hard to
imagine perceiving a world that the brain
hasn’t
organized into
objects. There’s something very clever going on here; we are taking in very little
information (only how the pattern changes over time), yet, in combination with an
assumption that accidental correlations of visual patterns are
unlikely, we construct a compelling perception of an object. In these demos,
you just can’t ignore the object. You are utterly unable to make yourself see a moving
collection of dots instead of the shape in motion because the construction of the object
is happening before the level of consciousness.
this, it’s invisible
Common fate can lead to some sophisticated inferences. “Kinetic Depth” (
http://www.lifesci.sussex.ac.uk/home/George_Mather/Motion/KDE.HTML
; a QuickTime video), just from a collection of moving lights, allows you to
see an object with three-dimensional depth moving in a particular way. In this case, the
pattern of dots causes you to see a sphere rotating on its axis.
What’s really cute about this video is that there’s an ambiguity in the visual
information — you can see the sphere rotating in one of two ways. Your visual system makes a
choice for you, and you see some of the dots moving behind some of the others, which move
in the opposite direction. The set you see as “in front” determines the direction in which
you see the sphere
rotating. If you watch for a while, your perception will switch and you’ll see
it reverse. You don’t need to make any effort to for this to happen; it occurs naturally,
probably due to some kind of adaptation process. Since you see the sphere rotating in one
particular direction, the neurons that represent that perception will be active. Over
time, they actively tune down their response, and the neurons that code for the other
apparent rotation can now dominate. This kind of gain control
[
Get Adjusted
]
plays a similar role in motion
aftereffects
[
See Movement When All Is Still
]
, in which neurons that are active for particular directions of movement
down-regulate after being consistently stimulated and neurons active for the opposing
direction take over and dominate our perception when the consistent moving stimulus is
removed.
All these demonstrations show just how effective correlations over time are in molding
our perception. And not just perception — synchronizing stimuli can actually alter your body
image, where your brain believes your hands are
[
Mold Your Body Schema
]
, for instance. The heart of the thing is
similar — if two things correlate exactly, our perception treats them as part of the same
object. For our brains, isolated inside the skull, perceived correlation is the only way
we’ve ever had for deducing what sensations should be associated together as part of the
same object.
Common fate can also draw inferences from points of light moving in much more complex
ways than rotating spheres. For the case of biological motion
[
See a Person in Moving Lights
]
, the visual system is
specifically prepared to fit moving points into a schema based upon the human body to help
perception of the human form. Alais et al. have suggested that the importance of common
fate reflects a deeper principle of the brain’s organization.
2
Neuroscientists talk about the
binding problem
, the
question of how the brain correctly connects together all the information it is dealing
with: all the things that are happening in different parts of the world, detected by
different senses, whose component parts have properties represented in different cortical
areas (such as color, contrast, sounds, and so on), all of which have to be knitted
together into a coherent perception. The suggestion is that common fate reflects
synchronization of neuron firing — and that is this same mechanism that may underlie the
brain’s solution to the binding problem.
- Part of Jim Levin’s “Gestalt Principles & Web Design” (
http://tepserver.ucsd.edu:16080/~jlevin/gp
). Applet developed by Adam Doppelt. - Alais, D., Blake, R., & Lee, S. (1998). Visual features that
vary together over time group together over space.
Nature Neuroscience.
1
(2), 160–164.