Thursday, November 24, 2011

Motivated Religious Reasoning

This is rough, but here's a first pass at something:  Let us consider some features of the human cognitive system that have emerged in our empirical investigations in recent decades. 

First, humans possess cognitive systems that are profligate with regard to religious, supernatural or spiritual beliefs.  That is, given the two systematic errors that a doxastic system might fall into, erring on the side of false positives or erring towards false negatives, humans are prone to believe more supernatural, religious, or spiritual claims than are true or well justified.  We are not, by contrast, the sorts of beings that systematically err because we are too reluctant, too skeptical, or too stingy with our assent when it comes to these sorts of beliefs.  The near universal subscription to religious beliefs now and in human history, and the abundance and variety of those beliefs show that, if nothing else, we are too eager to believe in such matters.  Too many of those beliefs are incompatible with each other, and too many of them have long since been shown to be false.  Even the ardent religious believer, such as someone who endorses the central theses of Christianity, will have have to concur that the vast majority of other religious beliefs in history which are incompatible with Christianity, have been mistaken.  If Jesus is the Son of God and the only path to salvation, then Allah cannot be the one true God.  If the resurrection of Jesus provides us with spiritual salvation, then the ancient Egyptian believers in Anubis must have been mistaken.  Therefore, given the abundance of religious beliefs, and their incompatibility with the facts (the world was not created, as the ancient Egyptians thought, by Atum out of the swirling chaotic waters of Nu), and their incompatibilities  with each other, we must conclude that the human religious error rate is very high.  And it is very high on the side of believing too much instead of believing too little. 

Second, the human cognitive system has a powerful tendency towards motivated reasoning.  What is motivated reasoning?  We can think of two models of reasoning.  First, a person might have a prefered belief already in mind and then go find some post hoc reasoning, evidence, or justification that would appear to support it.  Here, the reasoning is being steered by the goal:  defending the belief, and believing the conclusion that is best supported by an objective assessment of the evidence whatever that may be has been eclipsed by the concern for truth.  Second, a person might strive to conduct a broad, unbiased search for evidence that is open to all outcomes, and then engage in an evaluation of that evidence that is not driven towards any particular outcome.  Motivated reasoning is the first model. 

Here’s Dan Kahan’s summary: 

Motivated reasoning refers to the unconscious tendency of individuals to process information in a manner that suits some end or goal extrinsic to the formation of accurate beliefs.  “The Case for Motivated Reasoning,”  Ziva Kunda.  Psychological Bulletin 1990, Vol. 108, No. 3, 480-498.  They Saw a Game,  a classic psychology article from the  1950s, illustrates the dynamic.  Experimental subjects, students from two Ivy League colleges, were instructed to watch a  film that featured a set of controversial officiating calls made during a football game between teams from their respective schools.  Albert H. Hastorf & Hadley Cantril, They Saw a Game: A Case Study, 49 J. ABNORMAL & SOC. PSYCHOL. 129 (1954).   What best predicted the students’ agreement or disagreement with a disputed call, the researchers found, was whether it favored or disfavored their school’s team.  The researchers attributed this result to motivated reasoning: the students’ emotional stake in affirming their commitments to their respective institutions shaped what they saw on the tape.


And here, Mercier and Sperber summarize some of the research on motivated reasoning: 

A series of experiments by Ditto and his colleagues, involving reasoning in the context of a fake medical result, illustrate the notion of motivated reasoning (Ditto & Lopez 1992; Ditto et al. 1998; 2003). Participants had to put some saliva on a strip of paper and were told that, if the strip changed color or did not change color, depending on the condition, this would be an indication of an unhealthy enzyme deficiency.  Participants, being motivated to believe they were healthy, tried to garner arguments for this belief. In one version of the experiment, participants were told the rate of false positives, which varied across conditions. The use they made of this information reflects motivated reasoning. When the rate of false positives was high, participants who were motivated to reject the conclusion used it to undermine the validity of the test. This same high rate of false positives was discounted by participants who were motivated to accept the conclusion. In another version of the experiment participants were asked to mention events in their medical history that could have affected the results of the test, which gave them an opportunity to discount these results. Participants motivated to reject the conclusion listed more such events, and the number of events listed was negatively correlated with the evaluation of the test. In these experiments, the very fact that the participant’s health is being tested indicates that it cannot be taken for granted. The reliability of the test itself is being discussed. This experiment, and many others to be reviewed in this article, demonstrate also that motivated reasoning is not mere wishful thinking (a form of thinking that, if it were common, would in any case be quite deleterious to fitness and would not be coherent with the present theory). If desires did directly affect beliefs in this way, then participants would simply ignore or dismiss the test. Instead, what they do is look for evidence and arguments to show that they are healthy or at least for reasons to question the value of the test.

Other studies have demonstrated the use of motivated reasoning to support various beliefs that others might challenge. Participants dig in and occasionally alter their memories to preserve a positive view of themselves (Dunning et al. 1989; Ross et al. 1981; Sanitioso et al. 1990). They modify their causal theories to defend some favored belief (Kunda 1987). When they are told the outcome of a game on which they had made a bet, they use events in the game to explain why they should have won when they lost (Gilovich 1983). Political experts use similar strategies to explain away their failed predictions and bolster their theories (Tetlock 1998). Reviewers fall prey to motivated reasoning and look for flaws in a paper in order to justify its rejection when they don’t agree with its conclusions (Koehler 1993; Mahoney 1977). In economic settings, people use information flexibly so as to be able to justify their preferred conclusions or arrive at the decision they favor (Boiney et al. 1997; Hsee 1995; 1996a; Schweitzer & Hsee 2002).

All these experiments demonstrate that people sometimes look for reasons to justify an opinion they are eager to uphold. From an argumentative perspective, they do this not to convince themselves of the truth of their opinion but to be ready to meet the challenges of others. If they find themselves unprepared to meet such challenges, they may become reluctant to express an opinion they are unable to defend and less favorable to the opinion itself, but this is an indirect individual effect of an effort that is aimed at others. In a classical framework, where reasoning is seen as geared to  achieving epistemic benefits, the fact that it may be used to justify an opinion already held is hard to explain, especially since, as we will now show, motivated reasoning can have dire epistemic consequences. Why do humans reason?


Kahan also offers these ideas about how motivated reasoning occurs: 

The mechanisms are also diverse. They include dynamics such as biased information search, which involves seeking out (or disproportionally attending to) evidence that is congruent rather than incongruent with the motivating goal; biased assimilation, which refers to the tendency to credit and discredit evidence selectively in patterns that promote rather than frustrate the goal; and identity-protective cognition, which reflects the tendency of people to react dismissively to information when accepting it would cause them to experience dissonance or anxiety.  “What is motivatedreasoning and how does it work?”  Science and Religion Today, May 4, 2011. 
 

Ziva Kunda suggests that the mental route whereby motivated reasoning occurs can be even more subtle: 
  
I have proposed that when one wants to draw a particular conclusion, one feels obligated to construct a justification for that conclusion that would be plausible to a dispassionate observer. In doing so, one accesses only a biased subset of the relevant beliefs and rules. The notion that motivated reasoning is mediated by biased memory search and belief construction can account for all of the phenomena reviewed earlier, but the evidence for this process is mostly indirect. The most prevalent form of indirect evidence lies in the constraints that prior knowledge imposes on motivational biases, a pervasive finding obtained in several paradigms. Kunda, Ziva.  “TheCase for Motivated Reasoning,” Psychological Bulletin  1990, Vol. 108, No. 3, 480-498


So we’ve got cognitive and doxastic systems in humans that 1) have a high error rate with regard to religious beliefs, 2) are highly disposed to produce and believe religious claims, and 3) are also strongly motivated to construct reasoning towards those conclusions or beliefs that they favor through a variety of subtle and elusive biases.  These dispositions should give us pause about religious claims coming from humans.  Considering the source, we should have a high prima facie level of skepticism about religious claims coming from humans.  Most of those claims are mistaken, people readily and easily produce religious beliefs, and they will resort to a variety of reasoning gymnastics to construct reasonings that appear to substantiate them. 

The question of how best to address motivated reasoning is one that I will have to investigate over several posts in the future.  For now, we can consider one popular answer that comes up when the believer is confronted with the motivated reasoning problem.  It’s common for the advocate of a religious claim to defend the integrity of his reasoning by some appeal to his sense of how careful he has been in thinking about the question:  “I used to be an atheist, but then I became convinced that God was real by reading the Bible,” “I am a very skeptical person and I am not easily duped into believing something that isn’t justified,” and so on.  That is, I am a reliable judge of my own errors, and I am a reliable detector of the presence of motivated reasoning in my own judgments about religious beliefs.  If it feels to me that my religious beliefs are legitimately justified and as if they arise as the product of dispassionate reasoning, then they are. 

Once we make these assertions explicit, it is clear how suspect they are.  Since Descartes, the view that I am a reliable source of information about what I believe and  why I believe it has persisted, as have the views that I am know when I change my mind, I know why I changed it, and I am aware of those causal factors in my environment that influenced those changes.  This optimism about introspectionism has been thoroughly undermined by recent psychological studies.  See Nisbett and Wilson’s frequently cited survey:  Telling More Than We Know:  Verbal Reports on Mental Processes 

We cannot trust the religious believer to be a trustworthy judge of the reliability of his own cognitive processes; his subjective feelings that he is not guilty of motivated reasoning are no more reliable than his subjective feelings that God is real.  The religious urge is too powerful for us to simply take him at his word that he has been sufficiently skeptical.  See my recent lecture here for details about the anti-introspectionism research:  http://www.csus.edu/cppe/symposium/nammour_2011_beingwrong.html

Peter van Inwagen is one of the most widely respected philosophers of religion today.  In the light of what we’ve seen about motivated reasoning and the powerful drive to be religious, consider this passage where van Inwagen constructs a story about human pre-history that favors the story of Genesis by creating a sort of Genesis God of the gaps. 

“The following story is consistent with what we know of human pre-history.  Our current knowledge of human evolution, in fact, presents with no particular reason to believe that this story is false: 
For millions of years, perhaps for thousands of millions of years, God guided the course of evolution so as eventually to produce certain very clever primates, the immediate predecessors of Homo sapiens.  At some time in the last few thousand years, the whole population of our pre-human ancestors formed as a small breeding community—a few thousand or a few hundred or even a few score.  . . . In the fullness of time, God took the members of this breeding group and miraculously raised them to rationality.  That is, he gave them the gifts of language, abstract thought, and disinterested love—and, of course, the gift of free will. . . God not only raised these primates to rationality—not only made them what we call human beings—but also took them into a kind of mystical union with himself, the sort of union that Christians hope for in Heaven and call the Beautific vision.  Being in union with God, these new human beings, these primates who had become human beings at a certain piont in their lives, lived together in the harmony of perfect love and also possessed what theologians used to call preternatural powers—something like what people who believe in them today call “paranormal abilities.”  Because they lived in the harmony of perfect love, none of them did any harm to the others.  Because of their preternatural powers, they were able to somehow protect themselves from wild beasts (which they were able to tame with a look), from disease (which they were able to cure with a touch), and from random, destructive natural events (like earthquakes), which they knew about in advance and were able to escape.  There was thus no evil in their world.  And it was God’s intention that they should never become decrepit with age or die, as their primate forebears had.  But, somehow, in some way that must be mysterious to us, they were not content with this paradisal state.  They abused the gift of free will and separated themselves from their union with God.  Van Inwagen, Peter.  The Problem of Evil, Gifford lectures.  “The Global Argument Continued. 
 
We do not, as far as I know, have any substantial evidence to think of any of this story about miraculous primates being given magical powers is true.  Motivated reasoning is the best explanation that I can see for why van Inwagen or anyone else who struggles through such contortions and logical gymnastics to devise a way to make implausible Biblical stories cohere with our knowledge of human evolutionary history or cosmology.  Van Inwagen is correct; for all we know, something like this did happen in human evolution.  There are also an indefinitely long list of other mythologies that could be rendered similarly “compatible” with our current anthropological evidence with enough ingenuity and determination.  In every case, there are simpler natural explanations for why ancient people believed such stories that are much simpler and better justified than the suggestion that what Genesis describes actually happened.  Van Inwagen’s story is an illustration of just how far motivated reasoning can propel otherwise thoughtful and reasonable people out the spiral of silliness. 

Conclusion:

So, humans have a high error rate with regard to religious beliefs.  They are profligate producers of religious claims, making them highly suspect sources of reliable information about God.  Furthermore, their cognitive systems are strongly disposed to engage in motivated reasoning in favor of their prior held religious beliefs.  The mechanisms whereby motivated reasoning are subtle and difficult to detect.  The believers own assertions that he is not guilty of committing it with regard to his cherished religious views are not reliable.  We should be highly suspect of religious claims and the ostensive justifications that are offered for them unless we have substantial reason to think that motivated reasoning is not at work.  We’ve also seen that even some of the best philosophers of religion, like van Inwagen, can be swept up by the religious urge.  The results are silly and elaborate rationalizations.






  





Thursday, November 17, 2011

More on Motivated Reasoning

Mercier and Sperber give an impressive argument here:  Why Do Humans Reason?  Arguments for an argumentative theory  The standard view of reasoning is that its primary function is correct cognitive functions and find the truth.  They argue that it is better understood as facilitating persuasion in social or communication contexts.  Their thesis, they maintain, better explains the available evidence that shows how bad humans are at reasoning.


Their section on Motivated Reasoning contains this nice summary of some of the literature.  The applications to the sorts of reasoning we frequently see coming from religious believers seeking to defend the God/Jesus conclusion at all costs are striking.  For now, all I have time to do is offer a long quote:


A series of experiments by Ditto and his colleagues, involving reasoning in the context of a fake medical result, illustrate the notion of motivated reasoning (Ditto & Lopez 1992; Ditto et al. 1998; 2003). Participants had to put some saliva on a strip of paper and were told that, if the strip changed color or did not change color, depending on the condition, this would be an indication of an unhealthy enzyme deficiency.  Participants, being motivated to believe they were healthy, tried to garner arguments for this belief. In one version of the experiment, participants were told the rate of false positives, which varied across conditions. The use they made of this information reflects motivated reasoning. When the rate of false positives was high, participants who were motivated to reject the conclusion used it to undermine the validity of the test. This same high rate of false positives was discounted by participants who were motivated to accept the conclusion. In another version of the experiment participants were asked to mention events in their medical history that could have affected the results of the test, which gave them an opportunity to discount these results. Participants motivated to reject the conclusion listed more such events, and the number of events listed was negatively correlated with the evaluation of the test. In these experiments, the very fact that the participant’s health is being tested indicates that it cannot be taken for granted. The reliability of the test itself is being discussed. This experiment, and many others to be reviewed in this article, demonstrate also that motivated reasoning is not mere wishful thinking (a form of thinking that, if it were common, would in any case be quite deleterious to fitness and would not be coherent with the present theory). If desires did directly affect beliefs in this way, then participants would simply ignore or dismiss the test. Instead, what they do is look for evidence and arguments to show that they are healthy or at least for reasons to question the value of the test.

Other studies have demonstrated the use of motivated reasoning to support various beliefs that others might challenge. Participants dig in and occasionally alter their memories to preserve a positive view of themselves (Dunning et al. 1989; Ross et al. 1981; Sanitioso et al. 1990). They modify their causal theories to defend some favored belief (Kunda 1987). When they are told the outcome of a game on which they had made a bet, they use events in the game to explain why they should have won when they lost (Gilovich 1983). Political experts use similar strategies to explain away their failed predictions and bolster their theories (Tetlock 1998). Reviewers fall prey to motivated reasoning and look for flaws in a paper in order to justify its rejection when they don’t agree with its conclusions (Koehler 1993; Mahoney 1977). In economic settings, people use information flexibly so as to be able to justify their preferred conclusions or arrive at the decision they favor (Boiney et al. 1997; Hsee 1995; 1996a; Schweitzer & Hsee 2002).

All these experiments demonstrate that people sometimes look for reasons to justify an opinion they are eager to uphold. From an argumentative perspective, they do this not to convince themselves of the truth of their opinion but to be ready to meet the challenges of others. If they find themselves unprepared to meet such challenges, they may become reluctant to express an opinion they are unable to defend and less favorable to the opinion itself, but this is an indirect individual effect of an effort that is aimed at others. In a classical framework, where reasoning is seen as geared to  achieving epistemic benefits, the fact that it may be used to justify an opinion already held is hard to explain, especially since, as we will now show, motivated reasoning can have dire epistemic consequences.