Beige & in Florence

The Danger of Expert Mindset

In the midst of this clash of opinions, a great war of words and tumult of opinions arose. Some were crying, ‘Lo, here!’ and others, ‘Lo, there!’ Some were contending for the Methodist faith, some for the Presbyterian, and some for the Baptist.

I recently did a silly survey on rating abnormal conversation starters: here. The only question that came out of the ANOVA (n = 12) in a positive "significant" way was "Do dreams have an evolutionary advantage?" It's a fantastic mystery that apparently has popular interest (i.e. 12 of Jadens friends). A huge upside of using this question to get to know people is no expert has monopolized a solution to it.

Reddit's confident

I love this reddit thread discussing my same evolutionary dream question

While some hedging was given as token gestures to the uncertainty of the solution, most everyone seemed very confident in their own pet theory. Here are some examples of their opening statements:

Evolution will not explain why it is that animals dream.

Dreams are probably a way to save the memory of experiences the animal had during the day

It's competition for brain resources

But my favorite is this intro

So, do bear in mind that I'm not an expert. An expert will be able to give a more definitive answer than I can...

The genuine experts seem confused on this question, not more definitive than redditors!

A cherry-picked personal anecdote

I was chatting with a biology grad student and I asked her if there was an evolutionary advantage to dreams. She said that dreams are experienced most vividly during the beginning of sleep during REM sleep. Additionally some people have more vivid dreams compared to others.

I liked the info, I thanked her for telling me something new. I patiently asked the question again rephrased, "is there any expert consensus on whether dreams make us more reproductively fit?"

her response

"If we have dreams we are hitting REM sleep which is the goal. Bc if you don't hit REM sleep, you aren't getting proper sleep. But you don't have to have active vivid dreams to have good sleep. Some of us are just wired to have more dreams than others"

What she's saying isn't wrong, but we've clearly derailed here. My cynical explanation is that I proposed a genuine mystery, but if a questions asked it implies there's a possible answer. And who else is more likely to have that answer than an expert. The experts mind queries itself like a LLM with chain of thought reasoning: "What's the answer to dreams?" The query successfully returns: "Dreams occur during by REM sleep, REM sleep helps parts of the mind fully rebuild to be more productive." The expert mind does a couple of quick validation checks.

I think her answer obviously did not solve the mystery. But I want to make my thought process transparent, so here I'll go through the possible arguments in defense of her answer (all of this is my hypothetical extrapolation, this part of the convo did not occur):

In favor of dreams being an advantage: REM is necessary for a well-functioning brain, which is obviously an evolutionary advantage and dreams are correlated with REM

Spandrel adaptations can be correlated with useful things, that doesn't make them useful. For example the appendix only appears with when the digestion track is present. It's not obvious that the appendix is an evolutionary advantage.

Against dreams being an advantage: Consider the variation in how some people never dream while others have vivid dreams every night. Surely this is an argument against it being a necessary part of the human reproduction cycle.

It might be evidence in this direction, but it's not obvious how strong of evidence this is. There's variation in humans between people with autism and those without, we still think its evolutionary advantageous not to have autism.

Ok I've already written 10x what was actually said in our convo. I hope it's clear that there are interesting implications to the REM correlation and variation thing, but it's not obvious how it resolves the evolutionary advantage question in one way or the other.

My issue is "experts" acting like they've solved a mystery when all they did was discuss details that are related to the mystery. Like Sherlock investigating a murder, pulling out hundreds different facts about how the crime was committed, but then failing to use those facts to deduce who did it. Saying "ah elementary my dear Watson", if we ask again who committed the murder.

My Takeaway

Beware calling yourself an expert, or thinking you have expertise! It probably makes you feel obligated to have answers when you should be uncertain.

I have screwed up many times at work because I thought I had special training to answer statistical questions. I often couldn't explain my proposed solution to problems. When I did give explanations it was a hundred facts related to the issue, but none resolved the issue.

I wonder if 50% of good problem solving is stripping away the unnecessary details. Elon's successful mantra "solve from first principles" seems to support this idea.

The problem with becoming an expert is you have delved into a world of details. Worse yet your identity as an expert is tightly attached to those details. To cut through to the actual solution you must say, "almost all those things I have learned and worked hard to understand and memorize are useless here"