From TED: Using exam­ples from vaca­tions to colono­scopies, Nobel laureate and founder of behav­ioral economics Daniel Kahneman reveals how our “expe­ri­encing selves” and our “remem­bering selves” perceive happi­ness differ­ently. This new insight has profound impli­ca­tions for economics, public policy — and our own self-aware­ness.


Every­body talks about happi­ness these days. I had some­body count the number of books with “happi­ness” in the title published in the last five years and they gave up after about 40, and there were many more. There is a huge wave of interest in happi­ness, among researchers. There is a lot of happi­ness coaching. Every­body would like to make people happier. But in spite of all this flood of work, there are several cogni­tive traps that sort of make it almost impos­sible to think straight about happi­ness.

And my talk today will be mostly about these cogni­tive traps. This applies to laypeople thinking about their own happi­ness, and it applies to scholars thinking about happi­ness, because it turns out we’re just as messed up as anybody else is. The first of these traps is a reluc­tance to admit complexity. It turns out that the word happi­ness is just not a useful word anymore because we apply it to too many different things. I think there is one partic­ular meaning to which we might restrict it but, by and large, this is some­thing that we’ll have to give up and we’ll have to adopt the more compli­cated view of what well-being is. The second trap is a confu­sion between expe­ri­ence and memory: basi­cally it’s between being happy in your life and being happy about your life or happy with your life. And those are two very different concepts, and they’re both lumped in the notion of happi­ness. And the third is the focusing illu­sion, and it’s the unfor­tu­nate fact that we can’t think about any circum­stance that affects well-being without distorting its impor­tance. I mean, this is a real cogni­tive trap. There’s just no way of getting it right.

Now, I’d like to start with an example of some­body who had a ques­tion and answer session after one of my lectures reported a story.

[unclear …] He said he’d been listening to the symphony and it was absolutely glorious music and at the very end of the recording, there was a dreadful screeching sound. And then he added, really quite emotion­ally, it ruined the whole expe­ri­ence. But it hadn’t. What it had ruined were the memo­ries of the expe­ri­ence. He had had the expe­ri­ence. He had had 20 minutes of glorious music. They counted for nothing because he was left with a memory; the memory was ruined, and the memory was all that he had gotten to keep.

What this is telling us, really, is that we might be thinking of ourselves and of other people in terms of two selves. There is an expe­ri­encing self, who lives in the present and knows the present, is capable of re-living the past, but basi­cally it has only the present. It’s the expe­ri­encing self that the doctor approaches — you know, when the doctor asks, “Does it hurt now when I touch you here?” And then there is a remem­bering self, and the remem­bering self is the one that keeps score, and main­tains the story of our life, and it’s the one that the doctor approaches in asking the ques­tion, “How have you been feeling lately?” or “How was your trip to Albania?” or some­thing like that. Those are two very different enti­ties, the expe­ri­encing self and the remem­bering self and getting confused between them is part of the mess of the notion of happi­ness.

Now, the remem­bering self is a story­teller. And that really starts with a basic response of our memo­ries — it starts imme­di­ately. We don’t only tell stories when we set out to tell stories. Our memory tells us stories, that is, what we get to keep from our expe­ri­ences is a story. And let me begin with one example. This is an old study. Those are actual patients under­going a painful proce­dure. I won’t go into detail. It’s no longer painful these days, but it was painful when this study was run in the 1990s. They were asked to report on their pain every 60 seconds. And here are two patients. Those are their record­ings. And you are asked, “Who of them suffered more?” And it’s a very easy ques­tion. Clearly, Patient B suffered more. His colonoscopy was longer, and every minute of pain that Patient A had Patient B had and more.

But now there is another ques­tion: “How much did these patients think they suffered?” And here is a surprise: And the surprise is that Patient A had a much worse memory of the colonoscopy than Patient B. The stories of the colono­scopies were different and because a very crit­ical part of the story is how it ends — and neither of these stories is very inspiring or great — but one of them is this distinct … (Laughter) but one of them is distinctly worse than the other. And the one that is worse was the one where pain was at its peak at the very end. It’s a bad story. How do we know that? Because we asked these people after their colonoscopy, and much later, too, “How bad was the whole thing, in total?” and it was much worse for A than for B in memory.

Now this is a direct conflict between the expe­ri­encing self and the remem­bering self. From the point of view of the expe­ri­encing self, clearly, B had a worse time. Now, what you could do with patient A, and we actu­ally ran clin­ical exper­i­ments, and it has been done, and it does work, you could actu­ally extend the colonoscopy of Patient A by just keeping the tube in without jiggling it too much. That will cause the patient to suffer, but just a little and much less than before. And if you do that for a couple of minutes, you have made the expe­ri­encing self of Patient A worse off, and you have the remem­bering self of Patient A and lot better off, because now you have endowed Patient A with a better story about his expe­ri­ence. What defines a story? And that is true of the stories that memory delivers for us, and it’s also true of the stories that we make up. What defines a story are changes, signif­i­cant moments and endings. Endings are very, very impor­tant and, in this case, the ending domi­nated.

Now, the expe­ri­encing self lives its life contin­u­ously. It has moments of expe­ri­ence, one after the other. And you ask: What happens to these moments? And the answer is really straight­for­ward. They are lost forever. I mean, most of the moments of our life — and I calcu­lated — you know, the psycho­log­ical present is said to be about three seconds long. Which means that, you know, in a life there, are about 600 million of them. In a month, there are about 600,000. Most of them don’t leave a trace. Most of them are completely ignored by the remem­bering self. And yet, some how you get the sense that they should count, that what happens during these moments of expe­ri­ence is our life. It’s the finite resource that we’re spending while we’re on this earth. And how to spend it, would seem to be rele­vant, but that is not the story that the remem­bering self keeps for us.

So we have the remem­bering self and the expe­ri­encing self, and they’re really quite distinct. The biggest differ­ence between them is in the handling of time. From the point of view of the expe­ri­encing self, if you have a vaca­tion, and the second week is just as good as the first, then the two week vaca­tion is twice as good as the one week vaca­tion. That’s not the way it works at all for the remem­bering self. For the remem­bering self, a two week vaca­tion is barely better than the one week vaca­tion because there are no new memo­ries added. You have not changed the story. And in this way, time is actu­ally the crit­ical vari­able that distin­guishes a remem­bering self from an expe­ri­encing self. Time has very little impact on this story.

Now, the remem­bering self does more than remember and tell stories. It is actu­ally the one that makes deci­sions because, if you have a patient who has had, say, two colono­scopies with two different surgeons and is deciding which of them to choose, then the one that chooses is the one that has the memory that is less bad, and that’s the surgeon that will be chosen. The expe­ri­encing self has no voice in this choice. We actu­ally don’t choose between expe­ri­ences. we choose between memo­ries of expe­ri­ences. And, even when we think about the future, we don’t think of our future normally as expe­ri­ences. We think of our future as antic­i­pated memo­ries. And basi­cally you can look at this, you know, as a tyranny of the remem­bering self, and you can think of the remem­bering self sort of drag­ging the expe­ri­encing self through expe­ri­ences that the expe­ri­encing self doesn’t need.

I have that sense that when we go on vaca­tions this is very frequently the case, that is, we go on vaca­tions, to a very large extent, in the service of our remem­bering self. And this is a bit hard to justify I think. I mean, how much do we consume our memo­ries? That is one of the expla­na­tions that is given for the domi­nance of the remem­bering self. And when I think about that, I think about a vaca­tion we had in Antarc­tica a few years ago, which was clearly the best vaca­tion I’ve ever had, and I think of it rela­tively often, rela­tive to how much I think of other vaca­tions. And I prob­ably have consumed my memo­ries of that three week trip, I would say, for about 25 minutes in the last four years. Now, if I had ever opened the folder with the 600 pictures in it, I would have spent another hour. Now, that is three weeks, and that is at most an hour and a half. There seems to be a discrep­ancy. Now, I may be a bit extreme, you know, in how little appetite I have for consuming memo­ries, but even if you do more of this, there is a genuine ques­tion. Why do we put so much weight on memory rela­tive to the weight that we put on expe­ri­ences?

So I want you to think about a thought exper­i­ment. Imagine that your next vaca­tion you know that at the end of the vaca­tion all your pictures will be destroyed, and you’ll get an amnesic drug so that you won’t remember anything. Now, would you choose the same vaca­tion? (Laughter) And if you would choose a different vaca­tion, there is a conflict between your two selves, and you need to think about how to adju­di­cate that conflict, and it’s actu­ally not at all obvious because, if you think in terms of time, then you get one answer. And if you think in terms of memo­ries, you might get another answer. Why do we pick the vaca­tions we do, is a problem that confronts us with a choice between the two selves.

Now, the two selves bring up two notions of happi­ness. There are really two concepts of happi­ness that we can apply, one per self. So you can ask: How happy is the expe­ri­encing self? And then you would ask: How happy are the moments in the expe­ri­encing self’s life? And they’re all — happi­ness for moments is a fairly compli­cated process. What are the emotions that can be measured? And, by the way, now we are capable of getting a pretty good idea of the happi­ness of the expe­ri­encing self over time. If you ask for the happi­ness of the remem­bering self, it’s a completely different thing. This is not about how happily a person lives. It is about how satis­fied or pleased the person is when that person thinks about her life. Very different notion. Anyone who doesn’t distin­guish those notions, is going to mess up the study of happi­ness, and I belong to a crowd of students of well-being, who’ve been messing up the study of happi­ness for a long time in precisely this way.

The distinc­tion between the happi­ness of the expe­ri­encing self and the satis­fac­tion of the remem­bering self has been recog­nized in recent years, and there are now efforts to measure the two sepa­rately, the Gallup Orga­ni­za­tion has a world poll with more that half a million people have been asked ques­tions about what they think of their life and about their expe­ri­ences. And there have been other efforts along those lines. So in recent years, we have begun to learn about the happi­ness of the two selves. And the main lesson I think that we have learned, is they are really different. You can know how satis­fied some­body is with their life, and that really doesn’t teach you much about how happily they’re living their life, and vice versa. Just to give you a sense of the corre­la­tion, the corre­la­tion is about .5. What that means is if you met some­body, and you were told, oh his father is six feet tall, how much would you know about his height? Well, you would know some­thing about his height, but there’s a lot of uncer­tainty. You have that much uncer­tainty. If I tell you that some­body ranked their life eight on a scale of ten, you have a lot of uncer­tainty about how happy they are with their expe­ri­encing self. So the corre­la­tion is low.

We know some­thing about what controls satis­fac­tion of the happi­ness self. We know that money is very impor­tant, goals are very impor­tant. We know that happi­ness is mainly being satis­fied with people that we like, spending time with people that we like. There are other plea­sures, but this is domi­nant. So if you want to maxi­mize the happi­ness of the two selves, you are going to end up doing very different things. The bottom line of what I’ve said here is that we really should not think of happi­ness as a substi­tute for well-being. It is a completely different notion.

Now, very quickly, another reason we cannot think straight about happi­ness is that we do not attend to the same things when we think about life, and we actu­ally live. So, if you ask the simple ques­tion of how happy people are in Cali­fornia, you are not going to get to the correct answer. When you ask that ques­tion, you think people must be happier in Cali­fornia, if, say, you live in Ohio. (Laughter) And what happens is when you think about living in Cali­fornia, you are thinking of the contrast between Cali­fornia and other places, and that contrast, say, is in climate. Well, it turns out that climate is not very impor­tant to the expe­ri­encing self and is not even very impor­tant to the reflec­tive self that decides how happy people are. But now, because the reflec­tive self is in charge, you may end up — some people may end up moving to Cali­fornia. And it’s sort of inter­esting to trace what is going to happen to people who move to Cali­fornia in the hope of getting happier. Well, their expe­ri­encing self is not going to get happier. We know that. But one thing will happen. They will think they are happier, because, when they think about it, they’ll be reminded of how horrible the weather was in Ohio. And they will feel they made the right deci­sion.

It is very diffi­cult to think straight about well-being, and I hope I have given you a sense of how diffi­cult it is.

Thank you.

Chris Anderson: Thank you. I’ve got a ques­tion for you. Thank you so much. Now, when we were on the phone a few weeks ago, you mentioned to me that there was quite an inter­esting result came out of that Gallup survey. Is that some­thing you can share since you do have a few moments left now?

Daniel Kahneman: Sure. I think the most inter­esting result that we found in the Gallup survey is a number, which we absolutely did not expect to find. We found that with respect to the happi­ness of the expe­ri­encing self. When we looked at how feel­ings vary with income. And it turns out that, below an income of 60,000 dollars a year, for Amer­i­cans, and that’s a very large sample of Amer­i­cans, like 600,000, but it’s a large repre­sen­ta­tive sample, below an income of 600,000 dollars a year…

CA: 60,000.

DK: 60,000. (Laughter) 60,000 dollars a year, people are unhappy, and they get progres­sively unhap­pier the poorer they get. Above that, we get an absolutely flat line. I mean I’ve rarely seen lines so flat. Clearly, what is happening is money does not buy you expe­ri­en­tial happi­ness, but lack of money certainly buys you misery, and we can measure that misery very, very clearly. In terms of the other self, the remem­bering self, you get a different story. The more money you earn the more satis­fied you are. That does not hold for emotions.

CA: But Danny, the whole Amer­ican endeavor is about life, liberty, the pursuit of happi­ness. If people took seri­ously that finding, I mean, it seems to turn upside down every­thing we believe about, say for example, taxa­tion policy and so forth. Is there any chance that politi­cians, that the country gener­ally, would take a finding like that seri­ously and run public policy based on it?

DK: You know I think that there is recog­ni­tion of the role of happi­ness research in public policy. The recog­ni­tion is going to be slow in the United States, no ques­tion about that, but in the UK, it is happening, and in other coun­tries it is happening. People are recog­nizing that they ought to be thinking of happi­ness when they think of public policy. It’s going to take awhile, and people are going to debate whether they want to study expe­ri­ence happi­ness, or whether they want to study life eval­u­a­tion, so we need to have that debate fairly soon, How to enhance happi­ness, goes very different ways depending on how you think, and whether you think of the remem­bering self or you think of the expe­ri­encing self. This is going to influ­ence policy, I think, in years to come. In the United States, efforts are being made to measure the expe­ri­ence happi­ness of the popu­la­tion. This is going to be, I think, within the next decade or two, part of national statis­tics.

CA: Well, it seems to me, this issue will, or at least should be, the most inter­esting policy discus­sion to track over the next few years. Thank you so much for inventing behav­ioral economics. Thank you Danny Kahneman.

· Go to Daniel Kahneman: The riddle of expe­ri­ence vs. memory →