Hacker News new | past | comments | ask | show | jobs | submit login
Eight studies reveal a bias in human imagination (experimentalhistory.substack.com)
109 points by saikatsg on Dec 25, 2022 | hide | past | favorite | 79 comments



But what about the mundane explanation? It's not about imagination but motivation. When people are asked a question, they generally assume that a "stupid" answer is not what is desired.

Where is the nearest grocery store? "Earth."

How can your phone be worse or better? "It could not work."

There are endless ways, most of them uninteresting, to make something worse. Making something better is interesting.


You might be curious to know that philosophers and linguists would see your (good) examples as violations of the Gricean maxim of quantity. More: https://en.wikipedia.org/wiki/Cooperative_principle


The way to test this might be to ask about something that's so bad there are fewer (hence more interesting) ways to make it worse than to make it better?


My understanding is that all perceptions are done by temporal diff. So comparison of what has changed since the last sample.

So using comparison is probably our default way of looking at things. Which is what the studies say.


Exactly. When asked about how something is, people generally don't think anyone wants ideas on how to make it worse.


They eventually rephrased the question to specifically say "... better or worse". It didn't change anything.


Of course it didn't. Why would it ? Making stuff better is more interesting, unless you're holding "pick the most comical way to make something bad".


In fact, I love driving my gf up the wall with these types of answers :)

"What do you want for dinner?"

"Food"

"What KIND of food?"

"The good kind"


Before all the nitpicking starts I want to point out this part, which I found most interesting:

> Everybody knows about the hedonic treadmill: once you’re moderately happy, it’s hard to get happier. But nobody has ever really explained why this happens. People say things like, “oh, you get used to good things,” but that’s just a description, not an explanation. Why do people get used to good things?

> Now we might have an answer: people get used to good things because they’re always imagining how things could be better. So even if things get better, you might not feel better. When you live in a cramped apartment, you dream of getting a house. When you get a house, you dream of a second house. Or you dream of lower property taxes. Or a hot tub. Or two hot tubs. And so on, forever.


Anecdotally, my brain is pretty wired for making plans. When I didn't have a house and security my plan was that. Every move I made was connected to those plans in some way.

When I got a house, I spent a bunch of money and it was during a recession, so my two adjacent plans then became derivative. I already have plans for when I'm secure, but they're mostly around optimizing for cost, efficiency, and reliability.

I used to not have a lot of money, and for a time, was stuck in the poor cycle. Plan making let me escape that thinking/action cycle.

Now, I don't think I'd be less happy without optimizations to my house unless they violated some ideal I had like living frugally. I was definitely unhappy watching my chances to get a house slip away but I turned that into some happiness by realizing that every time I walked away from a deal I had a little more money, which increased my odds of not only getting a house, but getting the exact one I wanted. By the time I bought a house I basically dictated the terms because I had cash in a recession.


> Before all the nitpicking starts...

...because we can all imagine how the article could be better!


Touché


Because there's an evolutionary advantage to thinking you could be happy, not for actually being happy.


Can you expand on why you think that?

(Btw, I agree with you but want to check my priors)


I'll let the depression turtle elaborate on my behalf. Now You Happy Always Maybe

https://www.youtube.com/watch?v=yWkq7btSQvs


If that video is accurate does that imply tool use in other mammals is because they weren’t happy? For example: ravens use sticks, does that imply that they will evolve with unhappiness?

Then how come other animals like orcas and dolphins that show signs of a mood didn’t keep evolving their general intelligence?

Kinda hand wavy video but maybe that’s the idea? Not sure.


Are ravens ever happy with what they have? I legit don't know. What I do know is that the innovation of a permanently unsatisfied brain is an incredibly recent fork. So close we don't even separate it from modern humans. So close that you could reasonably argue if its actually biological evolution or if its the principle of natural selection applied on a sociological level instead.

The industrial revolution and the great windfall of wealth that came with it didn't make people happier (to the extent we can measure it). But there is one group that does seem to find happiness: the small pockets of civilization which took a different road, never bothering to switch from hunter-gatherer to farming.


First, thank you for actually discussing this because I’m interested in this topic and appreciate your point of view.

I do not know if any animal is “happy” but this[0] is the paper (and there are quite a few that are similar in nature) that have empirical evidence that “happy chemicals” aren’t only there for happiness.

Another point is that the “happiest” and longest living pockets of civilization are actually farmers[1]. Costa Rica and CA should be discounted as there is a 20 year average age discrepancy between Japan/Italy/Greece(90s) and CA/Costa Rica (70s).

In my opinion, I would say that happiness is evolutionary and it is driven by desire to get better/explore, etc. This fits your thesis but I fork at the notion that it’s a driver. My hunch is that happiness is kinda like candy or weekends — intended in small doses as reward for survival. Since “happy chemicals” are responsible for such a large swath of human bodily functions[2], a high release can have detrimental effects (dopamine is responsible for movement and memory, for example, in addition to motivation).

Basically, I agree with the video that we’re not suppose to be always happy and our search is actually a by product of understanding “happiness”. Kinda how pigs will purposefully knock apples to the ground and eat them a week or two later to get drunk[3].

[0]: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2720267/

[1]: https://time.com/5160475/blue-zones-healthy-long-lives/?amp=...

[2]: https://my.clevelandclinic.org/health/articles/22581-dopamin...

[3]: https://modernfarmer.com/2014/03/drunk-pigs/


Emotions in general are hard to define, let alone pin to a particular chemical. If I were wearing my contrarian cap I would question the extrapolation from "happiness correlates" in humans to other animals.

Sadly, that's about all the time for in depth effort posting I have today. Maybe if I had more time I'd be happy...


The mice died without dopamine by starving themselves to death...

Also, anti depressants have a strong history of effectiveness and their mechanism is dopamine re-absorption.

Regardless, I appreciate the conversation and happy holidays.


I mean, we have a physiological answer to the question, dopaminergic circuits control desire and goals, putting us on the treadmill, whereas serotonin circuits avoid that and bring contentment.


> For instance, why do people hate Congress and love their phones? Obviously the answer is “Congress is bad and my phone is good,” but what’s actually happening in people’s heads when they say that?

Congress is a clique filled with old inside traders that hoard money like dragons while the phone is a device which enables me to even know about that. That's what's going in my head.


It seems to me that very wealthy people hoard money like dragons. Congresspeople generally aren’t very wealthy. Instead it seems like prostitution is a better analogy. Congresspeople quite often have their votes swayed for relatively small amounts of money.

I’m being pedantic on the analogy because I think it’s important for people to have the right framing on a major problem with American politics. Namely, too much money going to campaigns and PACs influencing politicians. Congresspeople generally aren’t super rich.

Your point about what goes on in your head still stands.


Congresspeople regularly outperform the market with their investment which is called "insider trading" and is forbidden for anyone else but them. I'm not saying that's the biggest issue with them, but that was what came to my mind first.


Right…why is this nonsense becoming normalized?

Congress folk are highly likely to have an investment professional managing their money, who usually outperform “the market” (whatever that means, considering most don’t benchmark S&P for obvious reasons).

What I really don’t get is that most insider trades make less than 70k before going to jail and even with inside information, a large percentage of insider trades don’t make money (but you still go to jail). Yet folks think it’s some kind of cheat code.

Folks need to stop reading WSB.


Are you under the impression that it is illegal for members of Congress to engage in insider trading, so therefore the absence of prosecution proves it is not a frequent occurrence?

The laws are literally different for them: https://www.nbcnews.com/politics/congress/it-illegal-lawmake...


It is illegal: https://www.congress.gov/112/plaws/publ105/PLAW-112publ105.h...

SEC, that prosecutes insider trading, is a civil regulatory body and cannot send people to jail, which the article mentions for some odd reason. And I’d put the odds at 100:1 that any prosecutor in the SEC would love to go after a Congress person to further their career.

Lastly, not sure most are aware, insider trading is usually spotted by SECs internal AI/algo that is connected to all brokers, clearing houses, etc. they are very good at finding folks.

Edit: another thing that is missing from this discussion is the SEC awards a portion of the proceeds to the whistleblower of any case. I just can’t believe no one ratted out a Congress member for retirement money and all members are security masterminds.


I understand and agree that this is a great corruption but they don’t have Smaug levels of wealth and so your analogy fails. They are prostitutes that sell their votes/souls for money in order to influence outcomes.


>Congresspeople regularly outperform the market with their investment which is called "insider trading"

Have you considered the selection bias going on as an alternative explanation. Maybe the people with the competence and skills to get to Congress are just smarter than the average person with an investment account.


It’s a fair theory but given the long track record of proven insider trading, their connections to people with insider information, and their repeated refusal to enact prohibitions on insider trading (it is in fact perfectly legal, despite efforts to apply the same rules that everyone else plays by), I think there is a lot of evidence for systemic corruption and against “they’re just better investors”.


By your theory, auditors should be put into jail on a much higher rate as they have WAY MORE insider information, yet they don’t.

Also, there is a law prohibiting senators from trading on insider information, which kinda goes against your whole point.

https://www.congress.gov/112/plaws/publ105/PLAW-112publ105.h...


There are also laws against rape, murder, and jaywalking, yet they happen daily.


At first glance you seem to be comparing events that have to be witnessed vs database entries that are centralized, heavily audited, and very accurate.

It’s a lot easier to match patterns in a db vs. real life crime.


They're pointing out that something being illegal doesn't stop it from happening.

But more to the point, you're wrong here and we know you to be wrong because otherwise white collar crime wouldn't exist, but it does.


I don’t understand why you have such a stance and how you “know”…SEC publishes their enforcement data on an annual basis, which supports my argument.

https://www.sec.gov/news/press-release/2022-206


Maybe they're smarter and more knowledgeable than the average trader?


This doesn't pass the sniff test.

There's nothing about being a Congressperson that lends itself better to the kinds of knowledge or intelligence that would increase the chances of betting well on the stock market...

...except the knowledge that comes from a) knowing what legislation will be ahead of time, and b) getting "paid" in insider knowledge from companies that want favorable treatment. (And possibly other kinds of insider-trading-type knowledge; I'm not particularly familiar with the specific methods they use.)


> There's nothing about being a Congressperson that lends itself better to the kinds of knowledge or intelligence that would increase the chances of betting well on the stock market

How do you know that? Their outcomes indicate otherwise.

They're clearly in some regards quite exceptional people.


> Their outcomes indicate otherwise.

This is assuming the consequent.

There are two possible explanations being proposed for the observed outcome "Congresspeople do better than average at the stock market."

I am proposing the explanation "they are engaging in insider trading," and proposing evidence to support that.

You are proposing the explanation "they're just better at it", and giving as evidence solely the outcome we are attempting to explain. It's circular.


Sure, we both lack evidence. However, my view is that people, even congresspeople, deserve the benefit of doubt.


How do their exceptional investment insights fare after they leave Congress (and, for completeness, the Senate.)


I don't know. It'd be interesting to know. But I imagine there's less public information about their investments after their terms end, no?

In any case. I'm not saying there is no insider trading going on in Congress. After all, there are hundreds of representatives. Of course some are greedy idiots, and some even criminals. But the majority, and I believe thats a large majority, are honest people who care about the people they represent.


Or they use a professional that routinely outperforms their specific benchmark.


while I don't agree that they're better at trading, your response is clearly wrong here.

A congress person is able to constantly make decisions with imperfect information and be successful doing it, one can imagine that skill could give them an edge when trading.

But the reason I don't agree with it is because they're using 3rd part managers. It's far more likely that they're advising.

However, the way to put this to bed is to compare their accounts with those of others being managed by the same team. do they out-perform, under-perform, or roughly match those other accounts? There's your answer.


At glance making money at non-engineering is more correlated with sociopathy than above average intelligence...


Congress and its foibles are organizational behavior writ large and expensive.

We've reverted to a quasi-aristocracy and there is scant will to leverage tech in order to realize the ideals of the Constitution.


Reverted? That seems to be the inherent arrangement in representative democracy without an imperative mandate.


We could probably replace Congress with an app. That's what is going through my head.


I kinda wonder now... if you took a language model like ChatGPT, and additionally trained it on, say, public votes, statements, speeches, publications etc of some particular politician, how reliable would it be at predicting future votes of that politician?


For sure, I nominate TikTok so we can accelerate the inevitable app based government debacle.


There is a simple explanation for why humans tend to say how things could be 'better', when you ask them how things could be 'different'. It's even one of the core beliefs of buddhism: life is suffering.

To unpack that statement, think about someone that does nothing with their life. Literally nothing. They'll quickly suffer hunger, starvation and finally a slow death. These are the defaults of life, you cannot sit still. It is not a choice. Even us who feed ourselves, and live a life, will still suffer old age and death. All one can do is take action in their life to reduce their suffering now, and in the future. You can apply this model to all facets of life, from relationships, work, gym, healthy eating, art and so on.

So really the study has just thought about it from the wrong angle. It's not people thinking 'how things could be better', its the brain naturally thinking 'how can I reduce my suffering further'.


Somewhat counterintuitively, people are often more creative when they have more restrictions in place. Being limited in what you can do can make people more likely to find creative solutions.

I suspect that this is an example. There are a huge number of ways things could be worse, and a far smaller number of ways things could be better. So it may actually be easier to come up with examples from the latter set, since your options are more restricted.


That's interesting to read. I find that I do imagine how things could be worse in specific situations, the usual example being right before I get what I might want.

But in that case I'm almost always comparing hypothetical to hypothetical, i.e. bracing myself before trying someone's first homecooked meal. I imagine that I'm about to eat literal dirt, and this makes even lightly-seasoned sludge taste somewhat fancy.


Side info - here is the major recent revelation about human nature that is backed by the largest social study of its kind:

https://www.scientificamerican.com/article/scientists-probe-...


This isn't new - ethologists have been making similar conclusions about altruism for a long time now, including studies on very young children (which show that it kicks in somewhere around 2-3 years of age).

The catch is that said altruism is rather strongly affected by parochiality, meaning that it's weaker towards the outgroup, and conflict of interests between the ingroup and the outgroup tends to override it.


> ethologists have been making similar conclusions about altruism for a long time now

However those conclusions were not backed by such a large and objective, replicable study that had a good control group.

> The catch is that said altruism is rather strongly affected by parochiality, meaning that it's weaker towards the outgroup, and conflict of interests between the ingroup

Nope. If you read the details, you will note that such divisions do not have the importance people think they have. Any group they created from random people behaved the same way.

The even more reveling concluson was this: When they gave people time to think, education, upbringing started to kick in and people started making selfish decisions. Which points out that the selfishness that is seen in our society is not something natural or instinctive - its man-made.


This one study doesn't show parochiality, but plenty of those preceding it did, including those for young children. So I wouldn't rush to conclusions here.


Actually, I would. This is the only objective study that has this size and uses actually measurable criteria and looks into the issue in a fundamental level - even while using adults.


The 'possibly universal' came from also conducting a survey with English speaking Polish and Mandarin speaking mainland Chinese people. But what might the results have been if the experiment was done with paranoid people which likely wouldn't have been in these samples? 'Cross-cultural' would be a better description.


I was heading to my mom's for Christmas when my kid said "we forgot grandma's present". I said I'm glad you remembered now and not later. I could have imagined it differently but worse "wish you remembered before we left". So in this case different was worse for me.


Interesting observation. There are some people who would have responded in the latter fashion. They're generally quite unpleasant to be around, and perhaps that points to another "cause" for this apparent positivity bias. Persistent negativity might be selected against in both cultural and genetic evolution.


It may be that I wanted to make my daughter feel better in spite of her (and my) mistake, rather than blame her.


There might be a reason why "build a better mousetrap" is a phrase while "build a worse mousetrap" isn't.

Although the middle managers who have destroyed things like the e.g. the Discovery channel are kind of doing the latter, but they'd argue they're making it better by increasing profits through cheap paranormal ghost shows.


Not directly related to the conclusion but I did find it interesting how they claimed the Mandarin study was identical but also clearly omitted some of the most initially “negatively” viewed topics like coronavirus, government, and politics (probably for obvious reasons, but also seems like it might affect the results too keeping things squarely in the positive category to start).


Almost all cultures lay emphasis on using positive words than saying negative words. Long ago there was a superstition that spoken words might become true. It is not completely superstition as words do have psychological effects on the listener or thinker. So when encouraged to talk in certain way, it encourages to think the same way.


It might be interesting to check if this effect correlates (inversely) to gratefulness / religious practice. I mean, if you only think about how things could be better you might not be so grateful for the things you have, but if you think at least sometimes about how they could be worse, then you might!


The universe is highly biased to make most things worse, baring intervention.

We call it thermodynamics, Murphy’s Law, and other things.

So evolution countered that bias by focusing our attention wherever we might conceivably be able to improve things. Or push others to improve things.

That’s my theory



> How could X be different?

This is great prompt for ChatGPT. I had fun exploring its internal biases. Examples:

guns -> add safety & tracking features

drug abuse -> decriminalization, safe injection sites

national defense policy -> fewer nukes, more cyber & space defense

interrogation techniques -> more ethics & sensitivity training, more psychological & non-coercive techniques

environmental regulation -> prevention, penalties, accountability, global approach

cryptocurrency -> scale up, security features, regulation, integration

abortion -> increase access, funding, legal protection

racial justice -> more academic, employment & leadership opportunities, intersectionality

It clearly has a bias towards lefty, pro-government, pro-regulation positions. I would assume because its training inputs came more from academia, less from right-wing rants.


I suspect its fitness function was not only explicitly rewarding lefty rants, but also punishing righty rants.


It seems more pertinent to ask it about more general or even neutral items, like the study did, rather than inherently political and controversial ones.


This is about language, not thinking. When you use the word could, it usually implies a positive spin on things.


This is not mysterious. A preference for imagining improvements is evolutionary beneficial.


It's not a bias, 99.999999999% of things in the Universe could have been better.


Or worse!

So it is definitely a bias, and I respectfully submit your comment is a strongly expressed example.


I really doubt the claim that this is independent of the phrasing of the question.

I assume strongly that if you phrase the question as “how might your life be worse or better” people will also thing of negative alternatives.

The thing is when you only ask “how might your life be different” people will pull on what they have already thought about and most people have plans and dreams for improvements already in their mind.


"... worse or better" was one of the phrasings that they've tried, per the article, and didn't see any changes in their studies.


I don't think you can get possibly universal measurements on humanity by relying on services like mechanical turk for results.


This is why they deliberately went out of their way to get some non-WEIRD[0] subjects for the later studies.

[0] https://en.wikipedia.org/wiki/Psychology#WEIRD_bias




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: