Professor Paul Bloom: Just to review, here's where we left off. The discussion from last lecture and for about half of this lecture is going to be social psychology. And so, we started off by talking about certain fundamental biases in how we see ourselves. We then turned to talk about a bias and how we see other people, the fundamental attribution error. And now we're talking a little bit about some aspects of how we see other people. So, we quickly talked about certain aspects of why we like other people including proximity, similarity, and attractiveness, and where we left off was a discussion of the Matthew effect, which is basically that good things tend to compound. If you're rich you'll get a better education, if you're smart people will like you more, if you're attractive and so on. Nobody bring up their papers at this point. They'll collect them at the end of class. What I want to talk to-- [laughter] Okay, except for you. Just hand me it now. [laughter] I'm going to ask the teaching fellows to stop anybody from approaching that area.心理学空间}6ny-H\2uQe
0UFVB8I0I want to begin by talking about [laughter] impression formation, how we form impressions of others, and tell you a couple of interesting things about impression formation. The first one is, first impressions matter a lot. They matter a lot for different reasons. They might matter a lot because humans have, in general, a confirmation bias such that once you believe something other information is then encoded along the likes to support what you believe. So, the classic study here was done by Kelley where a guest speaker comes in and some of the students received a bio describing the speaker as very warm, the other as--do not bring your paper up if you're coming in late. Just--at the end of class, yeah. [laughter] Others got a bio saying--thanks, Erik--the speaker was rather cold and then it turned out later on [laughter] when they're asked for their impressions of the speaker people are very much biased by what they first assumed. If I'm described to you as a vivacious and creative person and you see me and I'm all kind of bouncing around and everything, you could then confirm this as, "Look how vivacious and creative he is." If I'm described as somebody who drinks too much, you might think he's an alcoholic. If he's described as somebody who's insecure and nervous, you could interpret my activity as nervous twitches. Your first impression sets a framework from which you interpret everything else.
pC#DAIr_!V0This was the theme of an excellent movie calledBeing There starring Peter Sellers. And the running joke of the movie "Being There" was that the main character, the character Chauncey Gardner, somehow through accident had the reputation for being a genius but while, in reality, he was actually mildly retarded. But he would go around and people would ask him his opinions on politics and he would say things like "Well, I like being in the garden." And because of his reputation as a genius people said, "Wow. That's very profound. I wonder what he means." And--or people would talk to him and he'd just stare at them and say--and people would say--would be intimidated by his bold and impetuous stare when actually he just totally didn't know anything. So, first impressions can shape subsequent impressions not just when dealing with people.
d B+sr1?B0A little while ago there was a sniper, actually a pair of snipers killing people in Washington and the one thing everybody knew about it was there was a white van involved. It turned out there was no white van at all but in the first incident somebody saw a white van, this was reported in all the newspapers, then every other incident people started seeing the white van. So, they started looking for them and they started to attending--attend to them. So, first impressions matter hugely when dealing with people because it sets the stage for how we interpret everything else.
A second finding building on the first is that we form impressions very fast, very quickly, and this is a literature known as "thin slices." The idea is you don't have to see much of a person to get an impression of what they are. The first studies done on this were actually done on teachers, on university professors. So, university professors have teaching evaluations and you could use this as a rough and ready approximation of what students think of them. So, what you do then is--the question that these people were interested in, Rosenthal and Ambady, two social psychologists, were how long do you have to look at a professor to guess how popular a teacher he is? So, they showed these clips for a full class. Do you have to see them for a full class? Do you have to see them for two classes? Do you have to see them for a half hour? How long do you have to be around a person to see him, to estimate how good a lecturer that person is? And the answer is five seconds. So, after clips of five seconds people are pretty good at predicting what sort of evaluations that person will have.
w"I(_d(E$k1l0Remember "The Big Five," how we evaluate people on "The Big Five?" Well, you have a roommate and your roommate you could evaluate on "The Big Five." You've had a lot of experience with him or her. How much time do you need to evaluate somebody on the five dimensions of personality? The answer is, again, not much time at all. After very brief exposures to people, people are very accurate at identifying them on "The Big Five." One of the more surprising findings is--concerns sexual orientation or "gaydar." That's not a scientific term [laughter] but the same psychologists were interested in studying how quickly you can--if at all how long does it take to figure out somebody's sexual orientation?
LD yo]u||O0Now, what they did was--they were clever psychologists so they set it up in a study where the people did not know sexual orientation was at issue. So, for instance, they may be people like you who filled in a form, one question along a very long form was your sexual orientation, and then you're sitting down being interviewed by somebody and your interview is being filmed, and then other people are shown--who don't know you are shown the film. And the finding is that people based on thin slices are quite good at detecting sexual orientation. Everybody's good at it, gay people are better at it than straight people, and, again, you don't need much time. You just need about a second. You see somebody for about a second, you could make a guess. You're far from always right. In fact, you're just a bit better than chance but you are better than chance at telling sexual orientation. So, these two facts taken together, thin slices and the power of first impressions, means that just by a brief exposure to somebody it shapes so much of how you're going to think about them in the future.心理学空间M,?\'w,]B5c
Now, we can look at this from the other direction. We're talking about the perceptions of other people, how we perceive other people, but social psychologists are also interested in the question of what happens to other people as a result of being perceived in a certain way. So, one question is, "What would cause me to perceive somebody as intelligent or stupid, gay or straight, anxious or level-headed?" A second question is, "What are the effects of being judged that way?" And psychologists have coined a term, talk about self-fulfilling prophesies, and the claim here more specifically is what's known as "the Pygmalion effect." And the Pygmalion effect is if I believe you have a certain characteristic this might cause you to behave as if you have that characteristic.
The name comes from the play by George Bernard ShawPygmalion, and the quote here is "The difference between a lady and a flower girl is not how she behaves but how she's treated. I shall always be a flower girl to Professor Higgins because he always treats me as a flower girl and always will," made into a better known movie,My Fair Lady. But I think that the same theme is better exemplified in a far better movie,La Femme Nikita, where a cold-blooded killer is treated with respect and affection and then she becomes a much more warm and accessible person, and then she kills a lot of people but that--[laughter] but still it illustrates the point.心理学空间tN:u)GO'`
And this point has tons of empirical validity. The classic experiment was by Rosenthal and Jackson where they told teachers that some of their kids were really smart and other kids were less--were not really smart, they weren't expected to show a huge jump or spurt in their IQ, and this was of course trickery. The children were chosen at random but the children who were described as showing--as expected to show a jump in IQ, in fact, did show a jump in their IQ scores and this isn't magic. It's basically--if I am told that you're a genius and your genius is about to be in full-flower throughout this class and it's a small class as these classes were, I'll focus more on you, I'll give you more of my attention. If I'm told "not so much for you," you'll suffer relative to him. And so the Pygmalion effect shows how our expectations can really matter.心理学空间;Z5I.sJ&Gmx
&Sq7e~N#r0This brings us to the final--the issue of expectations and how we judge people is a story that could be told about individuals but it's also a story that could be told about groups. And this is where I want to end this section on social psychology by talking about groups. A lot of social psychology is concerned with the question of how we think about human groups and we've already discussed this in the lecture on morality when we talked about the human dynamic pushing us to think in terms of "us" versus "them" as shown in the Robber's Cave study and also shown in the minimal group research by Tajfel showing that from a motivational, emotional standpoint it's not difficult for us to think in terms of "my group" versus "your group." And this way of thinking has real consequences for our emotional life, our affective life, and how we choose to distribute resources. What I want to talk about here though is a different aspect of how we think about human groups. I want to talk a little bit about stereotypes.
Now, "stereotypes" in English often just is a bad word. To have a stereotype is to be--is to have something wrong with you. You might say it's not good to have stereotypes. Psychologists tend to use the term in a broader sense. We tend to use the term to refer to information we have about categories and intuitions we have about the typicality, our frequency of certain features of categories. And it turns out that collecting information about categories is essential to our survival. We see novel things all the time and if we were not capable of learning and making guesses, educated guesses, about these novel things we would not be able to survive. So, when you see this object over here you categorize it as a chair and you recognize that you could probably sit on it. This apple is probably edible, this dog probably barks and has a tail and eat me--eats me and doesn't speak English. These are all stereotypes about chairs and about apples and about dogs. It doesn't mean they're logically true. This could be a vegetarian dog, a poison apple, an explosive chair, but [laughter] they're typically true. And if you were suddenly stripped of your ability to make generalizations, you'd be at a loss. You wouldn't know what to eat, how to interact. So, some sort of ability to record information and make generalizations is absolutely essential to making it through life.
What's interesting though is we also categorize types of people. So, we have stereotypes in our heads about men and women, about children, adolescents or adults, whites, blacks, Asians and so on. Now, this is not essentially a bad thing for a couple of reasons. First, some of these stereotypes are positive. You might have positive stereotypes about certain groups. You might believe some groups are unusually creative or intelligent. You might have a particularly positive stereotype about your own group even if your own group is Yale students or your own group is people from France or your own group is people from such and so college. You might have positive stereotypes. More importantly, we collect stereotypes about groups of people through much the same way we collect stereotypes about categories like chairs and apples and dogs. And so they're pretty often accurate.心理学空间l)Gr `C(`
When there are studies which ask people who is more likely to be a lawyer, someone who's Jewish or someone who is Hispanic, who is likely to be taller, somebody from Japan or somebody from Sweden, people can answer these things. They have their stereotypes that guide their answers, and the answers are not arbitrary or random. Their answers are often correct and often possessing stereotypes lets us make reasonable and correct generalizations about the world.
:o0\'z.bA7K"e0That's the sort of good news about stereotypes but there's also bad news. One problem is that they're not always accurate and there's a couple of factors that could lead them away from accuracy. One is what we talked about before regarding first impressions, which is a confirmation bias. If you believe that homosexuals are effeminate, that gay men are effeminate, then this is going to shape how you see future gay men. If you see an effeminate gay man, you'll probably say, "Ah, more evidence for my theory." If you see a man who is not effeminate, you might ignore it or say maybe he's not really gay after all. If you believe black men are criminals, then when you see a black man who is a criminal you'll chalk it down as support but you'll pay less attention to evidence that white men are criminals and some black men are not criminals. You won't look at this as a scientist objectively scanning data. Rather, you'll be biased in certain ways. You'll be biased to put extra weight on the cases that support your theory and diminish cases that refute it.心理学空间1\*SN3R&k9G"` j)b
[7JCi)V0Furthermore, our data is not always reliable. So--oh, and this is actually an example of this at work. It turns out in the world of classical music there's a stereotype of women being simply less proficient than men: they play smaller than men, they don't have the same force and they have smaller techniques, they're more temperamental and so on. If you asked somebody who was a judge, the judge would say, "Look. This is just the way things are. I'm not being biased at all." The test of this then is to have blind auditions where people do their auditions behind a screen so you can't tell whether they're man or a woman, or for that matter, white or black or Asian or whatever. It turns out when you do that women get hired far more suggesting that the stereotype is A, incorrect and B, has a real negative and unfair effect on people getting hired.心理学空间"~nlx#L[Y-^dy4O
2j@s&^!~:n |y0A second problem is – what I was talking about immediately before this – is some of our data are misleading so we get a lot of the information about the world from the media. The media would include television and movies but would also include plays and books and stories. And to the extent these portray an unrealistic or unfair or biased perception of the world we could construct stereotypes that are faithful to the data we're getting but the data is not representative. And so people, for instance, object to the fact that when there's Italian Americans on TV they're often members of the Sopranos, a mobster family. Throughout history Jews have been upset at the portrayal of Shylock in "Merchant of Venice," not a very nice guy. And often in response people who want to foster more positive views will often try to--will often put in representatives from other groups in unusual ways to make that point. Anybody here ever see the television showBattlestar Galactica? Okay. Who's that? He's the star of "Battlestar Galactica." You don't know because you're too young. In the original "Battlestar"--[laughter] and I hate you. [laughter] In the original "Battlestar Galactica," this was the star. This was the main character known as "Starbuck," who got transformed into a woman in the more recent one, a sort of example of how portrayals are shifting in interesting ways.心理学空间J%c7m8^)H'r T
^VL#Q Q7q0There's also, of course, moral problems over stereotypes. So, it's fine to judge chairs and apples and dogs based on the stereotypes. It's even fine to judge breeds of dogs. If I told you that I decided to buy a greyhound instead of a pit bull because I wanted a dog of a gentle temperament, nobody would scream that I'm a dog racist [laughter] involving--and--but honestly, it's a stereotype. Greyhounds are supposed to be more passive and gentle than pit bulls. I think it's a true stereotype but it's a stereotype nonetheless. But we have no problems when it comes to things like breeds of dogs with stereotypes. We have serious problems judging people this way. So, for instance, it's a moral principle that some of us would hold to that even if stereotypes are correct it is still immoral to apply them in day to day life. The term for this would be "profiling."心理学空间NyXG%@ZG
Q.r@8X"H@7VE(@0Now, it gets complicated because there are some cases where we do allow stereotypes to play a role. When you all go and get driver's licenses or when you did get driver's licenses you have to pay higher auto insurance premiums than I do. I think this is perfectly fair because young people like you get into a lot more accidents with your reefer and your alcohol [laughter] and so it is--now, some of you are saying "that's a stereotype." And it is a stereotype but it's a statistically robust one and nobody lines up to protest this. It's an acceptable stereotype to make a generalization from. On the other hand, what if insurance companies determined that people from Asia got into more accidents than people from Europe? Would people be equally comfortable charging people from Asia higher rates of insurance? Almost certainly not. So, the issues are complicated as to what sort of generalizations we're--are reasonable to make and what aren't.
There's also a second problem. Stereotypes have all sorts of effects. Now, some of them are obvious effects. If people--for instance, if people pull you over while you're driving because you're black, this could have a huge effect on how you feel welcome in this society on race relations and so on. But some of the effects are more subtle and more interesting and you might not expect this. And this is some work done by the psychologist Claude Steele and his colleagues at Stanford. And the issue is called "stereotype threat." Imagine you have a math test and this is the front of the math test. Claude Steele made an interesting discovery. Here is how to make black people do worse on this math test. It's very simple. The finding is that if your race or your group has a negative stereotype associated with it in any particular domain, being reminded of it serves as a stereotype threat and hence damages your performance in all sorts of domains. If the stereotype is "your group doesn't do good in this," if I remind you that you're a member of that group immediately before doing it, your performance will drop. Now, you know how to make women do worse on math tests too, like that, and this has a demonstrative effect.
/P{*K0ZI r)C0So, stereotypes are complicated and morally fraught things. When people study stereotypes they often make certain distinctions between three levels of stereotypes and this is nicely summarized here. There's "public." If I asked you: One of the people running for the Democratic nominee for president is black, another one is female--if I asked people to raise their hands for who thinks that because being black or because being female they should be automatically disqualified for being president, few of you would raise your hands. Those are your public presentations of stereotypes. Even if I was to ask you on a sheet of paper, you might deny it because you might be afraid that it's not anonymous. Then there is private. Private is what you really think but you don't tell people. Some of you--some of the population of the United States are not going to vote for somebody because he or she is black but they won't tell you but they know it to be true. That's common sense. What's more interesting is below even that there may be unconscious associations that work that people don't actually know about but affects their thoughts about race, gender and other social groups.
@d%c!{h,owp0So, here are some data about what people publicly say. This is the proportion of people who say they will vote--they would vote for an African American president. What's interesting is when--around when I was born the answer in the United States was about half. Now, it is as close to one hundred percent as you can get it. Here is another one. This is also public stereotypes of blacks, proportion of white respondents endorsing each trait. Now, it's infinitesimal. These numbers are so low they could be dismissed as people filling in the wrong things or making jokes or just being confused compared to [laughter] stunningly superstitious high rankings. And so, there's been a profound change in public presentations, public views, on race but what about implicit views? This gets more complicated.心理学空间4K+^Md8]
6~+^W[7wxc!\N@0Here is a simple study. This is the sort of study that you might do here at Yale. What you might do, for instance, is be sitting at a computer screen and you'll be given incomplete words to fill out like "hos-" and you have to fill out this word. What you don't know is that pictures of black faces or pictures of white faces are being flashed on the screen but they're being flashed on the screen subliminally so fast you don't even know you're seeing them. Still this has an effect. When you see black faces subjects are more likely to fill this with words like "hostile" while whites more likely to fill it with words like "hospital."
[;MbwF I0I will now welcome you to participate in an experiment on implicit attitudes. This was developed by Mahzarin Banaji who used to be at Yale and now is in an inferior university in Boston [laughter] and it's called implicit attitudes test and it's the biggest psychology experiment ever done in terms of people. I don't know. A million people have participated in this and you could just go online, implicit.harvard.edu and then you could do it yourself. But we'll do it now as a group. If you did it in the lab or on your computer screen, you would do it by pushing buttons. We'll do it by speaking. And it's very simple. You're going to see things over here and they're either going to be words or they're going to be pictures. If it's an African American or a bad word, a negative word, I want you to shout out "right," this side, "right." If it's a white American or a good word, I want you to shout out "left." People ready. Try to do it as fast as possible without making any mistakes. [audience response] [laughter] Because of the very loud wrong person we're going to try that again. [laughter] Are you ready? [audience response]心理学空间0I:j O9{5v5X`$\-H
w%A(I*vd[.` Y0_0Good. That is "congruent," congruent according to a theory that says that people, both African Americans and white Americans, have biases to favor white Americans over African Americans. How do we know by this? Well, we compare it. That's "congruent." Now, it's different. If it's a white American or a bad word, say "right." If it's an African American or a good word, say "left." [audience response] Okay. For all I could tell, people did equally well but this experiment has been done tens of thousands of times and you could do it yourself on a computer screen. And this is one way of doing it but they'll alternate and they'll give you different ones to shift around and everything. And it turns out that this version people are slower at than the other version suggesting that their associations run one way and not the other. And this work has been extended for all sorts of ways looking at for example at gender, looking at the connection between women and English and men and math, looking at age, attitudes towards people who are obese versus people who are thin, attitudes towards people who are straight versus people who are gay, and you could go online and do these studies and it'll give you some feeling for the sort of implicit attitudes that we have within us.心理学空间?0mJ`!BR7j
3L'? L V e J1k0Well, a legitimate question is "Who cares?" I mean, if you do the--If you look at the results for the study, it turns out that there is an association as bias to view white Americans as positive and African Americans as negative but it shows up in half a second difference. Who cares? Well, there's two answers to this. One answer is there are times in your life where half a second can matter a lot. So, studies with police officers using reaction time in split-second choices on who to shoot find that your stereotypical attitudes play a huge role in who you're likely to shoot when they're holding an object in their hand that's unclear. Also, more generally, it could be that these implicit attitudes play a role in judgment calls. In cases where you have a hard decision to make, you know you're not racist. You have no explicit racist attitudes, honestly you really don't, but the argument is that these stereotypes can affect your behavior in all sorts of subtle ways.心理学空间6V8m2^ r?6@&sW
!d3pt6s,d$vC1^%c0Here is one example. What they do is they do an experiment where somebody is in trouble. You hear a scream from outside either from a black person or a white person. In one condition you're the only person around. In another condition there's other people around you. Now, we know from the work in the Bystander effect that in general which one are we more likely to help in, when we're the only person or multiple? [audience response] "Only," exactly. And in fact, when it's the only person just about everybody helps regardless of the color of the person in trouble but when you're with other people there's a big difference. Now, again this isn't--these things are not done with members of the KKK. They're done with the standard university undergraduates like you and these--and if you were in this group you wouldn't say, "Oh. I didn't want to help because the person's black." Rather, what you would say is, "Well, I didn't think it was worth helping. There were other people around. Someone else would help," but we know by looking at it that this difference makes a real difference.心理学空间 P4P/p#d M{E v(g @
A final study, and this was done by my colleague who just got hired here, Jack Dovidio, who's done some wonderful work, looked at how people judge to hire somebody based on their recommendations. This is a little bit of a confusing thing. The green bars are the African Americans. The blue bars are the white Americans. And in some cases these people have strong recommendations. When they have strong recommendations in 1989 you're willing to hire everybody, and this [the level of hiring for African versus white Americans, depicted on a slide] is not a difference at all. But when their recommendations are so-so, when it's a judgment call, the subjects are significantly more likely to hire the white American than the African American. Nineteen eighty-nine was a long time ago but the same results showed up in 1999. The same results also showed up about a year and a half ago. And again, this doesn't show that people are explicit terrible racists. It does show that people possess these stereotypes that make a difference in their real-world behavior.
0f gDy0nD)B0A way to put this all together is in Trish Devine's automaticity theory, which goes something like this. The idea is that everybody holds stereotypes. These are automatically activated when we come into contact with individuals. In order to not act in a stereotyped fashion, we have to consciously push them down, we have to consciously override them, and that's possible, but it takes work. It takes work both at the individual level and it takes work at the group level. Any questions or thoughts about this? Yes.
{ k7ac?/A2D2`0Student:[inaudible]
5_ zaOI c@n0Professor Paul Bloom: It's a good question. These results are surprising and disturbing and I said there is work to be done at an individual and group level and this young man challenged me to be more explicit about that. Here is one case. We have job searches and sometimes job--senior job searches involve the faculty sitting in--around in a room and tossing out names and we use these names as a basis for further discussion. "Hey, I wonder--What about that person? That person does great work." What we do now in the psychology department is we make a special point of trying to get names from disadvantaged groups, not with an eye towards an affirmative action policy but rather because there's a lot of evidence suggesting that people who are just as qualified don't come to mind unless you make some sort of procedure to do it.心理学空间WSL;M GCw
Here's another example. A lot of journals do blind reviewing now because of the evidence I talked about before regarding sexual stereotypes, that whether it has a male name or a female name makes a difference. So those are not--those are group level in that they're not saying, "You get rid of your prejudices by trying harder." It's rather, "Let's set up a system so that your prejudices can't work," the blind auditions, for instance, being a beautiful example of that. At an individual level, your question's a harder one and I'm not exactly sure what we could do but I think what we should do is be conscious of these things and know that it's not enough to say, "Well, I'm explicitly not racist so I'll give everybody a fair shake," but to recognize we might have biases and to work hard to overcome them, not by overcorrecting in some random way but trying to--if you're interested for instance in qualifications setting up some system that these qualifications can be observed absent knowledge of race or sex. Yeah, in back.
Student:[inaudible]心理学空间y/ZX5a Z
Professor Paul Bloom:Yes. The question's a good one. How much of this is due to stereotypes and how much of this is due to an own-group bias? So, the fact that--so, the experiments as I have described them are to some extent ambiguous. The fact that white Americans favor white Americans over black Americans might be because of stereotypes but they also might be of in-group favoritism. I think the answer is that both play a role but some of the effects are due to stereotypes above and beyond in-group favoritism. And one reason why we know that is in studies like the IAT, the Implicit Attitudes Test. African Americans show much the same effect as white Americans. So, African Americans also are biased against African Americans and in favor of white Americans, showing it doesn't reduce to group favoritism though that probably plays a big role.心理学空间!jK\;y)x#s
!X'^^O(Z$W cx0Okay. I'm going to shift and spend the rest of this class on a couple of mysteries. Here's a summary. The first one is a minor mystery. I think we have some progress in explaining it. The second one is a total mystery. The first one's sleep. Sleep is a motivation. It is a motivation like food or drink. It is a form of torture. I won't get into definitions of what torture is but it would cause somebody tremendous pain and anguish to keep them from sleeping. When you're really tired sleep is what you want to do like when you're really hungry you want to eat. How many people here on average--from the beginning of the semester until now get on average more than eight hours of sleep a night? That's good. Good.
There is a sort of school of sleep macho. A sleep macho used to be "I only get forty-two minutes of sleep a night." Now sleep macho is "I sleep eleven hours." [laughter] Who gets on average more than seven? On average more than six? Who here has been making it since the beginning of the semester on under six hours a night? Okay. Anybody of you that has been getting under five hours a night? Okay. There is big individual differences in how much sleep people need and sleep itself is what we spend a lot of our life on but it's very hard to study and we didn't used to know much of it because you can't ask people what's happening during it because they're sleeping so you need sort of clever methods. One such clever method is an EEG. You bring somebody in the sleep lab, you put electrodes on their scalp and you see what these--what sort of electrical activities you get in the brain. Right now many of you are showing irregular beta waves suggesting intense comprehension and great intellectual focus. [laughter] Some of you are awake but non attentive and your brain's giving you these large, regular alpha waves. Some of you are sound asleep, deep in delta. [laughter]心理学空间o:a)N p!V6xA"mQ|
bi4~7IA$BO(\0When you sleep you get the following stages. You start off with a transition period when you're falling asleep. We call that stage I. Then you get successively deeper, II through IV, slow, irregular, high amplitude delta waves, and then once you reach stage IV you start going up again, up through stage III and II. Then REM sleep emerges, rapid eye movement sleep. REM sleep is neat because your brain looks like it's wide awake but – and I'm going to talk a little bit more in detail about this later – you're relaxed, your rapid eye movements occur; that's where your--the name comes from and dreams occur and then on a good night's sleep you've got four to five sleep cycles and it looks like this. You start off and you go down, down, down, and then you come up, then you get your first REM cycle, again, again, again, again, again, and then you wake up.