Read on Twitter
Here's something you should know about yourself. When given options, you tend to pick the one that is easiest to justify, not the one that is "best." Seems bad, but this is good when we all do it within groups, but terrible when done in isolation. Allow me to explain. 1/25
Justification is the primary output of your reasoning system, a collection of "mental modules" that come online when you feel motivated to explain yourself to yourself or others – not to be confused with REASON, the philosophical concept of human intellect and rationality. 2/25
Two experiments demonstrate this system well. In one, as a parting gift, scientists gave subjects the choice between two treats. One was an attractive, small, but cheap, chocolate heart. The other was a giant, expensive, realistic, chocolate cockroach. 3/25
Without asking for their reasoning, most people said they would rather have the heart. But when asked to justify their decision, most chose the roach, even though they knew they would hate it. 4/25
Unsure which was the better option, they employed their reasoning. They began creating justifications for each conclusion – “I should get the heart,” or “I should get the roach.” 5/25
The list of reasons they could bring to mind for the roach that seemed justifiable to others – bigger, more expensive – simply outweighed those for the heart. They picked something they knew they would not enjoy because they couldn’t justify doing otherwise. 6/25
In another run of they experiment subjects could perform one two tasks to receive either 50 points or 90, and were told they could spend those points one of two albums, a Beatles Greatest Hits for 50 points or a Barbara Streisand Greatest Hits for 90.
When asked which they'd prefer, most said the Beatles album. But when actually given the chance to earn the reward, the same people did the harder task so they could get the Streisand album. More expensive = easier to justify, even though they wanted something else. 8/25
It gets weirder. Tversky and Shafir asked subjects to imagine they had flipped a coin. For some it came up in their favor, and they won $200. for others it didn't, and they lost $100. They asked, "You are now offered a second identical gamble" do you take it? 9/25
When people won, they tended to think, “Yeah, I’d like to play again. I'm ahead, so I can risk it." If they didn’t, they tended to think, “Yeah, I would like to play again. I need to win back what I just lost.” Win or loss, they found a reason to take the second gamble. 10/25
But here is the crazy part: even though most would take the bet either way, no matter the outcome, if they DID NOT tell people the outcome, most people DID NOT flip again. Why? Because without that information they couldn't find a reason to justify their decisions. 11/25
We do this because, psychologically speaking, reasoning ain't logic. Reasoning is coming up with justifications – plausible explanations for what you think, feel, and believe, and plausible means that which you intuit your trusted peers will accept as reason-able. 12/25
In other words, the function of reasoning is to argue your case in a group setting. In psychology, this is called "the interactionist approach," a model that posits reasoning evolved to facilitate communication between peers in an environment where misinformation is unavoidable.
Imagine three people on a hill, each one looking in a different direction. The ability to pool those perspectives into a shared worldview is an incredible evolutionary advantage. Scale it up, add abstractions, and you've got yourself the vital intellectual gumbo we call culture.
The only problem is others can be wrong, or may mislead. Thus we must be, as psychologists say, "epistemically vigilant" to get the most out of the evolved mechanisms that allow for interpersonal communication, learning, and trading in innovations both physical and ideological.
This is why reasoning and arguing evolved as psychological mechanisms. Natural selection favored groups that produced reasons which they could then use to deliberate and argue as they worked together toward shared goals within a consensus reality. 16/25
Research shows people are incredibly good at picking apart other people's justifications to identify poor arguments and flimsy reasoning, but terrible at picking apart their own in the same way. 17/25
In research by @hugoreasoning, subjects read word problems, selected answers, and then wrote out arguments defending their choices. Subjects then reviewed the problems a second time along with the reasoning of subjects who chose different answers. 18/25
What the experimenters didn’t reveal was that they had actually hidden in those answers some switcheroos. For one of the problems answered incorrectly, the supposed justification from another person was actually the subject’s own justification. 19/25
They found that when people knew the justifications were their own, they defended them, but when they thought it was someone else's, 69 percent rejected their own justifications and then switched to the correct answer. 20/25
According to the interactionist model, this is the whole point of reasoning. In a biased and lazy manner (fast), we add plausible justifications for our inferences and conclusions to the marketplace of ideas then offload to others the cognitive labor of picking them apart (slow).
In many experiments where individuals predictably and routinely reach faulty conclusions, just allowing people to compare their answers with others and hear out their justifications corrects those mistakes, and over time it brings the majority into agreement on "the truth." 22/25
It would seem like the internet would be great for this. In some places it is, but in many more, biased and lazy, we are reasoning alone AS IF we are reasoning together, trapped on justification hamster wheels which often lead to polarization and mass, shared, delusions. 23/25
As @acroll said, "On the Internet, when you say I want grilled cheese you're not actually presenting an argument for grilled cheese. You're saying, ‘Help me find the Grilled Cheese Room, because that's what I want, and I'm going to go find like-minded grilled cheese enthusiasts.’
“You're helping the machines search for the group you want to be a part of as opposed to using the machine to have a conversation with the group you're with.”
