follow CCP

Recent blog entries
« The ideological symmetry of motivated reasoning, round 15 | Main | Do people with higher levels of "science aptitude" see more risk -- or less -- in climate change? »

Whoa, slow down: public conflict over climate change is more complicated than "thinking fast, slow"

With the (deserved) popularity of Kahneman's accessible and fun synthesis "Thinking Fast and Slow" has come a (predictable) proliferation of popular commentaries attributing public dissensus over climate change to Kahneman's particular conceptualization of dual process reasoning.

Scientists, the argument goes, determine risk using the tools and habits of mind associated with "slow," System 2 thinking, which puts a premium on conscious reflection.

Lacking the time and technical acumen to make sense of complicated technical information, ordindary citizens (it's said) use visceral, affect-driven associations--system 1. Well, climate change provokes images -- melting ice, swimming polar bears -- that just aren't as compelling, as scary as, say, terrorism (fiery skyscrapers with the ends of planes sticking out of them, etc.). Accordingly, they underestimate the risks of climate change relative to a host of more gripping threats to health and safety that scientific assessment reveals to be smaller in magnitude. 

This is not a new argument. Scholars on risk perception have been advancing it for years (and reiterating/amplifying it as time passes).

The problem is that it is wrong.  Empirically demonstrably false.


  • Variance in the disposition to use "fast" (heuristic, affect-driven, system 1) as opposed to "slow" (conscious, reflective, deliberate system 2) modes of reasoning explains essentially none of the variance in public perception of climate change risks. In fact, when one correlates climate change risk perceptions with these dispositions, one finds that the tendency to rely on system 2 (slow) rather than 1 (fast) is associated with less concern, but the impact is so small as to be practically irrelevant. 

  • What does explain variance in climate change risk perception -- evidence shows, and has for years -- are cultural or ideological dispositions. There is a huge gulf between citizens subscribing to a hierarchical and individualistic worldview, who attach high symbolic and material value to commerce and industry and who discount all manner of environmental and technological risk, and citizens subscribing to an egalitarian and communitarian worldview, who associate commerce and industry with unjust social disparities.


  • Because climate change divides members of the public on cultural grounds, it must be the case that ordinary individuals who use system 1 ("fast") modes of reasoning form opposing intuitive or affective reactions to climate change -- "scary" for egalitarians and communitarians, "enh" for hierarchical individualists. Again, evidence bears this out! (Ellen Peters, a psychologist who studies the contribution that affect, numeracy, and cultural worldviews make to risk perception has done the best study on how cultural worldviews orient system 1/affective perceptions of risk, in my view.)

  • Individuals who are disposed to use system 2 ("slow") are not more likely to hold beliefs in line with the scientific consensus on climate change. Instead, they are even more culturally polarized than individuals who are more disposed to use "fast," system 1 reasoning. This is a reflection of the (long-established but recently forgotten) impact of motivated reasoning on system 2 forms of reasoning (i.e., conscious, deliberate, reflective forms). 


So why do so many commentators keep attributing the climate change controversy to system 1/2 or "fast/slow"?

The answer is  system 1/2 or "fast/slow": that framework recommends itself -- is intuitively and emotionally appealing (especially to people frustrated over the failure of scientific consensus to make greater inroads in generating public consensus) and ultimately a lot easier to get than the empirically supported findings.

This is in fact part of the explanation for the "story telling" abuse of decision science mechanisms that I discussed in an earlier post.

There's only one remedy for that: genuinely scientific thinking.

Just as we are destined not to solve the problems associated with climate change without availing ourselves of the best available science on how the climate works, so we are destined to continue floundering in addressing the pathologies that generate public dissensus over climate change and a host of other issues unless we attend in a systematic, reflective, deliberate way to the science of science communication.

PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments (3)

You seem to be measuring the tendency to use system-2 with numeracy. I do not think that is even intended as a measure of system-2. The CRT is, but it is only 3 items. Perhaps the closest alternatives that are longer are the measures of actively open-minded thinking (Stanovich, also me). It would be interesting to look at these.

Of course, I completely agree that consistency with other beliefs (which you call culture) is extremely relevant.

February 10, 2012 | Unregistered CommenterJon Baron

Hi, Jon.

Thanks again for comments.

I am treating numeracy as a valid indicator for the disposition to use system 2 reasoning— yes. Some elaboration of why I am doing that is certainly in order.

As you know (other readers might not; I address them too), the issue of how to measure system 1/2 dispositions — or the closely related earlier constructs that figured in dual processing reasoning theory (such as “heuristic” and “systematic” reasoning)—is not well worked out. There has been a tendency to assume that self-reporting measures (like “need for cognition”) are valid—but many researchers seem to question the assumption.

Presumably the best way to validate a measure of a disposition to use System 1 vs. System 2 is to show that that measure predicts the likelihood that individuals will display one or another of the cognitive biases that Kahneman and other decision theorists view as the hallmark of heuristic reasoning gone bad.

I very much agree with you that CRT is a valid measure precisely because it does well under that criterion — better , in fact, than “need for cognition” and various other measures traditionally used to test whether individuals are using or are disposed to use “effortful” or “systematic” or otherwise “higher level” cognition. See Toplak, M., West, R. & Stanovich, K. The Cognitive Reflection Test as a predictor of performance on heuristics-and-biases tasks. Memory & Cognition 39, 1275-1289 (2011).

You say the numeracy scale was not “intended” to be a test of system 2 or systematic reasoning. Perhaps that’s right; as you know, some leading scholars in this area criticize the numeracy scale on the ground that the development of it wasn’t sufficiently informed by any particular theory.

Nevertheless, numeracy is definitely treated in the psychological literature as a proxy of system 2 or systematic reasoning . See, e.g., Peters, E., et al. Numeracy and Decision Making. Psychol. Sci. 17, 407-413 (2006).

The reason is that numeracy, like CRT, predicts how vulnerable people are to various of the cognitive biases that psychologists treat as the hallmarks of system 1 or biased heuristic reasoning. See Reyna, V.F., Nelson, W.L., Han, P.K. & Dieckmann, N.F. How numeracy influences risk comprehension and medical decision making. Pscyh. Bulletin 135, 943-973 (2009).

Indeed, numeracy seems to do better than CRT with respect to some of those biases (e.g., “conjunction fallacy” & “ratio bias”). See Liberali, J.M., Reyna, V.F., Furlan, S., Stein, L.M. & Pardo, S.T. Individual Differences in Numeracy and Cognitive Reflection, with Implications for Biases and Fallacies in Probability Judgment. Journal of Behavioral Decision Making (2011).

But I don’t think one could say that numeracy does better than CRT because numeracy is a more “valid” indicator of a System 2- or a systematic-reasoning disposition; more likely it does better simply because it is has many more items and thus enables more reliable measurements of whatever disposition they are both indicators of.

The numeracy scale used in the study that forms the basis of this post in fact combined CRT and the non-CRT numeracy items, resulting in an even more discerning or reliable scale.

So I think researchers are on firm ground — as firm as any ground that anyone is on in trying to measure the disposition to use higher-level cognitive processing — with numeracy. And I think in particular that we (my coauthors in this study, who included Ellen Peters, & Paul Slovic, and I) were on even more solid ground in treating our numeracy/CRT composite as an indicator of that disposition.

Of course, I agree that it would be very interesting to include AOT measures in studies that look at whether higher-quality cognitive processing mitigate or aggravate cultural polarization! I am aware, too, that CRT and AOT are correlated and both seem to indicate a disposition consciously to reflect on alternatives before committing to a particular inference (Campitelli, G. & Labollita, M. Correlations of cognitive reflection with judgments and choices. Judgment and Decision Making 5, 182-191 (2010)).

Thanks again, again.


p.s. On what to call the value predispositions — I think here it is just a question of “what to call” something that we know is measured by a variety of related constructs, including the Cultural Theory of Risk worldviews (which we use) and “liberal-conservative” ideology, partisan self-identification, etc. For posts on why we use CTR worldviews, see here, here, & here.

February 10, 2012 | Registered CommenterDan Kahan

Well, you picked out some of the flaws in that proposition, but there's also a big difference in the kinds of questions that abstract and intuitive thinking respond to. I have to study the different ways people recognize organization in the world around them, in my work as a natural systems scientist.

The most curious difference I've run smack into a number of times is that abstract thinkers only refer to the objects of nature as contained in their abstractions.. Intuitive thinkers more often think of objects of nature as self-defining realities, and can separate their abstract thinking from the thing in the world being discussed. I find scientists quite often are just befuddled by being asked to do that.

My blog ( discusses a bunch of other interesting conundrums like that too, picked up along the way in my studies of how nature manages to work by itself (...while I'm not watching to project it from my consciousness! :-)

February 13, 2012 | Unregistered CommenterJessie Henshaw

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>