Part 5: Post 7: Lies, Statistics, and Illusion

The Lies, Statistics, and Illusions class has now met five times, so I can talk about it with some sense

Classic Illusion

that I know what I'm talking about. The semester is only in its fourth week, but the semester is ten weeks long and these classes all have twelve meetings, so they each take turns meeting twice a week.

This class isn't required--none of the classes I'm taking this semester are, but they are kind of the default options for meeting our credit requirements in specific areas. For example, this class carries three credits, two in psychology and one in what they call earth science literacy, which is kind of a catchall. We need more credits in all those areas, but we don't have to take this particular class to get them. We could take a whole series of workshops and seminars or design an independent study instead. Or an ally might teach a different but comparable elective at some point. But I actually think everyone who can is taking this course--people say it's fun.

I didn't know, at the beginning of the semester, how a class that includes statistical literacy could be fun. I'm starting to take these kinds of things on faith, so I signed up for it, I just didn't know how Allen was going to pull it off. I shouldn't have worried; Allen's classes are almost always fun, which is one reason he's the second most popular teacher here, after Kit.

The basic principle of the class is that human brains are not very good at certain kinds of tasks, and that means we make certain predictable mistakes that are exploitable by both illusionists and less honest manipulators. Or, as Allen said the first day;

"Basically, you're stupid. But I'm stupid, too, so don't worry too much about it. Even the smartest people are stupid sometimes. But the problem is that we're predictably stupid. That's how stage magic works. If I show you a card," and he made a card appear in his hand, "I can be sure that none of you saw where it came from. Not that some of you didn't see it, or that most of you didn't see it, but that every single one of you didn't see me pick up the card and get ready to show it to you. Every single one of you made exactly the same perceptual mistake at exactly the same time, and you'll keep on making it, even if you make up your minds not to. Here, I'll do it slowly for you--"

And he made the card appear and disappear for us several times, but none of us ever saw where it went or where it came from. We tried guessing--asking if it was up his sleeve and so forth, but he showed us the insides of his sleeves and both his empty hands, and everywhere else we thought he might be hiding a card. No card. Then the card reappeared.

"But what makes you think you can actually trust your senses to find the hidden card when you know you couldn't see it when I moved it?" he asked, after a few minutes of this. We were all silent.

He wouldn't tell us how he had hidden the card. He doesn't give away professional secrets except to students actually learning to be magicians. He did show us several non-professional illusions, just to convince us of the limitations of our perceptions. One I'd actually seen before.

It turns out that the retinas in our eyes are not actually photo-sensitive across their entire surface--there's a blind spot in each eye where the optic nerve passes through. To notice this blind spot draw two small crosses on an otherwise blank piece of paper a few inches apart. Hold the paper close to your face so that each eye can look straight ahead to the cross right in front of it. Then, without turning your head at all, close your right eye and look at the right cross with your left eye. The left cross vanishes.

Here, it works on a computer screen, too. Try it;




                                     x                                                                      x






I did this in school in fourth grade and I thought it was cool then. I think it's cool now. It's kind of neat to see something vanish like that. But Allen asked a question none of us thought to ask in fourth grade;

"Why don't we see the blind spots under regular conditions? Why does our field of vision look continuous when we know for sure that it isn't?"

We guessed that maybe one eye compensates for the other, since one eye can see where the other one cannot. Allen conceded that this guess what reasonable, but then asked why the blind spot doesn't show up when you shut one eye. And it doesn't; we all tried it. Even through just one eye, the visual field looks continuous unless you do the exercise with the crosses.

"I'll give you a hint," Allen offered. "Does the exercise have to be done on a blank background?"

It does, more or less. If you place a fingertip over the crosses above and try the exercise, one fingertip vanishes. If you try again with your fingertips held up in the air, so you can see the rest of the world behind them, the trick doesn't work. Your finger-tip looks kind of fuzzy and vague, but it doesn't vanish. We couldn't figure it out. Allen explained that the brain fills in the blind spots based on what it can see. Using a blank background forces the brain to fill in the blind spots with blankness, which is why we can finally notice the blind spots that way.

"We don't actually perceive what our senses pick up. We are not computers recording incoming audio, video, and olfactory feeds," Allen explained. "Instead, we perceive a model of the world constructed by our brains based on our sensory input. Usually the filters and shortcuts our brains use produce a manageable and reasonably accurate model. It doesn't matter that we have those blind spots, for example, and being able to perceive them all the time would be pointlessly distracting. Occasionally, however, the model fails. And we make a mistake to pretend it doesn't--to rely too much on our subjective certainty."

From then on, that became the rallying cry of the class--don't rely too much on subjective certainty. Allen has been using every class to trick us, one way or another, and then explain what quirk of the mind made us vulnerable to the trick. And we've been trying to avoid getting tricked and it never works. He always gets us. Sometimes the tricks are not so much about us being wrong as predictable. He asks us questions and records how many people gave which answer--and then shows us a poster he'd obviously made up ahead of time showing how many people gave each answer. And the poster is always right--the poster usually lists a range, but the real answer is always inside the range and the range is narrow enough to be embarrassing. It's like he knows what we're thinking better than we do.

At first, the tricks were mostly about perception or memory.  For example, he showed us a video of people in black shirts and white shirts passing basketballs around and had us count the number of times the people in the white shirts passed the ball. Then he asked us another question about what we remembered from the video, which half of us got wrong. Click on the link to see it yourself--I won't ruin it for you. It's not an optical illusion, it's a selective attention illusion that results in about half of all people remembering what they see wrong.

In more recent class meetings he's been shifting over to tricks based on emotional reactions--I don't mean overt emotional reactions; you don't think you're reacting emotionally at all. I mean tricks that make you come to what feels like a rational conclusion that is completely wrong and completely predictable because your subconscious mind has been hooked. It's funny, but it's also scary, to see how easily we can all be manipulated. And as Allen points out, con artists, liars, advertisers, and propaganda producers know all of these tricks and use them all the time. There's a reason why they call this class "Defense Against the Dark Arts."

And in each class he's been talking about how to respond to these tricks. Because there is no way to not be tricked. You can't turn the vulnerabilities off. But he's been teaching us to recognize where our weaknesses are, to step back and think clearly, to withhold snap judgements, or even just admit we don't understand or don't know something. But he's also been asking how we might try finding out the truth behind these illusions and mistakes. He's brainstorming with us. And this week he started explaining how to find out. This is where mathematical and statistical literacy come in.

Mathematical literacy--I'm good at math, though I'm aware that not everybody did. I did not realize that a third of my fellow students didn't know how many thousands make a billion and half of those don't know how many millions make a billion. Now, what I want to know is how does a person know how many thousands make a million and how many millions make a billion, without also knowing how many thousands make a billion? But a lot of us don't. And even I got caught in the trick, a complicated word program equivalent to making six and half a dozen seem like different amounts. Makes you wonder what the point is of news reports about the national debt, whether they mean anything to anyone. Statistical literacy is the same kind of deal--like noticing that ten percent and one in ten are exactly the same, even though they hook the subconscious mind differently. Or hearing a news report about some product that doubles the risk of cancer or something and remembering to ask what the risk was to begin with--because if risk doubles from 0.000002% to 0.000004% there's really no reason to worry about it.

But it's more than that. Because as Allen says we can be really wrong without knowing it--while actually being completely sure we're not wrong. Confirmation bias, for example, makes it easier to remember events that seem to confirm things you already think are true. So if you think your magic lucky baseball cards works you'll remember all the times things went well while you were carrying the baseball card and forget about most of the times things didn't go well, even if the card doesn't actually change your luck at all. Statistical tests are the equivalent of looking at the crosses on a blank background--the trick that reveals the illusion, the way to force the mind to see past its vulnerabilities.

Allen's daughter, Alexis

"People tend to think that science and magic are different, that scientific method undoes magic somehow," Allen said. "I happen to think magic is more powerful than that. Science as we know it was invented by philosophers and alchemists--people who were practicing magic. And what we can do today, in terms of harnessing and manipulating natural forces and the human mind, dwarfs anything that could be done in ancient times. The fact of the matter is that when magicians invented the scientific method, they got a lot better at magic."                      

[Next Post: Monday August 26 Environmentalism for Dummies]