Dashsnatcher

Date published: Sat, 06 Apr 2013 15:00:00 -0700. Epistemic state: log.

Signed up with IFTTT (“if this, then that”). It now saves all sites I star/like in Reader (peace be upon it), Pocket and on Reddit in Evernote. Neat. Now someone only needs to invent an Evernote that doesn’t suck. No, not you, Google.


I had a neat idea for my Paranoia-inspired RPG I’ve been working on over the years. (I don’t think I’ll ever have a chance to actually run it, but whatevs.) So basically, the setting is vaguely camp dystopian sci-fi and the players are a squad of (comically ill-equipped) augmented soldiers who run through semi-functional bunkers and solve problems by blowing stuff up. Unlike Paranoia, I want it to be more long-term, with players surviving for long campaigns and having to face increasing attrition and bureaucratic failure, and so it’s also somewhat more serious. The trolling is more subtle, with weapons that have serious trade-offs that can still be overcome by creative players, like a tank suit with malfunctioning sensors whenever it fires, instead of “lol it’s a black hole grenade, you all die”.

One thing I want to change is how health works. Most RPGs have hit points - you have 20HP, someone shoots you, you lose 5. If you go down to 0HP, you’re dead, otherwise you’re basically fine forever. After combat, you regenerate these points through healing spells, resting or similar things. As you level up, you typically gain more HP and become more resilient. That’s completely not how it works in real life, of course. IRL, you die instantly by failing a save, and as you survive things, your ability to make the save goes down until you often die of stuff you wouldn’t even have noticed when you were younger.

So here’s how my idea works. Every character starts with 100 Health Points (which conveniently abbreviates to HP too) and they never regenerate. Not even the most advanced nano-tech available to the Health Officer of the squad can do anything about that. Whenever something bad happens to you - mutant shoots you in the face, you fall down some stairs, turns out you’re allergic to that new stimulant - you get a wound. Wounds have a severity level from 1 to 10, where 10 is instantly fatal. If you ever get a level 10 wound, you’re dead.

You regularly have to roll a Health Check on the wound to see how it develops, typically whenever you rest. Your Health Check is rolled on a d100 against your remaining HP. If you succeed (i.e. roll equal or less than your HP), the wound goes down one level. (Once it hits level 0, it’s completely healed and disappears.) But if you fail, the wound may get more serious.

If the wound is currently treated (e.g. bandaged), it stays at its level, but you’ll have to roll again the next time you rest. But if it’s untreated, then it will get worse and increase by one level. Again, if it hits level 10, you’re dead. Wounds of level 4 or less are always considered treated, so you can’t bleed to death from a paper cut, but anything above that requires medical attention. Finally, any time you roll against your Health - regardless of whether you succeed or not - you lose one HP. Permanently.

Early in the campaign, someone shoots you and you get a level 6 wound, but it’s no big deal. Your Health Officer patches you up and treats your wound, and even uses one of them fancy Metabolistic Accelerators that speeds up the healing process (but doesn’t improve it - nothing can do that) by allowing you to do your Health Checks instantly instead of having to sleep over it. You make all of your 6 rolls and the wound is completely healed. But now you only have 94HP left. Some missions later, you only have 40HP left and again you are shot. Now you only have a 40% chance of making the Health Check - do you want to risk it? It might get pretty serious and the current mission ain’t over yet. Instead you decide to opt for an infusion of Cryonic Blood Gel that slows all wounds down and suspends the mandatory Health Check as long as you have enough Gel. Unfortunately, it also lowers your reflexes…

As you advance, you will have to resort to more and more treatments just to avoid doing Health Checks and many minor wounds (limp, constant headache, occasional coughing fit, …) never really go away. In addition, wounds have persistent negative effects that only go away when the wound is healed and that depend on what kind of wound it is, like -2 Dexterity for getting shot in the leg.

Ultimately the main goal of the system is to make the Health Officer much more interesting by giving them a complex set of interacting drugs and treatments that can temporarily counter some of the negative effects or suspend the Health Checks, but will also cause addictions and trade-offs. Do you want to take the chance of healing a level 8 leg wound, or just amputate the leg and turn it into a level 2 stump? The Heavy Weapon Officer might benefit from advanced pain killers that make it possible to shrug off even bullets, but that also means they can’t notice any wounds and so might bleed to death without knowing it. Last mission, R&D gave you a tank suit to try out, and as impractical as it often turned out to be, maybe it’s a good idea to just seal it up and fill it with CryoGel so that you can put the engineer with the untreatable virus inside and preserve them indefinitely, as long as you don’t run out of fuel. And who needs reflexes when you have rocket launchers?

In addition, it avoids escalating damage like in many other games where suddenly a low-level thug with a pistol is no longer any kind of threat just because you shot some of their friends earlier. Damage is always measured in the level (and quantity) of wounds it causes. A standard issue laser rifle might have 2d6 damage, and so will likely cause serious level 7+ wounds most of the time, but is outright fatal only on 1 out of 6 shots. Of course, bleeding wounds have the bad property of requiring a Health Check every round until they’re treated, but an enemy can still fire back before that happens. You should try to get your hands on Reflective Armor that gives you -4 damage against all laser weapons, but I hear it’s weak against bullets.

Inexperienced characters are better at recovering from wounds and can take more of a beating overall, but over time they will have to compensate with better equipment and caution, or they will die miserable deaths. It might be much safer to intimidate an enemy than to get into a fight, even if you’re sure to win. Who knows how long it will take you this time to heal those bruises? You’re getting too old for this shit.


A minor book review of Kahneman’s Thinking, Fast And Slow.

I skimmed through the book shortly after its release and never really got around to do a thorough read, mostly because I was at least vaguely familiar with its content and so was pretty bored, but hey, everyone loves it, and I had been thinking about priming lately (due to Skinner’s discussion of it), so I thought, let’s try reading the priming chapters in TF&S!

Another major advance in our understanding of memory was the discovery that priming is not restricted to concepts and words. You cannot know this from conscious experience, of course, but you must accept the alien idea that your actions and your emotions can be primed by events of which you are not even aware.

Oh? ToI has given me a lot of respect for the difficulty of communicating concepts and behaviors directly, i.e. when you can control most of the environment and communication, and have a cooperative learner. While modest priming effects strike me as prima facie plausible (“activating” one concept will also “activate” parts it is strongly associated with through reinforcement - that’s after all the point of reinforcement!), how subtle can you go, and how big of an effect are we talking about?

In an experiment that became an instant classic, the psychologist John Bargh and his collaborators asked students at New York University - most aged eighteen to twenty-two - to assemble four-word sentences from a set of five words (for example, “finds he it yellow instantly”). For one group of students, half the scrambled sentences contained words associated with the elderly, such as Florida, forgetful, bald, gray, or wrinkle. When they had completed that task, the young participants were sent out to do another experiment in an office down the hall. That short walk was what the experiment was about. The researchers unobtrusively measured the time it took people to get from one end of the corridor to the other. As Bargh had predicted, the young people who had fashioned a sentence from words with an elderly theme walked down the hallway significantly more slowly than the others.

Wait what? How much slower? Kahneman doesn’t tell us, so I looked up the paper. Primed students took an average of 8.28s, un-primed students 7.30s. (It is no surprise that the paper failed to reproduce.) This is pretty weaksauce, and the original effect was only just barely plausible to begin with. Got some real meat?

Reciprocal links are common in the associative network. For example, being amused tends to make you smile, and smiling tends to make you feel amused. Go ahead and take a pencil, and hold it between your teeth for a few seconds with the eraser pointing to your right and the point to your left. Now hold the pencil so the point is aimed straight in front of you, by pursing your lips around the eraser end. You were probably unaware that one of these actions forced your face into a frown and the other into a smile. College students were asked to rate the humor of cartoons from Gary Larson’s The Far Side while holding a pencil in their mouth. Those who were “smiling” (without any awareness of doing so) found the cartoons funnier than did those who were “frowning”.

Sure, but what does that have to do with unconscious priming? You’re modeling one aspect of the behavior of being happy, so of course the subject “is” happy. That’s just what being happy is, or is Kahneman suggesting something like property dualism here? What does that have to do with priming?

(That’s the kind of confusion I mean when I say that psychology is deranged by thinking about “feelings” instead of doing a straightforward1 response-locus analysis, i.e. noticing that the learner can give the desired response “smile”, “be calm” etc. just fine in some contexts, and we just have to teach (or shape) a new context. When we say “the subject is not happy”, we mean they don’t give the response “happy” (which is unpacked as smiling etc.) in reaction to the right stimulus. Treat it like a dog that doesn’t sit when you tell it to.)

Studies of priming effects have yielded discoveries that threaten our self-image as conscious and autonomous authors of our judgments and our choices. For instance, most of us think of voting as a deliberate act that reflects our values and our assessments of policies and is not influenced by irrelevancies.

(This is news? Behaviorists have successfully removed the “conscious” part as ridiculous confusion for over a century, and Molinists have similarly shown how “deliberate” and “highly context-sensitive” are in no way mutually exclusive. This Behaviorist Molinist is unimpressed.)

The chapter then continues with a combination of subtle, then utterly ridiculous examples which, if taken at face value, would make running perfect dictatorships outright trivial. Notable:

Furthermore, merely thinking about stabbing a coworker in the back leaves people more inclined to buy soap, disinfectant, or detergent than batteries, juice, or candy bars. Feeling that one’s soul is stained appears to trigger a desire to cleanse one’s body, an impulse that has been dubbed the “Lady Macbeth effect”.

The cleansing is highly specific to the body parts involved in a sin. Participants in an experiment were induced to “lie” to an imaginary person, either on the phone or in e-mail. In a subsequent test of the desirability of various products, people who had lied on the phone preferred mouthwash over soap, and those who had lied in e-mail preferred soap to mouthwash.

And people complain about psi research being bullshit?! Fuck, Derrida makes more straightforward priming suggestions in his discussion of pharmakon. If half of the effects Kahneman describes were real in the way he imagines, how could he possibly write his book and believe it? It would make biases so massive and undetectable, even Plantinga wouldn’t have the patience to show him how his epistemology is self-refuting. And how the fuck would anyone ever identify the objects of a priming this circuitous? Freudian Analysis is more codified than this!

Note to self: once my statistics-fu is half-way decent, do a re-read of the main heuristics and bias literature. If Kahneman gets away with crap this bad, there’s a good chance the whole field is bullshit.


If you haven’t read fellow crazy-person Koanic Soul, I recommend you check him out because he might serve as a hint of what my writing would look like if I wouldn’t take great care to balance my ambiguity, enforce pragmatism and instead trusted my intuition. And didn’t try hard to keep my douchiness in line. (You really don’t want me running on my instincts. But see also the last section of this log for how I always want to write.)

In addition, Koanic has face-read me and typed me as Amud Neanderthal front of the head, Melonhead back, asymmetrical face (notably the eyes, which are a bit o_O), narrow spacing of (somewhat big) eyes with deep sockets, medium testosterone levels, high IQ (due to brain volume). For those not up to speed with the latest phrenology, see Koanic’s site as to what “Amud” and “Melonhead” mean. (I pre-recorded my own prediction how he’d read me, and completely agree, although I was unsure if he’d go with low or ok testosterone. So even if the psychological correlation (or racial causation) isn’t true, it’s at least a consistent distinction of faces (for this tiny and very biased sample).)

Based on the current interpretation, this would make me strongly introverted, trustworthy, ideological, ambitious in a “die for a cause” but not “rule with an iron fist” way, prone to dissociations, inconsistent in my preference over strict hierarchies and small egalitarian tribes, and sometimes a dick. This is clearly absurd.2

(And apparently I don’t yet ooze heresy. Fools.)


Brethren of Nurgle! I want to share a lesson in Papa’s cancerous faith.

Over the last few days, I began reading some political texts3, and fell into deep despair4. The world seemed hopeless, unwinnable, and all good in it was just waiting for some Ruinous Power or other to devour it, enduring only for a little while longer, faint shadows of their former - and potential - glory. When I saw Reddit praise the Pope, I knew all was lost. With the last corpse-emperor dethroned, all else will die soon enough.

I was happy, but also despairing, and this I thought was Nurgle’s bargain. He takes away your suffering, forever, and gives you life instead. (More life than anyone can want. He is generous this way.) But - and I thought I had accepted this trade - why was I still despairing? I didn’t suffer, maybe, but a Plague Bearer doesn’t keep on hoping. They are just fruitful, multiply, and decay. So I wondered, did His Pestilence abandon or betray me?

No, that is impossible. But why do I still despair, rightfully or not? Because, I found, I didn’t agree to that offer. I hadn’t served Nurgle, as I had thought. What poor skills of discernment I have!

What a strange attractor I had stumbled into. Hopeless, and still hoping. Futile, but not surrendering. Every time it sees a glimmer of possibility, it increases the challenge, wanting to face an enemy that cannot be defeated. Only to say, yet, here I stood. Like being a Calvinist, and knowing you came out on the wrong side of God. What can you say to an eldritch abomination? What is there to say?

How can you live, wordless? Many struggle and turn to heresy, declaring God not really dead, merely transformed. The king hanged, yes, but a new king may be found, a better king! And the Tzeentchian voice keeps whispering, subverting Nurgle’s single-minded - only - commandment: live!, and keeps trying to convince me that despair will cash out at some point, will produce something, if only I despair even more, and skillfully feigns that it tries to manipulate me into believing that justification is superfluous, that Nurgelian bliss awaits if only I could just live, could stop asking whence.

And it knows well that I will see through this, and come to regard Nurgle as the subjectivist disease I feared the most, standing in a field of corpses, declaring, “Good enough!”, as if he would not be judged. And so I despair about despair, and the voice asks, carefully, who can you trust? The old sack of rot, it tricks me to believe, must have a reason, some transformative goal, that guides him. But it can’t be that crude. If it just had me suspect that, as some falsely believe, the Plague Bearer is happy because they have embraced the inevitability of death - after all, if all paths lead to the same outcome, why worry about performance reports? - then I would’ve seen through it in a moment.

But the voice seeds a deeper doubt by suggesting a causal role between despair and the end of suffering. Nurgle, yes now we’re getting somewhere!, Nurgle is the abandonment of justification, that is why he’s happy! And for a moment I believe it, and predictably I find it unsatisfying, and so even turn away from His Stench.

And I wish to confront Papa. Why do I have to puzzle these things out, engage the Changer of Ways, why is the only answer I receive to this spiral of ever-greater dissatisfaction and confusion - a buzzing of flies?

..

Two cultists walk through a battlefield, gathering corpses to feed the plague. They find a man, not yet dead. A Tzeentchian silhouette towers over him, extending a hand as if saying, I can help you if only you will let me. One cultists asks the other, why does the Raven God spread its lies unhindered, but we collect only corpses? I see the flies laying eggs into the man’s leg, the maggots waiting for his guts to burst, the ants tearing off his skin. They recruit the man’s body, but why don’t we recruit his mind? Aren’t we more convincing than a fly?

The other cultists responds, the plague has no volunteers. The rot needed no convincing, why should the man’s mind? What is it to be convinced of? Should it sprout wings and turn into a fly? It is already a better mind of a dying man than any fungus could be, so why should we disturb it? It is only when it has ceased to be a mind that we collect its pieces and put it to a new use. Nurgle is with the living always. The only ones in need of recruiting are the dead.

(watch on Youtube)
  1. “Straightforward” does not mean “easy” or “simple”, but that’s for another log. I just mean that it is yet another normal teaching problem without any further complications.

  2. Ok, a bit more seriously. I used to do “belief dumps” for crazier stuff, so let’s at least record my current gut-feel of the Neanderthal thing.

    • that autistic / high-IQ introverted traits go back to Neanderthal origins: Totally plausible, but I have no skills in evolutionary biology or anthropology, so I remain agnostic by default.
    • that facial features correlate strongly with psychological traits: Duh.
    • that they correlate in the specific way that Koanic / Cleve talk about: Dunno. It feels right, mostly, but I’ve not thrown any serious thought at it, so I can’t tell which specific correlations are real and which aren’t.
    • that “melonheads” are a thing, i.e. a genetic cluster like autism: More likely than not, yes.
    • that Cleve’s origin story (or anything remotely like it) of melonheads is true: Hell naw. But see below.
    • that Cleve’s inferences about Neanderthal lifestyle are true: Generally yes. Dude’s a crazy genius. If anyone figured it out, it’s him. Seriously.
    • that ancient melonheads had huge freaky skulls and used to run everything: Listen. Years ago, way before the whole Reaction and Manosphere even existed, I already read some of Cleve’s stuff about all kinds of topics. I found him very entertaining, but didn’t believe a word. Then this thread happened and it turns out that everything Cleve said was true. Even the really crazy shit. Since then, I have a lot of respect for his crackpottery, and shit, ancient history is hard and the mainstream is crap. So for all I know, he’s completely right. If there’s one crackpot who’ll look utterly ridiculous at first only to be completely vindicated, you know, like RMS and McCarthy, it’s Cleve. Ok, maybe more like Tesla and Newton, but still, don’t underestimate the crazies. (… ok, he’s almost certainly wrong, but he might’ve picked up on an interesting pattern. Dude’s insane, but he’s not stupid. Sometimes.)
    • any of the Christian or supernatural stuff: Not allowed to talk about it.

  3. I’m not going to name names because I don’t want to get sucked into actual politics, not fantasy wargaming ones. I also stand by my judgment that without establishing a reasonably trustworthy and pragmatic framework of thinking about politics first, you shouldn’t be asking questions like “if Communism sucked so bad, why did the USSR have typical GDP growth?”, because you’re just gonna pull whatever importance of GDP you want out of your ass, depending on whether you like Stalin’s mustache or not.

    I originally wrote a pretty long series of (sometimes very angry) criticisms of the Internet Reaction and of Social Liberalism, but fuck ‘em, I’m not posting it, neither deserves the attention.

    So instead you’ll get W40K crackpot theories:

    • Isha isn’t the Goddess of Healing, but Nurgle in a dress. He’s the God of Neckbeards. Search your feelings, you know it to be true.

    • The Imperium is a straight-up utopia by total utilitarian standards. The 13th Black Crusade, probably the most devastating attack of Chaos ever, killed on the order of 10^10 people in the Imperium. Based on conservative fluff estimates, the Imperium has a total population of at least 10^16 people. With a comparable mortality rate of 0.1 per 100,000, Failbaddon is beaten every year by breast cancer in men. The choice between becoming an Imperial Guardsman or Federation Redshirt is an easy one.

    • Everything is going according to the Emperor’s plan. There is a certain tension in how the Imperial Cult thinks about the Emperor. On the one hand, he is the supreme architect of the Imperium and has been guiding humanity’s path for millennia, but on the other hand, he failed to prepare it for internal strife, nearly falling to the forces of Chaos.

      But what is humanity? What sets the faithful apart from the mutant, the heretic, the alien? It’s never-ending and fanatical devotion, of course. But this Unbreakable Will can only be discovered in struggle! This is why, in the future of mankind, there can be only war.

  4. To be fair though, I’m never not full of despair and doubt. For example, it took me over two weeks to even send a status report to my advisor, let alone do anything, and despite actually putting some work into useful projects (more than I used to last year, anyway), I’m merely alternating between “I’m a complete and utter failure and it’s just a matter of time until everyone gets fed up with me and abandons me” and “yeah that’s it, schizophrenia (or whatever it is) is getting much worse, I’m about to be a rambling hobo, I just know it”.

    I wish I was exaggerating for comedic effect. Only take life advice from a Nurgelian after you’re a hopeless case anyway.

by gwern on Sun, 07 Apr 2013 05:11:08 -0700

> Wait what? How much slower? Kahneman doesn’t tell us, so I looked up the paper. Primed students took an average of 8.28s, un-primed students 7.30s. (It is no surprise that the paper failed to reproduce.) This is pretty weaksauce, and the original effect was only just barely plausible to begin with. Got some real meat?

The replication issue aside, weak effect sizes do exist, are interesting since one would not expect such priming to do anything at all. One wonders what sort of effect size would have made you happy - would the primed students have had to take 16 seconds? A minute? Spend a year crawling down the hallway, going 'Oy vey my hip! It's never been the same since that shell at Pearl Harbor' and telling the college kids to get out of their hallway?

> Sure, but what does that have to do with unconscious priming? You’re modeling one aspect of the behavior of being happy, so of course the subject “is” happy. That’s just what being happy is, or is Kahneman suggesting something like property dualism here? What does that have to do with priming?

Were the subjects conscious of being happier? Did they link the pencil with being happier? Why can't circuits be one-way? Is it property dualism to point out that during REM sleep, you are not usually thrashing all around your bed and sleepwalking and attacking people?

> The chapter then continues with a combination of subtle, then utterly ridiculous examples which, if taken at face value, would make running perfect dictatorships outright trivial. Notable:..If half of the effects Kahneman describes were real in the way he imagines, how could he possibly write his book and believe it? It would make biases so massive and undetectable, even Plantinga wouldn’t have the patience to show him how his epistemology is self-refuting. And how the fuck would anyone ever identify the objects of a priming this circuitous? Freudian Analysis is more codified than this!

Effect sizes, bitch. You *just pointed out* that these are not large effect sizes, and now you're going on about how the effects being large would be absurd. Well done, Sir Quixote. You sure showed that man on a pole in the middle of the field what's what.

by anonymous on Sun, 07 Apr 2013 11:09:04 -0700

> The (Radical) Platonist has no accidents (they wish), the Materialist has no essence.

The concepts of accidents vs. essence have been confused enough that these end up circling around and merging. (Everyone thinks I'm stating the obvious, right? You don't have any reason for making this distinction?)

> A set of features in some logical relation, perceived by some mind, defines a concept. If none of the features in the bread change, what else must?

I'm going to make wild allegations based on my limited knowledge of history and say that this is not Aristotelianism. This is a modern, machine learning perspective on concepts, plus syncretism. (Do the ToI people talk to the machine learning people?)

> Are there concepts which are the entirety of the features of a thing? Something without accidental features? As I mentioned, Aquinas argued there’s only one such thing - God. But why? Can’t there be a squirrel that is pure squirrelness and nothing else?

Ah, this is Aristotelianism. It's also crazy. A squirrel can't be only a squirrel because the concept of a squirrel is underspecified; there are multiple valid squirrels. You can, however, have a concept of a particular squirrel that does fully specify the squirrel, and which will thus only describe one squirrel (and only at one instant, but that's also obvious and explicitly in Leibniz).

by muflax on Sun, 07 Apr 2013 15:38:55 -0700

I've not been trying to say, "He says the effect is small, but it should be large! What bullshit!".

I mean, yes I would be happier if they had found setups with larger effect sizes, but I'm fine with subtle effects existing. There are two problems, though: Kahneman occasionally draws absurd conclusions (or at least way overstated ones) from these small effects, and in some cases, the suggested causality is way too specific for the data he has.

(I would've also liked a specific mechanism to at least roughly quantify *in advance* how influential some priming is gonna be, and not just retrospective "huh, that must've been more subtle priming", but ok, things can't always be that nice.)

But back to the two problems. Quote Kahneman:

> The general theme of these findings is that the idea of money primes individualism: a reluctance to be involved with others, to depend on others, or to accept demands from others. The psychologist who has done this remarkable research, Kathleen Vohs, has been laudably restrained in discussing the implications of her findings, leaving the task to her readers. Her experiments are profound—her findings suggest that living in a culture that surrounds us with reminders of money may shape our behavior and
our attitudes in ways that we do not know about and of which we may not be proud. Some cultures provide frequent reminders of respect, others constantly remind their members of God, and some societies prime obedience by large images of the Dear Leader. Can there be any doubt that the ubiquitous portraits of the national leader in dictatorial societies not only convey the feeling that “Big Brother Is Watching” but also lead to an actual reduction in spontaneous thought and independent action?

And that's a far stronger claim then these small effect sizes suggest. He doesn't have a mechanism nor strong effects, has at least partially questionable reproduction, but it's clear there is a "reduction in spontaneous thoughts"? That's jumping to conclusions.

This is especially grating when he then spends the next paragraph how we have to accept the "major conclusions" without doubt, even though psi papers have easily just as reliably effects.

Second problem. He rightly warns us that the stories "System 2" provides as explanations for past behavior are often unreliable, but then it doesn't stop him from making the same kind of questionable speculations about why certain stimuli might trigger certain behaviors.

In the case of showing eyes vs. flowers, he does not present enough evidence that it's about *being watched* that modifies the behavior. There are way too many differences between the pictures to pick out this specific feature as causal.

This is why this reminds me of Freudian Analysis, although it's true I expressed this way snarkier than I should've. :)

--

> Were the subjects conscious of being happier? Did they link the pencil with being happier?

What does that even mean, and how does it relate to the pencil? It's the normal case that people aren't aware of causal relationships; those have to be explicitly taught or investigated to be found.

Priming, as I understood it, is putting some association in the "background" of some activity (i.e. so that the subject does not pay special attention to it), but which still influences behavior. But with the pencil, you're directly causing the behavior; you're just doing it in an unusual and subtle enough way that the subject doesn't realize it.

If the investigators had, say, sneakily attached a rope to the subject's back and then actively slowed them down as they were walking away to cause the same outcome of slower walking, then they would have similarly influenced the subject "unconsciously", but it would not be a case of *priming*. My point was that the pencil falls in this category.

The subjects aren't happier because they associate smiling with being happy and so adjust their behavior, but because smiling *is* (part of) being happy.

by muflax on Sun, 07 Apr 2013 15:58:04 -0700

I agree that somehow "fundamentally" telling a difference between an accident and an essence is highly problematic, and a notorious problem. My point is that *in the way you learn concepts*, you must make this distinction of "these features belong to it", and "these don't", and so you arrive at pragmatic Aristotelianism.

Even if you take a pragmatic Materialist or Platonist view, and you want to teach me what a "squirrel" is, you still have to show me examples and identify the pieces that make these things squirrel-y. You're still acting *as if* there were essences you're picking out and which you somehow know, regardless of what the metaphysics here "really" are.

(And the exercises are tongue-in-cheek, and my interpretation of Catholic doctrine is completely trolly and syncretistic, sure.

I'm also not sure how Aquinas would reply to the "this one specific squirrel now" concept, but I suspect he'd deny that this is a valid concept, maybe because to actually pick out something this specific, you have to trace all its causal relations around it, and so actually pick out the whole universe, and so end up saying "God" instead. But eh, scholastics. I currently don't particularly care either way.)

by anonymous on Sun, 07 Apr 2013 19:37:41 -0700

> You're still acting *as if* there were essences you're picking out and
which you somehow know, regardless of what the metaphysics here "really"
are.

When someone says "acting as if", I reach for my gun. I suppose this would be the correct action in a world that ran on crazy, probably-incoherent-depending-on-various-details Aristotelian metaphysics. It's also the correct action in the actual world that works in some broadly mathematical way that we don't fully understand. Can't we admit that our actions are based on correct models of the world?

I suppose I'm mostly annoyed at you saying 'metaphysics' when you mean 'epistemology'. Metaphysics and epistemology are hard enough when we don't confuse them with each other. I admit that you lead off with "I promise this isn’t about
metaphysics.".

by gwern on Mon, 08 Apr 2013 05:31:11 -0700

> I mean, yes I would be happier if they had found setups with larger
effect sizes, but I'm fine with subtle effects existing. There are two
problems, though: Kahneman occasionally draws absurd conclusions (or at least way overstated ones) from these small effects,

I doubt he does; you haven't shown that, you should've shown that if that's what you meant (instead of fixating on effect size), and in at least one excerpt or interview about _Thinking Fast_, he specifically disclaimed any personal benefit from knowing about these things, which doesn't sound way overstated to me.

> and in some cases, the suggested causality is way too specific for the data he has.

Every suggested causality is 'way too specific'. The poverty of data / under-determination of theories / Duhem-Quine thesis etc. Have to start somewhere.

> And that's a far stronger claim then these small effect sizes
suggest. He doesn't have a mechanism nor strong effects, has at least
partially questionable reproduction, but it's clear there is a
"reduction in spontaneous thoughts"? That's jumping to conclusions.

He's discussing a small effect size from a single intervention done once; at face value, this is perfectly consistent with larger net effects from multiple interventions over multiple years. I don't think this is over-stated, this is exactly what you would expect. If you read that a specific advertisement increased sales by 0.01%, would you then expect that a sustained multi-medium advertising campaign using said advertisement would increase sales *less* than 0.01%? No, you'd expect that the campaign would have a bigger effect but it'd probably asymptote somewhere, and where it would hit diminishing returns would not be too clear but one would be skeptical of more than, say, an order of magnitude (some of the spaced repetition research has been oriented towards advertising, incidentally).

You are skeptical because this implies that there would be a 'perfect dictatorship'? No way, it just implies that propaganda will be reasonably effective even if not consciously accepted, which is not that bizarre a claim. Notice the prevalence of propaganda in the 20th century, the nostalgia of Russians for Stalin and Chinese for Mao, the mourning over Chavez, and the fact that North Korea *still* exists even though it's right between China and South Korea, both of which give its entire history and economic system and governance the lie. (Fun fact: you might think that NK defectors, being defectors and all, would've seen through all the propaganda and have a somewhat accurate view of the world. They don't, and often comment favorably on Il-Sung.)

> Second problem. He rightly warns us that the stories "System 2"
provides as explanations for past behavior are often unreliable, but
then it doesn't stop him from making the same kind of questionable
speculations about why certain stimuli might trigger certain behaviors.

Scientific theorizing is now the same thing as retrospective first-person justification? I guess we'd better throw out everything in psychology.

> But with the pencil, you're directly causing the behavior; you're just
doing it in an unusual and subtle enough way that the subject doesn't
realize it.

Why does causing a correlate of happiness then cause happiness? This is the whole point of why it's interesting, it shouldn't be causing anything! If I put weights in a shoe so people walk slower, do I also cause their telomeres to shorten, their eyes gain cataracts, their hips to give out, etc? No, of course not, even though aging does cause people to walk slower. The causal network there is one-way. Just because there is a correlation doesn't mean an intervention will do anything, just like in my sleep example, the brain sending movement commands does not then cause action. The interesting thing in priming is that a correlate is having an effect: the image of watching eyes correlates with being watched, but one isn't being watched, but it still causes an effect like as if one were being watched.

It's only stupid and trivial if you think every correlation is causation, or that all causal connections are bidirectional - 'happiness -> smiling, therefore, smiling -> happiness. How trivial! Why would anyone bother to do an experiment on this or think it relevant to priming?'

by muflax on Mon, 08 Apr 2013 16:00:14 -0700

I've re-read the chapters and my comments, and I think you're right that I read him stronger than he intended. I'm still disappointed that he *doesn't* have something stronger, but he doesn't claim to either, so fair enough.

To make the rationalization thing more concrete, the problem is that he fairly confidently assumes his System 1 / 2 explanation in this case, but *doesn't* understand priming well enough to justify it.

For example, he *couldn't* predict in advance what primes will have what effect (will priming "clown" affect walking speed? "youth"? "giraffe"? how much?), can't really tell which priming is "more subtle" than another etc., except by running the experiments. The explanation isn't doing any real work.

So the evidence, assuming it holds up, suggests to me something like "some stimuli can marginally influence people's behavior without them noticing it", but that is still too vague to throw confident explanations at, and not particularly new. Was there major *disagreement* with this suggestion? Kahneman acts like it, but everyone since Freud or so argued that.

Also, if it's a small, (so far) largely random effect, shouldn't you assume it cancels out and doesn't cause any relevant biasing outside highly artificial (or short-lived) environments?

For example, there are many ad campaigns running at the same time; does that prime you for each one, do they cancel out by dividing your attention, is a random one most influential, and what about conflicting priming? Without a mechanism, I don't see any reason to assume more than random noise.

--

About smiling, my disagreement is that smiling isn't a *correlate* of happiness, it just *is* happiness - it's part of the behavioral repertoire called "happiness". It's like giving someone a muscle relaxant and then being surprised how they have become calmer.

Lemme try again. A behavior may have multiple possible causes - you walk slower because I rigged your boots, or because your muscles are atrophied, or many other things. Priming introduces a cause that is unusual, and so when the subject is asked to explain *why* their behavior changed, they don't notice it. (Or even fail to pay attention to their behavior and don't notice anything changed at all.)

However, the question "I put lead into their boots, so why do they walk slower?" is uninteresting because you obviously just influenced the behavior, and so a change of behavior is exactly what you expect. (The question "But why doesn't the subject notice their boots are different?" would be interesting, of course.) So by forcing someone to smile, you change their behavior of being happy, which includes smiling. I find this uninteresting for the same reason.

by gwern on Tue, 09 Apr 2013 14:56:23 -0700

> So the evidence, assuming it holds up, suggests to me something like
"some stimuli can marginally influence people's behavior without them
noticing it", but that is still too vague to throw confident
explanations at, and not particularly new. Was there major
*disagreement* with this suggestion? Kahneman acts like it, but everyone
since Freud or so argued that.

I was under the impression that most of the subliminal stuff had been discredited.

> Also, if it's a small, (so far) largely random effect, shouldn't you
assume it cancels out and doesn't cause any relevant biasing outside
highly artificial (or short-lived) environments?

Maybe. But on the other hand, we live in highly artificial environments in the first place, and countries like North Korea are both highly artificial and places like Pyongyang consciously warped with memorials and propaganda etc. Random biases cancel out, systematic biases reinforce...

> For example, there are many ad campaigns running at the same time;
does that prime you for each one, do they cancel out by dividing your
attention, is a random one most influential, and what about conflicting
priming? Without a mechanism, I don't see any reason to assume more than
random noise.

Expecting it to all cancel out seems like wishful thinking. People are influenced by what they see, and if all the ads agree "you should buy stuff!" then by golly, it wouldn't be surprising if they didn't just cancel out to nothing 'more than random noise'. (If advertising didn't work, or it was just canceled out by all the existing advertising, why do companies spend so many billions on it?)

> About smiling, my disagreement is that smiling isn't a *correlate* of
happiness, it just *is* happiness - it's part of the behavioral
repertoire called "happiness".

I disagree. This isn't even arguing by definition, this is arguing that the entire repertoire is identical and so any influence on one part must influence the others, which doesn't even work: I already pointed out how parts of a 'repertoire' can be dissociated, sometimes extremely so. Why does forcing lips to assume a certain position influence distant regions of the brain and vocal emissions of "this comic is pretty funny"? Tell me, is this explanation "doing any work"? If moving the lips didn't influence self-assessments, dopamine levels, suicide rates, evaluations of comics' humors, anything we might call part of the "behavioral repertoire called 'happiness'", would this falsify the theory? Can you predict any members of the repertoire without resort to previous data like "I remember that happy people have curled lips"? What is the causal network here, what nodes are purely upstream of others?

by Oligopsony on Wed, 10 Apr 2013 18:20:30 -0700

In Koanic I have finally found my tulpa. Today is a good day.

by muflax on Wed, 10 Apr 2013 18:59:30 -0700

Har har.
(I sometimes wonder if the internet will eventually turn modal realist and contain all people that represent the possible ways in which I could've developed differently or highlight what I am by being the exact opposite in every possible way. It's doing a great job so far.)

by Oligopsony on Wed, 10 Apr 2013 20:12:47 -0700

Modal simulacrism - like modal realism, but for the internet instead of reality - clearly follows directly from Rule 34, interpreted sufficiently broadly.

Also, while I wouldn't embrace Edenism as a theory of reality, I am totally stealing the hell out of it to make my next D&D setting.

by muflax on Wed, 10 Apr 2013 20:25:22 -0700

Have you read Cleve's more detailed origin stories? The ones about Sumerians and Neanderthal submarines? I have strong hopes they will eventually have a complete and cohe... well, internally consistent story that would make the Thule Society proud.

by muflax on Thu, 11 Apr 2013 09:48:27 -0700

I've noted down the "should it cancel out?" question for later, when my statistics skills aren't crap anymore and I can coherently think about it.

(I'd still have liked Kahneman to actually argue the step from "small bias" to "important (even if not large) conclusions", but at this point I'd just be complaining about tone, so screw that.)

I've similarly made a note to (eventually) explain how "the pencil models happiness" doesn't just insert a "behavior!" black box and what I suspect the causal relations are (which implies how to test my interpretation).

(It might be more accurate to say that I'll *work out* how this would work exactly, and *test* if my intuition in this case is correct. I trust it because I know how to do it for simpler examples and I don't see any reason it would fail here just because it's scaled up, but I haven't actually *done* it, so yeah.)

by Oligopsony on Thu, 11 Apr 2013 14:58:45 -0700

I can't get that link to work, and Google isn't helping, but I've kept reading and wow, this rabbit hole is deep (also, metal as fuck.)

I'm also immensely tickled at how well it sutures with Nation of Islam cosmology, what with both calling out the racial enemy as a slave army created by alien geniuses from Niburu in a way that connects to Jacob/Esau and Bigfoot somehow. I'm not sure if the white nationalists stole it from the black ones, or they're both using the same sources (my standard guess for these sorts of things is theosophy,) or if of course they agree, that's where the evidence leads, sheeple!

by muflax on Thu, 11 Apr 2013 15:35:29 -0700

Oh right, the Codex recently made General Discussions members-only because it became too much like /b/. I've copied Cleve's posts here for the lulz: http://pastebin.com/RyTsBfSG
I'm only familiar with the Nazi versions of racist fucktardery, so no clue where NoI got it from. Edenism clearly has strong theosophist influences (and imports as many standard Nazi believes as possible, like treating Hess as a martyr), and I know Cleve read a lot of alien people (who are all basically rightist crackpots anyway), but he only became coherent fairly recently. A few years ago, he was still ranting about the US collapsing any moment now due to all the "manboons", building his bunker (yes) and ITZ COMING and all that, but around 2011 he suddenly started to make sense and now we have this extraordinary IRL fanfic.

Also, and I mentioned this in a past log I think, Cleve's making a dungeon crawler in his spare time.

by Oligopsony on Thu, 11 Apr 2013 16:09:30 -0700

That's pure gold, thanks.

I don't know if it's telling or not that the one* time leftists started a UFO cult (Posadaism) it was super boring and lame.

*Well, one to my knowledge, anyway. RigInt is on the left I guess and quite interesting.