Abstract. A solution for improving institutional decision-making, so that motivated reasoning does not influence the decision-making process, is presented. The solution involves the use of a committee, that will serve as the vehicle for determining whether a decision made by an institution was influenced by motivated reasoning.
Motivated reasoning is a form of bias in which individuals arrive at conclusions based on their own subjective impressions. This is a major problem facing institutions like governments and corporations that carry out decision-making duties. Decisions influenced by motivated reasoning could lead to major errors in judgement, which then potentially results in negative consequences for members of society. The lives of many people from society being affected by the miscalculated assessment of a few, can seem very unjust.
A proposed solution to possibly countering motivated reasoning is a checks and balances type of system; a system in which a committee of non-partisan individuals could proctor whether a decision made by an institution was influenced by motivated reasoning. The members of this committee obviously need to be well versed in the issues(s) pertaining to the institution’s decision, and they should not be beholden to any special interest groups. The non-partisan nature of the committee would ensure that there is no bias in proctoring the institution’s decision.
In the trivial case that motivated reasoning has not played a role in the institution’s decision-making, the proctoring committee could simply approve the decision for going forward. If motivated reasoning played a role in the decision-making process on the other hand, then the members of the committee have the authority to override the decision made by the institution. The institution would have to then make modifications to the decision that they made. After making these modifications, the proctoring committee will have to again review the decision to make sure that there is no bias.
The non-partisan committee could use a ledger for keeping a record of all the decisions that they have had to proctor for the institution. For each decision made by the institution, they (the institution) can be allotted a fixed number of chances to revise that decision based on whether motivated reasoning contributed to that decision. If the institution made a decision (on one issue) involving no motivated reasoning in its first go, then they could be rewarded with an extra revision chance for the next issue they make a decision on and present to the committee. Failure to avoid motivated reasoning on a given decision will result in no extra revision chance for the next issue they make a decision on. In the case that the institution fails to avoid motivated reasoning on all of its allotted chances, then the proctoring committee would have to make the final decision on the issue.
I like this general idea and in theory it sounds great. In implementation, how would people who do not practice motivated reasoning be found and identified?
Is there a more objective way of determining whether motivated reasoning is behind a decision, to counteract the possibility that the "nonpartisan" people may not actually be as unaffected by motivated reasoning as we would hope?
Would requiring argument mapping for major decisions be a viable method to aid in this process?
Thanks for your feedback eaphilosophy. You have brought up some good questions, and I am not entirely sure how to go about fixing these issues, but could the following be possible solutions to these issues:
In regards to how it can be implemented, is there perhaps some sort of screening process/background check that can be carried out to identify individuals who do not practice motivated reasoning?
Also, perhaps the ledger system I mentioned could be a blockchain system?
Did you mean to post something after "possible solutions to these issues: " or is the paragraph that begins with "in regards to" your list of possible solutions? I assume the former because I also don't see any mention of a ledger system.
With regards to a screening process, maybe something like the Harvard Implicit Bias test but for motivated reasoning? Asking people to quickly make decisions in situations susceptible to motivated reasoning? This test has been criticized but also has some research in support so it may be the best option.
This may also be a good question for the Less Wrong forum. If you post there, let me know and I'll follow it because I am genuinely curious about this.
My paragraph beginning with "in regards to" is a possible solution to your first question "how would people who do not practice motivated reasoning be found and identified?"
My next statement "perhaps the ledger system I mentioned could be a blockchain system?" is a possible solution to your next question "Is there a more objective way of determining whether motivated reasoning is behind a decision, to counteract the possibility that the "nonpartisan" people may not actually be as unaffected by motivated reasoning as we would hope?" ( I was referring to the ledger system mentioned in the final paragraph of my Checks and Balances writeup. Correct me if I'm wrong, but are blockchains supposed to be unbiased? (https://dailyfintech.com/2018/06/04/blockchain-gives-hope-for-unbiased-news/) ).
You mentioned that I should try the Less Wrong Forum, should I post my whole Checks and Balances writeup there as well, or only ask them about the Harvard Implicit Bias test for motivated reasoning?
Ah, thank you for your clarifications. Completely missed that you referenced the ledger in your original post, makes more sense now. Yes, I believe blockchain are unbiased, but frankly I don't know enough about them to offer any input on that end.
For Less Wrong, I was referring to your Checks and Balances write-up, but the Implicit Bias Test would be good also as they do seem to like objective evidence. Take my advice with a grain of salt though, as I am a relatively new user to that site. However, you have little to lose and a fair amount to gain.
Thanks again eaphilosophy for your reply. I will post my write-up to the Less Wrong Forum. I just have one more question for now, are there any organizations out there who are looking for ideas such as this?
None that I know of, but on the 80000hours.org effective altruism website they mention that one of the top 5 problems to solve is institutional decision making, and mention that they can put you in touch with people who are in those areas, but I'm not sure if that's only for people in their mentorship program or any of their readers.
Please post (or pm) the link when you do!
Besides governments and corporations, what other decision-making institutions could possibly benefit from this approach?