Thursday, November 16, 2017

Partisan Gullibility: Two System Thinking and the Affect Heuristic

Jimmy Kimmel recently had some fun duping Trump supporters, first by promoting the Healthcare.gov site as "Trumpcare" and then by asking Trump supporters if they think Hillary Clinton should be impeached. In the first case, he was simply calling Obamacare by a different name, and in the second he was asking about impeaching someone who doesn't hold public office. When his targets fell for, much hilarity ensued.

But, in a way, this is somewhat unfair as he was taking advantage of a psychological blindspot that everyone has when thinking about things with which we are emotionally involved. I addressed the Two system model of thinking and the Affect Heuristics in the excerpt from a conference paper posted below. This excerpt explains the Two System model at some length (which should be generally useful for my students) before discussing the affect heuristic and how it makes people vulnerable to misinformation:

Kahneman introduces the two systems of thinking with a simple demonstration which is worth repeating here.  First, he asks you to look at the picture shown in figure 2 below
Figure 2:

(Kahneman, 2011, 19)

He notes that a single glance at the photo was probably enough for you to notice that the person in the photo was a woman with dark hair who appeared to be angry and was probably about to say something unkind. He goes on to point out that you probably deduced that without even intended to do so. Thus, you employed an automatic thought process that constitutes System 1, which he informally calls fast thinking.  In general, he notes that “System 1 operates automatically and quickly, with little or no effort and no sense of voluntary control. (Kahneman, 2011, 20)

He then asks you to look at a multiplication problem (17 x 24).  Here, he surmises that at first glance you were probably able to identify it as a multiplication problem that it involved 17 and 24. However, he suggests that most readers probably could not provide the correct answer to the problem (568) without spending some time computing the answer. Indeed, doing so would have involved some mental strain as most people would have to recall the steps of the process learned in school in a “deliberate, effortful and orderly” process that he notes is the prototype of System 2 or slow thinking. (Kahneman, 19-20)  He summarizes System 2 as follows:
System 2 allocates attention to the effort full metal activities that demand it, including complex computations.  The operations of System 2 are often associated with a subjective experience of agency, choice, and concentration. (Kahneman, 2011, 20-21)
While System 2 may be commonly associated with agency and choice, Kahneman points out that people use System 1 to do most of the things they do in life, such as carrying on a conversation, driving a car, or simple arithmetic (e.g., 2+2 =  ?).  Most of things that people do repeatedly in life, they do without ‘stopping to think’ about doing so. System 1 is what allows them to do so. In contrast, ‘stopping to think’ requires an exercise in self-control to stop other activities and devote cognitive effort to thinking, both of which are requirements of using System 2.
The seeming effortlessness of System 1 belies its complexity as can be seen in Kahneman’s description of its main function:
“The main function of System 1 is to maintain and update a model of your personal world, which represents what is normal in it.  The model is constructed by associations that link ideas of circumstances, events, actions, and outcomes that co-occur with some regularity, either at the same time are within a relatively short interval.  As these links are formed and strengthened, the pattern of associated ideas comes to represent the structure of events in your life, and it determines your interpretation of the present as well as your expectations of the future.” (Kahneman, 2011, 71)
So, System 1 is doing a lot more than just memorizing that 2+2 = 4. Rather, it is modeling and monitoring the world as one knows it, as well as interpreting and predicting events. Most importantly, it is doing so continuously and automatically with little effort.  Thus, it is doing the kinds of things specified by Burke’s model of identity verification and Gidden’s conception of practical consciousness.

As remarkable as System 1 is, Kahneman points out that it has many quirks. First among these is an ability, that tends towards bias, for finding causal connections. To demonstrate this, Kahneman offers the following brief statement:
 “Fred’s parents arrived late. The caterers were expected soon. Fred was angry.” (Kahneman, 2011, 74)

He notes that most people will automatically understand that Fred was angry because his parents arrived late, not because the caterers were expected soon. In so doing, System 1 is identifying a causal connection between Fred’s anger and one thing (the parents’ late arrival) and not another (the imminent arrival of caterers). While this is very useful for understanding the passage, Kahneman notes that System 1 tends to jump to conclusions. It is prone to impressions of causality and therefore offers a causal link between Fred’s anger and one of the other statements in the passage. It does this based on the information provided without questioning the information’s completeness or quality. As Kahneman puts it:
 “System 1 excels at constructing the best possible story that incorporates ideas currently activated, but it does not (cannot) allow for information that it does not have…  The measure of success for System 1 is the coherence of the story it manages to create.  The amount and quality of the data on which the story is based are largely irrelevant.  When information is scarce, which is a common occurrence, System 1 operates as a machine for jumping to conclusions.”  (Kahneman, 2011, 85)

Of course, System 2 is available to question the conclusions to which System 1 jumps, but there are several obstacles to System 2 doing so. First, using System 2 requires concentration and effort. System 2 does one thing at a time, which means it must block out other activities to work and cannot work if distracted. Maintaining the self-control also consumes mental energy and can led to ego depletion which impairs the functioning of all mental activity. As a result, Kahneman describes System 2 as “sometimes busy and often lazy.” (Kahneman, 2011, 81 )

Kahneman notes that both System 1 and System 2 have a confirmation bias resulting from a tendency to employ a positive test strategy that searches for confirming evidence rather than disconfirming evidence. For instance, if asked whether an acquaintance is friendly, people will tend to search their minds for instances of that person being friendly instead of instances in which the person was unfriendly. While people can learn to employ other strategies, and those individuals that do so repeatedly on a day-to-day basis may even train System 1 to do so (just as a few people might be able to automatically give the answer to 17 x 24 or a chess master can unhesitating play against several lessor opponents), Kahneman argues that most people do not have an intuitive grasp of the statistical principles used to test empirical hypotheses.(Kahneman, 2011).

More importantly, for the purposes of this paper, System 2 is susceptible to an affect heuristic through which their emotional likes and dislikes affect their beliefs about the world. While there are many instances when System 2 is able to resist accepting the conclusions to which System 1 jumps and apply logical analysis, System 2 is less able to do so in situations in which the affective attitudes and beliefs are involved. In such cases, Kahneman describes System 2’s functioning as follows:
“…System 2 is more an apologist for the emotions of System 1 than a critic of those emotions – an endorser rather than an enforcer. Its search for information and arguments is mostly constrained to information that is consistent with existing beliefs, not with an intention to examine them. An active coherence-seeking System 1 suggests solutions to an undemanding System 2.” (Kahneman, 2011, 103 - 104)

To see this in action, consider the photo in figure 2, which made the rounds in forwarded emails circa 2002 (in the days before Twitter and Facebook. The photo seems to show then-Senate Majority Leader Tom Daschle using the wrong hand to salute during the pledge of allegiance. The accompanying text usually interpreted the photo as evidence of how clueless and/or unpatriotic Daschle and Democrats were.
Figure 2

(Source: https://www.truthorfiction.com/daschle/)

Of course, it is patently absurd that a Senator would not know which hand to use in the pledge, or for that matter where his own heart was, but it takes close and careful examination of the photo to determine that it is photo-shopped. Specifically, one needs to notice that Daschle’s suit jacket is buttoned the wrong way and that hand he is using does not have a wedding ring or wristwatch on it as it would if he was saluting with his left hand. To do that kind of methodical analysis requires using System 2 and, therefore, requires an individual to devote some concentrated cognitive effort to the task. Most importantly, this requires someone to choose to devote that effort because System 2 does not function automatically.

The implication of the affect heuristic is that people who have a positive emotional response to the image, or more precisely the conclusion to which System 1 jumps when interpreting the image, will be unlikely to engage System 2 to question it. Indeed, if system 2 is engaged in such a case, it is likely to search for confirming rather than disconfirming evidence for System 1’s conclusion. Therefore, someone who is emotionally committed to being a partisan Republican, or to the non-partisan view that all politicians are idiots, would have a very hard time not believing that this photo is authentic.  Conversely, someone who is a partisan Democrat would be motivated to use System 2 to reject the implied message of the photo (not to mention the explicit message that accompanied it in the forwarded emails).  It is important to note that this appears to be the ‘correct’ response only because we know that the photo is a fake.

PS- It is vital to keep in mind that the affect heuristic is not unique to republicans but rather is something to which we are all vulnerable.


Monday, November 13, 2017

Dusting Off the Old Blog

It seems like most blogs have a post in them somewhere in which the blogger makes groveling apologies for neglecting it. This is one of those.

I pushed back from the blogging table a couple years ago as I became concerned that I was overwhelming students with too many posts (which were being assigned as course readings).

Then along came Trump and all the oxygen got sucked out of the daily news cycle. I didn't want to be constantly commenting on or fact-checking Trump, but I was nevertheless caught up in exactly that.

But now I am pushing past that and trying to get back on track. As part of that effort, I have established a Facebook page at https://www.facebook.com/ThinkingAboutIR  This will constitute my frontpage, if you will, and I will be posting links to and comments on news items there.

This blog will continue to function as place for me to, as the name implies, think out loud about various issues in IR. In particular, I am going to be discussing the link between identity driven behavior and populism. Of course, other topics are likely to come up as well. 

Wednesday, March 22, 2017

Innovation and Creative Destruction

Mark Perry recently has a useful post on the economics lessons from the demise of Gander Mountain. This post is particularly useful for explaining the ideas of creative destruction and consumer sovereignty.

How the mighty are fallen: Back in 2015, Mark Perry noted that only 12% of the 1955 Fortune 500 companies were still on the list in 2015.  He gives American Motors, Brown Shoe, Studebaker, Collins Radio, Detroit Steel, Zenith Electronics, and National Sugar Refining as examples of corporations that were on the list in 1955 but in 2015. Newcomers to the list since 1955 include Facebook, eBay, Home Depot, Microsoft, Google, Netflix, Office Depot and Target.

Innovation and Growth: In another AEI blog, James Pethokoukis reports on research  by Leonid Kogan, Demitris Papanikolaou, Amit Seru, and Noah Stoffman that measures innovation in terms of patents issued to companies from 1926 to 2010 and the stock market response to those patents. The combination gives the a way to measure the economic significance of  the innovation contained in the patent.


Creative Destruction: The face of innovation only an economist could love: All the above is fine and good, but it should be obvious that creative destruction is destruction nonetheless and the creativity of it is probably lost on the people having their livelihoods destroyed.