<ul><li>Kanheman: we're all trying to create a positive narrative of our life story</li><li>Identity protective cognition</li><li>With the pain / loss, our identity gets attacked if we attribute it to our poor decision making</li><li>So instead, we attribute it to luck</li></ul>

We process the world in a way that protects our identity

<ul><li>Reasoning to be right affirms your priors</li><li>Reasoning to be accurate helps develop a more accurate mental model of the world</li></ul>

Don't reason to be right, reason to be accurate

<ul><li>People are willing to take a discount to preserve their self narrative</li><li>They attribute the poor outcomes to luck</li><li>Socialised the result to the world, ie it wasn't my fault, the outcome wasn't in my control</li><li>You don't need to update your identity in anyway - you need to update your beliefs, or made poor choices or the outcome was something you caused</li><li>This choice is devastating to learning</li></ul>

Most people choose not to feel pain now

<ul><li>Getting a respected peer to think you're good, worthwhile to talk to, an equal --> that feels really good</li><li>The goal of each person is to make better decisions and increase the probability of better outcomes</li><li>Example: I won this hand of poker, but I think I played badly</li><li>The group reinforces learning behavior</li><li>Rewards in the long-term are poor motivators (ie knowing you're a better decision maker in the future isn't a strong reward).</li><li>The short term reward becomes gratification / approval / admiration from the group</li></ul>

A social group can be a strong motivator to get better

<ul><li>We've agreed that our beliefs are in progress and under construction</li><li>You are no longer certain about your beliefs</li><li>You have to be willing to hear the other side in order to make progress</li><li>Your identity becomes about acknowledging uncertainty, how good are you at listening to the other side, how good are you at analysing</li><li>Stops you from having that defensive reaction</li><li>You become agreeable in disagreement because you view the disagreement as helpful</li></ul>

Once you've agreed your beliefs are in progress, they aren't entrenched as your identity

<ul><li>It's much easier to spot someone elses' bias vs your own</li></ul>

You need to be in a group to overcome the bias

<ul><li>Annie uses the example of poker, but it applies to other areas as well. There are uncertain systems where outcomes and decisions are loosely linked.</li><li>In poker, you can have the best hand and get crushed or a poor hand and win</li><li>How do you separate yourself from the outcomes and sift through the noisey feedback?</li><li>You have to figure it out after the fact and after memory has gotten in the way</li></ul>

Outcomes and decisions are loosely linked in some games

<ul><li>I tell a "bad luck" story</li><li>You sympathise and then share your own</li><li>We both agree it was bad luck that was the cause</li><li>Neither one of us learns</li></ul>

Most people enable the second choice

<ul><li>how are you constructing a model of your opponent?</li><li>how are you figuring out what the right strategy against them is?</li><li>how are you picking the signal out?</li><li></li></ul>

An effective approach to uncertain systems is based on cognitive psychology

<ul><li>If you're in a group of people who attribute outcomes to bad luck, you won't learn</li><li>Surrounding yourself with people who constantly seek to update their mental models puts you on a very different learning trajectory</li></ul>

Peer groups are highly influential on your learning

<ul><li>You know a conversation about the decisions you made is going to happen </li><li>You want to have interesting things to ask later</li></ul>

Having this type of group incentivises you to pay more attention to decisions as you make them

<ul><li>Even if you are aware of the biases, you're still susceptible to them</li><li>Smarter people who know about them are even worse because:</li></ul><ol><li>Once they know about the bias, they're overconfident in thinking they can avoid them</li><li>The better you are at slicing data, the better you are at slicing it in a way that tells a story that doesn't sound biased</li></ol>

On our own, we have a very natural tendency to process the world in a way that supports our prior view of the world

<ul><li>I'm willing to take this pain that doesn't feel very good because in the long run it's going to help me learn</li><li>I'm going to go in, examine my decision and see where I could have improved that would have increased the likelihood of a better outcome</li></ul>

Humans have a choice: I can take the pain in the short term in order to learn in the long term or attribute it to luck and not learn

<ul><li>When someone disagrees with us, we can often feel it as an attack on our identity</li><li>That can cause disagreeableness because the information being told is perceived as a threat</li><li>Cause you to:</li><li>be defensive or angry</li><li>be dismissive</li><li>discredit the other person or decide they aren't worth listening to or they have ill intent</li></ul>

The learning pod helps you disagree better (without being disagreeable)

<ul><li>If you really lost because of bad luck, what's the point of telling a story about it</li><li>There's nothing you can learn from it</li></ul>

Bad luck stories have no point

<ul><li>The idea that you're different from other people is an important feeling you get from your tribe</li><li>Since most people have confirmatory thinking, having a group that you identify with that values exploratory thinking helps you feel different and that you're doing something difficult</li><li>It makes your tribe special and you special by association </li></ul>

The group can help create a feeling of tribalism

<ul><li>We want to focus on accuracy, not on being right</li><li>We want to hold each other accountable</li><li>We want to disagree without being disagreeable</li><li>No one is defensive</li></ul>

Create a charter for your group

<ul><li>There's 100 learning opportunities and left to your own devices, you catch 5 of them</li><li>The learning pod helps you catch 5 more</li><li>That has a huge impact on your learning</li></ul>

The learning pod helps you catch more mistakes

<ul><li>Knowing you're going to be accountable to a group helps you catch mistakes in the moment or right after making them</li><li>Gets you to recognize mistakes in a window that you can actually do something about it</li></ul>

The learning pod helps you catch mistakes faster

<ul><li>Decisions --> actions --> outcomes --> rewards or pain (ie positive or negative consequences)</li></ul>

Learning occurs when there's a lot of feedback tied closely in time to decisions and actions

<ul><li>You need the pain in order for it to matter to you to learn</li><li>but pain can get in the way of learning</li><li>But when you're losing, your limbic system lights up and gets in the way of learning</li><li>It does not feel good to lose</li><li>It particularly gets in the way when you're playing a game or doing something where you can attribute learning to your own decisions</li><li>These are decisions you've made based on your mental model of the world and attributing loss to it is perceived by you as an attack on your identity</li></ul>

Pain gets in the way of learning

<ul><li>Groups that have a confirmatory style of thought enable one another to not learn and preserve identity</li><li>Instead, create a group that has an exploratory style of thought</li><li>The goal of the group is to create a more accurate mental model of the world</li><li>The group agrees:</li></ul><ol><li>there's an objective truth in the world</li><li>our beliefs are in progress and under construction</li><li>our goal is to help each other to construct better models of the world</li><li>We may disagree with each and that's ok. We may have different values, we may have different strategies that work for us, etc. We may come to different conclusions as a result</li></ol>

Create a group that consciously decides to spot biases for each other

<ul><li>We all have some sort of model(s) of how the world works and what's true and what's not</li><li>That's what we're always doing - building models so we can act in it with intention</li></ul>

We all use mental models to help us predict the world around us

Annie Duke discusses how to disagree without being disagreeable, spotting self-sabotage, decision-making models for high stakes, high-pressure situations, and so much more.

An effective approach to uncertain systems is based on cognitive psychology

Similar ideas to An effective approach to uncertain systems is based on cognitive psychology

We need an exit strategy

The business model

How to respond to Gish gallops