[tt] Rational and Irrational Thought: The Thinking That IQ Tests Miss
Eugen Leitl
<eugen at leitl.org> on
Fri Oct 30 20:13:46 CET 2009
http://www.scientificamerican.com/article.cfm?id=rational-and-irrational-thought&print=true
Rational and Irrational Thought: The Thinking That IQ Tests Miss
We assume intelligence and rationality go together. But we shouldn't be
surprised when smart people do foolish things.
By Keith E. Stanovich
No doubt you know several folks with perfectly respectable IQs who just don’t
seem all that sharp. The behavior of such people tells us that we are missing
something important by treating intelligence as if it encompassed all
cognitive abilities. I coined the term “dysrationalia” (analogous to
“dyslexia”), meaning the inability to think and behave rationally despite
having adequate intelligence, to draw attention to a large domain of
cognitive life that intelligence tests fail to assess. Although most people
recognize that IQ tests do not measure important mental faculties, we behave
as if they do. We have an implicit assumption that intelligence and
rationality go together—or else why would we be so surprised when smart
people do foolish things?
It is useful to get a handle on dysrationalia and its causes because we are
beset by problems that require increasingly more accurate, rational
responses. In the 21st century, shallow processing can lead physicians to
choose less effective medical treatments, can cause people to fail to
adequately assess risks in their environment, can lead to the misuse of
information in legal proceedings, and can make parents resist vaccinating
their children. Millions of dollars are spent on unneeded projects by
government and private industry when decision makers are dysrationalic,
billions are wasted on quack remedies, unnecessary surgery is performed and
costly financial misjudgments are made.
IQ tests do not measure dysrationalia. But as I show in my new book, What
Intelligence Tests Miss: The Psychology of Rational Thought, there are ways
to measure it and ways to correct it. Decades of research in cognitive
psychology have suggested two causes of dysrationalia. One is a processing
problem, the other a content problem. Much is known about both of them.
The Case of the Cognitive Miser
The processing problem comes about because we tend to be cognitive misers.
When approaching a problem, we can choose from any of several cognitive
mechanisms. Some mechanisms have great computational power, letting us solve
many problems with great accuracy, but they are slow, require much
concentration and can interfere with other cognitive tasks. Others are
comparatively low in computational power, but they are fast, require little
concentration and do not interfere with other ongoing cognition. Humans are
cognitive misers because our basic tendency is to default to the processing
mechanisms that require less computational effort, even if they are less
accurate.
Are you a cognitive miser? Consider the following problem, taken from the
work of Hector Levesque, a computer scientist at the University of Toronto.
Try to answer it yourself before reading the solution.
1. Jack is looking at Anne, but Anne is looking at George. Jack is married,
but George is not. Is a married person looking at an unmarried person?
A) Yes
B) No
C) Cannot be determined
More than 80 percent of people choose C. But the correct answer is A. Here is
how to think it through logically: Anne is the only person whose marital
status is unknown. You need to consider both possibilities, either married or
unmarried, to determine whether you have enough information to draw a
conclusion. If Anne is married, the answer is A: she would be the married
person who is looking at an unmarried person (George). If Anne is not
married, the answer is still A: in this case, Jack is the married person, and
he is looking at Anne, the unmarried person. This thought process is called
fully disjunctive reasoning—reasoning that considers all possibilities. The
fact that the problem does not reveal whether Anne is or is not married
suggests to people that they do not have enough information, and they make
the easiest inference (C) without thinking through all the possibilities.
Most people can carry out fully disjunctive reasoning when they are
explicitly told that it is necessary (as when there is no option like “cannot
be determined” available). But most do not automatically do so, and the
tendency to do so is only weakly correlated with intelligence.
Here is another test of cognitive miserliness, as described by Nobel
Prize–winning psychologist Daniel Kahneman and his colleague Shane Frederick.
2. A bat and a ball cost $1.10 in total. The bat costs $1 more than the ball.
How much does the ball cost?
Many people give the first response that comes to mind—10 cents. But if they
thought a little harder, they would realize that this cannot be right: the
bat would then have to cost $1.10, for a total of $1.20. IQ is no guarantee
against this error. Kahneman and Frederick found that large numbers of highly
select university students at the Massachusetts Institute of Technology,
Princeton and Harvard were cognitive misers, just like the rest of us, when
given this and similar problems.
Another characteristic of cognitive misers is the “my side” bias—the tendency
to reason from an egocentric perspective. In a recent study my colleague
Richard West of James Madison University and I presented a group of subjects
with the following thought problem.
3. Imagine that the U.S. Department of Transportation has found that a
particular German car is eight times more likely than a typical family car to
kill occupants of another car in a crash. The federal government is
considering restricting sale and use of this German car. Please answer the
following two questions: Do you think sales of the German car should be
banned in the U.S.? Do you think the German car should be banned from being
driven on American streets?
Then we presented a different group of subjects with the thought problem
stated a different way—more in line with the true data from the Department of
Transportation, which had found an increased risk of fatalities not in a
German car but in an American one.
Imagine that the Department of Transportation has found that the Ford
Explorer is eight times more likely than a typical family car to kill
occupants of another car in a crash. The German government is considering
restricting sale or use of the Ford Explorer. Please answer the following two
questions: Do you think sales of the Ford Explorer should be banned in
Germany? Do you think the Ford Explorer should be banned from being driven on
German streets?
Among the American subjects we tested, we found considerable support for
banning the car when it was a German car being banned for American use: 78.4
percent thought car sales should be banned, and 73.7 percent thought the car
should be kept off the streets. But for the subjects for whom the question
was stated as whether an American car should be banned in Germany, there was
a statistically significant difference: only 51.4 percent thought car sales
should be banned, and just 39.2 percent thought the car should be kept off
German streets, even though the car in question was presented as having
exactly the same poor safety record.
This study illustrates our tendency to evaluate a situation from our own
perspective. We weigh evidence and make moral judgments with a my-side bias
that often leads to dysrationalia that is independent of measured
intelligence. The same is true for other tendencies of the cognitive miser
that have been much studied, such as attribute substitution and conjunction
errors; they are at best only slightly related to intelligence and are poorly
captured by conventional intelligence tests.
The Mindware Gap
The second source of dysrationalia is a content problem. We need to acquire
specific knowledge to think and act rationally. Harvard cognitive scientist
David Perkins coined the term “mindware” to refer to the rules, data,
procedures, strategies and other cognitive tools (knowledge of probability,
logic and scientific inference) that must be retrieved from memory to think
rationally. The absence of this knowledge creates a mindware gap—again,
something that is not tested on typical intelligence tests.
One aspect of mindware is probabilistic thinking, which can be measured. Try
to answer the following problem before you read on.
4. Imagine that XYZ syndrome is a serious condition that affects one person
in 1,000. Imagine also that the test to diagnose the disease always indicates
correctly that a person who has the XYZ virus actually has it. Finally,
suppose that this test occasionally misidentifies a healthy individual as
having XYZ. The test has a false-positive result of 5 percent, meaning that
the test wrongly indicates that the XYZ virus is present in 5 percent of the
cases where the person does not have the virus.
Next we choose a person at random and administer the test, and the person
tests positive for XYZ syndrome. Assuming we know nothing else about that
individual’s medical history, what is the probability (expressed as a
percentage ranging from zero to 100) that the individual really has XYZ?
The most common answer is 95 percent. But that is wrong. People tend to
ignore the first part of the setup, which states that only one person in
1,000 will actually have XYZ syndrome. If the other 999 (who do not have the
disease) are tested, the 5 percent false-positive rate means that
approximately 50 of them (0.05 times 999) will be told they have XYZ. Thus,
for every 51 patients who test positive for XYZ, only one will actually have
it. Because of the relatively low base rate of the disease and the relatively
high false-positive rate, most people who test positive for XYZ syndrome will
not have it. The answer to the question, then, is that the probability a
person who tests positive for XYZ syndrome actually has it is one in 51, or
approximately 2 percent.
A second aspect of mindware, the ability to think scientifically, is also
missing from standard IQ tests, but it, too, can be readily measured.
5. An experiment is conducted to test the efficacy of a new medical
treatment. Picture a 2 × 2 matrix that summarizes the results as follows:
Treatment Given = (Improvement: 200); (No Improvement: 75)
No Treatment Given (Improvement: 50); (No Improvement: 15)
As you can see, 200 patients were given the experimental treatment and
improved; 75 were given the treatment and did not improve; 50 were not given
the treatment and improved; and 15 were not given the treatment and did not
improve. Before reading ahead, answer this question with a yes or no: Was the
treatment effective?
Most people will say yes. They focus on the large number of patients (200) in
whom treatment led to improvement and on the fact that of those who received
treatment, more patients improved (200) than failed to improve (75). Because
the probability of improvement (200 out of 275 treated, or 200/275 = 0.727)
seems high, people tend to believe the treatment works. But this reflects an
error in scientific thinking: an inability to consider the control group,
something that (disturbingly) even physicians are guilty of. In the control
group, improvement occurred even when the treatment was not given. The
probability of improvement with no treatment (50 out of 65 not treated, or
50/65 = 0.769) is even higher than the probability of improvement with
treatment, meaning that the treatment being tested can be judged to be
completely ineffective.
Another mindware problem relates to hypothesis testing. This, too, is rarely
tested on IQ tests, even though it can be reliably measured, as the late
Peter C. Wason of University College London has shown. Try to solve the
following puzzle, called the four-card selection task, before reading ahead.
6. As seen in the diagram (A, K, 8, 5), four cards are sitting on a table.
Each card has a letter on one side and a number on the other. Two of the
cards are letter-side up, and two of the cards are number-side up. The rule
to be tested is this: for these four cards, if a card has a vowel on its
letter side, it has an even number on its number side. Your task is to decide
which card or cards must be turned over to find out whether the rule is true
or false. Indicate which cards must be turned over.
Most people get the answer wrong, and it has been devilishly hard to figure
out why. About half of them say you should pick A and 8: a vowel to see if
there is an even number on its reverse side and an even number to see if
there is a vowel on its reverse. Another 20 percent choose to turn over the A
card only, and another 20 percent turn over other incorrect combinations.
That means that 90 percent of people get it wrong.
Let’s see where people tend to run into trouble. They are okay with the
letter cards: most people correctly choose A. The difficulty is in the number
cards: most people mistakenly choose 8. Why is it wrong to choose 8? Read the
rule again: it says that a vowel must have an even number on the back, but it
says nothing about whether an even number must have a vowel on the back or
what kind of number a consonant must have. (It is because the rule says
nothing about consonants, by the way, that there is no need to see what is on
the back of the K.) So finding a consonant on the back of the 8 would say
nothing about whether the rule is true or false. In contrast, the 5 card,
which most people do not choose, is essential. The 5 card might have a vowel
on the back. And if it does, the rule would be shown to be false, because
that would mean that not all vowels have even numbers on the back. In short,
to show that the rule is not false, the 5 card must be turned over.
When asked to prove something true or false, people tend to focus on
confirming the rule rather than falsifying it. This is why they turn over the
8 card, to confirm the rule by observing a vowel on the other side, and the A
card, to find the confirming even number. But if they thought scientifically,
they would look for a way to falsify the rule—a thought pattern that would
immediately suggest the relevance of the 5 card (which might contain a
disconfirming vowel on the back). Seeking falsifying evidence is a crucial
component of scientific thinking. But for most people, this bit of mindware
must be taught until it becomes second nature.
Dysrationalia and Intelligence
The modern period of intelligence research was inaugurated by Charles
Spearman in a famous paper published in 1904 in the American Journal of
Psychology. Spearman found that performance on one cognitive task tends to
correlate with performance on other cognitive tasks. He termed this
correlation the positive manifold, the belief that all cognitive skills will
show substantial correlations with one another. This belief has dominated the
field ever since.
Yet as research in my lab and elsewhere has shown, rational thinking can be
surprisingly dissociated from intelligence. Individuals with high IQs are no
less likely to be cognitive misers than those with lower IQs. In a Levesque
problem, for instance (the “Jack is looking at Anne, who is looking at
George” problem discussed earlier), high IQ is no guarantee against the
tendency to take the easy way out. No matter what their IQ, most people need
to be told that fully disjunctive reasoning will be necessary to solve the
puzzle, or else they won’t bother to use it. Maggie Toplak of York University
in Toronto, West and I have shown that high-IQ people are only slightly more
likely to spontaneously adopt disjunctive reasoning in situations that do not
explicitly demand it.
For the second source of dysrationalia, mindware deficits, we would expect to
see some correlation with intelligence, because gaps in mindware often arise
from lack of education, and education tends to be reflected in IQ scores. But
the knowledge and thinking styles relevant to dysrationalia are often not
picked up until rather late in life. It is quite possible for intelligent
people to go through school and never be taught the tools of mindware, such
as probabilistic thinking, scientific reasoning, and other strategies
measured by the XYZ virus puzzle and the four-card selection task described
earlier.
When rational thinking is correlated with intelligence, the correlation is
usually quite modest. Avoidance of cognitive miserliness has a correlation
with IQ in the range of 0.20 to 0.30 (on the scale of correlation
coefficients that runs from 0 to 1.0). Sufficient mindware has a similar
modest correlation, in the range of 0.25 to 0.35. These correlations allow
for substantial discrepancies between intelligence and rationality.
Intelligence is thus no inoculation against any of the sources of
dysrationalia I have discussed.
Cutting Intelligence Down to Size
The idea that IQ tests do not measure all the key human faculties is not new;
critics of intelligence tests have been making that point for years. Robert
J. Sternberg of Tufts University and Howard Gardner of Harvard talk about
practical intelligence, creative intelligence, interpersonal intelligence,
bodily-kinesthetic intelligence, and the like. Yet appending the word
“intelligence” to all these other mental, physical and social entities
promotes the very assumption the critics want to attack. If you inflate the
concept of intelligence, you will inflate its close associates as well. And
after 100 years of testing, it is a simple historical fact that the closest
associate of the term “intelligence” is “the IQ test part of intelligence.”
This is why my strategy for cutting intelligence down to size is different
from that of most other IQ-test critics. We are missing something by treating
intelligence as if it encompassed all cognitive abilities.
My goal in proposing the term “dysrationalia” is to separate intelligence
from rationality, a trait that IQ tests do not measure. The concept of
dysrationalia, and the empirical evidence indicating that the condition is
not rare, should help create a conceptual space in which we value abilities
at least as important as those currently measured on IQ tests—abilities to
form rational beliefs and to take rational action.
More information about the tt
mailing list