Thinking, Fast and Slow - Two Selves

Created: February 16, 2021

This part of [[Thinking, Fast and Slow, Kahneman]]N explores recent research that draws the destinction between two selves, the remebering self and the experiencing self. It looks at how the conflict between these two gives rise to interesting consequences for subjects like mental well-being.

Two selves mindmap

Notes

Two Selves

This first chapter introduces the two selves: the remembering self and the experiencing self.

  • The remembering self: The self that deals with evaluating past experiences from memories: “How was it, on the whole”.
  • The experiencing self: The self that exists right now, “Does it hurt now?”.

The main effect of this is that we confuse experience with the memory. We learn from the past to maximize the qualities of our future memories, not necessarily of our future experience. This means we cannot fully trust our own preferences to reflect our interests even when we’re drawing from personal experience!

Decision Utility vs Experienced Utility

  • Decision utility is the “wantability” of a certain outcome. This is the utility that people refer to when talking about economics and the one that governs Expected Utility Theory.
  • Experienced utility is the amount of pain versus the amount of pleasure that we experience as a result of something.

In the rational agent model of economics, these things are the same: people want what they enjoy and enjoy what they choose for themselves. Rational agents know their tastes and try to make decisions that will maximise their interests.

Experienced utility is the quantity measured by questions like:

  • How much pain did Helen suffer during the medical procedure?
  • How much enjoyment did Helen get from 20 minutes on the beach?

Measuring Experienced Utility

One way that you could measure experienced utility would be with a “hedonimeter”, a fictional instrument like the equipment at a weather station, that will measure the level of pain or pleasure that someone is experiencing at a given moment. You could use this device to plot a “pleasure vs time” graph, and the total experienced utility would be the area under the curve, i.e. the integral.

Another way you could measure experienced utility would be to simply ask someone at the end of an experience “What was the total amount of pain/pleasure?”. This would be like someone trying to approximate the integral of the pleasure vs time curve.

Duration Neglect and the Peak-end Rule

In experiments measuring experienced utility in those two ways found that there were some systematic findings in the differences between hedonimeter totals and retrospective assessment:

  • Duration Neglect: The duration of an experience often has no effect on the overall rating.
  • Peak-end Rule: The rating is well predicted by the average of the pain/pleasure of the worst/best moment of the experience and the pain/pleasure at the end.

So, duration neglect would say that a 20 minute operation would be considered just as bad as a 40 minute operation if the level of pain was the same. The peak-end rule would say that if you have a painful experience followed by a gentle ending, you’re likely to favour it over the same painful experience without the ending at all, even though you have a worse “hedonimeter total” with a longer experience.

Experienced Utility and The Two Selves

The experiencing self and the remembering self are the reasons for these findings.

  • The experiencing self is what gives the hedonimeter readings, “how do I feel right now?”.
  • The remembering self is what gives the retrospective assessment, “how was it overall?”.

Memories are all we can get from our experience of living and so the only perspective from which we can reflect on our lives is from the perspective of the remembering self.

Pain/Pleasure as a Sum-like Variable

An explanation for duration neglect and the peak end rule is that System 1 struggles to deal with sum-like variables.

System 1 represents sets by averages, norms and prototypes. Memories of some events are stored as prototypical moments, which are often the moment that it was the worst or the best, and the moment at the end. This means we fail to account for the rest of the information about the experience, like the duration or all the other moments not at the extremes.

Pain and pleasure being a sum-like variable means there is conflict. Decisions are not correctly attuned to experience.

Examples of Experience Conflict

  • Participants in a study that had a painful experience keeping their hands in cold water opted to repeat a longer duration since the pain reduced slowly at the end rather than stopping abruptly.
  • A painful end to a marriage makes the whole experience seem terrible.
  • A amazing musical masterpiece that ends with a record scratch seems like it ruins the whole experience despite only ruining the end.
Wireheading

Rats hooked up to electrodes that stimulate the pleasure centres of their brains will repeatedly press a lever until they die of starvation. The stimulation can be delviered in bursts that vary in intensity and duration.

Experiments have shown only the intensity matters, the duration is neglected. Increasing the duration of a burst of stimulation does not increase the eagerness of an animal to seek the stimulation.

Life as a Story

When we evaluate a story, we tend to do so by the memorable moments. In most cases, this is a mixture of particularly noteworthy moments and the end (which is often noteworthy in itself). Thinking about life as a story that we construct provides a nice way to think about how the remembering self influences our behaviour.

As we live, we compose a story about our life’s events and maintain them in memory for future reference. We all intensly care about the narrative of our own life, not neccessarily the day-to-day experience.

Holidays

Broadly speaking, there are two types of holiday categorised by these two examples:

  • A nice relaxing week at a familiar beach
  • A holiday that enriches your store of memories

Tourism is an industry that tries to help people construct stories and create memories. We use words like memorable to describe the best parts of a touristic vacation, and it’s these moments that shape our retrospective feelings about it, not the day-to-day experience.

Research has shown that our prospects for the next holiday are based on the remembering self’s retrospective assessment rather than the “hedonimeter totals” from the experiencing self.

Life as a Story Examples

  • Participants were asked to evaluate the desirability of someone’s life. In one example, the person had a mostly happy life and died swiftly in a car crash. In the other example, the person had a mostly happy life apart from the last 5 years which were slightly less happy. Participants rated the former as more desirable, much like the cold-hand experiment.
  • A play will be rated by the best moments and by the ending.
  • Some anesthesia means that you still experience all the pain but have no memory of the event. Do you feel pity for yourself?

Experienced Well-Being

This chapter is about general conclusions that can be drawn from studies which compare “hedonimeter totals” and a retrospective assessment.

Measuring Experienced Utility, Again

One way to measure experienced utility is through retrospective assessments like these:

All things considered, how satisfied are you with your life as a whole these days?

...or...

Please imagine a ladder with steps numbered from zero at the bottom to 10 at the top. THe top of the ladder represents the best possible life for you and the bottom of the ladder represents the worst possible life for you.
On which step of the ladder would you say you personally feel you stand at this time?

Another way is using the method Csikszentmihalyi pioneered in [[Flow, Csikzentmilhalyi]]N called experience sampling. At random intervals during the day, a participant’s phone will beep or vibrate and presents a brief menu of questions about how they are currently feeling.

The way that Daniel Kahneman conducted his research was through the Day Reconstruction Method, or DRM for short. This asked participants at the end of the day to mentally break down all their activities like they were scenes in a film and answer questions about each episode. These included questions about what types of emotions and also how intense the emotions were.

These assessments can then be paired with the times each episode started or ended in order to create a pleasure vs time graph that was mentioned earlier, and the integral can be used to compute a measure of how the day was.

The time spent in an unpleasant state is called the U-index. If there were 4 unpleasant hours in a day, the U-index would be 25%.

What Does it Mean to be Happy?

Being happy means:

  • Spending yout time engaged in activities you would rather continue than stop
  • Little time in situations you don’t want to be in
  • Not much time in a neutral state where you don’t care either way, life is short

A high U-index indicated you’re not happy since you’re spending a large majority of the time in unpleasant experiences.

Findings from the U-Index

For a study of 1,000 women in America:

  • 29% for a morning commute
  • 27% for work
  • 24% for childcare
  • 18% for housework
  • 12% for socialising
  • 12% for watching TV
  • 5% for sex

There are also cultural differences. Another study found that French women have a lower U-index for childcare.

The U-index findings show that the mood of the moment depend almost entirely on the current situation. Mood at work, for example, is not influenced by factors like the amount of leisure days or your salary. There are a few outliers, like when someone feels in love.

In other words, we primarily draw pain and pleasure from what is happening in the moment.

Furthermore, the more we notice we are doing something enjoyable or appreciate it, the lower the U-index. French people spend the same amount of time eating as American people, but rate eating as higher since they are less likely to pair it with other activities.

Being in [[Flow, Csikzentmilhalyi]]N also means a lower U-index. Switching from passive leisure like TV watching to something like socialising can be beneficial.

On a social level:

  • Improving transportation (reducing commutes)
  • Increasing the availability of child care
  • Improving socialising opportunities

Some other interesting findings:

  • Educational attainment means people evaluate their life overall as higher but it doesn’t affect their day-to-day experience, highly educated people tend to report feeling stressed.
  • Ill health has a large effect on experienced well-being but not so much an effect on overall evaluation.
  • Living with children has a large effect on experienced well-being but not so much an effect on overall evaluation – more angry, stressed. Kids are annoying.
  • Religion has an effect on both (that’s definitely not a reason for being religious though)
  • Money does not buy happiness… all the time. If you live in poverty, negative emotions are made worse and so increased wealth will improve the U-index. The effect on the U-index of money trails off around a 75,000 dollar sallary.

Thinking about Life

This final section is about errors we make when asked to consider our lives.

Affective Forecasting

Affective forecasting is the process by which we predict our future mood and happiness. It’s often wrong in systematic ways.

  • We extrapolate our current mood rather than considering how it will vary.
  • We give to much weight to specific factors that we think will correspond to happiness. (The focusing illusion).
  • Duration neglect means we don’t take in to account how long a specific factor of our happiness will have an effect on our mood.

Miswanting something is when errors in affective forecasting mean we make wrong decisions.

We might miswant a fancy new car rather than to dedicate time going to a social club because errors in affective forecasting make us think we are going to spend the whole time we drive the car appreciating the car, when in reality the car is just a tool. When at a social club on the other hand, we focus on the experience.

Heuristics and Biases in Affective Forecasting

The errors in affective forecasting can be explained in terms of heuristics and biases.

When asked to consider our lives, we rarely come up with a carefully weighted reasoning and instead substitute an easier question. This might be something like “What good things have happened to me recently?”, “How am I feeling right now?” or “What have I been thinking about recently?”. An example was in the [[Thinking, Fast and Slow - Heuristics and Biases]]N section, where students prompted about their dating life gave more positive answers to life satisfaction questions.

Because of this, attention and availability pays a large role in self-reporting our happiness and explains the discrepancies between hedonimeter totals and retrospective assessments.

Life satisfaction before an after marriage often has a triangle shape; rising at the beginning, peaking in the middle before falling at the end. You might conclude that this is because married life becomes more and more mundane as it becomes routine, but you could also explain it through a focusing effect.

Shortly before a marriage, most people will have the happy day on their minds. Shortly after a marriage, people will still reflect on the experience often and feel happier than normal. As time passes, marriage no longer becomes a focus and so people don’t give as much weight to the happiness they recieve from marriage.

Again, the actual day-to-day experience of well-being comes from the current situation and so marriage won’t have a huge effect on experienced well-being – although it changes some aspects of life for the better, it also makes some worse.

Taking this focusing illusion genrally, you could summarise it as:

Nothing in life is as important as you think it is when you are thinking about it.

WYSIATI exacerbates this effect, you don’t actively search for evidence that doesn’t automatically come to mind.

More Examples of the Focusing Illusion
  • People think Californians are happier than average when studies have shown that they are the same as the rest of the general population. Thinking of Californians tends to make people focus on the sunny weather in contrast to their own weather. People who have recently moved there seem happier, but their happiness slowly comes back down as the weather fades into the background.
  • What proportion of the day do paraplegics spend in a bad mood? Turns out it is roughly the same as everyone else’s. When asked about their life satisfaction as a whole though, they inevitably focus on their disability.
    • This was nicely summarised as “Adaptation to a new situation, whether good or bad, consists in large part of thinking less and less about it.”
Things You Can’t Get Used To
  • Chronic pain
  • Loud noise
  • Severe depression

There’s no adapation to these conditions. Chronic pain and loud noises are both biologically set signals that are used to attract attention and depression is a self-reinforcing cycle of miserable thoughts.

Goals and Life Satisfaction

Achieving goals also improves life satisfaction.

Time

Pleasure is a sum-like variable, and the mind isn’t good with sum-like variables. Doubling the time spent doing something pleasureable should double the amount of pleasure we percieve. This is how the experiencing self would like things, but not how the rememberings self would.

The remembering self, the part that reflects on past experiences, represents “episodes” in our life by the critical moments: the beginning, the peak and the end. There is little concept of how time plays a role.

Flashcards



Related posts