Jamie Hale

Jamie Hale

Thursday, October 28, 2010

Dysrationalia: Intelligent people behaving irrationally

by Jamie Hale

The following interview features the Stanovich, West, Toplak Research Lab.

Your research shows that intelligence does not imply rationality. Could you please briefly explain your general findings?

Those findings are easy to summarize briefly. They are simply that the correlations between measures of intelligence and various tasks from the cognitive psychology literature that measure aspects of rationality are surprisingly low. We use the term “surprisingly” here, because for many years it has been known that virtually all-cognitive ability tasks correlate with each other. Indeed many show quite high correlations. So, being psychologists, the surprise is in the context of this wide and vast cognitive ability literature, which has the technical name “Spearman’s positive manifold.” This positive manifold--that performance on cognitive tasks tends to correlate, and often quite highly--is more than 100 years old.

Thus, it was in this particular context, when we started observing fairly modest or low correlations between measures of intelligence and rational thought, that we thought this quite startling. Indeed, in restricted samples of educated adults this correlation can be virtually zero on certain tasks in the literature. Most often the correlation is positive, but, again, in light of 100 years of correlations between cognitive ability tasks, the correlations are often surprisingly low.

Of course one of the implications of this is that it will not be uncommon to find people whose intelligence and rationality are dissociated. That is, it will not be uncommon to find people with high levels of intelligence and low levels of rationality, and, to some extent, the converse. Or, another way to put it is that we should not necessarily expect the two mental characteristics to go together. The correlations are low enough--or moderate enough--that discrepancies between intelligence and rationality should not be uncommon. For one type of discrepancy, that is for people whose rationality is markedly below their intelligence, we have coined the term dysrationalia by analogy to many of the disabilities identified in the learning disability literature:


What is the definition of rationality?

Dictionary definitions of rationality tend to be of a weak sort—often seeming quite lame and unspecific. For example, a typical dictionary definition of rationality is: “the state or quality of being in accord with reason”. The meaning of rationality in modern cognitive science has a much stronger sense, it is much more specific and prescriptive than typical dictionary definitions. The weak definitions of rationality derive from a categorical notion of rationality tracing to Aristotle, who defined “man as the rational animal”. As de Sousa (2007) has pointed out, such a notion of rationality as “based on reason” has as its opposite not irrationality but arationality. Aristotle’s characterization is categorical—the behavior of entities is either based on thought or it is not. Animals are either rational or arational.

In its stronger sense, the sense employed in cognitive science and in this book by de Sousa (2007), rational thought is a normative notion. Its opposite is irrationality, and irrationality comes in degrees. Normative models of optimal judgment and decision making define perfect rationality in the noncategorical view employed in cognitive science. Rationality and irrationality come in degrees defined by the distance of the thought or behavior from the optimum defined by a normative model. This stronger sense is consistent with what recent cognitive science studies have been demonstrating about rational thought in humans.
We would also warn that some critics who wish to downplay the importance of rationality have been perpetuating a caricature of rationality that involves restricting its definition to the ability to do the syllogistic reasoning problems that are encountered in Philosophy 101. The meaning of rationality in modern cognitive science is, in contrast, much more robust and important. Syllogistic reasoning and logic problems are one small part of rational thinking.
Cognitive scientists recognize two types of rationality: instrumental and epistemic. The simplest definition of instrumental rationality, the one that is strongly grounded in the practical world, is: Behaving in the world so that you get exactly what you most want, given the resources (physical and mental) available to you. Somewhat more technically, we could characterize instrumental rationality as the optimization of the individual’s goal fulfillment.

The other aspect of rationality studied by cognitive scientists is termed epistemic rationality. This aspect of rationality concerns how well beliefs map onto the actual structure of the world. The two types of rationality are related. In order to take actions that fulfill our goals, we need to base those actions on beliefs that are properly calibrated to the world.

Although many people feel that they could do without the ability to solve textbook logic problems, virtually no person wishes to eschew epistemic rationality and instrumental rationality, when properly defined. Virtually all people want their beliefs to be in some correspondence with reality, and they also want to act to maximize the achievement of their goals. Psychologist Ken Manktelow (2004) has emphasized the practicality of both types of rationality by noting that they concern two critical things: What is true and what to do.

Epistemic rationality is about what is true and instrumental rationality is about what to do. For our beliefs to be rational they must correspond to the way the world is—they must be true. For our actions to be rational they must be the best means toward our goals—they must be the best things to do.

De Sousa, R. (2007). Why think? Evolution and the rational mind. Oxford: Oxford University Press.

Manktelow, K. I. (2004). Reasoning and rationality: The pure and the practical. In K. I. Manktelow & M. C. Chung (Eds.), Psychology of reasoning: Theoretical and historical perspectives (pp. 157-177). Hove, England: Psychology Press.

What are some of the rational thinking skills that are positively associated with intelligence? How about rational thinking skills that are not associated with intelligence?

Various probabilistic reasoning tasks have moderate correlations with intelligence. However, myside bias (the tendency to view evidence from one’s own side) is pretty much independent of intelligence in university samples. There are many, many domains of rational thinking measures and they each have important characteristics that will impact whether they are associated with intelligence. Stanovich’s Yale book contains a theoretical explanation of why some rational thinking tasks correlate with intelligence and others do not:

Stanovich, K. E. (2009). What intelligence tests miss: The psychology of rational thought. New Haven, CT: Yale University Press.

In a TV interview you (Toplak) mentioned the need for RQ testing. Do you think we can expect to see RQ testing within the public domain, in the near future?

Yes, this would be a great thing, but it is not likely to happen in the near future. The development of such an instrument would be a logistically daunting task, partly because rational thinking is such a big construct with so many parts. We use the term “multifarious” to describe this, and a metaphor we use is that it is like going to your family doctor for a check-up: there is not one test that will tell you that your health is good, rather the doctor checks multiple things to make this assessment.

The purpose of our work, and many of our recent publications, has been to speed the development of an RQ test along. We have done this by showing that there is no impediment, theoretically, to designing such a measure. The tasks that would be on such a measure have been introduced into the recent literature. In several recent publications we have been working on bringing them together into a coherent structure. Of course there are many, many, more steps that are needed before one has an actual standardized test. Standardization samples would need to be run and items would need to be piloted. In terms of the corporations that produce mental tests, it’s an endeavor that, if one were to measure it in dollars, would be millions of dollars.

Again, the purpose of some of our recent work has been to sketch out what such an endeavor would look like, to show that there is no theoretical or empirical impediment to such a thing, and to recruit others into this endeavor of working on such an instrument. We would like to include others in this endeavor, because we believe that it is way beyond the capabilities of a single laboratory. Our hope is that such an instrument might someday stand in parallel to the intelligence tests. This has been one of the motivations in our recent books and chapters, such as the following:

Stanovich, K. E., West, R. F., & Toplak, M. E. (in press). Intelligence and rationality. In R. J. Sternberg & S. B. Kaufman (Eds.), Cambridge handbook of intelligence (3rd Edition), Cambridge,UK: Cambridge University Press.

We need to emphasize, however, that there is no reason for this to be an all or nothing, rather than an incremental, process. There clearly would be immediate practical uses of less all-encompassing instruments that focused on important components of rational thinking (e.g., economic thinking, probabilistic thinking, scientific thinking, reduced myside biased thinking).

Is rationality more important than intelligence?

No, we would never make such a blanket statement. We would only say that the magnitude of its importance at least approaches that of intelligence. Differences in rational thought have real world consequences that cash out in terms of important outcomes in peoples lives. We don’t want to get into a contest of which is more important. We acknowledge that intelligence, as assessed by standardized tests, is one of the most important psychological constructs that has ever been discovered. But outlining the nature of rational thought, how to theoretically conceive it, and how to measure it empirically, is certainly up there with intelligence in terms of the most important five or six mental constructs that psychologists have investigated.

Can a person be highly rational, but rank low in intelligence?

Yes. This was addressed in our response to question number 1, that the whole point of our research showing that the correlation between the two is not excessively high is that you can have discrepancies, and that one can be high on one and low in the other.

Tell our readers how they can improve their rational thinking skills.

A good first start is education, which readers have already started here by reading this blog entry. Having an understanding of how cognitive scientists have expanded what is meant by rationality is important, namely that rationality is about two critical things: What is true and what to do.

There are numerous books that deal with rational thinking. Some of the chapters and books in our own research lab have contributed to this, and we will list them at the bottom of this entry.

Do you think a good starting point would be becoming educated on basic logic?

Basic logic would be part of a rational thinking skills curriculum, but not necessarily the first part. Again, rational thinking in cognitive science encompasses decision theory, epistemic rationality, and many areas beyond simply the study of basic logic in philosophy 101. It is very important to understand that rational thinking in cognitive science is rooted in good decision-making. Good decision making skills and good skills of knowledge acquisition do have logical thinking as one subcomponent. But there are many subskills that are even more important than logic. The subskills of scientific thinking, statistical thinking, and probabilistic reasoning, for example. Many of these are listed in the books that we will recommend here.

Baron, J. (2008). Thinking and deciding (Fourth Edition). Cambridge, MA: Cambridge University Press.

Hastie, R., & Dawes, R. M. (2001). Rational choice in an uncertain world. Thousand Oaks, CA: Sage. (a new 2010 edition is just out)

A recent chapter of ours contains a large number of citations to successful attempts to teach the skills of rational thought:

Toplak, M. E., West, R. F., & Stanovich, K. E. (2011). Education for rational thought. In M. J. Lawson & J. R. Kirby (Eds.), The quality of learning. New York: Cambridge University Press.

Is there a particular book that you recommend- for people interested in increasing their rationality- for the lay public?

Yes, some of the books that we have already mentioned. We will be so immodest as to recommend a small textbook of our own.

Stanovich, K. E. (2010). Decision making and rationality in the modern world. New York: Oxford University Press.

Weblinks with bios and further information:

Video- Stanovich Grawemeyer Lecture- Third link from the top of the page

Friday, October 22, 2010

Good Thinking: More Than Just Intelligence

by Jamie Hale

Are intelligent people good thinkers? Some are, some are not. Society is replete with examples of intelligent people doing foolish things. There is a plethora of scientific data showing intelligence does not necessarily predict rationality. Intelligence shows a low to moderate association with some critical thinking / rational thinking skills, while showing little to no association with other rational thinking skills. A study published in 2008, in Thinking & Reasoning (Stanovich & West), investigated two key critical thinking skills- avoidance of myside bias and avoidance of one side bias.

2008, 14 (2), 129 – 167

On the failure of cognitive ability to predict myside and one-sided thinking biases
Keith E. Stanovich
University of Toronto, Canada
Richard F. West
James Madison University, Harrisonburg, VA, USA

Two critical thinking skills—the tendency to avoid myside bias and to avoid one-sided thinking—were examined in three different experiments involving over 1200 participants and across two different paradigms. Robust indications of myside bias were observed in all three experiments. Participants gave higher evaluations to arguments that supported their opinions than those that refuted their prior positions. Likewise, substantial one-side bias was observed—participants were more likely to prefer a one-sided to a balanced argument. There was substantial variation in both types of bias, but we failed to find that participants of higher cognitive ability displayed less myside bias or less oneside bias. Although cognitive ability failed to associate with the magnitude of the myside bias, the strength and content of the prior opinion did predict the degree of myside bias shown. Our results indicate that cognitive ability—as defined by traditional psychometric indicators—turns out to be surprisingly independent of two of the most important critical thinking tendencies discussed in the literature.

Key cognitive skills required for critical thinking are the ability to evaluate evidence in an objective manner, and the ability to consider multiple points of view when solving a problem, or coming to a conclusion. Most people fail to demonstrate these critical thinking tendencies. Myside bias is displayed when people evaluate evidence and come to conclusions that are biased towards their own beliefs and opinions. One side bias is demonstrated when people prefer one sided arguments over arguments presenting multiple perspectives. Intelligent people are just as likely as less intelligent people to demonstrate these thinking biases. Before going further, it is important to mention that intelligence in this context refers to cognitive abilities measured by popular intelligence tests and their proxies. These tests do a good job assessing computational power and certain types of declarative knowledge. But, they do not adequately assess critical thinking skills. Avoidance of myside bias and one side bias are not measured on intelligence tests. It seems that intelligence tests are missing an important element of good thinking- evaluating evidence in a unbiased manner, and considering a multitude of perspectives when problem solving. I don't think any sane person would argue that these skills are not important.

In a series of experiments Stanovich and West examined the association between cognitive ability and two cardinal critical thinking skills- avoidance of myside bias and avoidance of one side bias. In Experiment 1 natural myside bias was investigated in 15 different propositions. In Experiment 2 myside bias and one-sided bias was studied. In Experiment 3 associations between thinking dispositions- in addition to cognitive ability- and one-sided and myside bias were investigated.

In Experiment 1, the researchers concluded, there was "no evidence at all that myside bias effects are smaller for students of higher cognitive ability" (p.140). The main purpose of Experiment 2 was to investigate the association of cognitive abilities with myside and one side bias. "The results... were quite clear cut. SAT total scores displayed a nonsignificant 7.03 correlation with the degree of myside bias and a correlation of .09 with the degree of one-side bias (onebias1), which just missed significance on a twotailed test but in any case was in the unexpected direction" (p.147). It was also revealed that stronger beliefs usually imply heavier myside bias. In Experiment 3 "the degree of myside bias was uncorrelated with SAT scores", and "[t]he degree of one-side bias was uncorrelated with SAT scores" (p.156). Myside bias was weakly correlated with thinking dispositions. One side bias showed no correlation with thinking dispositions.

The final two sentences or the research report read: "Our results thus indicate that intelligence—as defined by traditional psychometric indicators—turns out to be surprisingly independent of critical thinking tendencies. Cognitive ability measures such as the SAT thus miss entirely an important quality of good thinking" (p.161). The good news is critical thinking abilities are malleable, and in fact, probably more malleable than intelligence

Stanovich, K. & West, R. (2008). On the failure of cognitive ability to predict myside and one-sided thinking biases. Thinking & Reasoning, 14 (2), 129-167.

Additional Sources
What IQ Tests Miss- Dr. Toplak Interview

Stanovich, K. E. (2009, Nov/Dec). The thinking that IQ tests miss. Scientific American Mind, 20(6), 34-39. First Link at the top of this page -

Next week I will be publishing an interview with the Stanovich, West, Toplak Research Lab.