Against Automaticity | Hacker News Reader

Against Automaticity

carcinisation.com

134 pointssimonsarris38 comments

Early in my career, essentially a lifetime ago, we were launching a new line of eco-friendly kitchenware. My team was divided on two messaging approaches. One camp believed in the emotional appeal: serene images of pristine forests, juxtaposed with families using the product, essentially proposing that using our product was akin to saving the environment. The tagline was something pithy to the effect of "embrace nature, one meal at a time."

The other approach was more socio-culturally driven. In particular, it focused on communities that had traditions rooted in nature conservation. the advertisements showcased local figures, elders if you will, using our kitchenware while sharing age-old wisdom about respecting the environment. the tagline was (something like) "honoring traditions, preserving tomorrow."

We ran both campaigns in a split test across multiple regions. to our surprise, the socio-cultural campaign outperformed the emotional one by a substantial margin, especially in regions with strong cultural ties to nature, like certain areas in the Pacific Northwest and parts of New England. Follow-up interviews revealed that people felt a deep, personal connection to the community-focused ads. They felt that the product wasn't just another item but a bridge to their roots.

All that said, even with that and other similar experiences over the course of my career, I'm bearish about completely sidelining the more automatic responses ingrained in us. While cultural imprinting and rational processes play significant roles in decision-making, it's probably a mistake to entirely dismiss our primal instincts and inherent biases. They've evolved over millennia and shape much of our intuitive reactions, long before conscious thought even enters the equation.

I'm not surprised by that, but for a slightly different reasoning: the "buy this product now to be good to nature" feels extremely fake, a bit dishonest and perhaps most importantly: unoriginal (every company seems to be pulling the "buy this to save nature/climate" kind of card).

The only product/consumption that could be saving nature is no product/consumption. To keep using your old kitchen stuff feels obviously better to nature than trashing it and buying more new stuff. No matter how "nature like" the new product tries to promote itself. Maybe this notion is obvious to most people? Maybe not...

The link to the past/elder/ancestors definitely feels less fake/bs to me.

Is this more than just pattern recognition? I mean, what you're saying is that we're wising up to the fact that "buy this for nature/sustainability/etc" is just another way capitalism gobbles up and resells criticisms against it. Capitalism is destroying the planet? Well, just make more plastic stuff and slap a "sustainable" label on it, we'll sell the good conscience AND the plastic stuff all in one.

The past/elder angle is equally fake, it just sells traditionalism, another anti-capitalist current, as a product, only that.. it's not as played-out yet?

But I was very much of the understanding that cultural pressures are among the strongest influences there are. You don't exert cultural pressure by showing pictures of lakes, you exert cultural pressure by having people of a shared culture exerting pressure.

Also there are some classical studies in advertising, which show that humans are much more effective than plants, and landscapes.

I appreciate the flaws pointed out in specific studies in this article, but the detour into phenomenology at the end (a philosophy on the subjectiveness of reality) gave me a bit of whiplash. It seems that the author here writes "we are not so irrational that a little nudge suddenly changes our whole character" and derives "so everyone's own reality is their definition of rational."

A more logical antecedent in my opinion is "we are influenced mostly by concrete priors, rather than minute nudges in behavior."

Agreed. Rational agents are not less mechanistic than automatic prime-able agents.

“Those who will accuse us of irrationalism have an extremely narrow view of rationalism, a view that scientists have abandoned a long time ago. Ever since quantum mechanics came to be, and quantum mechanics is probably the greatest scientific idea of this century (n.b. the 20th -ed.), you can kiss goodbye to causal explanation.“

There we go! Relativist anti-science nonsense masquerading as rational thought.

Yes, psychology and other “soft” sciences have serious issues. But I suggest to the author (and Rota) to understand physics (and the level at which quantum effects become viable) before quoting physicists.

Most of this article revolves around semantics. Pity! It had potential.

Check out the fraud committed by Francesca Gino. If it takes a decade to catch the smart Harvard professor (earning a couple of million dollars per year) who made zero effort to hide her data fabrication/manipulation, ya gotta wonder how pervasive the practice really is.

> An important motivation of the rationalist movement, as I saw it, was that we were all very irrational beings, and had to struggle to become more rational. My argument in this essay is that we are actually very rational, but managed to convince ourselves, for a variety of (perfectly rational) reasons using a variety of tactics, that we were helpless idiots.

Exactly my reaction when I saw how much energy LessWrong wasted on Torture vs. Dust Specks.

If I accept everything about the studies in here, that really only means people can be rational if they try to be and in put in the effort to get the right information for what they want to do. That's also limited by what you can remember to think about at the time or successfully encode into plans and notes. The problem is, lots of people don't put major effort into doing so. They drink alcohol, they get addicted to opiates, they have massively high time preferences (possibly due to previously mentioned), and they take massive risks. People may generally be rational, but running up to someone you think is insulting you, attacking them, and having no escape plan is probably not helpful for achieving your goals, but a lot of people do it every day.

There was a section about ego depletion. I know the science that supports ego depletion was horribly done, but nobody who has done a lot of climbing can doubt that there's something to it. You spend all of your spoons on the hard part, get past it, and then do something dumb for the easy part that follows. Happens all the time.

The ego depletion literature is sort of a minefield. The lack of replicability of the classic ego depletion phenomenon is pretty clear but (like a lot of psychological phenomena) there tends to be nitpicking about definitions, and strawmen disproofs.

The ego depletion idea specifically refers to the idea that exercising self-control leads to self-control fatigue, which leads to a rebound effect.

No doubt there's self-control and attention in climbing, but I think another phenomenon of "attentional focus leads to fatigue which leads to poor decision" or "physical exertion leads to fatigue which leads to poor decision" is not specifically an ego depletion phenomenon.

I'm not really trying to be pedantic, more thinking out loud about how phenomena that seem closely related and similar can get dismissed because of a very narrow definition of something.

There's also something to be said for lab versus real-world tests of ego depletion. I've always suspected that whatever you can do in a lab ethically is not ever going to be a real-world test of ego depletion. If nothing else, it happening in the context of some study suggests that people know there's an "out", and anything that would really induce fatigue to moral temptation probably wouldn't be approved by an IRB, at least easily.

I've always been skeptical of the original ego depletion phenomena as outlined in the original studies, in part because they never seemed to rule out alternative explanations well, but also have guessed that there's some variant phenomenon that is similar and explains similar things better. I also don't agree with the author that the ego depletion idea is "silly". Everyone knows about fatigue, and the idea it could be applied to temptation isn't unreasonable (on the face of it, this is how a lot of people often explain binging). Something can be incorrect but reasonable.

> then do something dumb for the easy part that follows. Happens all the time

An old rule I learned doing extreme sports in high school: Never say "just one more". That's when you get hurt.

I don't think you need ego depletion, brain metabolism, or any other grand science to explain why this happens. You relax, you pay less attention, you think you got this, and then you don't. Error rate is lower when you focus.

A corollary for this might be great rule of thumb I’ve got from mountain biking and skiing: “when you start to mess up the easy stuff is when you back it off and call it”.

Start coming up short on the easy little tabletops? Start finding yourself in the backseat on ski turns a bit too much? Time to have a break and reevaluate before you axe yourself because your skill for recovering might be saving you on the easy stuff, but if you’re not nailing that, you’re almost certainly going to screw up something with more serious consequences.

> “when you start to mess up the easy stuff is when you back it off and call it”.

A few nights ago I was parking my car parallel to the side of a street, the usual retro maneuver. I failed it badly, unusual but it happens. I got out and did it again. Failed again. That was when I understood that I was really tired. I probably made too much sport in the previous days. I got out and made it right, then took a couple of days of rest.

The brain is an organ like any other and requires nutrition for proper function. Nothing grand about noting proper nutrient flow helps with cognition. It's intuitively obvious that hungry and tired people make more mistakes.

> hungry and tired people make more mistakes

This is true. But even non hungry rested people make more mistakes when they underestimate a task. Which is what happens when you attempt a hard skill with a “eh this is the easy part” attitude.

It's not an overconfidence that gets me, it's a pessimistic extrapolation. I assume it's going to continue to be crux-level hard even past the crux and when it's not I still end up taxed as if it was.

The article seemed so overly biased against the studies even the concepts discussed in the earlier studies that didn't 'replicate', that I was immediately put on guard and didn't trust the statements about any current studies sited in opposition.

And, the last half kind of went off the rails. It almost turned into a rant against any hint that humans might be 'mechanistic'. But, there was no opposing theory around human thought or actions presented except "Its Quantum Mechanics" and that came off as "Woo".

So I was left with a kind of whiplash between the start with kind of reasonable questioning (paraphrase) 'all these studies showing humans can be influenced are wrong and woo', then the end of the article (paraphrase) 'humans are completely rational, there is no un-conscious influences, but we don't know how but maybe Quantum Mechanics".

At the beginning of the article, there was a lot of time spent discussing "Replication".

Then by the end it's "Science doesn’t work that way anymore, but few have gotten the message. ... Ever since quantum mechanics came to be, and quantum mechanics is probably the greatest scientific idea of this century (n.b. the 20th -ed.), you can kiss goodbye to causal explanation. "

So? Just a really long argument against Physics in the sense of Mechanistic/Clockwork Universe, but then turning around and using Quantum Physics to call upon Quantum Strangeness to bypass Causes/Effect.

All to get rid of the 'clockwork view'. But for what? To imbue humans with something more 'mystical', like a 'soul' or something? Cause and effect are tossed out, but nothing put forth to replace what is being torn down.

That's some extremely vigorous hand-waving in TFA!

Some automaticity theories were oversold, yes, but TFA, in typical SV-style "rationality" manner, manages to hand wave away large parts of reality.

It’s not at all clear automaticity even exists in a way that improves your understanding of the world.

You can assume people do things because they’re basically rational with consideration for their values, which may be axiomatic and their information, which may be incomplete or wrong - or you can assume people have a head full of fuzz.

Or we could not assume a priori whether people are rational or irrational, and instead observe (including our own behaviors) and try to fit them what we see in an explanatory framework.

Are our purchasing decisions all rational "with consideration for our values"?

(TFA on that aspect weasely includes class signalling as a "rational" choice, even though such signalling is mediated through implicit, not rationally arrived at and analysed, influences, not to mention people performing the same choices even when their purchases are not visible to third parties anyway).

Or, do we see rational behavior on Black Friday stampedes? Are those rational actors, trying to maximize their money's worth, and are those rational purchases (even with class signalling included)? Is sleeping outside an Apple Store for the night to get the new model first when you're not doing it to resell it? Or any number of similar phenomena?

If you extend rationality to also consider that even if the rationality of a circumstance is computable, you may not personally have the computational resources to determine it (what may be rational for a human may not seem straightforward to a macaque, or a goldfish) then yes, all of these things are rational.

The problem with calling things irrational is that it doesn’t allow you to make any predictions. You’re saying “there’s no logic to this thing” which may or may not be true but it is fairly bold to look at something you don’t have an explanation for and say “that’s irrational”.

>The problem with calling things irrational is that it doesn’t allow you to make any predictions. You’re saying “there’s no logic to this thing” which may or may not be true but it is fairly bold to look at something you don’t have an explanation for and say “that’s irrational”.

That's absolutely true.

But it's not the kind of irrational that's the article tries to refute. That one comes with mechanisms and predictions, it's just not conscious (or not fully conscious) to the subject.

The kinds of non-rationality ("automaticity") discussed are e.g. as if we noticed that every time we play somber music at work, a subject gets more moody. There's a mechanism to invoke it, and we can predict the reaction of us playing it vs more energetic music, but the subject would not consciously know that it does it - and it wont be a "rational" decision they made to be more moody.

Basically, the article tries to refute the existence of a whole class of unconscious automatic responses, bring back the crude man as some fully "rational agent" that economics threw into the dustbin decades ago.

Also, I imagine you've noticed that if two people get involved in a more in-depth discussion about particulars, there's typically no common agreed upon definition (or interpretation of "facts") for the word "rational", or many other of the words we use. Yet we carry on oblivious.

Row row row your boat...

I found the linked article about ads more interesting than the post itself. It's an insight that makes a lot of sense but I hadn't considered at all before.

Interesting! I would have thought that the ego depletion idea would be the only plausible bit from the entire school of research that the author (aptly) criticizes. (Who among us has not forced themselves to go to the gym and thereby been unable to resist the pizza?)

I think for it to be a test of pure ego depletion, the thing you were unable to resist would have to be unrelated to the thing you forced yourself to do.

Whereas 'hunger' or 'I just did something healthy' could both lead one to choose post-gym pizza without needing an ego depletion concept to explain it.

> If I’m going to bring Corona to a party or backyard barbecue...

This sure sounds different when you read it after the Covid pandemic...

Not really, everyone I knows calls Covid, "Covid", like you did.

I have a lot of mixed feelings about this essay, probably too many to put in a comment. I really enjoyed it and they raise a lot of good points to think about, and appreciated the bit at the end about causal mechanism, even though it came out of nowhere a bit.

I think they're making a bit of a strawman argument in an attempt to make an untenable argument for homo economicus, though.

This shows up in a few places.

For example, the author sets up this argument that ego depletion "is the foundation for priming-style automaticity and in fact most of the phenomena that I label as examples of automaticity." This makes no logical or theoretical sense at all. Ego depletion is a resource depletion theory of cognitive and behavioral control; priming and automaticity has to do with dual-process and fast-slow processing theories. They intersect but are not dependent on one another at all.

The author might be interested to know that there is a respectable literature arguing against dual-process / automaticity models on more sophisticated grounds. But they don't argue against the idea that humans are biased or that decisions can be made outside of conscious awareness, they argue against the idea that automatic and controlled processes are distinct, that it makes sense at all to separate the two types of processes.

There's several other examples of this. Evidence for cognitive biases is massive; the priming literature is just a small slice of it. There's all sorts of biases that operate outside of primes, and might not even be thought of as automatic, but are thought of as operating outside of awareness. EEG/psychohysiological studies have also documented pretty clearly that decisions can occur before we are aware of them. It's pretty clear that humans are not always rational and unbiased, unless you stretch the definition of rational to some point where it ceases to have any meaning outside of "not automatic".

I had trouble with the placebo effect section also. The meta-analysis on the magnitude of the placebo effect on depression the author cites itself recognizes the difference between active and passive placebo, which provides evidence the placebo effect isn't a simple regression to the mean or reversion via homeostasis. There have also been studies suggesting that iatrogenic placebos do maybe happen (in waitlist conditions for example), but they happen in such a way that outcomes are worse than passive placebo. Most people know they will be given treatment at the end of a trial regardless of their condition, so the explanation that they have to "fake bad" to get treatment seems plausible but just as likely as someone with the treatment who believes they've been assigned to control. Are placebo effects small? It depends on the placebo effect. That average is just an average: depending on the nature of the control, and the patient population, it can be small or large. Sometimes, the placebo effect is largest for precisely the patients where it matters the most.

Anyway, this all sounds like I disliked the essay which strangely I didn't. I think it comes across to me as raising some strong arguments (that maybe placebo effects are smaller than we think, or maybe automaticity maybe needs to be revisited), and I agree about the causal mechanism part. But a lot of it comes across as sort of hand-wavy; I think it could have left half of it out, or even aquiesced on some points, and it still would have been persuasive.

The phantom in the room is the lack of a definition of "rational" provided by the author. It seems as if a lot of the essay hinges on somewhat loose definitions: that goes for attacks on automaticity, but it could go the other way too. What's "rational" versus "biased"? Can someone learn something rational in one scenario, but have it become irrational due to circumstances changing in a way they don't recognize? Is that irrational?

Are there peer-reviewed studies supporting phenomenology at five sigma? Seems incongruous to lump phenomenology in with quantum mechanics if not.

I think the huge blindspot in American understanding of human behavior is individualism, the idea that you there is a model of human behavior that makes sense in some complete way even if that human were living perpetually alone on and island.

To me, that's like trying to understand a single worker ant without knowing anything about ant colonies. How can this creature be sterile yet survive as a species? Why is it constantly gathering food it does not eat? How does it defend itself from predators?

So much of human behavior that is perplexing or seemingly irrational makes obvious sense once you consider that humans are a species that evolved to live in an environment primarily made of other humans, some of whom are friend and some of whom are foe. And, further, a species that can't survive without a tribe of friends, and whose greatest threat to survival is members of other tribes. And, most vexing of all, a species where distinguishing friend or foe is next to impossible just from surface attributes.

Once you imagine the evolutionary pressure of a species in that environment, a whole lot of human behavior falls into place.

That's evo psych though, and I'm pretty sure general skepticism of evo psych was a lot higher than the skepticism of priming effects was. Call what you wrote a weak prior if you want, I guess.

Reply on news.ycombinator.com