Thinking Fast and Slow

This is a phenomenal book. One of the best I've ever read, and the first book that for me coherently ties psychology to every day applications. It changed my assumptions about myself, and transformed the way I approach thinking and decision making.

Below are some notes on my main takeaways. 

System 1 and System 2

The overarching metaphor in the book is that our brain operates at one of two speeds: fast (System 1) or slow (System 2) depending on the task at hand.

System 1 is responsible for quick judgments and fast decision making (what is someone's mood?), System 2 is responsible for tasks where concentration and focus are required (what is the answer to a complicated math question?). 

System 2 can be thought of as "the rational" system, and is generally how humans think of themselves and their decision making (as well as how they expect other humans to behave). System 1 can be thought of as "instinctive" or "emotional" system, and is in reality the one used to make the vast majority of decisions. Its primary purpose is to come up with an answer quickly, and it relies on biases, heuristics, instincts, and other shortcuts to do so. 

System 1 is far more time and energy efficient, and it is generally good (or at least sufficient) for coming up with good decisions. System 2 is not possible to invoke for all our decision making, it is too intensive and would overwhelm our brains if used all the time. An optimum solution would have us rely on our System 1 in situations where it is safe to do so, and invoke our System 2 where it is required. The problem with this, is that we are not always good at recognizing which situations require our attention, and which don't. Furthermore, we don't instinctively recognize how we've made specific decisions, whether we've properly thought them out using System 2, or instinctively jumped to conclusions with System 1. Furthermore, while we would like to believe the strength of our beliefs about something corresponds to how carefully we thought about it, in truth the strength of our convictions is not related to which system was used to arrive at those beliefs. 

Telling Ourselves A Good Story

The most important thing to a person's psychological comfort, is that their views of the world, in their own mind are consistent. It does not matter much if individual views are in themselves rationally justifiable, if they are logical, or where they come from - it matters more whether they fit in consistently with other views we hold. The strength of our conviction about a specific idea depends on how well the idea fits in with our existing views. 

We are very bad at detecting when we are lacking relevant information about something. Our mind tends to evaluate things based on the "What You See Is All There Is" principle, and we tend to make judgments about situations as if we have a full unbiased sets of facts, even when we are explicitly made aware beforehand that this is not the case. 

Loss Aversion

Loss, and potential loss evokes a very powerful emotional response in us. It has been shown that the negative emotional impact of a quantifiable loss, is around twice as strong as the positive emotional impact of a gain of equal magnitude. When we evaluate risks, we tend to weigh risk/reward based on the emotional value of each outcome, not based on the economic (or other quantifiable utility) value. This may be fine for one-off decisions (it is understandable a person will choose something based on how it will make them feel, as opposed to the measurable value of that decision - the two things do not have to be strictly related), but this can get us in trouble when are facing a system or series of decisions to make. 

Consider a series of choice, which have to be taken simultaneously:

1) Choose between:
A. a sure gain of $240
B. 25% chance to gain $1000 and 75% to gain nothing

2) Choose between:
C. sure loss of $750
D. 75% chance to lose $1000 and 25% chance to lose nothing


Stop and consider carefully each choice above and your preference before continuing.


Most people due to their risk aversion for gains, and risk taking to avoid losses will choose the combination AD. This however is obviously far inferior when taken together and weighed against BC.

AD. 25% chance to win $240 and 75% chance to lose $760
BC. 25% chance to win $250 and 75% chance to lose $750

We usually make systems of decisions by combining our preferred choices (which do not necessarily reflect rational expected value) instead of first computing the combinations and making the choices based on that.

Other Biases, Flaws, Heuristics, Instincts, Tendencies, etc...

Substitution: To answer complicated questions, our mind often subconsciously substitutes an easier question (ie Is Apple stock a good investment -> Do I like Apple products?)

Availability: We estimate the likelihood of something happening to be much greater if we think about it often or if it evokes a strong emotional response, than the underlying evidence would suggest, and vice versa. (ie Some people overestimate the likelihood of being killed in a terrorist attack from abroad, but underestimate the effect of loose gun regulation)

Experiencing Self vs Remembering Self

The way we remember our experiences is not the way we experience them. We form our impressions of experiences mostly based on the "peak moment", and the "last moment". The "average moment" or "duration" are largely negligible in our remembering self.

It has been shown that having gone through 2 experiences:

a) a painful experience for 1 minute
b) same as (a) followed by slightly less painful experience for additional 30 seconds

people retrospectively prefer to go through (b), because they remember just the tail end of the experience which was less painful, and they do not factor in duration.


The biggest takeaway is that we should be extremely skeptical of our own rationality. Where possible, we should be aware of what kinds of biases exist in certain situations, and understand how they influence us.  When we make decisions, make sure we are answering the right questions, and not their approximations. When we make a set of decisions, make them based on the combined outcome of the decisions, and not on the outcome of each decision in itself. When we make decisions based on past experiences, we need to understand that our remembering self is making decisions for our experiencing self, and we should reconsider if the way we remember things actually justifies that decision.