It depends on how likely your original opinion is true, how likely you are to change a wrong opinion to a correct opinion (using this exercise), and how likely you are to change correct opinion to wrong opinion (using this exercise).
If your opinions are good on average, and if the process of argumenting for a belief has approximately the same chance to move you in any direction, then it is a harmful experiment. If the process is more likely to move you in a correct direction than in a wrong direction, and your original opinions are not too great, then you would benefit from this process on average.
Unfortunately, I don’t know any of these numbers. But it should be possible to make an experiment where the correct answer is known to the researchers, but half of the population believes the incorrect answer anyway—whether the experiment would be more likely to move them in the correct direction. This would have to be repeated for differend kinds of topics, to make sure the effect is not specific for one of them.
It depends on how likely your original opinion is true, how likely you are to change a wrong opinion to a correct opinion (using this exercise), and how likely you are to change correct opinion to wrong opinion (using this exercise).
If your opinions are good on average, and if the process of argumenting for a belief has approximately the same chance to move you in any direction, then it is a harmful experiment. If the process is more likely to move you in a correct direction than in a wrong direction, and your original opinions are not too great, then you would benefit from this process on average.
Unfortunately, I don’t know any of these numbers. But it should be possible to make an experiment where the correct answer is known to the researchers, but half of the population believes the incorrect answer anyway—whether the experiment would be more likely to move them in the correct direction. This would have to be repeated for differend kinds of topics, to make sure the effect is not specific for one of them.