Happiness.

“If you don’t get what you want, you suffer; If you get what you don’t want, you suffer; Even when you get exactly what you want, you still suffer because you can’t hold on to it forever. Your mind is your predicament. It wants to be free of change. Free of pain, free of the obligations of life and death. But change is law and no amount of pretending will alter that reality.” - Socrates


Intro

The purpose of this essay is to explore and attempt to discover what happiness is, and form a model of what a life well spent looks like. It’s established that the reason we are here is to survive and reproduce. But humans managed to outsmart their evolutionary imperative and are now aware of our own mortality. Each of us has on average 36,000 days of life to do something worthwhile, and it is important that we spend this limited time in ways we value. I’ve done my best to explore the many angles that can be taken to answer this question - covering the history and science of happiness research, with some conceptualization and thoughts of my own. Hopefully an hour read can change how you spend some of your time.

sadf

Happiness means something different to different people, at different times, and in different places. This isn’t helpful because to be able to improve something, you need an invariable definition of what it is. What I’m referring to by the term happiness will become clear throughout but at its core it is simply a value system - a system of preferences about the past, the present, and the future.

Historically, through different points in time and different cultures, happiness actually meant good fortune (1, 2). It didn’t mean the emotional response to good fortune, but literally good fortune itself. It didn’t mean that a happy person is also a lucky person, it meant that a happy person is somebody who is lucky. This previously dominant definition has been displaced by modern culture and for an obvious reason - it’s not very applicable.

The importance of this analysis and modern happiness research is to uncover variables that we can modify, to empower our decisions which can alter our lives in large and small ways. To back trace through history a bit - some of the first figures to attempt to deviate from this good fortune viewpoint and take happiness within their control were some of the Ancient Greeks.

I’ll begin by covering what three schools of Ancient Greek thought had to say about happiness: the Epicureans, the Cyrenaics, and the Stoics. I’ll follow by categorizing what I think are the common themes extracted from the Greek thought - which I’ll use to define 4 main theoretical models which we can go about improving our happiness: Minimizing Pleasure, Maximizing Pain, Maximizing Pleasure, and Minimizing Pain. I’ll also give my definition of happiness here. I’ll give a quick summary of whether Minimizing Pleasure is conducive to our goal, followed by an in depth discussion of the research on maximizing happiness - from some techniques like savoring, to covering the three temporal models we use to measure happiness: remembering it, experiencing it, and predicting it, and the potential problems with each. I’ll follow by covering potential methods we can use to minimize pain. And finally I’ll conclude with how we can potentially modify our base level happiness as a potential countermeasure to the problem of adaptation.


Ancient Greek History

Note: This section can be skipped if you do not care for the history.

There is other happiness historical data I could cover, but I think a lot of the patterns that an be captured are highlighted well by the Ancient Greeks - in particular the three schools of thought. I think the three most influential are Epicureanism, Cyrenaicism, and Stoicism. The reason being that they were close in time to each other, and were in part responding to and opposing one another.

Besides attempting to extract themes from reviewing these schools, I also don’t think any account of happiness would be complete without covering them in basic detail.

Epicureanism

epicurus

“The acquisition of riches has been for many men, not an end, but a change, of troubles.” - Epicurus

Epicureanism (ref, ref, ref) is the school of thought founded on the writings of the individual Epicurus. He is considered a Hedonist, an important term which frames the position I’m arguing for. The pop culture usage of the word hedonism usually refers to decadent hedonism, which is more related to the indulgence of Cyrenaicism as you will see. A lot of scholars effortfully defend Epicureanism from falling under this same label, and have done so by calling him a negative hedonist - negative meaning a focus on the removal of pain rather than addition of pleasure.

There are two main principles to Epicureanism. The first principle is that the removal of basic physical needs and desires results in the happiest state possible. The second principle is the necessity of friendship to happiness.

In regards to the first principle, he noted that there is a small list of innate needs which are required for survival, and once you remove the pain that comes with those “basal desires”, it results in what he calls static pleasures (or what scholars call negative pleasure). This state of freedom from desire and discomfort results in what he calls ataraxia. This ataraxic state is the absence of any physical pains, or physical pleasure, and is the peak of happiness in his view. This is opposed to adding pleasure on top of a neutral state, or doing something that doesn’t remove an innate need (such as drinking alcohol), resulting in what he calls kinetic pleasure (or what scholars call positive pleasure). The simple needs which are needed to be removed include the need to sleep, to eat, to drink, and to avoid physical pains.

Secondly, although not a physical pain like his others, he emphasized the necessity of friendship as it is only through them that freedom from anxiety and “peace of mind” is possible.

His argument for why negative pleasures are more important than positive pleasures is because he thought positive pleasures leads to further desire for them, which if cannot be attained leads to pain. He wrote his views in large part as a reaction to the Cyrenaics which I mentioned emphasized the addition of physical pleasures. His argument is also based on there being natural desires necessary for the maintenance of life, which manifest as bodily discomforts, which when resolved result in happiness. He thought these were separate and more important than other unnecessary and trivial desires.

Whether or not this is true, what he understood is that desire only arises from a particular deficiency, and that pleasure is the result of satisfying said deficiency. This is simple but also sophisticated - he highlights that the removal of physical and mental pains is the goal of life, because once you remove them, you no longer desire pleasure or the removal of pain any more. Everything we are motivated to do is to achieve freedom from psychological pain (like fear) and physical pain (like hunger), and once we do this we don’t feel the need to continue seeking something which is absent to add to our lives. In concrete terms: We only have the need for pleasure when we feel a pain, but whenever we do not feel pain then we no longer need pleasure.

The Problems With Epicureanism

There are two major problems with Epicureanism: the first is that he fails to weight the duration and quantity of pleasure and pain in affecting our total happiness. The second is that he fails to acknowledge mental pleasures/pains.

So firstly, he didn’t believe that longer pleasures were superior to shorter ones. If you accept the premise that pleasure is good, and pain is bad (which I’ll argue for shortly), mathematically the more pleasure and the less pain you experience, the better. Failing to weigh the duration of an experience in assessing it, is a common bias which psychologists call duration neglect.

The other major problematic aspect is that he acknowledges physical pleasures and pains but neglects mental pleasures and pains. He actually does make an exception for friendship as you read, which is odd because if he admits that there exists mental pains such as anxiety which friendship can remove, then he must leave open the possibility for many other mental pains and that the removal of them is just as important as physical pains. More importantly, he argued that once you remove physical pains, then this is the peak of happiness. Obviously this is wrong because you can have the absence of physical pains but the presence of mental pains such as sadness, anger, boredom, fear, etc.

Our time needs to be used doing something and it can’t completely be used removing physical pains, because such needs are finite. What will result are a list of mental pains which have the ability to affect our total happiness.

What he did highlight is that one potential way to improve our happiness is by minimizing physical pains.

Cyrenaicism

In contrast to the negative hedonism of Epicurus is the Positive Hedonism of the Cyrenaics. Cyrenaicism is a school of thought which focused on the addition of physical pleasures, making these two diametrically opposed. It was a school of thought lead by the figure Aristippus.

aristippus

He believed that happiness was achieved by maximizing physical pleasures beyond a neutral sated state - such as by eating extravagant food, drinking alcohol, or indulging in sexual pleasures. And consuming as many of these as one could. This is the pop culture conception of hedonism - engaging in pleasures of the senses.

Unlike Epicurus who failed to acknowledge that mental pleasures (such as ‘knowledge’) existed, the Cyrenaics did. But he thought that physical pleasures were of more importance than mental pleasures due to them being more visceral and greater in intensity, which is hard to deny.

He also thought that physical pleasures should be more sought after due to their ease and certainty of attainment in comparison to the distance and difficult mental pleasures, caused by for example the mastery of a skill. This uncertainty is understandable once the writing is placed in his historical context - written at a time of war and short life expectancies, thus fostering a perspective of scarcity. When life expectancy is relatively secured by a modern lifestyle, this is less uncertain.

The Problems With Cyrenaicism

He commits a similar error to Epicurus which is that although he accurately acknowledges that physical pleasures are more intense than mental pleasures, he neglects the ability for mental pleasures to persist over time which can lead to a large pleasure total. Physical pleasures are only experienced in the present, whereas psychological ones are exist in the past, present, and future, even if less intense.

But what he did highlight is that one potential way to improve our happiness is to maximize physical pleasures, and we can also see that another way is to maximize mental pleasures.

Stoicism

zeno

The third and also negative hedonist school of thought which aims to minimize pain were the Stoics, lead by the individual Zeno.

The Stoics aimed to live a life of peace from pains, similar to the Epicureans. They thought that the presence of any physical pains or pleasures would disrupt their mental tranquility. But unlike the Epicureans who neglected mental pains, the Stoic form of Ataraxia includes both physical pains as well as mental pains such as desire, boredom, hatred, and guilt.

So the Stoics have the added advantage over the Epicureans of realizing that our mind itself may be a source of pleasure or pain, which is aptly emphasized in the quote:

“We are not troubled by things but the opinion which we have of things.” — Epictetus

The Problems With Stoicism

The same issue occurs with Stoicism as the two other schools - that is they have recognized some improvements to wellbeing - namely minimizing physical pains and mental pains. But they have neglected the possibility of physical pleasures and mental pleasures in mathematically increasing our total amount of pleasures/pains.

So just like the Epicureans neglected mental pleasures and pains, and just like the Cyrenaics neglected all mental pains while recognized physical pleasures, the Stoics recognize mental pains but fail to acknowledge physical pleasures.


What is Happiness (Theoretically)?

You can see how through unfolding what the basic principles of these three Greek schools are, we can extract out some common patterns for categorizing methods of improving happiness. The first theme is that there are physical pains and pleasures, and there are mental pains and pleasures. The other theme is that you can aim to maximize or minimize these various states. This leads us to a basic 4 by 4 matrix of possible ways to improve our happiness, or a 2 by 2 if we combine the physical and mental for the sake of simplicity:

  1. Minimising Pleasures
  2. Maximising Pains
  3. Minimising Pains
  4. Maximising Pleasures

There is a fifth way to improve happiness not highlighted by the Greeks, and that is by modifying our resting setpoint of happiness, which I will come back to later.

Before we take the assumption that pleasures are good, and pains are bad, for granted - let’s first explore that.

The question of how to be “happy” or live a happy life has been a question plaguing humans since time immemorial.

Interestingly, even the most simple of organisms, the Amoeba, seems to display a “pain-like” and “pleasure-like” response, moving closer and feeding off of nearby foods and light, and avoiding noxious chemicals and other objects (4). All that and the amoeba is believed to be related to the earliest forms of life 800 million years ago! So the motivation of organisms to be pleasure seekers and pain avoiders is not a new phenomena.

aomeba

Throughout most of animal history, animals would use their immediate environment as a guide for their behavior. They would live their life completely immersed in each moment, responding to threats of danger and seeking food to keep themselves alive long enough to spread their genes.

a

But at an unbeknown point, early humans developed enough sentience to begin asking how they could be better and what they should do with the finite amount of time they’re endowed?

When you look at how people today actually behave, the rewards they seek and punishments they avoid are not based on some complex forms of reasoning and analysis like here, but are usually based on mere emotional response shortly preceding each behavior in time - which is predicted by learning theory. That is - people have conditioned responses to increase or decrease certain behaviors which have been previously associated with rewards or punishments. Some would even strongly argue that they have no free will.

Although it seems like human experience is infinitely diverse, due to the basic universal biology and evolution we all share, there is a finite list of less than 50 types of pleasure or pain episodes/experiences that we can all have. Anything we can think of as pleasurable or painful can be captured under this list. If you can find any others let me know:

Physical

Type Pain Episodes Pleasure Episodes
Sensory Hunger / Thirst / Eating Bad Food / Food Aversion Eating Good Food / Hydrating
- Olfactory: Bad / Noxious Smells Pleasurable Smells
- Auditory: Loud / Annoying Sounds Sound Pleasures (eg Music)
- Visual Grotesquesness Visual Pleasures / Art / Vivid Colors
- Negation of a sensory organ - eg blind, blocked ears, blocked nose, burnt tongue
- |
Biological Ailments of the Body - eg vomiting, nausea, headache, diarrhoea Health
- Sexual Frustration Sexual Stimulation
- Thermal discomfort - extreme hot / cold Thermal Comfort - Cool weather/body during heat, and physical warmth during cool
- The need to urinate / defecate Urination / Defecation
- Tiredness / Exhaustion Approaching Sleep during tiredness / Rest / Feeling Energetic
- Itchiness Itching a scratch
- Tactile pains - eg damaged tissue, insect bites, clothing discomfort Tactile Pleasures/Comfort - eg receiving a massage, soft fabrics
- Shortness of breath / Inhaling Toxic Fumes Inhaling fresh air
- Internal bone aches or lactic acid burns
- Chemical Pains - Depressants (like alcohol), Chemical Imbalances Chemical Pleasures - eg runner’s high, stimulants (like caffeine), drugs

Mental

Type Pain Episodes Pleasure Episodes
Social Socializing with people you dislike / outgroups Socializing with people that you like
- Social rejection, embarrassment Social approval / (Accomplishment / Mastery / Recognition)
- Social Aggression / Relational Aggression / Being Verbally Attacked Games / Competition / Sports
- Loneliness, Social Isolation Love, Companionship
- Lack of personal space / proxemics
- Missing somebody
- Guilt and shame
- Self Loathing
- | Laughter / Humor
- | Experience of cuteness
- |
Psychological | Learning - reading, watching movies, consuming media
- Repetition, boredom Novelty - traveling to new locations, trying new things, browsing the web
- Financial Insecurity - debt, buyers remorse, lack of money, loss of property Financial security
- Feeling dirty Feeling clean
- Death anxiety / “sense of impending doom”
- Lack of meaning Presence of meaning
- Anger / Goal frustration, or sadness / misery / despair / helplessness / hopelessness Joy / Getting what you want
- | Being in nature
- Anxiety, fear
- Regret
- |
Control Dirtiness / Environmental disorder Cleanliness / order
- Job dissatisfaction Freedom to work on what you like
- Altered states of consciousness - eg drug consumption
- Authority / Oppression
- Forgetting
- Inundation / Time Pressure Freedom
- Distraction Focus
- Stress / Inundation (High Tasks, Short Deadline) Relaxation / Sunbathing

pleasure list

The Purpose of Pleasures and Pains

So basically there are sets of events which are rewards and give you pleasure, and sets of events which are threats and give you pain. But there is no hierachy here and it is slightly more complex because some events last longer than others, have more intensity, suffer diminishing returns, or change from pleasure to pain (or pain to pleasure) over time.

But given this list and our perspective on the pleasures and pains available, we can look over this list and see a trend. Even if not immediately evident, most if not all of these serve an evolutionary purpose - as Epicurus hinted at.

I’m not going to run through the whole list for the sake of brevity, but you can see how the obvious physical pleasures and pains serve our immediate biological needs for homeostasis: hydration, food, sleep, a stable temperature, cleanliness, freedom from physical pains of various sorts, indicating threats with visual disgust, loudness, and fear. The pain motivates us to avoid some situation and the pleasure is the reward when get for removing that pain.

Beyond the physical needs, the mental needs also fall into a handful of categories - we have a lot of needs for social cohesion and approval which serves at the group selection level - namely we have benefited from being in social groups for our survival. Love and cuteness are also bonding emotions - cuteness makes us care for babies, and love bonds partners to stick around long enough to raise offspring.

We are attracted to novelty and change because of the potential for discovering new rewards in the environment. Similarly, we are attracted to learning of various sorts because it allows us to observe others and acquire solutions to potential survival and reproduction problems without trying it first hand.

In regards to control, it has been shown that humans have a strong innate need for control, over themselves and their environment (ref, ref). Chemical pleasures are man made hacks we have into our biology which don’t really fit with the rest of the model, besides endogenously produced which usually offset pain - like runners high.

This gives us an interesting and intuitive model to think about individual episodes of pleasures and pains. The model is that we evolved to experienced pleasure because it serves an evolutionary function: to motivate and reward us for experiences that increase our odds at survival and reproduction. The same fundamental motivation for nearly everything else. In fact, a Norwegian biologist Bjørn Grinde wrote a whole book about trying to use evolution as a guide for living, called Darwinian Happiness.

In regards to basic emotions like anger at the frustration of a goal, or joy at the attainment of one, leading emotion psychologists such as Edwin Locke also support this (ref). Modern emotion researchers think that emotions are the form of taking automatic, subconscious value appraisals, with a specific value appraisal serving specific functions, some being:

  • Fear is experienced during a perceived threat to one’s life or well-being (or that of a loved one)
  • Anxiety is experienced during a perceived threat involving one’s self-esteem, and involves more uncertainty over the threat
  • Sadness is experienced at the loss of a value
  • Depression is a more intense form of sadness involving self-deprecation and a sense of hopelessness
  • Anger is experienced when a goal is blocked
  • Guilt is experienced when one violates their moral values
  • Pleasure is experienced when one has gained or possessed something of value
  • Pride is experienced during an achievement due to one’s own efforts
  • Admiration is experienced during the appraisal of another person’s positive achievements
  • Happiness is the state of non-contradictory joy, and more intense than pleasure

He thinks every emotion has the same pattern (excluding chemical imbalances): an object in the world (a thing, person, action, event, idea, memory, or previous emotion) -> cognition (perception of the object and associations) -> value appraisal (measuring the object by whether some value is being fulfilled or threatened in some way) -> resulting subconscious emotion (which is either a positive or negative state). He even acknowledges that emotions differ in intensity, eg: hate is stronger than disliking. He says the middle stages (cognition -> value appraisal) are automatic and subconscious so what is consciously experienced is object -> emotion. He makes the interesting point that a mood is an enduring emotional state, and it endures because the object (eg a bad boss) is always present, or because causes endure based on memory (eg losing a good job) or rumination (replaying triggering thoughts).

The modern Broaden-and-build theory is a more general theory on the behavioral function of positive emotions developed recently. It argues that positive emotions such as happiness, interest, and anticipation broaden one’s psychological awareness and encourages novel, varied, and exploratory thoughts and behaviors. E.g. curiosity about a new area becomes valuable navigational knowledge; pleasant interactions with strangers become supportive friendships; aimless physical play becomes exercise. In contrast, negative emotions prompt narrow and survival-oriented behaviors. E.g. anxiety leads to the fight or flight response for immediate survival.

Again, What is Happiness?

So what we know is that we evolved, and so did our basic emotions as well as our experiences of pleasure and pain. Those pleasure and pains most likely evolved to serve our evolutionary needs, which are finite. So we have a finite amount of pleasures and pains, some mental and some physical. Pleasures are subjectively experienced as good, rewarding, desirable, positive, and something we should approach and seek more of. Pains are subjectively experienced as bad, punishing, aversive, negative, and something which we should avoid.

In the sense, I am arguing for a form of hedonism with different premises.

The first premise is that pleasure is the only intrinsic values, and pain the only intrinsic disvalue. These are the fundamental dimensions that all human experience can be evaluated on. Anything else we can think of as valuable can be compared with pleasure and pain and seen to be measured by it, and inferior than it - they are the only intrinsic values. You can have other values such as knowledge, but consider a life full of knowledge without mental pleasure or pain - it’s neutral and not preferable over anything. Preference is essentially what value is! And value is pleasure and pain!

In fact, hedonistic philosopher Fred Feldman wrote a great book which does an extremely good job defending this view of hedonism against every counter argument that you can think of. He refers to mental pleasures/pains as attitudinal pleasures/pains because they are attitudes you have about the world - a very philosopher way to describe things (they often talk about ‘propositional attitudes‘). Besides that, he shows that any alternative to hedonism you can think of is simply nonsense when compared with it. Anything you describe as valuable, whether that be knowledge, love, aesthetics, etc, can be captured under the idea of positive and negative experiences and therefore it is only rational to be a hedonist, in the sense of wanting to maximize good and bad emotions. To demonstrate the positivity or negativity of a stimuli, all that is required is to do is to compare them side by side and choose which is more preferable, which will always be the positive (ref).

Feldman makes an interesting note that all physical pains and pleasures are psychological also, but not vice versa. For example, the positive experience of eating or the painful experience of being injured both have a physical nature and a psychological nature - if they didn’t have a psychological nature you wouldn’t experience them. It is only because an attitude is taken about a sensation of pleasure or pain that it is so. But that’s not the case for psychological pleasures and pains where it is possible to experience them without a corresponding physical sensation. This makes them interesting because we can have psychological pleasures or pains about a set of events in the past (via memory or the future (via mental forecasting, or even about a set of events perceived in the present — such as the pleasure taken at the perception of an artwork. Thus all physical pleasures and pains are based in the present but not so for psychological. This opens up a whole new avenue of pleasures such as masochism, social pleasures, romantic pleasures, of accomplishment, recognition, service, and fame.

green up arrow red down arrow

The second premise of hedonism is that pleasure is positive and pain is negative, as demonstrated above. This premise was reached by the Ancient Greeks paving way for later developments in happiness research - Although they disagree with each other in different ways, what they do agree on is this first premise. What they disagree on is the importance of physical vs psychological pleasures/pains, and whether duration is important.

There are many competing views on well-being, but one feature that they all share is the idea that some activities and mental states are considered better than others - whether it’s that pleasure is better than pain, intense absorption in contemplation or activity (aka flow) is better than lethargic emptiness, that intense communion with others is better than loneliness and hostility (ref). All else being equal, happiness can be increased by spending more time in the good states and less time in the bad. This holds regardless of what you consider the good states - spiritual pursuit, flow, pleasure. Time is the ultimate finite resource of life, and finding ways to spend it well is a worthy objective.

Anchoring the concept of happiness in approach or avoidance is appealing and clear: you can ask people to indicate whether they do or don’t like their current situation and whether they prefer for it to continue or to end. This identifies well-being with the extent to which individuals live their lives in a state of wishing for the present to extend, versus wishing they were somewhere else, or neither. This applies equally well to pleasure/pain, flow, boredom, or any other value (ref).

sadf

So we’ve established clearly that both physical and psychological pleasures/pains are important. This still falls short one glaring point, which is obvious to anybody living in the 21st century: pleasure is not happiness, and pain is not unhappiness. The reason this is so is mostly to do with duration. If you are eating an enjoyable meal and experiencing pleasure, you may say you are happy, but that is merely a report of your current experience. If you said this and then lived the rest of your life in pain, it would be silly to state that you lived a happy life. We will come back to this more in the section on remembered utility.

What duration highlights is the importance of quantity of pleasures/pains over time. As we’ve hinted, a life with more (intensity x duration) of pleasurable experiences, subtracting painful experiences, the better and happier the life.

So pleasures and pains are still key, the problem is that are short and not intense. What would potentially be more impactful is important life events like marriage, achievement, parenting, or avoiding injury. But note that this is still compatible with the premises of hedonism which I’ll summarize below. These events are just theoretically composed of a lot of small episodes of pleasure and pain which result in a larger quantities of pleasure and less pain, which is acknowledged by hedonism.

In summary, I argue that happiness is hedonism. The premises of hedonism are:

  1. Pleasures are the only intrinsic good
  2. Pains are the only intrinsic bad
  3. Pleasures and pains can be physical or mental
  4. Pleasures and pains by themselves don’t result in happiness/unhappiness
  5. Total happiness/unhappiness is over a lifetime by summing the (intensity x duration) of pleasures minus the (intensity x duration) of pains
  6. Minimizing physical and mental pains improves total happiness
  7. Maximizing physical and mental pleasures improves total happiness

Given this outline, the mathematically best way to live a better life is to increase the total amount of pleasures and reduce the total amount of pains. There are many important figures throughout history who have held a similar view, from the Greeks above, to Artistotle (and his Eudaimonia), to Jeremy Bentham (discussed below), to John Stuart Mill, to Sigmund Freud’s pleasure principle, to David Hume‘s informal account.

Philosophers call the atomic unit of pleasure a hedon and economists call them utils. I prefer using the utility version in particular in the tradition of Daniel Kahneman who iheavily influenced this essay.

I will be referring to the arbitrary time blocks that contain pleasures/pains as episodes which can vary from 3 seconds to a lifetime. Psychologists have shown the subjective experience of one slice of the “present moment” lasts approximately 3 seconds (ref). An individuals life could be described as a string of these 3 second moments - with approximately 20k moments in a day, or 500m in a 70-year life. Most of these moments simply disappear. But while these moments last, they are rich and multidimensional, and they have a quality of pleasure and pain which is important to us.

Methods of Improving Episodes

Jeremy Bentham

Extending on the view above, old mate Jeremy Bentham here came along in the 18th century and developed what he called Hedonic Calculus was a method for attempting to measure pleasure and pain. The criteria he used to judge pleasures and pains were similar to some we’ve already discussed, but added a few more:

  1. Intensity
  2. Duration
  3. Certainty/Uncertainty (of attainment)
  4. Proximity: How soon it will occur
  5. Repeatability
  6. Purity: Whether the pleasure is followed by a pain, or the pain by a pleasure

We’ve already seen the importance of intensity and duration - it is a crucial feature of our happiness calculation. The Cyrenaics already hinted at the importance of uncertainty and proximity. But repeatability and purity adds two new interesting aspects to our model of how we can improve happiness. I’ll cover purity below and come back to repeatability in the maximizing pleasure section.

Purity

Another interesting point which is obvious to most people, is whether or not the pleasure will be followed by pain, or vice versa. Bentham referred to this as Purity. Most of us have an intuitive grasp on purity of experience and is one of the biggest objections we have against decadent hedonism - because we infer that gorging ourselves with physical pleasures or hard drugs is followed by pains, therefore lacking purity.

a

This acknowledges an important fact about the pain and pleasure response: it isn’t a sinewave or binary, it’s a function. The value of the experience changes over time depending on what it is and the inputs. This gives us an interesting metric by which to judge various pleasures and pains - by averaging them out over time.

For example, if you take out a loan to go traveling, you might experience a chunk of pleasure upfront but you will suffer for it afterwards. Another very basic example is the sexual response cycle which is quite standard: an intensity of arousal/pleasure build from baseline to high, and then peaks at a climax before dropping back down to near neutral or just below.

Climax

I’ll be covering purity more in the section on maximizing pain.

Summary of Happiness Defined

So in summary, pleasures are intrinsically good, and pains are intrinsically bad. The greater the pleasure sum, and the lower the pain sum, the happier the life. Pleasures and pains primarily serve evolutionary functions and reward seeking/pain avoiding goes back to the beginning of evolution. There is a finite list of pleasures and pains in humans. Basic emotions also serve adaptive functions.

Episodes of time are composed of a mix of pleasures/pains. These quantity of pleasures/pains in these episodes is calculated by multiplying the duration by the intensity. Episodes also differ in the certainty/uncertainty of attainment, the proximity, the repeatability (which compounds into greater totals over time), and the purity.

All of this gives us a framework to improve our happiness - maximize the finite list of pleasures, and minimize the finite list of pains.


Minimizing Pleasures

This section is here for theoretical completeness.

But beyond the reasons given by the Greeks about how seeking physical pleasures not necessary for survival may lead to physical pain, which we discounted, there isn’t really any strong reason to minimize pleasures, or schools of thought that advocate doing so. This is obviously going to be the least developed section of our 4 as it’s difficult to imagine how a happy life could be arrived at by deliberately minimizing pleasures as much as possible.

minimize pleasure chart


Maximizing Pains

Just like it is unproductive to try to improve happiness by minimizing pleasures, it is also unproductive to try to improve happiness by maximizing pains.

a

Though there is one crucial difference between these two, and that is that pains are often impure. Remember that or purity or impurity is whether a pain/pleasure is followed by the opposite kind. Even if pain is immediately bad in the short term, they often have the possibility of being followed by pleasures which result in the episode being a net positive.

Granted, pleasures are still what is intrinsically good, and pains are still what’s intrinsically bad. But we have just acknowledged that sometimes pains can be followed by greater pleasures. So you can see that pleasures/pains are not binary, in episodes, they are functions that change over time.

a

The main figure that represents this school of thought was Friedrich Nietzsche, which is attributed with the famous quote we have all heard that aptly summarizes this succinctly:

“No pain, no gain”

Obviously pain is not a necessary condition for all physical or mental pleasures. Not every pleasure has a pain that motivates you to seek it - for example you can serendipitously stumble upon the aesthetic pleasure of a flower without planning to do so. But pain is a necessary condition for some pleasures.

Nietzsche thought that happiness didn’t come from escaping every pain we have, but instead cultivating them/using them to gain pleasure, because pains are “flagposts” which draw our attention to what needs to be done. If something is causing you pain, even though that pain is bad, that is a hint that you should change your behavior or otherwise in order to remove it. It is a very life affirmative philosophy which attempts to embrace consciousness and all of its constituents.

This fits with what we noted about the evolutionary function of the basic drives like hunger, thirst, heat, and tiredness, etc: that they are motivations which are seeking homeostasis.

But rather than pains simply being signals that we can use to remove them to return back to a neutral state like Epicurus thought, Nietzsche believed that the greatest pleasures were produced as a result of not avoiding pains but using them as guides to what needs to be solved. In this way he thinks that pains are trails to large pleasures.

Interesting and related to this is the concept in biology called Hormesis. Hormesis is when an organism gains a net positive response to a short term stressor. Examples include:

  • Exercising causes physiological stress in the short term but larger breathing capacity and improved cardiovascular system function as a result
  • Extreme heat/cold stress (ie Saunas/Cryotherapy) are stressful in the short term but have net benefits
  • Alcohol/Fermentation potentially has a role here, by being mildly toxic in the short term but beneficial in the long
  • Even vegetables are believed to exert their positive effects via hormesis. Plants contain phytonutrients which are toxic to small insects, but are present in such small amounts that humans are able to fight them off and develop a various of beneficial systems to offset the periodic consumption
  • Allergen immunotherapy
  • Radiation

These are just a few interesting examples that highlight how initially stressful / painful events can result in net gains. Actually the examples above of hormesis are related to biochemical stress, the more general term which captures psychological, and others is Eustress - meaning “good stress”, or for our purposes “good pain”. In exercise terminology, the positive response to exercise is called Supercompensation. Recently, the popular author Nassim Taleb came up with the term Antifragility which also attempts to describe the same phenomena unknowingly.

In support of this idea was Edwin Locke’s functional account of emotions in the previous section. He thought that all emotions, including negative ones, served a specific function. Whether that supports the idea that the greatest pleasures are impure pain episodes is questionable though.

There aren’t any extremely actionable lessons from this section. What we highlighted is that the curve of emotions is a tricky one, and sometimes you might want to strategically expose yourself to pains in the short term if they are followed by a greater amount of pleasure in the long term - or simply: delay gratification. Not all short term pains are net bad, and they are often a signal to a problem that needs to be solved which can remove the pain, and add pleasure (ref). But don’t do so at the exclusion of pleasure.

How you figure out what those pains are is the trickier question. Our list of pain episodes is probably a good place to start. To begin with, most of these are relatively minor intensity pleasures and pains, and are relatively pure. The best thing to do is to look at which seem to be relatively large episodes, and then try to work out whether they are impure. Mastery and achievement seems like a likely candidate here - usually requiring upfront suffering with the potential for a big payoff later. Obviously we are just predicting here and this is very imperfect, which we will come back to in the following section. Maybe the biggest take away is just that this allows us to update our model to include impure pains.


Maximizing Pleasures

So we’ve seen that there are two potential methods for increasing our happiness total. The first was by minimizing pain, which fails. The second was by strategically maximizing pain in the short term, which can result in net happiness in the long term, but this isn’t too effective. The two most obviously impactful methods are by directly maximizing pleasure and by minimizing pain. These are the two that most of us already focus all of our efforts on. We will come back to minimizing pains in a later section, but for now, let’s look at how we can improve our happiness by maximizing positive states.

So we’ve already noted that there’s a few theoretical methods that we can use to maximize pleasures:

  • Repeat small pleasures often
  • Focus on large pleasure events (like Nietzsche hinted at)
  • But a third potential way is by savoring or prolonging episodes of pleasure.

These are three theoretical methods that apply to all pleasure episodes which I’ll cover here first.

After that is where we get into the real meat of the modern research on happiness. We will be looking at the science of what makes people happy in the section on Subjective Wellbeing.

Repeatable Pleasures

“Humans don´t live in territories but in habits.” — Peter Sloterdijk

So the first theoretical method that we can use to increase pleasures is to increase the frequency of them - as per the Cyrenaics did for physical pleasures. They consumed as many of them as possible. We can also do that for the mental pleasures - by socializing, learning, loving, engaging in novel activities, being in nature, or whatever else is on our pleasure episodes list.

a

Repeatability is interesting because doing a small thing enough times can compound to have large effects. So even if there are a few events which result in the largest pleasure gains and pain removals, perhaps there are activities that we can do repeatedly which can improve our lives in small ways which add up to a large total over time. This is a smart view to take when you consider that our lives are a sum of the small activities that we do repeatedly - whether that be sleep, eat, socialize, work, play, our thoughts, the environment we are in, or our daily habits. They are recurring emotional experiences which compose the majority of our lives.

What this additional criteria suggests is that one way to improve your happiness is to find ways in which you can make repeatable small gains in pleasure and the removal of pain, which add up over time. I think why a lot of lay happiness theories get this premise wrong is because humans have a tendency to think globally in the timespan of a whole life and spotlight large events, rather think linearly and acknowledge that life is a secession of discrete events. This reminds me of an interesting point brought up in the movie Before Sunset where the main actor mentions how crazy it is that we are living our life right now.

“Tomorrow will be like today. Life wastes itself whilst we are preparing to live.” - Ralph Waldo Emerson

Most interestingly, the better predictor of your wellbeing and happiness isn’t the intensity of positive experiences, but rather the frequency and duration (ref, ref). But we will still cover the intensity below.

Large Pleasure Episodes

The second theoretical method that we can use to increase our total pleasures is by focusing on episodes of experience which include large quantities of pleasure - in intensity and duration. This is what most people spend their life chasing - those few big payoffs: whether that be having kids, buying a house, getting tenure, winning the lottery, becoming the world best, travelling, acquiring prestige, acquiring freedom, being famous, finding The One and getting married, or finding their dream job.

This is an appealing approach because increasing a lot of small pleasures in life doesn’t intuitively feel too impactful, and thus most of us are actually drawn away from that in the hopes of attaining some large pleasure episodes later.

a

And I’m not denying that these large pleasure episodes actually are larger than the smaller pleasures. The research supports that, and when I watch something trivial like the documentary The King of Kong) and see the joy on being the best at something trivial like a video game, it supports that conclusion.

I will come back to this more in the section on Subjective Wellbeing because before we can attempt to answer the theoretical question of which episodes actually result in the largest pleasures, first we need a method of measurement to answer it, which will we cover there.

Savoring

The third theoretical method that we can use to increase pleasures is to increase the duration of them. One way to do that is just to partake in pleasures which are naturally long, as the previous section hinted. Another way is to intentionally increase the duration of pleasures we are naturally already exposed to. This technique is called savoring in psychology, and it is where an experience smaller in intensity and time is taken and intentionally made longer and more intense.

savoring

The research shows that it can be achieved by:

  • Being mindful/present/attentive to your current positive experiences
  • Anticipating the positive experiences in advance
  • Reminiscing on positive memories
  • Avoiding dampening / “killjoy” thinking about positive experiences, and
  • Sharing your positive emotions with others while you have them.

It’s also shown that not only does savoring improve positive experiences directly, but also has a bunch of other positive effects: helps build stronger relationships with others, improves mental and physical health, reduce hopelessness, and increases creative problem solving.

Some researchers believe that savoring can be state based (ie about a particular experience) or trait based - a disposition for all experiences. What this suggests is that cultivating trait savoring can have large effects long term. Interestingly, there is also an opposing phenomena to savoring called dampening where you reduce the intensity and duration of positive emotions, which some believed is a personality trait. People with lower self esteem also savor less and are more likely to dampen.

Research also supports that one of the reasons wealthier people are less happy is due to lower savoring ability (ref), and that the more abundant life experiences one has had (eg the more one has travelled) the less likely they are to savor (ref). And Easterners tend to savor less than westerners (ref).

Subjective Wellbeing

Alright so let’s get into it. Thanked in part to the Positive Psychology movement, recent years have brought a surge of interest in the correlates of happiness.

As you would expect, Psychologists have tried operationalizing happiness and wellbeing] by making it as measurable as possible. This is useful for any construct because it means we can standardize it and try to find practical applications for the research.

The way they have done this is simply by asking people how happy they are, and getting them to subjectively tell you - since their happiness is a subjective experience. Self Reporting is a technical way of saying ‘the measurement of asking people’. Hence the title of this area of research - Subjective Wellbeing (SWB).

If you ask people how happy they are on a scale of 1–10, they can self-report a 7 for instance, and this can give a somewhat accurate picture of the state of affairs if you assume that people know how they feel (have the self awareness) and have the ability to attach a number to it. There’s no reason to think that we can’t do this since we can do a thing called intensity matching. Intensity matching is converting between different unrelated mediums. For example, if I ask you “If John 6’2 and Peter is 5’2, and John is as angry as he is tall, how angry is Peter?”, you’re likely to answer “less angry than John” or something of the sort. It turns out that humans can and do utilize this for all sorts of thinking. People can do a similar thing with happiness.

The field began in 1984 and was developed by Ed Diener which has done a lot of pioneering work in the area. He mentions that the word happiness tries to capture a single construct, whereas subjective wellbeing captures two - the cognitive and emotional account as shown below. It also emphasizes the subjective nature of happiness measurements - as opposed to the objective nature of measurement in Economics that occurred prior to that, under the term ‘decision utility’.

Classical Economists would infer user preferences based on the decisions that they made, based on the assumption that consumers were rational and tried to maximize their utility (ie happiness) at all times (ref). They could then make inferences about their happiness based on their decisions. Unfortunately this turned out to be wrong and consumers often exhibit a lot of biases in their decisions which show that they are not rational in maximizing their happiness at all.

Therefore the solution was to stop inferring their preferences from their behavior, and to ask them directly what made them happy.

So you can see that the crux of the issue is that humans experience a wavering of pleasures and pains over time, captured in a single episode. What we want to be able to do is accurately measure this - this is the real crux of all happiness research. Previously economists would infer these pleasure and pain utilities based on their behavior, which turns out to be inaccurate.

The two methods that Psychologists developed to measure the pleasure/pain over an episode, is (ref):

  1. Remembered Utility.
  2. Experienced Utility.

Remembered Utility is the cognitive account. It has been the primary method of measuring happiness in the SWB literature and has used by the majority of the studies. In these studies, subjects are asked to self-report their level of happiness on a scale such as 1-7, in an attempt to capture their amount of utilities over that episode. The measurement can be about a specific area of one’s life, such as work or home, in what’s called domain satisfaction reports. Or it can be a global judgement/evaluation about one’s whole life - which has been the primary measure of happiness, in what’s called Life Satisfaction Reports. Life Satisfaction reports are usually a 1-5 question survey with the questions almost always verbatim “all things considered, how happy are you with your life at the moment on a scale of 1–10?”. In answering these, they are making a cognitive retrospective evaluation of the episode after it has finished, whether that be a week later, years later, or a global evaluation about their whole life like in the life satisfaction reports.

Experienced Utility is the emotional account. It is a more recently developed form of measurement popularized by Mihaly Csikszentmihalyi (the creator of Flow psychology) and Daniel Kahneman. It is also referred to as ‘Affective Balance’ in the psychology research (Affect referring to positive and negative emotions). It also depends on self-report, just like Remembered Utility research, but rather than being measured after the episode is finished, and capture one’s level of net pleasure/pains in a single report, it tries to capture many self-reports during the episodes. It does so by what’s called the Experience Sampling Method where participants are messaged or paged at random intervals of the day to record their activities, who they are with, and their mood.

Experienced Utility has an added benefit in accuracy over Remembered Utility as you would expect. But first lets look at what the abundant research on Life Satisfaction has revealed, before we look at the drawbacks over this form of measurement.

Before we dig into that, the way this fits into our model is that if we can accurately measure what makes people happy - or what causes pleasures/pains over an episode, in a study of a large sample of people, firstly it can overcome biases and blind spots that we as individuals may have in measuring our own individual happiness. But secondly, even though we have a table of pleasure and pain episodes, we don’t know the hierarchy of importance. So this gives us a method we longed for in answering Nietzsche’s question of figuring out which pleasure episodes are the most important.

Unfortunately the research has not dug into all of the pleasure episodes we have laid out in our table, but it has covered a large surface area of what it considered to be the most important.


Remembered Utility - Life Satisfaction (LS)

remembered utility

By asking people to rate their satisfaction with their life on a scale of 1-10, a lot of findings have been made in the field of happiness research. The findings of these studies are quite interesting, and many commonsense notions about happiness appear to be inaccurate.

Here is a list of things that do and don’t correlate with Life Satisfaction:

Variables That Strongly Predict LS

  • Genetics (ref, ref) unfortunately account for about 30-40% of variance of LS (ref), which is a lot - more than anything else. How? By shaping one’s personality and temperament, which are known to be heritable. The best predictor of individuals future happiness is their past happiness. Some people seem destined to be happy and others unhappy, regardless of their triumphs or setbacks. The limited influence of life events is demonstrated with the hedonic treadmill (which I’ll talk about further below).
  • Personality (ref, ref, ref) - with the largest effect being due to extroversion, with agreeableness, conscientiousness, self-esteem, and optimism also being strongly related. (ref) An Optimistic mindset in explaining negative setbacks is also important (ref) - ie ones beliefs about themselves, others, and the world, and the absence of cognitive distortions/learned helplessness.
  • Mental Health is the one of strongest individual predictors of LS (ref)
  • Love and marriage (ref, ref). People consistently rate being in love as one of the most critical ingredients of happiness. Married people have higher LS than the single or divorced. However the relationship is unclear whether LS causes marital satisfaction, or marital satisfaction causes LS.
  • The Fulfillment of Basic Needs / Not Being in Poverty (ref, ref, ref, ref) - Basic needs include food and shelter. LS rises steeply from dire poverty to an income of 20k per year (depending on the country), and then a slow trend upwards to 50k/yr.
  • Job satisfaction (ref). People often complain about their work, so you wouldn’t expect it to be a key source of LS, but it is. It’s not as critical as love or marriage, but Job Satisfaction has a substantial association with LS. Studies also show that unemployment has strong negative effects on SWB (ref). Again, it’s difficult to sort out whether job satisfaction causes happiness or vice versa, but evidence suggests that causation flows both ways.
  • Sleep Quality (ref)
  • Being Sick or Disabled highly correlates with low LS.
  • Debt (ref, ref)

variables

Variables That Moderately Predict LS

  • Self Reported Health (.32) (ref, ref). Good health status by itself is only moderately correlated with LS, as many people tend to take good health for granted. Being sick on the other hand has a larger effect.
  • Social Activity / Sociability (ref). Humans are social animals, as as such interpersonal relations do appear to contribute to LS. Those who are socially active and are satisfied with their social support and friendship networks report above average levels of LS. Also, people who had deep and substantive conversations have higher LS than those who mostly engage in small talk (ref).
  • Health & Happiness Of Close Relatives (ref) (ref)
  • Religion (ref), though this may be due to religious attendance and not religious belief (ref)
  • Mindfulness (ref)
  • Participating in a Sport (ref)
  • Whether You Eat Alone Often (ref)
  • Sex life (ref)
  • Frequent Verbal Expressions of Positive Emotions (ref)
  • Internet users are more satisfied with their lives than non internet users (ref)
  • Chronotype: Morning Larks have higher LS than Night Owls (ref, ref)
  • Being able to regulate negative emotions (ref)
  • Height (ref)
  • Appreciation which includes being present, self/social comaparison, gratitude, and interpersonal appreciation (ref)
  • Living With Your Partner rather than alone (ref, ref)
  • The satisfaction of key needs, desires, and goals give rise to high life satisfaction, and the dissatisfaction of them gives rise to lower life satisfaction (ref).

Variables That Weakly Predict LS

  • Leisure Satisfaction: Being satisfied with how you spend your leisure time (ref)
  • Income: There is a positive correlation, but it’s weak: but its weak (between .12 and .20) (ref, ref). Though it’s high between poverty and low LS as we saw.
  • Video game use in young people improves LS slightly (ref)
  • Home Ownership (ref). It’s more important for females than males. However its effect on LS seems to depend on income level - with home ownership being less important for lower income groups, potentially due to ‘aspiration preferences’ (ref). Some seemingly better research suggests that homeownership improves housing satisfaction but not LS (ref) and that they might actually not be happier at all due to spending their time in worse ways (ref). Also having a home loan has a negative effect on LS (ref).
  • The Absence of Smoking Cigarettes - quitting smoking improves LS (ref)
  • Donating to Charity (ref, ref)
  • Your Country weakly predicts your LS (<.10) - ranging from 6.5 (for Brazil) to 7.5/10 (for Finland) (ref).
  • Working 4 days > 5 days increases life satisfaction by 5% (ref, ref)
  • The personality trait Zest due to living in the present (ref)
  • Healthy eating (ref, ref, ref)
  • Yoga (ref)

a

Variables Unrelated to LS

  • Physical Attractiveness (ref)
  • Intelligence (ref)
  • Gender (ref)
  • Age is completely related according to some research (<1% variance) (ref, ref), whereas others demonstrate a U curve, meaning a drop in happiness mid life before a raise in late life (ref)
  • Parenthood (ref, ref). Compared to childless couples, parents worry more and experience more marital problems. But parents seem to have around the same LS as the childless. Other research shows a sharp drop in LS post birth (ref).

As you can see, quite a number of factors that you might expect to be influential appear to bear little or no relationship to general happiness. And you can find study correlations between all sorts of things (ref, ref) - OurWorldInData did a great writeup. The list of factors that turn out to have fairly strong associations with happiness is surprisingly short. The key ingredients of high LS appear to involve having basic needs met, health, love, work, and genetic predisposition expressed through personality.

In summary, the main findings of the LS research has been highlighted by Ed Diener (the pioneer) in these 3 reviews: here, here, and here. His main findings are that life satisfaction reports tend to be stable over time. Genetics account for 30-40% of the variance in happiness ratings, caused by having a happy/unhappy personality - specifically extraversion being good, and neuroticism being bad. The cause of genes and personality on wellbeing are perhaps the only major long term causes. But 60-70% is also environmental and within our control. Those environmental variables are fulfilling our basic needs (food and shelter), psychological needs (eg autonomy), social needs, and experiencing respect from others. Income is important up until a point, because it helps us fulfill our basic needs.

With this knowledge, we are now more equipped to answer the question of which pleasure episodes are large.

Hedonic Adaptation

hedonic adaptation

Unfortunately there is not any perfect research measuring the curve or experience of all the different pleasures we laid out, but the LS research has given us data on which life events have the biggest effects.

What the data shows is that people tend to get a substantial drop in LS after unemployment, while approaching widowhood, after marriage, after the birth of a child (ref), at the ending of a relationship, or the death of a loved one or parent (ref).

And people tend to get a substantial boost in LS after divorce, layoff, starting a new relationship, gaining employment, starting pregnancy, starting a new educational course, and buying a new house (ref, ref.

a

So these changes in life circumstances have the potentially largest gains and losses on our well-being, but as the chart above demonstrates, for most life events, people quickly return back to a baseline of LS. For example, people report higher LS for around 2 years after marriage before returning to baseline. A lot of us have heard of the two most popular examples of this adaptation: the first is where a lottery winner (experiencing a large positive event) and a quadriplegic (experiencing a large negative event) both have the same level of happiness after a few years. Unfortunately or fortunately, this returning to baseline happens to basically all life events, except a couple of exceptions like (imprisonment? etc? from which paper?). Nearly all major life events are return back to baseline by the 5th year mark.

This returning back to baseline of LS is what’s known as Hedonic Adaptation, or Hedonic Treadmill, and the the odd fact that people return back to different levels of baseline is what’s called Set Point Theory, which is believed to be biologically determined, and which we will come back to later.

It’s referred to as a treadmill, because just like on a treadmill you return to the same spot, so too do you return to the same level of happiness. This happens for both LS and the more everyday experience of pleasure and pain which we are all aware of. This is one of the sad facts about human experience and pleasure: the first bite of a meal is great, the last bite not so much. Or in more general terms, the more we do of something pleasurable, the less enjoyable it becomes. Most of us have experienced this personally in the form of dreaming for something: a product or experience, and once attaining it moving on with life. We usually refer to this adaptation to experienced utility as diminishing returns, and it’s what Economists more precisely call Diminishing Marginal Utility.

diminishing returns

Though remember that this doesn’t make pleasure completely valueless, as you are still experiencing pleasure or LS prior to returning to baseline.

The reason that we adapt to pleasurable sensations as well as LS is likely due to the same reason that we adapt to all other sensory phenomena - because we adapt to irrelevant information. It’s known that what makes its way into human perception is heavily weighted on changes in the environment. For example, an abrupt change in sound relative to the baseline will be perceived, as well as visual movement, tactile sensations, and new smells, etc. The reason is because changes in the environment most require our attention and are potentially risky or rewarding.

This makes evolutionary sense as remember that pleasures are rewards for attaining a need, and obviously we cannot be in a constant state of bliss because (a) the need gets satisfied quickly, and (b) if we were constantly experiencing pleasure we wouldn’t seek out our needs. This is also similar to what the philosopher Arthur Schopenhauer in regards to critiquing Utopian societies.

This is also likely the reason that our evaluative responses to LS are weighted in changes to our circumstances - but which we also adapt to to (ref). For example, our LS undergoes sudden drops or gains soon after big life events - and our evaluation of our life changes briefly, but then it returns back to baseline. This has led some to believe that although our LS reports are adapting, perhaps the circumstances themselves are still ‘objectively’ better or worse, but it is our aspirations that have adapted.

Other research calls this return to baseline the Hedonic Adaptation Prevention Model, which leads us into looking at any ways that it can be prevented. The psychologist Sonja Lyubomirsky is the leading researcher in this area and covered a lot of it in her book The How of Happiness. One of the main ideas is that we can prevent adapting to positive experiences by (1) continuing to savor them, and (2) having continued variety of experiences (ref). Novelty is one of the pleasure events that we mentioned in our list, and it turns out one of the reasons it is so great is that it is the solution to adaptation.

A third solution can be found in the great book with a bad name Happy Money, which is simply to partake in the pleasurable experiences less frequently - this is another mode for preventing adaptation.

fighting adaptation

Issues with SWB / Remembered Utility

Bringing all of the above life satisfaction research into perspective, remember that this is one method that we use in researching happiness, and what this method relies upon is an individual trying to summarize an experience after the fact, into a single rating. This is equivalent to us having a memory (of an experience), which is then followed by an emotional response. And then we base our evaluation of that whole experience based on that emotional response.

You would think this would be extremely erroneous for many reasons, which have led many to doubt the validity of LS and remembered utility. For instance, researchers have manipulated the context and mood of the subjects in many ways to see if it influences their reports - from priming their environment (ref), to the weather, to being influenced by the order of the survey questions, to making them feel good beforehand by example finding a coin before filling out the survey (ref), and there is also room for the subject to interpret the meaning of the question differently. But it was found that although LS reports are influenced by current mood, it’s not a large effect (ref).

It turns out that given the large sample size of these studies, these anomalies basically balance themselves out. One promising finding of LS reports are that they are relatively stable across time - they correlate at 0.77 when measured again after 4 weeks which isn’t as high as some other measures in psychology but is still quite stable and reassuring (ref). And although current mood and context may fluctuate day to day which may alter individual reports, it might be reliable enough for many purposes (ref). So there are reasons to be skeptical of the complete reliability of the reports, but they are not grounds for dismissing it completely, yet. Though they do fall short as a measure of real-time experiences.

Diener has said that the reason they continue to study life satisfaction reports in face of these anomalies is the idea that “how each person thinks and feels about his or her life is important” (ref). Indeed, life satisfaction reports tend to correlate around .7 with reports of experienced emotions over time (ref) - at least in this short 3 week study.

But intuitively, there does seem to be some obvious deficits with measuring happiness like this (ref). And there is one big reason, and that big reason it is inaccurate is because it is a cognitive rather than an emotional account - meaning participants reflect and then rate, there is no reason to believe it’s very accurate or that humans are able to do this well.

If you ask somebody to rate their life on a scale of 1-10, this is a single retrospective judgment at one moment in time and it is impossible that participants are able to accurately remember every single emotion they have ever experienced, and sum them up in their head, do a calculation on them, and output an accurate reflection of their life. It is infeasible that we are able to conduct a comprehensive search of the relevant memories and aggregate them into a meaningful report (ref) - in fact the research shows that people make judgements too quickly for it to even be feasible that they are searching their at all (ref) - people make the judgments almost instantly.

So what are they doing then? It’s been suggested that perhaps people are recalling a LS judgment stored in their memory, or they are using quick heuristics that enable fast but still potentially accurate reports ((unlikely), or their current mood was influencing their LS judgment. Research demonstrates the latter does happen as mentioned, but unfortunately the sample sizes of the current studies demonstrating this are quite small, even if it seems likely. So although the current research shows the effect sizes of the latter are small (ref), these are small and imperfect studies.

The best argument I’ve found for what LS reports are doing is the following (ref). As we’ve established, pleasure and pain are approach/avoid signals. This makes a lot of sense in the context of experienced utility where our behavior is being changed in real-time by those signals. The core purpose of pleasure and pain is regulate our response to the current situation, immediately. But LS and remembered utility are different - they are not experienced real-time, they are retrospective. But like experienced utilities, remembered utilities share an evolutionary function: they determine whether an episode in the past should now be approached or avoided. So remembered utilities are also controlling our current and future behavior, but it is based on events from the past, rather than the present. Our brain is summarizing an episode of experience, with all of its pleasure and pain, and outputting a single rating which reflects whether overall it was something good which we should approach in the future, or avoid. This is what all of our evaluations of the past are doing. And our LS reports are very likely just an odd new construction that takes advantage of the human ability to do this cognitive evaluation of the past, with the researchers thinking that it was accurately doing so.

“The general conclusion is as clear for well-being as it was for colonoscopies: people’s evaluations of their lives and their actual experience may be related, but they are also different. Life satisfaction is not a flawed measure of their experienced well-being, as I thought some years ago. It is something else entirely.” - Kahneman

But besides the reasons mentioned earlier about the unlikelihood of it being accurate, there are real demonstrations which highlight the errors of remembered utility.

To demonstrate, it is first necessary to introduce you to the term instant utility which will be elaborated on more in the the section on experienced utility. Remember that in Remembered Utility, individuals try to summarize their series of positive and negative emotions over a time period? And as mentioned, Economists refer to these waning of pleasures and pains as utils. Well the term employed to represent the base unit of a 3 second “present” is Instant Utility - that is, the current experience of pleasure or pain at its most basic unit, which is immediately reported when asked how one is feeling in the moment. It is the bare unit of measurement of experience with the most accuracy.

Measuring these Instant Utilities over a time period and summing them up result in Experienced Utility - that is, many instant utilities that have been reported in real-time. An instant utility is singular. And so the way to measure whether Remembered Utility is accurate is by measuring Experienced Utility and comparing it with Remembered Utility to determine whether our evaluations of the past accurately reflect what we feel while that experience was happening - which is the closest source of truth about that experience.

A perfect way of measuring instant utilities moment to moment was fantasized as early as 1880 by Francis Edgeworth, in what he called the “Hedonometer“. Fortunately now we have developed instruments which could do this, and many studies have done so (ref). One way that instant utility is measured is by giving subjects a handle that they can move up or down which indicates their experience from “very unpleasant” to “very pleasant” in real-time.

The first noteworthy example is a study where subjects were given 16 short plotless videos, with half being pleasant (eg nature scenes), and the other half aversive (eg surgeries). There were two versions of each film: a 1 minute short version and a 2 minute longer version, with the same content. Each subject watched the short and long versions of 8 of the videos while controlling the sliding knob, to indicate their instant utility or disutility. Afterwards the subjects were asked “Overall how much pleasure or displeasure/discomfort did you experience during the film?” to measure remembered utility.

What the result of this among others demonstrated is the following: That the subjects based their remembered utilities on the instant utilities at the Peak (the most intense value) and the instant utilities at the End of the video - what’s come to be known as the Peak-End Rule in psychology that affects most evaluations. Basically - remembered utility is determined by the peak end rule. An implication of this is that the subjects completely neglected the duration of the experience or sum of Instant Utilities in making their Remembered Utility evaluations (remember “duration-neglect”?). The third is that when asked to make a choice about which future decisions to repeat - individuals generally choose the experiences from the past with the highest Remembered Utility, not Experienced Utility (which we will come back to). This supports our argument from above that Remembered Utilities have an evolutionary function - to guide our future actions. And it seems to be doing that on the assumption that the Peak-End Rule is a valid way of assessing experiences.

What this demonstrates is that Life Satisfaction measures which depend on Remembered Utility evaluations cannot be completely relied upon. In fact, Psychologists recognize this and is why Subjective Wellbeing research is split into Life Satisfaction and “Affective Balance” measures which aim to measure Experienced Utility (next section). Although these cognitive evaluations are doing something - informing our future decisions, it’s discrepancy with Experience Utility leads us to think that humans did not evolve to be happy, but to survive.

As a final demonstration of the above oddity, there are a lot of experiments that demonstrate something similar to the following (ref). If you get subjects to hold their hand in cold 14C water for 60 seconds, experiencing substantial pain, and also later a longer 90 seconds, with the first 60 seconds being the same 14C water with pain, but with the final 30 seconds the temperature is slowly reduced to a less painful 15C, subjects remembered the longer period as less aversive (due to it ending on a less negative note). And as a result they will choose to repeat this experience over the shorter again if given the choice.

This is terrible that an extra period of discomfort at the end can reduce the peak end average and make the remembered utility less bad, when the longer episode is experientially worse. What this demonstrates is that people are willing to endure more painful experiences if they remember them to be less painful. This is obviously not conducive to experienced happiness, which is what I’ll cover next. So it’s clear that evaluated and experienced well-being are two separate things, but as Kahneman has noted, evaluation is still important to measure because it plays a significant role in people’s decisions and also because people deeply care about the narrative of their life (ref).


Experienced Utility

experienced utility

So by now you should know that the goal of both Remembered Utility and Experienced Utility is to accurately measure and capture the wavering of positive and negative emotions or utilities over an episode of time. We just saw how remembered utility is flawed due to it neglecting duration, and it being prone to biases like the peak-end effect which can make us do things like endure more experienced pain. Both of these are big errors which it difficult to trust as a measure. What we now know is that pain and pleasure can be measured in real time, and that life satisfaction reports are not a good measure of the sum total of the actual experience.

We now also known that instant utility is the gold measure of instant well-being, it is when someone immediately reports their emotional state - which is difficult to doubt. Remember that a moment of the “present” lasts around 3 seconds psychologically. The advantage of real-time measures of instant utility is that they avoid the biases of memory and evaluation that affect retrospective judgments of wellbeing. But instant utility don’t give us anything, what we want is instant utilities over time which lead us to experienced utility. Experienced Utility captures a series of instant utilities over an episode of time - for example once per minute for an hour, or once per hour for a day, or every 3 seconds for a minute. Obviously this is much more granular and trustworthy data than remembered utility due to it being immediate in space and time, and also by taking many reports.

Researchers have also used more psychological terms to describe this, rather than the more economic term utility: measuring experienced utility has been measured under the terms positive affect, negative affect (respectively meaning positive and negative emotions) and thus affective balance (the balance of positive to negative emotions) - or net affect / emotional wellbeing (ref).

So we know that the experience of a moment or episode is not easily The experience of a moment or an episode is not easily represented by a single happiness value as we saw in the flaws of the Life Satisfaction research. The experienced utility research aims to solve that by measuring emotions over time and moment to moment, rather than retrospectively. It’s often said that this form of measurement is “right now, not later; and right there; not elsewhere. More clearly: If I were to ask you how you are currently feeling rather than how you felt a week ago, you’d give me a more accurate answer. That’s what this research aims to get at.

Kahneman’s conception of experiencing utility is that experience is a continuous flow of pleasure and pain, and that the happiness of an individual during a period of time is the sum of the momentary utilities over that time period.

Measurements

There are three possible methods of measuring experienced utility: instant utility reports, the Experience Sampling Method, and the Daily Reconstruction Method.

We saw earlier how the most accurate form of measurement was a sequence of instant utilities: where the subject holds a lever that when moved up or down in real-time moves markers on a scale from from “very pleasant” to “very unpleasant”. These studies result in an ideal form of happiness measurement: a moment to moment measurements of a subjects emotional state. It would be more than a godsend if we had a large sample size of people that were able to do this over many different activities and collect the data.

Unfortunately, although that works in specific lab settings (ref), to measure experienced utility over one’s whole life, a continuous recording of experience is unfortunately impossible - a person cannot live normally while constantly reporting their experiences.

The closest alternative is the Experience Sampling Method (ESM), which Mihaly Csikszentmihalyi (the dude that also created the theory of flow states) invented. What he did was give people diaries or pagers and then pinged them at several random times during the day (or more recently vibrated peoples phones). When he pinged them, they would report what activity they were currently doing when they were interrupted, who they were with, and how they felt (using a 0-6 rating scale of the presence/absence of various feelings - happiness, tension, anger, worry, pain, engagement, and more). Using this ingenious method we could get a direct measure of how people were currently feeling whilst doing various activities. With large enough datasets which can begin to find correlations which give support to the conclusion that certain activities are associated with pleasurable and painful states.

You can see how this is similar to the remembered utility measurements but it is far more accurate as you are no longer depending on the quality of the subject’s memory to accurately report their experience and it overcomes most of the issues bound up in framing effects.

Unfortunately ESM studies are expensive, involve high levels of effort from the subjects, and provides little information about uncommon or brief events which are rarely sampled. A new and more efficient mix approach between the ESM and Remembered Utility reports is the Daily Reconstruction Method (DRM) where subjects revive memories of the previous day and construct a sequence of episodes in a diary, then describing when the episodes began and ended, how they felt during each episode, who they were with, what they were doing, etc (ref). The ESM is still the gold standard to which the DRM must be compared.

Results suggest that despite relying on a day old memory, the DRM reproduces results of the ESM accurately and is an efficient approximation in place of the more tedious ESM (ref). So it turns out that remembered utilities are not completely inaccurate, but become more so the longer the episode of time evaluated. Although it is still not practical to apply this to your personal life as it took participants 45-75min to complete a DRM report.

Findings

Applications of the ESM and DRM have unfortunately been limited due to them being difficult to implement in large sample sizes, but some studies have been done. These gives some interesting results - as what you would expect is that how people feel moment to moment would be the same as how people remember feeling, but this is not the case. So the research finds the following (ref):

General

  • People report basically the same profiles for painful experiences (eg dipping their hands in cold water) (ref), which includes their duration neglect
  • Many people report themselves in at least a moderately good mood most of the time, negative affect is reported only 34% of the time (ref). Negative emotions are relatively infrequent and mild, and the total absence of positive emotions is rare.
  • Negative emotions are the most pronounced in the morning and fall for most of the day (ref), ref), with positive emotions reaching a peak around noon and evening (ref)
  • No suprise, people are happier on weekends (ref) (except the very poor)
  • People spend about 47% of waking hours ruminating, and that ruminating typically makes them unhappy (ref, ref), and people would actually prefer to electric shock themselves instead of being alone with their thoughts (ref)
  • Reported sleep quality explains a very large 56% effect of the difference in positive/negative affect btw people (.3 correlation), a strong effect also seen on life satisfaction - with those who reported ‘very good sleep’ having a mean net affect of 4.05/6 over the day vs 2.80/6 for those with ‘very bad’ sleep.
  • Personality and temperament also have similarly large effects, similar again to LS
  • As with life satisfaction, age and education have very little correlation with net affect (ref)

What We’re Doing

  • According to ESM research, people are happiest (in order) when making love, exercising, or socializing. Followed by taking a walk (ref), listening to music, eating, meditating, preparing food, childcare, relaxing, reading, and then watching TV (ref, ref). Some older DRM research puts the order as: making love, socializing, relaxing, meditating, eating, and exercising (ref). Some other research puts the order as making love, socializing after work, relaxing, dinner, lunch, exercising, meditation, socializing at work, and watching TV ref.
  • According to ESM research, people are unhappiest (in order) when working, using a home computer, commuting, grooming, listening to the news, and doing housework (ref, ref). Some older DRM research puts the order as: commuting being the worst, followed by working, housework, computer, childcare, using their phone, preparing food, shopping (ref. Some other research puts the order as morning commute, working, evening commute, childcare, housework, home computer, shopping, cooking, phone, and watching TV ref).

Who We’re With

  • Of all of our interaction partners, we experience positive affect from highest to lowest with: friends the highest, then relatives, spouse/SO, children, clients/customers, coworkers, boss, and alone. So being alone or with our boss is the worst, with our friends, family, and significant other the best (ref)), but either way socializing with our friends and family is imperative to our wellbeing ( ref, (ref.
  • The mean level of enjoyment on a 0-6 scale was 4.68 for episodes in which respondents socialized with friends, 2.97 when they were commuting alone, and 2.15 when the only activity was interacting with one’s boss (ref)
  • A big suprise is that for American Women the emotional experience of time spent with one’s children is slightly less enjoyable than doing housework (ref), though some research contests this (ref)
  • passively using social networking sites (fb, ig, twitter) has a negative relationship with affective states (by provoking social comparisons and envy), whereas actively using them has a positive relationship (by stimulating feelings of social connection) (ref)
  • Divorced women report slightly more positive affect (in comparison to slightly higher life satisfaction for married women)
  • Smartphone use diminishes the wellbeing benefits of socializing with friends and family, resulting in us feeling more socially disconnected, distracted, having a lower sense of meaning, and having less net affect. It also reduces the likelihood we will interact with strangers which has well-being benefits (ref). Though when the immediate social environment is uncomfortable, or hostile, or there are no opportunities for social interaction, smartphone use should on average increase wellbeing, by acting as a buffer against social exclusion.

Other

  • Facebook use predicts decline in both experienced utility and remembered (ref)
  • Household income have a modest influence on positive emotions which shrinks in controlled studies, much less of an effect than is seen with life satisfaction (ref, ref, ref, ref), with plateauing effects on positive affect at 60k USD (as opposed to 95k USD for life satisfaction), but note that this varies by country see the ref for more details (ref, ref). So the correlation for affect is weaker at .17 as opposed to a strong .44 for life satisfaction (ref). And the correlation btw income and net affect over the day was only 0.05 (ref)
  • The effects of religion is similar to above, but shrinks less
  • Not suprisingly, depression (measured by the indicator “trouble maintaining enthusiasm”) is also strongly correlated with daily affect
  • Worringly, most of the job satisfaction correlates (such as benefits, training, etc) have less of an effect on the affective experience of a working day, except a few characteristics which include features of the immediate environment, such as time pressure, and opportunity to talk with coworkers. (2.88/6 mean affect for high time pressure, vs 3.96/6 mean affect for low time pressure - ref). Oddly, even the risk of being laid off doesn’t effect affect at work.
  • Sadly, in opposition to the adaptive effects of life satisfaction reports for disability, there is no such adaptation for experienced utility and there is a strong negative association btw disability and experienced utility (ref)
  • Retirees have higher experiential wellbeing than older working adults (ref)
  • Although people think repeating experiences (such as visiting a museum or restaurant) will be less enjoyable than the first time, that’s not the case (ref)
  • Not suprisingly, a headache will make a person miserable
  • Business students report more negative affect than others due to more motivation to make money (ref)
  • On the days when people eat more fruit and vegetables they report feeling more positive emotions, as well as the following day (ref, ref)
  • Only life events that occurred during the previous 3 months influenced daily affect (ref)
  • Being exposed to offensive noise is extremely highly correlated (-.90) with net affect at work (ref)
  • Spending money on others increases positive affect (ref, ref, ref, ref, ref), but with more benefit when spending money on strong social ties (friends, lover, or family) than weak social tissues (acquantances) (ref).
  • People enjoyed their conversations more when they talked less (ref)
  • Individuals who prioritized time over money reported more frequent positive emotions (ref)

In summary, the results of experienced wellbeing studies are similar to the remembered wellbeing studies: Positive affect is strong influenced by the pervasive effects of personality, mild depression, and poor sleep quality. Features of the current situation such as our interaction partner and time pressure at work also exert powerful effects on affect. And lifestyle circumstances (unless they are locally salient - aka by thinking about it) such as income, education, and religion have little impact on the enjoyment of a regular day.

In terms of the effects of personality, a striking observation is that about half of the participants reported going through an entire day without experiencing an unpleasant episode, whereas a significant minority experienced considerable emotional distress and pain for much of the day ((ref)). It appears that a small fraction of the population does most of the suffering - whether because of physical or mental illness, an unhappy temperament, or the misfortunes and personal tragedies in their life.

In terms of time use, net affect (highest positive affect to lowest negative affect) is highest on average when individuals are engaged in leisure activities (such as socializing after work, watching tv, and relaxing), and lowest when they are engaged in work and personal maintenance activities (such as housecleaning). Commuting in the morning is particularly unpleasant. Social contact during an episode is associated with high positive emotions. Interestingly, the average net affect experienced during episodes of commuting rises to that of the overall average level if the commuter is accompanied by someone, indicating the influence of social contact.

Besides not suffering any physical pain, the second best predictor of the feelings of a day is whether a person did or did not have contacts with friends or relatives. It is only a slight exaggeration to say that happiness is the experience of spending time with people you love and who love you.

Although net affect and reported life satisfaction are positively correlated, there are differences worth emphasizing - some aspects of life have more effect on the evaluation of one’s life than on the experience of living. For example, more educational attainment is associated with a higher evaluation of one’s life, but not with greater experienced well-being. Income, marital status, ethnicity, and education predict life satisfaction more strongly than they predict affect. And time use predicts affect more than it predicts life satisfaction. Ill health has a much stronger adverse effect on experienced wellbeing than life evaluation, and also living with children (with high reports of stress and anger - but less adverse effects on life satisfaction). These contrasts suggest that net affect provides a window on people’s experience that is distinct from that captured by standard life satisfaction measures.

On a good note, hedonic adaptation applies less to daily affect than it does to life satisfaction reports, and the reason seems to be that when you are making a life satisfaction report you are focusing your attention on specific objective circumstances of your life and comparing them to others circumstances as well as your aspirations. And of course, these circumstances are more likely to come to mind for recent changes in circumstances (ref) which is why there seems to be an ‘adaptation’ to them which doesn’t apply to experienced wellbeing (as much (ref)). The small (or lack of) impact of life circumstances on affective experience is because thoughts of ones circumstances are less likely to come to mind during the routine course of a day, in comparison to an assessment of one’s life (ref).

This is good news because in contrast to the life satisfaction reports which showed that life circumstances are mostly irrelevant to well-being, the large variations in emotional states throughout the day highlights the importance of optimizing your allocation of your time across activities and situations. Measurements taken in real time connect wellbeing to something that matters a lot - how people spend their time. And time use is within our control. So the question becomes what to spend one’s time doing.

But affect does still show adaptation to external events the same as life satisfaction. People who undergo various external events end up adapting to the same daily net affect as prior. This highlights two potential ways that adaptation is occurring for both life satisfaction and net affect. The first is acknowledging that attention plays a key role in our happiness - in that even if we suffer a large negative event, it will only occupy a portion of our attention. A second is that we are potentially shifting activities - eg from playing sport before losing a job to strolling in the park after - but with the same result on our happiness. Measures of wellbeing connected to time use like this are potentially fruitful.

So we have two popular types of measurement used so far - we’ve got data on how people evaluate their lives to be, and what variables correlate with that life satisfaction. And we have data on how people actually report feeling moment to moment. Commonsensically I think people are referring something similar to life satisfaction reports when they talk about pleasure/pain not being happiness/happiness - that is, they think evaluating ones life gives a more accurate picture of happiness than pleasure/pain. On the other hand experienced utility captures a measurement of pleasure and pain in quite an accurate way - rather than people just verbally reporting on their memories, they are telling us what emotions they are actually feeling - positive or negative, pleasurable or painful. So hopefully this section made clear that there are two main types of happiness - the wellbeing that people experience as they live their lives, and the judgment they make when they evaluate their life.

This distinction is useful to know, because as we highlighted earlier, repeated events add up to compose your life over time. Your life as you will live it is a totality of thoughts and emotional experiences, of single events and episodes that you will do repeatedly. Your life doesn’t exist in a sort of time capsule where you will get to experience the Total Pleasures or Total Pains at each point. You will only experience the episodes serially one after the other and the closest you will get to experiencing the Totality at once is a remembered utility judgment, which lasts a few seconds. Obviously what’s more important is how you spend and control your time use, at each small interval.

“If today were the last day of my life, would I want to do what I am about to do today?” And whenever the answer has been “No” for too many days in a row, I know I need to change something.” - Steve Jobs

All of these points support Daniel Kahneman’s conclusion that he drew from this area of research, which was: “The easiest way to increase happiness is to control your use of time. Can you find more time to do the things you enjoy doing?” which is obvious in retrospect. It doesn’t objectively matter whether you are married, rich, poor, or any other [insert life circumstance]here for your happiness - because they are not full-time states - they are only things which are experienced when your mind or cognitive attention is brought to bear on them for those moments.

Although it sort of makes sense that in spite of all this, people continue to pursue their life narratives by ignoring experienced wellbeing and chasing remembered wellbeing, because that’s what they get to keep. What I find extremely interesting though, is that when people were asked to choose between experienced happiness and remembered happiness, most people choose experienced happiness for longer time frames (eg one’s lifetime), but not for shorter time frames (eg next hour). Since people typically live moment to moment, these findings imply that people may end up living a different version of happiness than what they believe is a happy life (ref).


Predicted Utility

predicted utility

So we have covered the two primary methods that are used to try to capture the wavering of positive and negative emotions over a time period. The first method of measurement was remembered utility (or life satisfaction) where after the episode we retrospectively and cognitively try to remember and summarize the episode into a single value. The second method of measurement was experienced utility where many measurements are taking during the episode which give us a more accurate picture.

But there is a third method called Expected Utility, or Predicted Utility (ref) which I’ll cover here for completion. Although not used formally to try to accurately measure well-being over an episode like we were interested in with the other two, it is used a frequently by people everyday to make choices about the future.

Predicted Utility is where we predict before an episode how much positive affect and negative affect it will result in. As you would expect, this is basically what we are doing most of the time when we think about events in the future and whether we should pursue them or not, and it is how we make a lot of major decisions.

A lot of research in this area came about due to the efforts of the Psychologist Daniel Gilbert, who aptly labeled the field Affective Forecasting, and covered a lot of the research in his fascinating book Stumbling on Happiness (SOH) which a most of the research below is derived from.

As we hinted earlier, the affective states of pleasure and pain are signals to Approach or Avoid rewards and threats in the environment. In the experienced utility account - we experience these pleasures and pains as an immediate response to our environment (or thoughts). In the remembered utility account - our evolve capacity for memory allows us to store the most intense and latest (peak-end) pleasures and pains of an episode and save it for later.

But in the predicted utility account, the most recently evolved region of our brain - the prefrontal cortex - allows us to simulate, visualize, forecast, or predict what the pleasure and pain of a future event will be, without having done it (ref). This is obviously useful for avoiding dangerous situations without having to physically learn from those mistakes, and it also allows us to simulate ideal futures. Unfortunately this capacity didn’t evolve to be perfectly accurate, and has biases similar to the devastating peak-end bias of remembered utility.

We spend most of our time constructing lives that we think will make our future selves happy. We don’t live each moment as if it was the last, but rather we go easy on the unhealthy but pleasurable activities in expectation that our lives will continue.

Unfortunately, although we spend almost all of our time planning for our future selves, the mistakes we make when we try to predict our futures are as systematic and biased as optical illusions (ref).

Humans can think about the future in a way that no other animal can or ever has, and this is a defining feature of our humanity (ref). We have the capacity for abstract thought - the ability to construct mental perceptions about phenomena which isn’t immediately perceived by the senses. These mental predictions about the future arrive in our consciousness regularly and without choice - making up about 12% of our daily thoughts (SOH P17).

Thinking about the future can be a source of pleasure - indeed most thoughts about the future are people imagining themselves succeeding (SOH P18) and can also minimize the negative impact of pains that we encounter. So it can be a source of pleasure and prevent pain.

But the most important reason that our brains constantly churn out these predictions is so that we can spare ourselves the hard lessons of experience, whereas other animals must experience an event in order to learn about its pleasures and pains but by predicting. By doing this - we can control our future. And the reason is because humans crave control over their future (ref), and if they lose their ability to control things at any point in their lives they become unhappy, helpless, hopeless, and depressed (ref).

We should want to control our experiences because some futures are better than others, and we try to predict which. Unfortunately we are very bad at predicting which, and the future always turns out to be fundamentally different than it appears for the predicting mind. And these biases in predicting the future are why we often fail to know what will make us happy in the future. These failures apply to small decisions, as well as big such as whom to marry, where to work, when to reproduce, where to retire (SOH P85) - and these decisions are based to a large extent on our predicted utility for them (ref).

The main biases that we exhibit when predicting that future reality are the following:

  1. Focalism: We don’t predict the future as it actually will be.
  2. We simplify it to a particular image based on a particular memory. In the process - we add in details that weren’t in the real memory (“filling in trick”`), and exclude important details that were.
  3. Your estimated enjoyment of the future is based on the simulated image, not in the reality. We do the predictions so quickly and effortlessly and assume that the predictions are accurate representations of reality when they are made up and usually based on simplified memories
  4. Our predictions based on memories only include things that DID happen and not the things that DIDN’T happen - which can happen in the future and drastically affect our predicted response
  5. Presentism: We predicts that an event in the future will make us feel the way it does while we predict it in the present
  6. We find it difficult to imagine that we will ever think, want, or feel differently than we do today (ref)
  7. We compare the future to the past instead of generating new possibilities, but even if we do that we still make errors because we compare it to our current situation rather than our future situation
  8. Impact Bias: Positive and negative events don’t last as long or effect us as much as we predict them to. This is related to the psychoanalyst Lacan’s concept of Jouissance - where we get a milder “jouissance obtained” in place of the full satisfaction of a “jouissance expected” - so attaining pleasures or fulfilling our desire is never as great as we expect.

Some interesting research that demonstrates these 3 biases are:

  • People seem to overestimate the effects of past events on their happiness (ref), which may explain errors in prediction
  • As mentioned, we base our predictions on our memories, but the memories we base them on are the atypical memorable ones rather than the representative (ref)
  • Losses do not actually loom larger than gains like we predict (ref)
  • We receive greater utility from segregating a pleasure into smaller pleasures (eg spreading it over time), but only to an extent (ref)
  • People enjoy interacting with outgroups more than they expect (ref)
  • Thinking about the absence of a positive event from one’s life (eg imagining that they never met their romantic partner) improved their affect and made them appreciate more than the presence of a positive event (eg thinking about how they met their partner) (ref)
  • People expect to be pleased by punishing/revenging against an offender, but they feel worse (ref)
  • Having extraordinary experiences are enjoyable in the moment but leave you feeling worse after than only having had ordinary experiences instead (ref)
  • People predict to be more happy with changeable decisions than unchangeable, but it’s the opposite (ref)
  • People place greater weight on the physical features rather than social features of future housing, demonstrating the impact bias (ref)
  • People fail to predict how much they will enjoy exercise due to the unpleasant beginning of it (ref)
  • People fail to predict the mood benefits of positive self representation (eg putting your best face forward) in social interactions, especially with their romantic partner, and also how good they would feel after being judged positively by others (ref)
  • People fail to predict the effect of their own personalities on their future affect (ref)
  • And people enjoy thinking about the absence of positive life events more than the absence of negative events, but failed to predict this (ref)

Fortunately, Daniel Gilbert proposes a solution which we’ve already been relying on: rather than trying to predict what will make us happy in the future which is prone to biases and error, we should instead use others and surrogates and look at those that are currently having the experiences we are predicting about, and see how they feel (ref). Doing this avoids the faults of prediction, and allows people to make remarkably accurate predictions.

Thankfully we’ve already covered a lot of how people feel when doing various activities in the previous section on experienced utility, and the from Gilbert is to base your decisions on this research. Unfortunately people don’t use this technique, even when showed it’s accuracy. The reason is because we don’t see ourselves as average, but rather as unique and different, and because we think we have a unique viewpoint into our emotions that others can’t see.

So in summary of predicted utility, there is a lot of evidence that shows how people overestimate the affective impact of future events - overestimating the intensity and duration of both pleasures and pains (ref). We imagine the future will be less similar to the present than it will be, we exclude relevant details and add in made up details to our forecasts. All in all, this research leads us to the conclusion that forecasts about how we will feel in the future are unreliable and we should instead rely on experienced utility reports.


Minimizing Pains

minimizing pains

So up until this point, the primary focus has been on the various ways that we can go about maximizing pleasure and positive emotions in our life. The two primary ways we can do so are by improving our cognitive reports of life satisfaction, or our lived experienced of emotions. Fortunately, a lot of that research also implicitly gives us a way to minimize pain. By maximizing net affect we will also be lowering negative affect. But I’ll cover a few other methods here that I think are particularly relevant for minimizing pains.

As we know, pains can either be physical or mental in nature. Fortunately, the ways in which to minimize physical pains (the pains listed in the chart earlier) are quite obvious, so I’m not going to beat a dead horse. And remember that Epicurus had a focus on minimizing physical pains as well. Most of the pains which are physical are usually about maintaining physical homeostasis - by removing hunger, thirst, tiredness, and avoiding tissue damage. What’s nicer about these pains than the mental is that although they can be more intense, they are usually only experienced in the present and are fleeting, resulting in a less cumulative effect on our happiness.

The pains that are of more concern are the mental pains, what the Stoics and Buddhists wanted to eliminate. These can be unfulfilled desires, obstructed goals, an array of negative emotional states, social pains, or any of the other in the large list mentioned earlier.

I think it’s very easy to go down the path of the philosophers Arthur Schopenhauer or Peter Zapffe and proclaim what the Buddhists did - that life is constant suffering. It can even be elaborately argued that logically we experience more pains than pleasures, and how pleasures are fleeting but pains quite consistent, or that even if pleasures are good and pains are bad that we always have more pain, etc. This can lead to viewpoints like Antinatalism which aims to extinguish human sentience to remove the possibility of suffering, or that it’s better to never have been born.

“There’s nothing left to do but find it humorous” - Emil Cioran

In fact, this asymmetry of pleasure/pain probably fits with most people’s intuition. Philosopher Robert Nozick had a well-known thought experiment where he prompted people on whether they would get inside a “pleasure (maximizing) machine” - a machine that gives you everything you want and simulates happiness. There’s nuances for and against the argument, but the main idea was to critique hedonism as a worthy pursuit. Most people say that they wouldn’t get in the machine because it lacks meaning, or isn’t real, or some other thing. But I think what might be a more question to ask is whether people would get in a pain minimizing machine - that is, it removes all suffering and the ability to suffer. I would guess that this is likely to be a more attractive alternative to most (ref)

I used to even actually empathize with some of the logic in these arguments - in particular I agreed that humans had a negativity bias where humans had a fundamental asymmetry in how they weighted positive and negative emotions and that it would be more evolutionarily adaptive to experience more negative emotions than positive, as negative emotions help us avoid severe threats to our survival, and that humans evolved to be adaptation executers (ref) not happiness maximizers. But although this argument seems philosophically persuading, the data consistently demonstrates the humans display a positivity bias rather than a negativity bias (ref). Regardless of life circumstances and what is happening throughout a day, except under extreme circumstances, individuals generally report high life satisfaction and more positive affect than negative affect. We also remember positive memories longer than negative memories (ref) - not all negativity, but most, but particularly negative experiences can be magnified in memory.

Even if that were the case, as we continue to live, some lives will continue to be better and happier than others - containing more pleasure and less pain, and the question is what those lives are rather than fall into hopelessness and a self-defeating mode of living. We can’t completely do away with thoughts. Even if we were to get a lobotomy to remove our ability to have mental pains, we would also be removing our ability for mental pleasures.

The one standout and recurring point across these pessimistic philosophers though which isn’t invalidated by the point above is their emphasis on death anxiety. As we saw, humans are the only animals that have the capacity to think about the future. The horrific result of this is that they also have the ability to contemplate their own mortality, and thus their insignificance. This was Zapffe’s main point - that because humans can suffer in their mind like no other animal can (and especially about their mortality), we would be better off to be sub-primates. We did acknowledge this as a pain in our list of pleasures/pains under “death anxiety”. He thinks we have an ‘overdeveloped consciousness’ which leads us to be in a constant state of mental pain about things that don’t exist in the present.

Unfortunately the research actually supports this as we saw - people spend around 47% of waking hours ruminating, and that rumination makes them unhappy (ref, ref). That holds even if people are thinking about positive things - it makes them less happy than if we were to be in the present. People would actually prefer to electric shock themselves instead of being alone with their thoughts (ref). “How often our minds leave the present and where they tend to go is a better predictor of our happiness than the activities in which we are engaged”.

So we know that pains are bad. We know that pains can be physical or mental, but the physical are relatively less frequent or enduring. And we know that we spend about 47% of hour waking ours in pain due to ruminating and not being in the present. So far this is quite a bleak view that confirms what the Buddhists have been saying for a couple thousand years - that existence is mostly suffering. What can we do about it?

Zapffe argued that there are 4 ways that we can counteract our ruminating mind (ref):

  1. Isolation - Isolation is a voluntary dismissal of all conscious thought. This is essentially ascenticism - or what the buddhists and most eastern philosopher’s focused on in the form of mindfulness / transcendental meditation. It also looks like this is what Alcohol or depressants do by being a thought-remover in a bottle or pill.
  2. Anchoring - Anchoring is whereby we consistently focus our attention on an ideal or goal - whether it be god, the state, morality, travelling, or sport. They’re all things that we can anchor our attention to and gives us something to strive towards.
  3. Distraction - He thinks that distraction is the most popular in western society as it’s the easiest. It’s where we limit our attention on thought by bombarding it with information - whether that’s consuming music, film, books, socializing, overeating, visual arts, etc.
  4. Sublimation - Lastly is sublimation, which is what me writing this essay would be categorized as. It refocuses our mental energy from inward rumination to outward expression in writing, painting, or creating. It requires the most effort and is the most ‘productive’ per se, but isn’t superior to the others and has the same function of eliminating thought.

Unfortunately this model is not evidence based at all and so makes it essentially useless. But it does give us an interesting model. The model is the following: that cognitive reflection is mostly bad for our wellbeing, and that we should go about trying to reduce it. Schopenhauer actually talks about this same thing and calls it the Sublime - which is the loss of self-reflective consciousness (ref).

I will say that his premise about the psychological conflict to our death has been supported. Sociologists have research into what’s called Terror Management Theory which has a lot of findings showing that humans have a basic psychological conflict resulting from having a desire to live but realizing that death is inevitable. There’s also many ways that humans safeguard against this terror - by ignoring it, or building cultural institutions like politics, religion, nations, and sport to counteract the conflict. It supports Zapffe’s point that people attempt to avoid their psychological conflict by doing various things.

I don’t agree with Zapffe’s direct categorization but I do agree that there is a unifying theme here in how humans aim to eliminate mental pains. The instances of that theme are that people drink Alcohol to be in the moment, they meditate and practice mindfulness, they get into flow in various activities (covered further below), and other.

So granted that we spend most of our day suffering mentally, and that is bad, it is worth doing something about it. I think there are a handful of useful methods for doing so:

  • Situation Selection
  • Solving Problems
  • Challenging Negative Beliefs
  • Journaling
  • Mindfulness
  • Flow State

But before we get into all of those further, I want to mention that isolating as Zapffe puts it is not the way to go. Isolation in the direct sense of trying to voluntarily not think negative thoughts. Not only did this give happiness research a bad rep - ie the idea of choosing to feel happy. But there is also research on how attempts to suppress negative thoughts actually makes them more present (ref). So rather than directly trying to suppress them, a better method is to accept them or mitigate them indirectly as per the methods below.

Situation Selection

We have already noted how behavioral habits can be important because they can act as looping experiences which compound. Another thing you could do is modify your environmental triggers so you aren’t exposed to triggers of bad emotional states, and are exposed to good ones. For example, if you found that going on certain apps or websites made you unhappy, it would be wise to stop doing so. In fact, some cool research suggests that situation selection - doing activities and being places that make you feel good, and avoiding people who make you feel bad, is an effective intervention for reducing negative emotions. But we can also use it to increase the positive.

That is, we can reduce negative emotions by avoiding the environmental triggers for them.

situation selection

Unfortunately although this may work directly for environmental triggers of negative emotions, it isn’t very helpful for solving the main issue laid out here - that we spent 47% of our day in thought.

Solving Problems

This one is quite obvious but worth mentioning. Again, it’s not directly helpful for minimising negative thoughts, but if there are glaring problems in your life that are causing recurring negative emotional experiences - whether they be big or small, it’s probably wise to figure out a way to remove those pains or solve those problems.

In support of the deliberately focusing effort in removing pains from your life, research suggests that daily hassles exert a remarkable downward force on our happiness. Psychological distress over a 9 month period was predicted less by major life events than by day-to-day hassles, ranging from sexual problems to annoying neighbours, to filling out forms, to vacuuming (ref).

Challenging Negative Beliefs

“Pain is inevitable, but suffering is optional” - Haruki Murakami

Challenging negative beliefs won’t help us reduce all of the negative affect that results from 47% of our day spent ruminating - because remember that it makes us unhappy even when it is about positive things. But it does make us more unhappy when we are thinking about negative things.

And there is an abundant amount of research supporting Cognitive Behavioral Therapy as a method of reducing the intensity and frequency of negative thoughts - and how they effect our sense of efficacy in fixing a problem. In fact, it’s one of the most used and well supported forms of therapy for depression.

I’m sure you’ve all heard of learned helplessness at this point, well it turns out that there is also a thing called learned optimism where you can train yourself to overcome setbacks. And it works basically the same as CBT.

The idea of learned optimism is that recurring negative thought patterns are learned responses to the environment, and those learnings can be undone. It can be undone by deliberately challenging the accuracy of the thoughts, and how the subject feels about themselves, about others, and the future. It turns out that the main feature of helpless thoughts is that when a negative setback in our life occurs, it’s explained as Permanent, Pervasive, and Personal - ie forever, everywhere, and your fault. In opposition, the feature of hopeful explanations to these setbacks is that they are Temporary, Local, and External - ie short lived, to this one particular instance, and not your fault.

The main exercise to challenge the setbacks is that whenever they occur, you write them down on paper and do the ABCDE exercise - this is basically self-CBT. ABCDE stands for Adversity, Beliefs, Consequences, Disputing, and Effects. For Adversity, you write down the event that triggered the negative emotional response - not how you reacted to it but the event itself. For Beliefs, you write down your beliefs about the event - what you thought happened. For Consequences, you write down the behavioral and emotional responses to those beliefs. Disputing is the key step, and for disputing you actively challenge the accuracy of the thought - does it fit with reality? what is the evidence for your belief? is it possible that there’s another explanation? It turns out that most of the time the extremely negative thoughts are not actually based in physical reality but made up in our beliefs. And then observe the effects once disputed - the negative emotions should be dampened.

Journaling

As argued in the book Learned Optimism, journaling your negative thoughts is an effective way to prevent them from recurring. It won’t obliterate the 47% rumination figure, but it should help somewhat. The argument is that the reason negative thoughts (especially recurring ones) continually insert themselves into our consciousness is because they are reminders to us. But by simply writing them down on paper - we have for all intents and purposes “acted” on them and ‘told our brain’ that we have saved that information for later, and that there is no longer any need to remind us.

Mindfulness

As we know, a wondering mind is not a happy mind (ref). Being present in the moment has been argued to be the solution to this conundrum for a long time, across many cultures and individuals. But telling people to be present probably does close to nothing.

Mindfulness is the process of bringing one’s attention to the present moment. As we mentioned, telling people to be present is useless, so what we need to do is have a training system, and as we all know that training system is Meditation. Daily meditation is an effective method of improving the ability to be mindful.

It’s been confirmed that meditation is effective at reducing the intensity of pain (ref).

Mindfulness and Flow (below) are what I consider to be the 2 most effective methods of minimizing mental pains.

Flow

Flow (what some research calls Engagement) is the process of being so absorbed in a task that we lose self-reflective consciousness (ref). It is when we have intense and focused concentration on the present moment where we are intrinsically rewarded, and lose track of space and time (ref). It is being one with the music, with the interaction, with the activity, with the game. It is where time flows like water. It is where you are challenged and enjoy it (ref).

Besides Mindfulness in the previous section, flow is the most research-based and supported method of minimizing negative thoughts. If you are completely engaged in the task - then you don’t have the capacity for negative thought and it is effectively removed. It is different from mindfulness - flow is about the demands of the task being so high that it occupies all of our attention and leaves little attention left for thought, whereas mindfulness is the deliberate process of focusing our attention on the present. Mindfulness intuitively seems more difficult and less reliable, whereas controlling what tasks we undertake is more within our control.

Flow fits more in with the research of the modern happiness research we’ve laid out than the other methods. It’s highlighted in many papers and is actually the pet theory of the researcher Mihaly Csikszentmihalyo who also developed the experience sampling method.

The mechanism of flow is postulated to be that human attention can only process 110 bits of information per second. As a benchmark, processing speech takes about 60 bits of information per second. When we go over that limit, we lose awareness of other things as we have no more attention to be allocated to them (ref) - including our negative thoughts. This is obvious to anyone who has tried having two conversations at once or got lost in time while playing chess or another mentally exhausting game.

Flow theory stipulates 3 conditions that must be met in order to be in a flow state:

  1. The activity must have a clear set of goals and progress
  2. The task must have clear and immediate feedback on how well you are doing
  3. There must be a good balance between the perceived challenges of the task and their own perceived skills

flow chart

One big perk of flow is that it is infinitely renewable. As opposed to a physical pleasure like eating that reaches a threshold, or the diminishing returns that come with many of the other pleasures, there is no threshold or diminishing returns for flow state. You can continually focus your attention in a consistent manner.

For a full exploration of flow, I highly recommend the short (150pg) book by Mihaly called Finding Flow, where he covers every nuance of it and an interesting categorisation of time-use amongst humans.


Improving Hedonic Setpoint

As we saw for the life satisfaction, regardless of the majority of the life events that occur, most people return back to a “setpoint” of happiness. We return back to our setpoint relatively quickly after an even of pleasure or pain, or even major life events, and we spend a lot of time in that setpoint. But the set points that people return back to aren’t the same - different people have different setpoints (ref). What this hints at is that improving our setpoint is potentially a large-impact way of improving our happiness, as it will have huge cumulative gains over our life.

Unfortunately the setpoint seems to be set at a genetically determined level. This is why genetics plays such a crucial role in both life satisfaction reports and affect reports (along with depression, moods, and every other psychological process). We are biological creatures, and as such our biology has an effect on our moods and perception. Given this, it’s not an understatement to say that the happiness in one’s life is to a large extent determined by the balance of the relevant brain chemicals / neurotransmitters.

But what exactly is the role that our genes have on our moods and reports? All genes do is encode proteins, and so there must be a chemical difference between the people that have a higher setpoint versus a lower setpoint. The main culprit for the longest time has been the neurotransmitter serotonin. I’m not qualified enough to know whether this research still stands, but apparently serotonin was the culprit to the extent that when they found the a specific gene that encoded for serotonin, a variant of it was called the depression gene. More accurately, it’s a gene that transports serotonin (ie “serotonin transporter gene”) called SLC6A4. And even more accurately, it’s not the gene itself but the “5-HTTLPR polymorphrism in the promotor region of that gene”. Individuals who carried this polymorphism were more vulnerable to depression following exposure to stressful life events (ref), with the events being employment, financial, housing, health, and relationship stressors, along with stressful early life events. This gene X environment (ref) interaction explains why stressful events leads to depression in some people but not others.

By understanding the neurochemical differences in people with different setpoints, perhaps we can take it within our control to improve our setpoint. In fact, some researchers have even suggested that we give anti-depressant drugs to healthy individuals to do just that - in what is called ‘cosmetic pharmacology‘ or what I like to call “neurohedonism”. If it works for depressed people to get them from -6 to -3, why not use it on the non-depressed to get them from 3 to 6? This is just another case of performance enhancement - just like we use supplements to help our athletic performance, why not take supplements that improve our wellbeing?

We already know for a fact the importance of biology in various mood disorders and it’s usefulness for treatment. As a few examples: how much physical intimacy one receives as a child as a big predict for later life mood disorders (ref), the strong positive effects of engaging in regular exercise on depression (ref), the ability of excessive pornography to deplete one’s reward system (ref), the importance of delaying gratification and how it’s moderated by serotonin levels (ref), how an omega 3 deficiency can result in mood disorders (ref), how getting a massage increases serotonin and dopamine and decreases cortisol (ref), the role of sleep in mood disorders, serotonin can be increased with blue light / exercise / and meditation (ref).

So some brain chemicals are good, and we want to go about expressing particular neurotransmitters and reducing others. But which are they? Although there are 100 neurotransmitters, only a handful seem to be implicated in mood and depression:

  • Cortisol and Glucocorticoids are heavily implemented in the stress response (ref)
  • Dopamine is involved in arousal, motivation, positive reinforcement, and reward perception. Low levels result in depression and lack of motivation. It’s implicated in mood (ref), but more interestingly it is believed that high dopamine is associated with positive affect (ref, ref)
  • Serotonin is involved in sleep, appetite, mood, and feelings of wellbeing. Low levels result in depression. It’s implicated in mood (ref)
  • Norepinephrine and Epinephrine are responsible for alertness, wakefulness, arousal, and focus. Norepinephrine is implicated in mood (ref)
  • Opioids / Endorphins / Dynorphin
  • Substance P is a pain neurotransmitter
  • Acetylcholine is responsible for learning, memory, attention, and alertness upon waking
  • GABA is responsible for inhibition and calmness. Low levels result in anxiety and ADHD.

This is really oversimplified neurochemistry, and it would be stupid to suggest that you go about maximizing the “good ones” and minimizing the “bad ones”, as they have functional roles which would be detrimental if removed. But it does raise the interesting question of which are better/good and which are worse/bad, and how to improve and reduce them, and which neurotransmitters are implicated in well-being. After all, it is very real that you can take certain drugs and attain extremely high positive effect for short periods of time.

You could take ritalin, adderall, mdma, heroin, opioids, coffee (ref), dexamphetamines, cigarettes, you name it. Or you could take those anti-depressants like I mentioned in cosmetic pharmacology - wellbutrin (NDRI), prozac (SSRI), etc. And you’d feel great for a short period of time.

Beyond the safety of drugs, the problem is what we all know which is you build up a tolerance to drug use. The receptors in your brain become less sensitive to the drug and you continually need more and more of the drug to receive the same mood boost. Obviously this is not sustainable long-term to maximize happiness over a lifespan. The actual mechanism of this ‘downregulation‘ is that as you increase the signaling of a neurotransmitter (ie release), it binds to the receiving receptor which then decreases how much of the receptor is created. The way that it does this is three-fold: if there is too much signalling then the receiving cell can actually die (ref, ref), it can prevent more of the transmitter or receptor being made, or it can actively remove the existing receptors from the cell membrane.

Perhaps rather than taking drugs which increase the release/signaling of neurotransmitters, a more sustainable method is to increase the ingredients required by our body to make the neurotransmitters (ie biosynthesis), which can indirectly increase them. So besides our genes which determine the speed and quantity we make neurotransmitters, another bottleneck to their creation is the raw ingredients available. For instance, most of the neurotransmitters we mentioned are made in our cells with the ingredients being an amino acid retrieved from the breakdown from protein in our diet (“precursors”), along with some vitamins (“cofactors”):

  • Dopamine uses the amino acid l-phenylalanine as a precursor and b6, b12, SAMe, and vitamin C as cofactors which gets converted into -> l-tyrosine -> l-dopa -> and then into dopamine.
  • Serotonin uses the amino acid l-tryptophan as a precursor and b6 and b9 as cofactors which gets converted into 5-htp and then serotonin.
  • Norepinephrine uses dopamine as a precursor and vitamin c and b6 as cofactors.
  • Epinephrine uses norepinephrine as a precursor and SAME / b6 as cofactors.
  • Acetylcholine uses the nutrient choline as a precursor and b6 as a cofactor.
  • GABA uses the amino acid l-glutamine as a precursor and b6 as a cofactor

So perhaps we can shift the balance of our neurotransmitters by upping the precursors and cofactors needed to make the neurotransmitters that we desire more of? This will work for a short amount of time. It turns out that for neurotransmitter systems, the opioid, gamma-aminobutyric, and endocannibinoid systems in the nucleus accumbens, ventral pallidum, and orbitofrontal cortex are what mediate the perception of rewards (ref). It turns out that dopamine is more related to learning and motivation, but opiates are what actually constitute the reward. If you block opiate function in the brain, net affect greatly increases (ref). Your endogenous / internal levels of opiates are what determines your hedonic set point (ref, ref).

Obviously this doesn’t mean we should all become opioid users, we’re all aware of the tragic effects of the current opioid epidemic (ref) and how addictive it is. Now we know why - it is the biological definition of rewarding.

Unfortunately regardless of whether you directly increase the release of neurotransmitters with recreational drugs, whether you prevent the destroyal and reuptake of them with antidepressant drugs, or whether you promote the creation of more neuotransmitters, it turns out that your genes have still set an arbitrary threshold on your neurotransmitters and they will continually return to a homeostatic level (ref). One except may be that vitamin D levels regulate the speed of serotonin synthesis in your body (ref), and is assisted by omega 3s (ref).

There’s other factors besides neurotransmitters of potential importance. Another standout factor is brain structure. There’s a positive association between the volume of gray matter (in the right precuneus area) and subjective wellbeing (ref), with the prefrontal cortex (ref), orbitofrontal cortex and dorsolateral cortex being involved in the experience of pleasure (ref). Other evidence suggests that the presence of positive affect vs negative affect is determined by the balance of activity between the left and right frontal hemispheres of the brain (ref). Or that hedonic tone (basically setpoint) is associated with ‘left supero-lateral medial forebrain bundle microstructure’ (ref).

Unfortunately most of these structures aren’t easily modified as you would imagine. The one area where there is promise is the research showing how meditation can lead to a significant gray matter increase within the precuneus (ref, ref, ref, ref, ref, ref, ref). Not only that but meditation also improves some neurotransmitters we want and reduces others (ref, ref, ref, ref).

Other potential factors are: working memory has been shown to play an important role in emotional processing (ref) and low ‘hedonic tone’ (ref).

One other potential improvement to the brain structures and neurotransmitter systems is by electrically stimulating certain parts of the brain (“deep brain stimulation”) (ref). Obviously this is quite extreme and not applicable to most.


Summary

So what are we to make of all this? How can we go about taking this deluge of information and try to instill it into some actionable insights into how to life the best life?

The Psychologist Martin Seligman attempted to define a framework which captured the components of a happy life - called PERMA, standing for the presence of Positive Emotions, Engagement (ie flow), Positive Relationships, Meaning, and Achievement. He was trying to avoid being cornered into defining happiest as strictly ‘positive emotions’ - ‘negative emotions’ as we have for the most part. He tried taking a more multifaceted approach with the 5 criteria. His criteria for counting each of these as an element of “flourishing“ was that: they contribute to wellbeing, they are intrinsically motivating (people pursue them as an end and not a means), and they are defined and measured independently from the others. Positive emotions consistent of socializing, savoring, and mindfulness. Engagement consists of using your “signature strengths” in relationships, work, and play. Meaning is found in contributing to a cause greater than your individual self. But unfortunately the model has not been empirically validated. Either way, we have already covered all of these elements separately so we will not give any weight to his model which is quite an arbitrary grouping which excludes important factors. He does acknowledge that his model isn’t empirical but theoretical.

But I do think he was onto something which is that we must take a multifaceted approach to happiness, just like I’ve taken a multifaceted approach to job-satisfaction and sleep. Just as a satisfying isn’t merely one that pays well as people expect, but a mentally challenging one, with good coworkers, that suits your personality, that is close to home, in a small org, etc. And just as a good sleep isn’t merely getting enough hours but is also regular, cool, dark, quiet, etc. So too is happiness not just X but is also a multifaceted phenomena. The universe is complex and so too is your happiness.

But there is one thing that all inquiries into happiness have in common and that is that they all share the same common theme of trying to find an optimal set of outcomes.

The methods that most people use to solve this question are either by relying on their own experience (informally via trial and error or formally as self-experimentation) or looking at the happiness stories of other people (in stories, movies, art, conversation, observation, etc). As we’ve seen throughout this essay, they are wholly dubious methods and instead we should rely on the scientific method to figure out using large sample sizes and carefully defined terms and interventions what constitutes this slippery term happiness.

The scientific method uses a systematic set of observations as an attempt to quantify and arrive at answers which go beyond a single person, and in a more standardised manner. Such tools are valuable because they can overcome biases / blind spots individuals may have by using larger datasets of information. There are a handful of scientific tools that can be used - questionnaires, brain scans, facial expressions, and others. The simple questionnaire has dominated a lot of the happiness research on the assumption that people know how they feel when they say so. Obviously this is wrong in some cases, but it is correct in others. We are lucky that we now no longer need to rely on introspecting or making conjectures about happiness, but have the ability to peruse the large body of literature on the subject.

The main drawback of relying on our personal experience is that we often lack awareness into what makes us feel good and bad, and even if we know we don’t usually crystallize in language which is helpful for deciding we want more or less of it.

What we do know is that our lives are experienced as a continuous flow of pleasurable and painful emotions. By summing those up, we arrive at a happiness index of sorts - to some extent.

The nature of organisms including humans to be pain avoiders and pleasure seekers is an evolutionary imperative. As you look over the comprehensive pleasure/pain list in the definition section, you can see that nearly all of the pains either motivate the organism to avoid a threat, or reward the organism for removing it or improving its odds of survival or reproduction. That is true for all of the biological pleasure/pains which return the organism to homeostasis, all of the social pleasures/pains which are important at the group-selection level, or even for financial security / novelty / learning / health which ensures that the organism maintains its homeostatic state for the foreseeable future. This same function is true of our basic emotions.

So our experience of pleasures and pains has an evolutionary function. Pleasures are subjectively good, rewarding, desirable, positive, and something we should approach. Pains are subjectively bad, punishing, aversive, negative, and something we should avoid.

Hedonism takes a step up on this acknowledgement and tries to take these pleasures and pains within our control. It argues that pleasure is the only intrinsically valuable and good thing,and pain the only intrinsic bad. Regardless of what activities you think constitute happiness, what’s agreed on is that some emotional states are better than others, and all else being equal happiness can be increased by spending more time in the good states and less time in the bad. Time is a finite resource and we want to spend it in states that are valuable. In fact, the control we have over the use of our time is one of the biggest ways we can improve our happiness (ref, ref).

It turns out that humans subjectively experience the moment of the “present moment” in 3 second intervals. An individual life can be described as a string of these 3 moments - with approx 20k a day, or 500m in a 70 year life. Most of them simply disappear, but while they last they are rich and multidimensional, and are composed of pleasure and pain which are important to us.

So pleasure isn’t happiness and pain isn’t unhappiness, they are individual states. But the more pleasures we have and the less pains the better. What we care about isn’t the individual pleasures, but the sum total of pleasures/pains over a lifespan - the net. That net can be mathematically calculated by (intensity x duration x quantity of pleasure) - (intensity x duration x quantity of pain). These pleasures and pains can be physical or mental.

An episode is an arbitrary period of time that we sum pleasures and pains over, from 3 seconds to a lifespan. A Util is the basic basic unit of pleasure/pain. Instant Utilities are what people experience moment to moment, Remembered Utilities are what people remember their pleasure / pain utils to be over an episode, Experienced Utilities are what people report their utils to be in real-time when asked, and Predicted Utilities are what people expect their utils to be over an episode in the future.

With the introduction of episodes, not all pleasures/pains are equal. Pleasures/pains differ over an episode in their intensity, duration, certainty/uncertainty of attainment, proximity, repeatability, and purity.

Although not directly applicable, this model does give us 4 main theoretical methods of improving our happiness: by maximizing pains, minimizing pleasures, maximizing pleasures, and minimizing pains.

Minimizing pleasures is an obviously bad strategy.

Maximizing pain seems like an obviously bad strategy, and it directly us. But in some cases pains are impure and followed by pleasures in which case it can be wise to consider them. As we saw, pains are evolutionary flags to motivate us to remove them. So although pains are directly bad, they are also required for some rewards. The potential (low confidence) candidates for which pains may result in large gains later are mastery and achievement.

But the most effective routes to improving our happiness index is by maximizing pleasure and minimizing pain.

There are 3 conceptual methods to improve our happiness: by repeating small (short and low intensity) pleasures frequently (which results in a large net sum), by focusing on large pleasure events less frequently (intense and longer), or by savoring / prolonging episodes.

Repeatable small events is interesting, because after all our lives are the sum of the small events that we do repeatedly. And the better predictor of your wellbeing does turn out to not be the intensity of the pleasures but rather the frequency and duration.

We could also focus on the few big payoffs - buying the house, getting married, becoming the best, having the kid, etc. But they are less frequent and as we saw above the intensity is less important.

We can also intentionally savor our pleasures by being mindful, anticipating, reminiscing, sharing them socially, and not dampening them. This could add up to big gains over time. The rich, those with abundant life experience, and the east have a worsened savouring ability. Lacking this ability is lacking a stream of happiness.

But those are just theoretical. What does the science say?

Most of the science is on Remembered Utility. The major determinants are Genetics (which account for 30-40%) - likely via Personality (in particular Extroversion), Optimism, Mental Health, Love, The Fulfillment of Basic Needs (not being in poverty), Job Satisfaction, Sleep Quality, Lack of Disability, and Lack of Major Debt.

The moderate determinants of remembered utility are health, social activity, health of family, religion, mindfulness, sport participation, sex life, being a morning lark, emotional regulation, height, appreciation, and living with your partner.

Weak predictors of remembered utility are how you spend your leisure time, your income, video game usage, home ownership, being a non-smoker, donating, your country, doing yoga, and eating healthy.

Null predictors of remembered utility are your attractiveness, intelligence, gender, age, and parenthood.

So if you really cared about improving your cognitive evaluation of life as much as possible (which I don’t recommend), you’d want to focus on health, love, work, your extroversion, lack of neuroticism, autonomy, and social activity.

Regardless of most extremely positive life events, or extremely negative, most of us adapt back to a baseline setpoint soon after (which is biologically determined). You are not a paraplegic or lottery winner full-time, only insofar as your attention is allocated towards it. The 4 ways that adaptation can be prevented is by: continually savoring, by continually engaging in new and novel experiences, by partaking in pleasurable experiences less frequently, and by altering our biology (discussed below).

But all of these Remembered Utility findings are erroneous in that they are cognitive reports, not emotional. We do them quick and effortlessly, so quick that it’s infeasible that we’re accurately summing our experience. So what are we doing then? We’re likely retrieving a memory. Although it’s not an immediate report of pleasure and pain in our environment, our brain has saved a summary of the experience to decide whether an episode encountered in the past should be approached or avoided again in the future.

Unfortunately it turns out that during the process of summing the experience into a stored memory, we commit some biases in the process which make those memories untrustworthy. The main bias we commit is the Peak-End effect, where the Remembered Utility is a function of the intensity of the experience at it’s peak and towards the end, ignoring the duration (ie duration neglect). Our memories did not evolve to be accurate or to maximize our happiness, they evolved to maximize our odds of survival. And what’s more important to our survival is the intensity of the pleasure/pain which is a measure of the severity of a threat or reward, rather than a duration.

But obviously as creatures which live their lives moment to moment, the duration of our experiences is important to us. Our experiences are arguable more important to us than our memories. In fact, most people choose experienced happiness for longer time frames and remembered happiness for short. But then they chase remembered happiness for longer. Most people end up living a different version of happiness than what they believe is the happy life. Funnily enough, this reminds me of when you read stories of traditionally successful people who remark at ages 40-50 that they realize that there’s no moment that you reach of “complete happiness / grand cognitive satisfaction”. like they expected before they were successful, the ride just continues on. This is both terrifying and enlightening - it doesn’t really matter what big events we do or don’t get, and the smaller events are within our control.

So what we want is to maximizing pleasure that is experienced, not remembered. When we measure experienced happiness throughout the day, here’s what we find:

People are consistently the happiest (in order) when making love, exercising, socializing, eating, taking a walk, listening to music, meditating, preparing food, childcare, relaxing, reading, and watching TV. But regardless of the activity, people are always happier when doing it socially with people they like.

People are consistently unhappiest (in order) when: working, using their home computer, commuting, grooming, listening to the news, and doing housework.

Socially, we’re happiest with friends, then relatives, then spouse / SO, children, clients / customers, coworkers, boss, and alone. Oddly for mothers being with their children is less enjoyable than housework.

Other findings are that sleep quality explains a very large (56%) effect of the difference between people. Similarly large findings for personality and depression. Income only has a modest effect, and religion smaller. Divorced women are happier than married. Time pressure and loud noise has the biggest effect on happiness at work. Facebook makes us unhappy. Passively using social networking sites also, but actively using them less so. The disabled are less happy. Retirees are happier than the working. We’re happier in the morning, and on weekends. Most people report being in a happier state than negative, except a small subset who do most of the suffering. Repeated experiences are more enjoyable than we expect. A headache will make a person miserable. Business students are less happy. Eating fruits and vegetables makes us happier. Only life events from the prev 3 months effect our happiness. Age and education have no effect on happiness. Smartphones make us less happy when we’re with others, but happier alone or during social exclusion. Spending money on others make us happier.

In summary, what causes fluctuations in our positive and negative emotions is complex as you’d expect, but some findings are similar to remember happiness reports: the pervasive effects of personality, depression, and poor sleep quality. Experienced happiness reports are similar but diverge from remembered happiness reports in important ways - primarily that our experienced happiness is a response to our current situation, whereas our remembered happiness is a response to our aspirations. We should trust our experienced happiness much more.

But oddly, if you try to create a ‘perfect day’ based on a duration and intensity weighted assessment of the above findings, what you get is an odd and unrealistic day where you are doing 15 different activities, where you have sex for 2 hours, then socialize, relax, eat, pray, exercise, shop, and watch TV for 1h, and work for only 36 minutes (ref).

Obviously that’s not the take away. The more realistic take away is that we should try our hardest spend less time working, using our computers and phones, commuting, in childcare and house-work, shopping, and watching TV. And we should spend more time with our loved ones, socializing with friends and family, relaxing, savoring our meals, exercising, meditating, going for a walk, and listening to music. We should try to avoid being alone as much as possible. And our sleep is momumentally important.

And for Predicted Utility, since all the predictions depend on our memory which are faulty, obviously they will be even more faulty. Our ability to predict utility has been a huge evolutionary advantage, but again, it’s there to ensure our survival and not our happiness. The biases we commit in predicting are focalism - we simplify the future and base it off inaccurate memories, presentism - we think the future will be less similar to the present than it will be, and impact-bias - positive and negative events in the future don’t last as long and aren’t as intense as we expect them to be. How we think we will feel about the future isn’t reliable and we should depend on Experienced Utility.

So there are 3 sorts of phenomena that we can control in an attempt to improve the overall quality of our lives: our activities and environment, our thoughts, and our brain chemistry. We saw above what the science says on which activities we should spend more time doing and less time doing to be happy.

But the other finding of Experienced Utility was how we spend 47% of our day - 8 hours lost in thought. Regardless of whether those thoughts are about positive, negative, or neutral things, those thoughts make us happy. So it is clear that what we think about has a huge effect on our happiness, or thinking at all. Which leads us into Minimizing Pain.

Thoughts aside for a second, the ways that we can go about minimizing pains in our environment or thoughts are by: situation selection, solving our problems, journaling, and challenging our negative beliefs with CBT. But if we allow ourselves to spend 47% of our day suffering in rumination, we are cutting our happiness in half.

We can go about minimizing our thoughts in 2 primary ways: via mindfulness and flow. Mindfulness is is the intention to be present in each moment, which can be trained via daily meditation. And flow is being so challenged by the activities we are doing that we lose the capacity for thought. We can make sure the conditions for flow are right by always making sure that we are challenged by whatever task that we are doing.

And lastly, when we are not thinking, our happiness in each moment beyond the choice of activity is also controlled by our baseline happiness. We return back to this baseline regardless of most life events - and it’s genetically determined. We know that drugs or supplements could shift us a bit here, but it’s not sustainable long-term. And zapping our brains could work, but we’re not willing to do that. There is one notable thing we can control here which is our grey matter volume, that’s associated with wellbeing, and has been confirmed to increase with meditation.


Compressed Summary

Happiness is multifaceted, like everything else. We should depend on the scientific method to figure out what constitutes happiness. Our lives are composed of a continuous flow of pleasure and pain. We experience the present as 3 second “moments”, with 500m in a 70-year life. Most of them simply disappear, but while they last they are rich and are composed of pleasure and pain. This pleasure / pain dichotomy goes back to the beginning of evolution, and it serves the function of ensuring our survival and reproduction. Pleasures are good and signal approach, pains are bad and signal threat. They can be mental or physical. We want to spend as much of our time in these subjectively experienced ‘good’ states, and as little as possible in the ‘bad’ states. The more good the better, the less bad the worse. Therefore the two main strategies to improve our happiness is to maximize pleasure and minimize pain.

To maximize pleasure, we have a few methods: repeat small pleasures as often as possible, by seeking large pleasures, and by savoring experiences. It turns out that the frequency and duration of pleasures is a better predictor of happiness than the intensity - so seek to be happy a lot, rather than very happy, and savor those moments. To figure out what those moments are, we can measure what we remember them to be, what we experience them to be, or what we predict them to be. Our memories of them are faulty as they suffer the peak-end bias and duration neglect. And our predictions of them are also faulty as they depend on our memories and suffer the impact bias. So we should depend on reports of experience.

Those reports of experienced happiness show that we are most happy when making love, exercising, socializing, eating, meditating, and eating. And we are least happy when working, using our home computers, and commuting. But regardless of what we’re doing, it’s always better when we’re with others. It’s not an understatement to say that happiness is the experience of spending time with others that you like, and less time alone. If you want to be as happy as possible, you should aim to be as social as possible, work and commute as little as you can, practice meditation, get offline, and go be active and spend time eating and socializing with others.

Regardless of our life circumstances, we basically adapt back to a baseline of happiness, and we also adapt to pleasurable moments. This can be prevented by savoring, by partaking in them less frequently, by continually engaging in new and novel experiences, and by practicing mindfulness. You should aim to be as extroverted as possible, and because it makes you more social. Sleep also has an extremely large effect on your happiness.

To minimize pain, you should select your situations carefully, solve any glaring problems, practice journaling, and challenge negative thoughts with CBT on yourself. Outside of how we control our time and thus environment, the other major thing within our control is our thoughts. We suffer in thought for half of the day. If you want to be happy, you should take it seriously to do something about that. But note that pains are often flagposts to what needs our attention, and directly trying to suppress them is counterproductive.

The two most effective ways to minimize thoughts are by practicing mindfulness and being in flow. You can improve your ability to be mindful by meditating daily, and you can maximize your odds of being in flow by making sure you’re being challenged by your environment at all times.


Ultra Compressed Summary

To be happy: focus less on the ‘big pleasures’ in life, and more on the small frequent pleasures. Savor each moment. Spend less time working, commuting, and computing, and more time socializing, and being active. Get good sleep. Practice meditation and mindfulness, and aim to be challenged as much as you can by each moment to get into flow.