The Rationality of Confirmation Bias

Confirmation bias is a form of cognitive bias and while it is not considered a logical fallacy, it is considered irrational. But this is not necessarily true. Intuitively, one would imagine that rational behavior is the same as scientific behavior in that it excludes bias.

That is, if you wanted to act rationally, you should carefully weigh the arguments of both sides of an issue before you make your conclusion. And to make sure that you limit your biases over time, you would be prudent to constantly expose yourself to opposing points of view.

Unfortunately, confusion might be the unavoidable consequence of such behavior even though clarity might have been the initial goal. By trying to constantly update your world view, you will have less beliefs and more ideas. Your weakly held assumptions will change often.

While this degree of openness may be good if you want to become exposed to more ideas, it is necessary to consider the other side of the argument. And to more deeply appreciate the problem of confirmation bias, it would be helpful to understand why it is so pervasive. And I argue that it is pervasive precisely because it is rational behavior, not because it is irrational.

Consider living in a deeply religious and conservative community. Everyone you know has similar opinions about the world and your opinions, eclectic to their tastes, will not be received kindly. In which case, you are faced with two options. The first to defend your perspective while the second is to assume a more tolerant and conventional opinion. If your priority is to live in harmony within your community, then it would be harmful to think beyond the norms set by your community.

It is also worth considering that given the complexity of the world and your limited capacity to grasp it, a healthy skeptical attitude should impel you to believe that it is highly probable that whatever position you choose to hold is likely wrong. That is, given that your ignorance ā€“ the knowledge you do not have access to ā€“ is greater in magnitude than what you do know, it is likely that your position can be adequately argued against by a particular set of arguments or experiences that are unknown to you.

If you were reasonably skeptical about your subjective ability to accurately decipher reality, you should conclude that you are likely wrong about your current position. And if you were someone who was committed to the best practice of constantly gathering new data that covers all sides of the argument, then you are already with this degree of scepticism. It is in your best interest to seek social harmony since your well-being depends on your ability to function well in groups, both from an economic and emotional perspective.

Behavioral economics teaches us that a fundamental assumption of economics – that people act rationally – is false. Empirical evidence shows us that people often act against their best interest. But behavioral economists commit a different error. The problem with economics is not that it assumes that human beings are rational, it is that it assumes that it understands precisely what human rationality is. Specifically, that unbiased behavior is rational.

But biases are not necessarily irrational. There are many good reasons to be biased. Under certain conditions, being unbiased can be very costly. Consider a scenario where you are tasked to write a paper about any topic of your choosing with a fast approaching deadline. In this case, having a bias would be very useful. Since you already know which subjects you find interesting, you will save valuable time and are more likely to produce higher quality work.

If you are politically or religiously biased, you may find it easier to integrate into your community and increase your social standing within it. It might increase the number of opportunities you are exposed to professionally and even romantically. You may find that you are happier and have greater peace of mind knowing that you fit in.

The case to be made against these arguments would have to rest on the assumption that human individuality should be prized above everything else. And that the success of democratic nations rests on the ability of its individual members to think objectively about issues. But the foundation of liberal democracy is an ideal, it is not one that manifests itself in the real world.

In even the most democratic countries, groupthink is highly pervasive (group think would be the most feared consequence of cognitive biases). As pointed out in 21 Lessons for the 21st Century, sharing common fictional beliefs with your neighbour is more powerful than sharing no fictional beliefs. If you are willing to believe the same absurdity as he does, then you are likely to stand together in the face of adversity while a commitment to truth will yield no such advantage. Further, these shared and highly biased beliefs will likely lead to more a more stable political reality that will cater your shared world view.

The fact that scientists try to isolate bias is clearly beneficial to the world, but this same benefit does to apply to all aspects of the social world, and it would be irrational to assume that it does.

"A gilded No is more satisfactory than a dry yes" - Gracian