All theories within Psychology and Economics are based on people being 'rational'. Any thing contrary to their theory is branded 'irrational' and given a name. The name usually sounds like a 'disease/ailment'.
All theories within Psychology and Economics are based on people being 'rational'. Any thing contrary to their theory is branded 'irrational' and given a name. The name usually sounds like a 'disease/ailment'.
It works for perception research: when science needs to explain an illusion and why people are subjects for it, to be able to explain why people see or hear "wrong" things, or why this "wrong" things are not wrong at all, is a good science. Often times science use a very artificial setup for an experiment, the very setup is tuned so that people start to make mistakes. Take Ames Room as an example[1], it is artificially created environment when participant have not enough information to be sure, and his mind make a mistake. But mistake of the mind is a great achievement, if you try to do better with an AI, I suspect you'll end with the same result. Mind take into a consideration a lot of details, for Ames Room to work reliably, experimenter needs to draw skewed windows on the back wall, that would look as rectangles after projection. So the setup is highly unlikely a priori and mind makes a good bayesian decision that the most probable explanation is two people of different sizes.
For cognitive biases we also need to be wary, because the process of creating the right experimental setup could include a lot of tweaks to make people's decision process "to fail". Scientist needs an effect that could be shown with a statistics, so he/she tweaks setup until it works.
It leads to a conclusion, that if cognitive bias lacks explanation why it is a rational thing, we cannot say that this bias is a "disease/ailment".
Take Luce's coffee cup example as an illustration. You prefer black coffee to sweet coffee. Suppose you compare coffee with no sugar to coffee with one grain of sugar added. You're indifferent: a~b. Then add another grain, and so on. You will get comparisons a~b, b~c, c~d, e~f, ..., j~k, and then suddenly a>k, a violation of the supposed transitivity of equipreference (aka indifference, equally good). But that seems to be rational.
So people relaxed rationality requirements and now there is the problem what 'rationality' actually means.
Fast forward a few years and empirical studies found the following strange behavior: If you mention a high number before asking people for some fictional charity contribution, then people tend to be be willing to pay more than if you mention a low number before, and it does not matter in which way you mention the numbers. (Actual experiments were made by making people roll a rigged lottery wheel before doing some completely different task, for example.) You can even tell participants about the observed effect before, it will still be observed.
I see no way how this "anchoring effect" could be described as being rational.
But many people nowadays share your opinion, and there is a whole field called "ecological rationality" in which scholars try to re-interpret supposedly irrational biases as good and rational heuristics increasing e.g. evolutionary fitness. I don't think they're right in general, though. Some of the biases are just flaws. If I flash a number before your eyes and this affects your subsequent decision making, then that's not a useful heuristics, it's a flaw in your brain processing. My 2 cents, others disagree with me.
It's probably a trait that is (or was) advantageous in one context, that is disadvantageous in this new or less vital context.