Perception, inference and change blindness

So I’ve been suggesting that we have a systematic illusion about the world around us in our own minds, that we think we know a lot more than we do, and that as soon as we’re asked a question, we can cook up an answer to it. And that gives us the impression that there’s a vast store of knowledge and answers to all sorts of questions just lying there, waiting to be retrieved. But in fact, we’re making up the answers quite literally as we go along. Let’s quickly look at one interesting case of perception. So it turns out that, as you probably know, you’re very good at seeing things which are right in the center of your vision.

So for example, if you take an object about the size of your thumb, at an arm’s length, you can see that really very clearly. But it turns out that your ability to see decays very, very rapidly as you move away. So in fact, an interesting thing you might try is extend the tip of your thumb to an arm’s length, and just look at the thumbnail. You can see that thumbnail pretty clearly. Now think how much you can see of stuff around it. Keep looking at the thumb, though. Keep focusing on the thumbnail and trying to establish what you can make out in the surrounding scene. And the answer is not very much at all.

It turns out that if other objects change colour, for example, you won’t notice. It turns out that if a familiar face were visible in the periphery of your vision, relatively near to the thumb, in fact, you wouldn’t recognize it as a familiar face. It turns out that you can’t read a few degrees away from the thing you’re focusing on. In fact, interestingly, if you’re reading a book, you have a sense that there’s a whole page of text in front of you, a whole high definition page of which you’re just paying attention, especially to the words of interest.

But interestingly enough, very clever experiments have been done where you can take the text that you’re not looking at and just turn it into a string of X’s with, admittedly, white space in between so that it has a vaguely word-like look. And the trick is that as you flick your eyes to look at a particular point on the page, suddenly, at that very moment, the relevant letters are put there. And you can read quite happily like this, without noting anything amiss. So you have a sense that there’s this rich page of text, but there is no rich page of text. There’s just a little teeny window where you’re looking. And the same is true in vision quite generally.

You probably know that the colour vision cells are pretty much all in that little central area. Just switch focus on the area you’re looking at. So when I’m looking at my thumb now, I’m getting a very good sense of the colour of my thumbnail. And as you move away from the point I’m looking at, my ability to see colour diminishes dramatically. So almost everything, you’re not directly looking at is essentially black and white. You have no idea what colour it is. But that’s not the way we imagine the world. We think the world is high definition. It’s fully colourful. And we can see all of it equally clearly. This is complete nonsense.

On the other hand though, of course, if I ask you, oh and what colour is your hair or what’s the colour of a particular garment? Of course you can easily answer that question. You just flick your eye to look. Or what’s the word at the top of the page? Oh, I can answer that. But of course, I’m making up the answer at the time you ask me the question. If you ask me about the top of the page, I’ll tell you about that. If you ask me about the bottom of the page, I’ll tell you about that. If you ask me about the colour of some clothing, I’ll tell you about that.

If you ask me about the number of buttons, I’ll tell you about that. I don’t know the answers before you ask me the question. It turns out that this phenomenon that we’ve been talking about in general is extremely easy to see and extremely powerful in perception. Because perception is, in fact, incredibly limited. But because we can flick our perceptual attention around extremely fast, we just don’t notice. Surely it’s obvious if something is making you angry or sad or making you laugh or cry. Interestingly, when you observe other people, it can be quite difficult to know which emotion they’re experiencing. Famously and notably, laughter and crying often do seem quite difficult to distinguish.

So you can be in a situation where you’re laughing so much, you are kind of crying. And in fact, it can be quite painful. Now it turns out though that our own emotions are a little bit like the emotions of others in the sense that we have to figure out what they are too. So just as when I’m looking at someone else and I think are you happy or sad? Are you laughing or crying? I have to observe my own body, as it were, from the inside and figure out what am I feeling? This is a really strange idea.

The traditional, sort of common sense view would be that I go through the world, things happen to me, I have emotions about them, and the emotions lead to bodily signs. So if the emotion is a sad emotion, I might cry. If the emotion is amusement, I might laugh. If the emotion is an emotion of anger, that might cause me to look and act in a cross way. William James, an early philosopher and psychologist, suggested that just as we have to infer the emotions of other people, we have to infer our own emotions. So if you ask me what my emotions are, I have to figure this out for myself, just as I do for other people.

The standard, common-sensical view is that inside my head there are just emotions going on. There’s mental depth. I can’t see into that depth necessarily. I might have to think a while to think what my emotions are and why they’re there, but they’re inside me. So let’s consider a famous experiment from about 50 years ago which illustrates William James’s point– an experiment by Schacter and Singer. They had the idea that one of the physiological components that we use to infer our emotional state is a degree of arousal. And that’s something you can moderate quite a lot, just by affecting the amount of essentially adrenaline in your system.

If you have lots of adrenaline, you’re feeling sort of ready to go and ready to react. And if you have very little, you’re drowsy. Now their experiment involved giving some people shots of adrenaline and not others and in some cases, telling people that the adrenaline would have an adrenalising effect– would make them feel aroused and in other cases not. Now the way they did the experiment was rather clever. So you had your shot of adrenaline or not. And then you went to sit in a waiting room where there was another participant, as you thought– in reality, of course, a stooge of the experimenters. The stooge of the experimenters would then behave in a particular way.

It might, for example, be an irritating way. Or it might be a lightly amusing way. And it turned out that the way you interpret the stooge of the experimenters behaviour depends on your physiological state at the time. So for example, if you’ve been given a shot of adrenaline, but you don’t know it’s adrenaline, so in fact, you’re feeding aroused and jittery, but you don’t really know why, that it turns out that when you encounter a person who is slightly irritating, you find them really irritating. Why? Well, they’re being slightly irritating, and you’re feeling very strong physiological signals of arousal. What could this mean? Your brain’s thinking aha!

That arousal must be because I’m finding this person very threatening, annoying, generally difficult. So you assume that the arousal is being caused by the social behaviour of the other person. So you interpret them as being extremely irritating. And in fact, you may even be rather across and combative with them. Conversely, if that person is being mildly amusing, you may find him very funny. Not because they’re actually funny, or especially funny, but because you are feeling a strong sense of physiological arousal and your brain attributes that to the person being mildly amusing. So what’s happening is in the very same adrenaline is having opposing– opposite effects on your interpretation of your emotion. It’s generally giving a sense of heightened-ness.

Whatever you’re feeling, you feel it more. Now these effects, though– and this is really crucial– are very strongly reduced, if you tell the people who are getting the shots of adrenaline that the adrenaline is going to have this particular effect. So if I know I’m going to feel jittery, then when I sit in the waiting room and someone’s mildly irritating or mildly amusing, I think I’m feeling a bit jittery. Oh, but they told me that’s because I’ve just had the shot of adrenaline. No surprises there. And I do not, or do not to the same extent, attribute my feeling, my subjective physiological feeling, to the behaviour of the other person.

So the general point here is that if you want to figure out the emotions of another person, you have to make an inference from the way they’re behaving. If you want to figure out your own emotional state, you also have to make an inference from your behaviour. One aspect of that, of course, is your physiological level of activity, which you have, of course, access to and you don’t have the access to other people’s. But the same basic principle is applying. It’s not that the emotion is there inside you, just waiting to be reported on. Your brain is having to figure out the degree of emotional reaction.

And it’s figuring it out partly based on what it knows about your adrenaline levels. And if you trick someone and push their adrenaline up artificially, then they just feel they’re having stronger emotional reactions. If you tell them that why those jittery feelings are there, that explains everything, and the effect goes away. The general point is your emotions aren’t just given to you. You have to figure them out. And psychologists, as ever, can trick you into figuring them out wrong.