it's obvious .... isn't it?
July 6, 2018 7:31 AM   Subscribe

 
TL;DR: the scope of our attention is not infinite; contextually irrelevant data is excluded when it impacts the collection of relevant data.
posted by grumpybear69 at 7:39 AM on July 6, 2018 [12 favorites]


That is a very unkind way to describe Stan Van Gundy.
posted by Holy Zarquon's Singing Fish at 7:58 AM on July 6, 2018 [7 favorites]


What Gorilla?
posted by Liquidwolf at 8:01 AM on July 6, 2018 [5 favorites]


Watch this video of a gorilla very carefully and count how many steps the gorilla took.

Got it?

DID YOU NOTICE THE BASKETBALL GAME YOU DUMMY??????
posted by ian1977 at 8:06 AM on July 6, 2018 [33 favorites]


I like the fact that this article calls attention to the pessimistic ideology at the core of so much popular psychology. There are a number of dubious or nonreproduced psychological concepts that have come under recent scrutiny. The article cites the various "priming" experiments; the concept of ego depletion is another one. What these ideas have in common is a kind of self-undermining of our own capacity for thought and action. It's a secular flavor of total depravity, with a corresponding secular soteriology in the form of AI, "rationalism" (cf. LessWrong), and transhumanism. Unsurprisingly, these ideas are hot among the technocratic sages who would become the AI-Magisterium.

Believing IN ourselves (cautiously and within reason) is a subversive act.
posted by overeducated_alligator at 8:09 AM on July 6, 2018 [12 favorites]


The MetaFilter vision symposium continues!

Much of the surprising or counterintuitive nature of the kinds of perceptual and cognitive biases begins to fall away if you abandon the approach of assuming that the function of vision, our other sensory systems, and the human mind in general is to produce accurate, complete representations of reality, for a Cartesian agent to act upon. Instead, the function of these systems is to produce adaptive behavior, which entails a different set of requirements. The process of producing behavior begins almost immediately in every sensory system, with inputs from the brain modulating the nature of the responses to external stimuli on the basis of the demands of the current behavioral state.

As the article says, the failure of most people to see the gorilla in the basketball video is not due to a cognitive bias that prevents us from seeing the obvious. It is due to a highly efficient, optimized visual and attentional system for extracting behaviorally-relevant information from a complex, dynamic, information-dense world. The lesson of this and other studies is not that humans are bad at obvious things, it's that humans are really good at switching between tasks and disregarding information irrelevant to their current goals. To the extent that this manifests itself as errors in reasoning and cognitive biases, it is because humans are not always good at identifying what the relevant task is, or what their current goals ought to be. Address these issues, and much of the rest will follow. And, as a side note, it is exactly these problems that artificial intelligence still sucks at.
posted by biogeo at 8:11 AM on July 6, 2018 [86 favorites]


I didn't even see this post.
posted by kyrademon at 8:14 AM on July 6, 2018 [31 favorites]


If I didn't know better, I'd say the current run of FPPs has been a ploy to keep me from writing the various things I actually need to be writing. That said, talking about inattentional blindness (Simons and Chabris, 1999) is so far up my alley that the raccoons are dancing on the trash cans to get my attention.

For anyone who hasn't seen the original invisible gorilla video, go have a look (stimuli from Simons and Chabris, 1999). Your task is to report the number of times the white team passes the ball back and forth. About 80% of naive observers will fail to report the gorilla who walks through the scene (although, given how well-known this video is, there aren't many naive observes out there anymore. The fundamental issue here isn't that your visual system doesn't have the information (your retina sees the gorilla just fine), but that you don't notice the gorilla. Why don't you notice the gorilla? Because it's not task-relevant, and you're engaged in a relatively demanding attentional task. If you just watch the video without a task, you'll notice the gorilla pretty easily.

The basic inattentional blindness result here (which doesn't originate with Simons and Chabris, it goes back to Mack and Rock in the 1970s, at least) has been pretty extensively replicated, and is always mediated by task demand. A couple great versions are pedestrians who fail to notice unicycling clowns (Hyman et.al, 2010), or pedestrians who dodge a branch with money on it, and do not report that they dodged dollar bills hanging in their face (Hyman et.al 2014). The key here is that the implicit detection task in all of these (gorilla, clowns, money tree) is totally orthogonal to what the subject is actually doing, so there's no reason to detect these atypical events.

This is usually interpreted in an attentional framework, which is where that article really, really gets a lot of the field wrong, and I'm not too surprised, because a professor of business strategy isn't likely to either read the visual attention literature or the vision science literature more broadly. Interpreting it superficially, one might say "you don't notice the gorilla because you weren't attending to it." The problem is, that's kinda circular, and assumes you need to attend to things to perceive them, which rams into the problem of how you know to attend to things if the world is changing. If your awareness is totally gated by attention, how do you know where to attend? You've got to have preattentive processing of visual input in order to guide attention, and that's where the author's reading of the literature here falls apart.

Modern ideas of visual attention pretty much spring from Treisman and Gelade, 1980, which describes Feature Integration Theory. In a nutshell, it says "you use attention to bind features (orientation, shape, color; the elements which comprise objects you perceive) into a perceptual object, and in the absence of attention, you don't know what the object actually is, but you have a sea of unbound features." Lots of people just read it as "you need attention to bind features" (read: attention = perception), without realizing that Treisman considered what you had to have preattentively. In the 1980 paper, she's not particularly clear about how you go from unbound to bound, just that attention does it. There are a couple of her papers in the 1980s that push this forward (Treisman and Schmidt, 1982; Treisman 1988), describing what you perceive without attention (illusory conjunctions; your brain going "I've got these features, and something must be here" and starting to think about the question of spatial information inherent in features), but it takes a slightly different take on Feature Integration Theory to really push the question of how attention knows where things are in order to bind them.

That's Wolfe's Guided Search theory (Wolfe, Cave and Franzel, 1989; Wolfe 1994), which describes feature maps. It's a theory of visual search (as is Feature Integration Theory), so it describes how you find what you're looking for. Think of feature maps as "you know where all the red things are in the scene" and now imagine that across all features (e.g., you know where each color is, you know where each orientation is, etc.), so now attention knows where to go when you're looking for something, and does so based on the sum total of what's most likely the search target at each location. So, to go back to Simons and Chabris and their undergrad in a gorilla suit, you're tracking the basketball, which, as a target, is nothing like the gorilla.

The really critical thing to note about these theories of attention (rather than how they're interpreted by someone who doesn't seem to know this literature as well as they should) is that none of the people involved ever interpreted attention as a total gatekeeper on perceptual awareness. They particularly don't these days (Treisman last updated her theoretical work in 2006 to reflect the intervening several decades of work showing what the visual system is capable of without focal attention), and Wolfe is still improving Guided Search (see Wolfe and Horowitz, 2017).

That article screams "I know just enough perceptual science and psychophysics to be dangerous, and not enough to think about it outside my own preconceptions." What's even better here is that the article's author seems to think that Kahneman wouldn't have a good appreciation of modern attentional theory. That's supremely ironic, since his late wife was Anne Treisman (sadly, she passed away earlier this year, after a 50+ year career in psychology and vision science).
posted by Making You Bored For Science at 8:17 AM on July 6, 2018 [118 favorites]


grumpybear69: your TLDR says it perfectly.

One day a few years ago I'd just gotten home from work and Ms. nobeagle was telling me about her day. I don't remember the exact subject, but I think she might have been talking about her classes and future projects / assignments. In the middle of this, she said, "Oh, there's bread on the stairs so don't step on it." Perhaps she even gestured at the bread while saying this. She then continued talking to me about the same, non-bread topics. She finished talking and I stepped over the dog gate to go up the stairs and stepped right onto the bread.

She still says that I should have been able to avoid the bread since she warned me. I still say that putting a sentence of non-sequitor randomness in the middle of current and future planning is a great way to have something get ignored/forgotten. grumpybear69's summary will now give me the obvious allusion that I should have been using when trying to point out how easy it is to miss/dismiss something like that while one's consciousness is concentrating on building up and around something else.

And seriously; bread on the stairs?
posted by nobeagle at 8:26 AM on July 6, 2018 [10 favorites]


It would interesting to look at the role of saccades (related post here) in people's perception of the video.

Presumably, if someone is counting passes, their eye movements are going to be saccades, where the brain fills in everything from one point of attention to another without actually perceiving it. So the gorilla might actually be getting filtered out at a more basic neurological/visual level.
posted by pipeski at 8:28 AM on July 6, 2018 [7 favorites]


Attention is good for two kinds of things. One is goal-based: when you are hiking, you want to be able to look at the forked path ahead of you and then look back to your map to choose where you want to go. The second is stimulus-based: when something in the world is relevant to your survival, all of your brain's processing resources should divert to that.

Which is why when you're reading a map and then A WILD TIGER APPEARS your brain should be able to say OH MY GOD THERE IS A TIGER, I AM DONE WITH THIS MAP, IT IS TIME TO RUN.

What the FPP misses is the critical context that researchers in attention study the interaction between these two modes of attention. He doesn't think it's utterly fascinating that observers engaged in a task don't have their goal-based attention utterly disrupted by a freakin GORILLA? That goal-directed attention can be so efficient and powerful that we miss stimuli that should demand our attention? Alright. The rest of us are happy to continue to study how the human visual system works and to figure out how long it takes drivers to notice a moose in the road (shoutout to MYBFS), or figure out how I can be so engrossed in reading MeFi on my phone that I could literally step in front of a moving car, or figure out how we can design transit alerts so that passengers actually notice them.
posted by nicodine at 8:32 AM on July 6, 2018 [8 favorites]


In fact, pipeski, there is work on that (although it's not published; I know about it because I know the people who did the work). It doesn't seem to be a saccadic suppression issue, but the eccentricity of the gorilla (how far it is away from the current point of gaze) may be a factor. It's a lot harder to distinguish the gorilla from the black t-shirted players if it's not where you're looking, since your visual system has a harder time segmenting the two due to featural similarity (the gorilla suit and the shirts both being black, albeit with different textures).
posted by Making You Bored For Science at 8:33 AM on July 6, 2018 [8 favorites]


> shoutout to MYBFS

Oh my god how fantastic, I didn't need to shoutout, he is already here, summoned along with biogeo and the other vision scientists

There is no cabal. Look, count the basketball passes!
posted by nicodine at 8:34 AM on July 6, 2018 [7 favorites]


The trash-can dancing raccoons summoned me.

That, and the need to stop sweating before I could start thinking in lab this morning.
posted by Making You Bored For Science at 8:39 AM on July 6, 2018 [3 favorites]


MetaFilter: the eccentricity of the gorilla may be a factor
posted by biogeo at 8:41 AM on July 6, 2018 [9 favorites]


So, one can be eccentric, and one can also study eccentricity (in a visual sense; where images on the retina are distant from the fovea). It gets entertaining if you're a talk show producer and you're putting together a show about eccentric people, go looking for a psychologist who studies eccentricity, get someone who does the vision science version, but who is also quite eccentric in his own right.

Why yes, there is video.
posted by Making You Bored For Science at 8:45 AM on July 6, 2018 [11 favorites]


What these ideas have in common is a kind of self-undermining of our own capacity for thought and action. It's a secular flavor of total depravity, with a corresponding secular soteriology in the form of AI, "rationalism" (cf. LessWrong), and transhumanism.

As far as I'm aware (and I may be misremembering) the emphasis on our biases and the inexactness of our cognitive heuristics was in specific reaction against the computer metaphor for human cognition. So it was like

mid-century psychologists: Computers are cool and process information. Humans also process information. Are human brains essentially computers?
late-century psychologists: VERY NO
Transhumanists: But don't you wish you could be?
posted by a snickering nuthatch at 8:48 AM on July 6, 2018 [10 favorites]


Yes, very nice. If I could favourite some of these comments twice, I would. I had glanced at the article earlier, but didn't bother reading it, because the author seemed to have totally misinterpreted Kahneman.

Why yes, there is video

OMG, the one and only comment on the video is: "oh my goodness stuart had hair at some point?"
posted by tickingclock at 8:49 AM on July 6, 2018 [3 favorites]


Yep. He's always been about that weird, though.
posted by Making You Bored For Science at 8:54 AM on July 6, 2018


She still says that I should have been able to avoid the bread since she warned me. I still say that putting a sentence of non-sequitor randomness in the middle of current and future planning is a great way to have something get ignored/forgotten.

My wife does this too, but more importantly I also had a boss that did this, long ago. It was like a guessing game to figure out which task was the top priority, because he would excitedly talk for ten minutes about the pet projects we were working on, with one sentence in the middle about some completely unrelated project some other group had deemed the priority. He would deliver project requirements in the same way. Fortunately we have better processes now and he's an individual contributor again, not in charge of giving important instructions to anyone.
posted by The_Vegetables at 8:59 AM on July 6, 2018 [3 favorites]


I've never understood why anyone is surprised by the gorilla video thing. If you don't look directly at the gorilla - and of course you don't if you've been specifically instructed to look at something else - it's just another figure of roughly the same size, shape and color as the black-shirted players. Of course you don't notice it isn't one of them! Do it with a hot pink gorilla and I'll be impressed.
posted by waffleriot at 9:15 AM on July 6, 2018 [2 favorites]


What these ideas have in common is a kind of self-undermining of our own capacity for thought and action. It's a secular flavor of total depravity, with a corresponding secular soteriology in the form of AI, "rationalism" (cf. LessWrong), and transhumanism.

I disagree, a realistic view of what humans are capable of and how and why we fail is the single best tool we have for positive change and increased compassion for those who fail, make mistakes or are just disadvantaged. While the "all humans are gods in their ability to reason!" stance might be good for our own egos, it is terrible for building a just society.
posted by Infracanophile at 9:16 AM on July 6, 2018 [6 favorites]


Anyone who has gone hunting for mushrooms is aware of how "looking FOR a thing in order to see the thing" works.

Since "you see what you are looking for" is a pretty solid concept, so also should be "you don't see or attend to what you are not looking for". Even, it seems, if what you are not looking for is someone in a gorilla suit. If we saw all the things, all the time, sleight of hand would not exist. I am not sure why people are amazed by this, but okay.

Anyway, along similar lines is the following video: https://www.youtube.com/watch?v=A8syQeFtBKc (Though I do think it's kind of cheating to give us a cute narrative to obscure the point of the thing. Of course people are going to follow the narrative about the block printer and the user of cursive.)
posted by which_chick at 9:25 AM on July 6, 2018 [1 favorite]


Can I just say that I am really digging all these recent FPPs on visual processing and the like and the informed commentary? It's an interesting topic, but I have no background and am very skeptical of pop neuroscience.
posted by praemunire at 9:25 AM on July 6, 2018 [6 favorites]


I actually enjoyed the Aeon article despite having misgivings about the way it characterized Kahneman's and some others' arguments. The second article by Jason Collins was really on-point, though. I have for many years had a serious problem with the way many popularizers of the sciences of decision making, cognitive biases, and behavioral economics frame their explanations. There's a strong bias towards "counterintuitivism" in the way these fields are presented to the public. Scientific findings that run contrary to conventional wisdom are hyped and presented as something that should fundamentally change the way you think about the world, without consideration of the actual quality of the evidence or with the fact that any single scientific study usually needs replication to be trustworthy. As the saying goes, extraordinary claims require extraordinary proof, but claims about human nature that seem extraordinary to most of us are often supported only by weak evidence at best.

Kahneman is far from the worst offender, but this example that Collins cites:
The idea you should focus on, however, is that disbelief is not an option. The results are not made up, nor are they statistical flukes. You have no choice but to accept that the major conclusions of these studies are true.
This is a thing of incredible hubris for a scientist to say. And indeed, it turns out that many of the results were in fact statistical flukes.
posted by biogeo at 9:28 AM on July 6, 2018 [13 favorites]


Can I just say that I am really digging all these recent FPPs on visual processing and the like and the informed commentary? It's an interesting topic, but I have no background and am very skeptical of pop neuroscience.

Seconded! I don't have the expertise to chime in, but I'm really enjoying all the comments. It's like getting to sit in on a particularly cool journal club in a field I don't share.
posted by sciatrix at 9:33 AM on July 6, 2018 [2 favorites]


also oh my god what is wrong with him? of course disbelief is an option, that's what science is all about!

How do you engage in science if you think you have to believe everything that any given experiment tells you? How do you construct any kind of sense from the noise?
posted by sciatrix at 9:35 AM on July 6, 2018 [7 favorites]


She still says that I should have been able to avoid the bread since she warned me.
In fairness to your brain, this is 100% normal. I remind myself about stuff like this all the time and still forget about it. And I literally got a PhD in avoiding distraction.
posted by nicodine at 9:40 AM on July 6, 2018 [6 favorites]


Agreed, it's hard to imagine that Kahneman really applies that to himself. This is a very uncharitable read, I know, but I wonder to what extent the "you" in that sentence was a true second person, rather than the generic "one". I have the sense that some of these counterintuitivists think that skepticism is something reserved for them and their peers; everyone else should just believe what they've concluded.
posted by biogeo at 9:41 AM on July 6, 2018


Agreed, it's hard to imagine that Kahneman really applies that to himself.

Counterpoint.
posted by asterix at 9:58 AM on July 6, 2018 [13 favorites]


Wow, seriously good for him. That reconsideration of one's position and self-critique of mistakes is very admirable.
posted by biogeo at 10:01 AM on July 6, 2018 [3 favorites]


mid-century psychologists: Computers are cool and process information. Humans also process information. Are human brains essentially computers?

Computer Scientists: "When we used words like 'memory' and 'computing' to describe electronic Turing machines, those were metaphors. We weren't literally saying that the machines are doing the same thing a human brain is doing when it computes or remembers things."
posted by straight at 10:12 AM on July 6, 2018 [4 favorites]


That reconsideration of one's position and self-critique of mistakes is very admirable.

I know that you're right, but I wish it weren't worthy of singling out for praise so much as, like, table stakes for adult humans.
posted by adamgreenfield at 10:18 AM on July 6, 2018 [1 favorite]


These are the wrong questions. The real question is, how many passes are there, and how do I compare in counting passes and noticing the gorilla to other people? I must know who I've beaten, get on it science!
posted by cui bono at 10:49 AM on July 6, 2018 [1 favorite]


I've seen/ read about this before, always fascinating. The think I think is kind of cool is that humans can be so good at focusing attention, they don't even notice a gorilla.

Cats don't pay much attention to prey that doesn't move. Move an object and the cat notices it.

You'd almost think there was some kind of evolution of brains that would explain this.
posted by theora55 at 11:15 AM on July 6, 2018


theora55, that one I do know, at least in part. It's not just the visual processing and attention to detail that's different between a human and a cat's visual system--it's also the eyes themselves. Cat retinas have more rod cells that are better at picking up motion, particularly compared to human retinas (which are jam-packed with cones, particularly at the center of the retina). Rod cells are better in low-light conditions, too, but they pay for it in spatial resolution and color processing (which is what cones are good for).

So a cat may not pay as much attention as a human to non-moving prey, but that's not the only reason a cat might not respond to a stationary mouse (or cat toy). Depending on how well the mouse can blend into the background, the cat might not even sense the mouse until it moves! I'm using sense here as distinct from perceive--we don't perceive or notice everything we sense.
posted by sciatrix at 11:30 AM on July 6, 2018 [2 favorites]


Theorists of analog computation: "Why doesn't anyone ever remember we exist?"
posted by biogeo at 11:56 AM on July 6, 2018


The Luck Factor

I don't know whether the research has held up or whether there's been more research, but the premise was that people who focus hard on not making mistakes miss positive opportunities.
posted by Nancy Lebovitz at 12:00 PM on July 6, 2018 [1 favorite]


It's been a long time since I thought about feline retinal anatomy, but to riff off sciatrix's comment for a sec... rod sensitivity is screamingly good. Like "single photon is enough to set off the photoreceptor cascade" good. The problem is, they're swamped in most lighting conditions, but if you're a cat, stalking a mouse in dim environments, it's really damn useful. Feline eyes are different in a bunch of cool ways from human eyes (not just the difference in pupil geometry; there are some weird pupils out there); they've also got the tapetum lucidum which is a retroreflective layer behind the photoreceptors, bouncing light back to the photoreceptors. It's also why your cat's eyes seem to glow.

I'm instantly skeptical of the cover claims of "The Luck Factor", because the idea of "attracting luck" makes no sense in the context of psychology or neuroscience. It feels more like positive psychology than anything else, although I can believe there are behavioral modifications lurking inside it that might give the perception of being luckier. It's not the kind of research I run into, but it's part and parcel of an area in psychology I give a pretty wide berth, because it's extremely hard to do rigorously, and even harder to interpret the results.
posted by Making You Bored For Science at 12:32 PM on July 6, 2018 [6 favorites]


It's also worth noting that the cat might or might not be leaning on vision as much as a human observer assumes. I have a little cat who has been blind more or less since birth, and he does a damn fine job of chasing moving objects based on, as far as I can tell, a mixture of hearing and tactile cues. I would like him to apply these skills to objects besides my moving feet under blankets, but c'est la vie. Humans also miss an awful lot by being so pitiful at chemosensory cues; we have vision that is so much better than many mammals at spatial and color resolution in particular that we tend to over-attribute observable perception in other animals to vision and audition and under-attribute chemosensation like taste and smell.

(Speaking of Incredibly Cool Shit, magnetoception in birds is mediated at least in part by the retina, too, which just blows my tiny little mind every time I try to think about it too hard.)
posted by sciatrix at 12:56 PM on July 6, 2018 [7 favorites]


The idea that magnetoception is mediated by the retina in birds just boggles me. That's amazingly cool. I'd love to know what the pathways look like; it's got to be incredibly cool... I thought the pulvinar pathways in human vision were weird enough (about 10% of retinal output dodges the LGN entirely, and makes it to visual cortex via the pulvinar, of all things).

The brain, because weird (and deeply cool) is always an option.
posted by Making You Bored For Science at 1:03 PM on July 6, 2018 [5 favorites]


Making You Bored for Science, The Luck Factor isn't about a psychic ability to make luck happen, it's about being able to notice opportunities. It doesn't seem weird to me that people's ability to notice opportunities might vary.
posted by Nancy Lebovitz at 1:06 PM on July 6, 2018


I know, right? Come to think of it, I should try a FPP on magnetoception one of these days. It's one of those fields with an awful lot of research in it that as someone not in the field I kind of go "hm" at because it's either completely revolutionary and worldchanging or else completely crackpot out to lunch, but I've never had the time to really dig into it properly. The bird stuff seems pretty solid; I'm much more skeptical of the general magnetoception research in mammals I've seen. Particularly since it seems to be cryptochrome in the UV-sensitive cones that the birds are co-opting, at least partly, I would guess that they're inputting that information along the same pathways as other visual information, but birds have weird-ass brains and I mostly do acoustic anyway and shrug emoji.

Anyway, here's a solid-looking review if you want to check it out, complete with a rough sketch of the pathways the information is traveling down in Section 8.12. (There's an alternate mechanism involving deposits of iron in the beaks and some evidence that that requires the trigeminal nerve? IDK, birds are weird. No cortex, either, since cortexes are mammalian-specific, so everything's organized a little bit strangely if you're used to humans.)
posted by sciatrix at 1:31 PM on July 6, 2018 [3 favorites]


The real question is, how many passes are there, and how do I compare in counting passes and noticing the gorilla to other people?

You lost. HTH.
posted by Greg_Ace at 1:34 PM on July 6, 2018 [1 favorite]


There are always individual differences in the ability to do any visual task (or really any task; I just know the vision literature here). So it makes sense that "opportunity detection" might vary across individuals. Deciding that it's "luck" as opposed to a trainable skill is a little weird to me, and feels like a weird jump in logic, but the idea that someone could look lucky to an observer if they knew to look elsewhere makes some sense. I think I'd be less averse to the idea if it was just "hey, if you look where other people don't, maybe you'll find something they've overlooked" rather than "The Luck Factor." There's some evidence I can think of in driving and aviation for training people to look where they wouldn't otherwise as a safety measure, so the idea, separated from luck qua luck works.

That said, I don't think this is an inattentional blindness problem. It feels more like an attention problem, or more specifically, a visual search and saccade planning problem. The problem is, if you don't know where to look to find what other people miss, you won't look there. There's work on this from a safety standpoint in driving going back to Mourant and Rockwell, 1973 (and probably earlier; they're just the classic citation for expert vs novice drivers and fixation patterns). There's also a lot of work on what's gotten called "Scene Grammar" in visual search; the idea that you look where it makes sense to look based on the content of the scene (here's a review by Draschkow & Võ, 2017 talking about their work in this area).

As a bit of an aside, Wiseman does have a recent paper in iPerception showing something somewhat like inattentional blindness in an old Python sketch, but it doesn't look like he used Simons and Chabris' approach to testing it, so I'm a little leery of calling it inattentional blindness. It doesn't sound like the subjects had a task when they watched the Python clip, so their implicit task was to just follow the scene, and the extra characters in it aren't relevant to that task, so they're ignored. That's certainly a decent demo of inattentional blindness, and worth knowing about. Another good demo is this video, although it skirts the difference between inattentional blindness and change blindness.
posted by Making You Bored For Science at 2:03 PM on July 6, 2018 [4 favorites]


*spends the rest of the day looking for bread on the stairs*
posted by It's Raining Florence Henderson at 2:47 PM on July 6, 2018 [2 favorites]


"Were you listening to me Neo? Or were you looking at the woman in the red dress?"
posted by gucci mane at 2:50 PM on July 6, 2018 [1 favorite]


*spends the rest of the day looking for bread on the stairs*

Username up for grabs...
posted by Greg_Ace at 3:00 PM on July 6, 2018 [1 favorite]


Just try to remember. There is no spoon gorilla in the red dress.
posted by It's Raining Florence Henderson at 3:01 PM on July 6, 2018


Can I just say that I have really enjoyed the comments I've read so far? Super informative. Thanks!
posted by Hutch at 3:05 PM on July 6, 2018


> skirts the difference between inattentional blindness and change blindness.

I can't seem to find it, but there was an interesting video of 'slow change blindness' that had a static landscape photo that had a feature very slowly change. This is the closest I could find to that.
posted by CheapB at 6:01 PM on July 6, 2018


I thought the pulvinar pathways in human vision were weird enough (about 10% of retinal output dodges the LGN entirely, and makes it to visual cortex via the pulvinar, of all things).

I know, right? I did a postbac with a researcher who was starting one of the first in-depth physiological studies of the pulvinar, and I asked him what it does. His response was: "I can tell you everything I know for sure about the pulvinar: it's name is Latin for pillow. That's it." (Obviously that was hyperbole but not by a whole lot.) These days I think of pulvinar as the parietal thalamus, but that's surely an oversimplification.

I should try a FPP on magnetoception one of these days.

Yes! You'd do a great job with it. If you don't do it, I might.

The neuroethology class I took as an undergrad was taught by someone who studies magnetoreception, polarized light reception, electric field reception, and other "exotic" (in an anthropocentric sense) sensory systems. It really tipped me over the edge to wanting to study animal behavior at a mechanistic level, even though I didn't end up studying those systems. Glancing briefly at that review, it looks like the main state of avian magnetoreception research hasn't shifted too dramatically since then. At the time, the two major hypotheses were that either 1) birds sense magnetic fields via a cryptochrome-mediated pathway for vision, or 2) birds sense magnetic fields via magnetite particles in their beaks. Looks from the review like both hypotheses are still viable, and both mechanisms might be in use.

For those who can't access through the paywall, the visually-mediated magnetoreception pathway in birds is totally crazy and fun so I'll elaborate a little beyond what sciatrix already said. Basically, the idea is that in the retina of birds, there are molecules called cryptochromes. Depending on whether these molecules are aligned with or at right angles to a magnetic field, their chemistry is slightly different. (Something to do with coupling between electron spin and the magnetic field, I think, which changes the energy levels of the electrons and thus the kinds of reactions they participate in.) The exact biochemistry of how this would work is not known, but the idea is that when the bird's eye is aligned with the Earth's magnetic field, there would essentially be a ring of photoreceptors around the retina which would become less efficient at responding to light of certain wavelengths. Conceivably, the birds could perceive this as a dark ring centered around the middle of their field of vision.

Evidence for the vision-mediated hypothesis includes that when there's no light, there also seems to be no magnetic field sensitivity. The use of light-mediated activity for magnetoreception in other animals, like newts (who do so using the pineal organ, aka the "third eye"), is also confirmed, so the mechanism is a plausible one. From the review article, it seems that more recently a specific visual pathway has been identified, which is awesome!

No cortex, either, since cortexes are mammalian-specific, so everything's organized a little bit strangely if you're used to humans.

This is true, but slightly misleading. Birds don't have a true cortex, but they do have a very large amount of homologous tissue called the pallium. As far as the basic architecture of the brain in terms of connectivity with other structures, the pallium looks pretty much exactly like cortex, and it also expresses homologous cortex-specific genes during development and in adulthood. I mention this only because there's a misconception on the part of lots of neuroscientists that because birds don't have cortex, they're not good models for learning about things relevant to human brains. I also have a pet theory, totally unsupported by any evidence, that the structure of the avian pallium is a weight-saving optimization for flight (nuclearly organized tissue requires fewer long-range projections and thus less white matter to support the same circuitry).
posted by biogeo at 6:06 PM on July 6, 2018 [6 favorites]


You lost. HTH.

Noooooooooooooooo!!!!!

Also, when presented with something like this, I am at this point trained to be suspicious that someone is out to trick me. How do they account for that?
posted by cui bono at 6:07 PM on July 6, 2018


I mention this only because there's a misconception on the part of lots of neuroscientists that because birds don't have cortex, they're not good models for learning about things relevant to human brains.

Oh! I didn't even consider that, so I'm glad you pointed it out. (You can see where my bias is, heh.) Bird brains are organized a little differently to human ones, which will throw someone who is extensively educated in and comfortable with human neuroanatomy, but that doesn't mean that avian studies can't teach us about people.

And yeah, give me a couple of days and I'll see what I can find on magnetoception that is popularly accessible and also interesting. It's that or electroception, which for my money is actually even cooler...
posted by sciatrix at 8:04 PM on July 6, 2018 [2 favorites]


Count me in for wanting to hear about what birds can teach us about how our brains work (my background is mostly human behavior and human neuroscience, hence the username).
posted by Making You Bored For Science at 8:30 PM on July 6, 2018 [1 favorite]


Your username's insistence that I'm bored is getting boring. (I hope that doesn't screw up the data).
posted by Greg_Ace at 8:47 PM on July 6, 2018 [1 favorite]


Not sure if I should put it here or in the implausible thread...

Anting is similar to mushroom hunting. Once you start paying attention you see that they are all around you, they were just invisible.

It is the middle of the anting season here, and so far I've managed to find a few queens of cool but common species.

Last week I was riding my bike in traffic, at about 20 kph, concentrated on cars and pedestrians. I have no fucking idea how, I spotted a queen ant on the sidewalk across 2 lanes of traffic. A 12 mm long ant.

I caught it, it's already laid eggs and is taking good care of them. I still need an expert for a final ID, but it's been narrowed down to one of 2 species of Camponotus, both extremely rare in my state.

This incident had me thinking about the gorilla video a lot. This ant is my antigorilla.
posted by Dr. Curare at 9:13 PM on July 6, 2018 [10 favorites]


One day a few years ago I'd just gotten home from work and Ms. nobeagle was telling me about her day. I don't remember the exact subject, but I think she might have been talking about her classes and future projects / assignments. In the middle of this, she said, "Oh, there's bread on the stairs so don't step on it." Perhaps she even gestured at the bread while saying this. She then continued talking to me about the same, non-bread topics. She finished talking and I stepped over the dog gate to go up the stairs and stepped right onto the bread.

If you had Tourette's, nobeagle, I wouldn't hesitate to say that telling you not to step on the bread in that way could very well have made it more probable that you would in fact step on the bread than if she'd failed to mention the bread altogether, because of the way Tourette's can sometimes make that which is explicitly forbidden compulsory instead.

That might seem irrelevant to the "Gorillas in our Midst" experiment (as the article seems to imply the experimenters themselves titled it), but I think it could actually be central to what really went on there.
Let’s start with a careful look at Simons and Chabris’s classic experiment, and see how it might suggest something different, and more positive, about human nature. In the experiment, subjects were asked to watch a short video and to count the basketball passes. The task seemed simple enough. But it was made more difficult by the fact that subjects had to count basketball passes by the team wearing white shirts, while a team wearing black shirts also passed a ball. This created a real distraction. (If you haven’t taken the test before, consider briefly taking it here before reading any further.)

The experiment came with a twist. While subjects try to count basketball passes, a person dressed in a gorilla suit walks slowly across the screen. The gorilla even stops in the middle of the screen, thumps its chest, and then walks off. The surprising fact is that some 70 per cent of subjects never see the gorilla. When they watch the clip a second time, they are dumbfounded by the fact that they missed something so obvious. The video of the surprising gorilla has been viewed millions of times on YouTube – remarkable for a scientific experiment. Different versions of the gorilla experiment, such as the ‘moonwalking bear’, have also received significant attention.

In 1999 basketball was dominated at the pro and college level by black athletes, and had been for a long time, but there was still kind of an issue about whether they was due to some kind of intrinsic superiority, or circumstances of upbringing, and at least one pro team had a reputation for preferring white players over equally available yet more talented black players.

And in the experiment, there were both players in white and players in black passing the ball, but the subjects were given the task of counting only the passes made by the white team, essentially, I think it's fair to say, affiliating them with the white team rather than the black team.

Then, a gorilla appears -- but calling black people gorillas is a vicious ethnic slur, and bad enough to cost people who used it against Michelle Obama their jobs, as I recall. And "Gorillas in the Mist" the title of a book by Jane Goodall, became notorious in 1991 during the Rodney King affair as a slur used against residents of black neighborhoods by the LA police.

So for naive subjects, suddenly seeing a Gorilla in the midst of a basketball game can be viewed as the visual equivalent (or the dream equivalent, if you will) of having an ethnic slur pop into one's head when one sees a black person, and I think a lot of them simply repressed it rather than ignored it.

It might be pretty easy to design an experiment which could distinguish between these two possibilities, though. All you might need to do is monitor heart rate, respiration, and skin resistance during viewing, and look for changes in those when the gorilla appears.
posted by jamjam at 9:27 PM on July 6, 2018 [2 favorites]


MYBFS, having been a pilot subject for numerous friends' psychophysics studies, I must say that if that's what you do, your username is indeed quite apt. I love psychophysics, but damn it is boring to be a subject for it.

Maybe if sciatrix is on magnetoreception, I'll see if I can find some fun links on bird brains. I riffed on avian vision a bit in this thread a couple months back, and there was a nice thread on avian cognition last year (sorry to cortex for possibly bringing up a sore spot there), but a collection of links on what we've learned from the study of bird brains could make for a nice FPP.
posted by biogeo at 10:00 PM on July 6, 2018 [1 favorite]


When a pickpocket meets a saint all he sees are his pockets.
posted by Obscure Reference at 5:42 AM on July 7, 2018 [1 favorite]


I've made experiments that were less boring (a recent one on hazard detection comes to mind; it was less boring than disconcerting, since pre-collision events aren't exactly common in daily driving. I cranked the prevalence through the roof in order to run the experiment in a timely fashion), but by and large, biogeo is totally right: psychophysics experiments are inherently boring. And, arguably, should be. The reason they work, methodologically, is that we're looking at a subject's responses across a range of levels (e.g., stimulus duration, physical orientation of the stimulus... something we can manipulate, and see how their responses change). Doing that within-subjects, because subjects vary dramatically, means you need a lot of trials per level. Hence, making my subjects bored for science is what I do. Anyone who wants to find out whether what I do is boring (and is local to Cambridge, MA) is welcome to MeMail me, and I'll make you bored for science!

My graduate lab measured experiments in kilotrials (kT) for this reason. We had a poster in lab that said "Comrade, Have You Done Your Kilotrial Today?" There wasn't actually a mandate for how much you did in a day, but the joke was that when you were collecting data, 1 kT/day wasn't unreasonable. More than that, and your data quality starts to dive, at least on most tasks.

jamjam, that's a element in Simons and Chabris 1999 that I haven't thought of. It's certainly worth considering, although I don't think it explains the effect, since they ran the fully balanced version (track white-shirt players passing the ball; track black-shirt players passing the ball), and find inattentional blindness in both conditions, although it's a much stronger effect when players are tracking the black-shirt players, and it's mediated by task difficulty (count all the passes, or separate out pass types while tracking). There's also the work that's followed their original paper here (the unicycling clown and money-dodging results from Hyman et.al in 2010 and 2014 respectively) which wouldn't, I think, be influenced this way. There's even a really interesting take on this in visual search: radiologists who fail to see a gorilla inserted in a chest x-ray (Drew et.al 2013).
posted by Making You Bored For Science at 1:16 PM on July 7, 2018 [3 favorites]


« Older Guo Pei Combines Fashion and Architecture   |   In Silicon Valley... Newer »


This thread has been archived and is closed to new comments