In today's fast-paced and complex world, making effective and efficient decisions is critical to success. The art of problem-solving requires a combination of intuition, experience, and analytical skills.
In this workshop, you will learn about heuristics, the mental shortcuts we use to make decisions. You'll explore the benefits and limitations of using heuristics and when they are most useful. By the end of this workshop, you'll have a deeper understanding of the power and pitfalls of heuristics, and how to make informed decisions that lead to better outcomes.
Cohort Activity
Introduction [5 minutes]
Which publicly available AI model has been trained on the largest dataset? Come up with an answer as a group, then check yourself against the key.
Do not reveal the following block until you've answered the AI question above.
Anchoring [20 minutes]
The anchoring heuristic is people's tendency to use the first number they come across as a reference point for a subsequent question. Then, when you realize that the anchor is too high or too low, you adjust repeatedly until you get something that feels right. This leaves you with a final answer that's still too high or low, because there's a wide range of answers that would feel about right.
Example. Suppose that:
- You start anchored on the number 50
- 25 is the true answer
- Numbers from 20-30 will feel about right intuitively
Then if you adjust downwards in increments of 1, you'll end up with 30 as your answer. You'll never end up with 20.
The anchoring heuristic is a special case of the primacy effect, where people over-weight the first piece of information they acquire.
Defending Against Anchoring [10 minutes]
Each member of the cohort, think of three times you fell victim to the anchoring heuristic. Share the three times with your cohort. Were any of the situations adversarial, such as a negotiation? How could you have avoided getting anchored?
Complete the above for each member of your cohort before moving to the next member.
Using Anchoring [10 minutes]
Each member of the cohort, think of one way you could use the anchoring heuristic to your advantage in your near future. Share your plan with your cohort and spend a couple minutes refining it.
Complete the above for each member of your cohort before moving to the next member.
Break [10 minutes]
Take a ten minute break.
Groupthink [20 minutes]
Anchoring affects groups as well as individuals. When a group faces a problem, the natural instinct is to propose a solution immediately, and then discuss the merits and issues with the proposal. This tendency to get locked into the first idea suggested hinders exploration of the full space of possible ideas, partly due to individual anchoring effects, but also because people get emotionally attached to the idea. By the time someone finds a better idea, they face an uphill battle getting it accepted.
Baseline [5 minutes]
This exercise is to get a baseline of how accurate your cohort is without anti-anchoring techniques.
As a group, answer the question below without external assistance. Try to minimize the amount of time you take to think about it before starting to discuss the problem. In other words, go with the instinct to immediately propose the first solution you think of.
Once a solution has been proposed, discuss the question among yourself as you would normally. If multiple solutions are proposed, you may argue over them as you would if they came up in a regular discussion.
Once you have your final answer, compare yourself against the answer key and see how close you were.
Question. What was the population of Tacoma, Washington in 2020, according to Wikipedia?
Hold Off On Proposing Solutions [10 minutes]
Now that you know how accurate you are by default, it's time to try using an anti-anchoring technique known as 'hold off on proposing solutions'.
As a group, answer the question below without external assistance. This time, instead of immediately discussing solutions, set a timer for five minutes. Spend this time discussing the problem, relevant information, and potential angles to approach the question from. Do not propose any answers until at least five minutes have passed.
Once you've thoroughly discussed the problem and nobody can think of anything else to say, you may begin discussing solutions. Proceed as you normally would until you have a single group answer, then compare yourself against the key and see how close you were. Was it closer than the baseline?
Question. What was the population of Rochester, New York in 2020, according to Wikipedia?
Mastermind [25 minutes]
The steps below should be completed for one person at a time. That is, you should fully discuss a single person's problem before the next person states their problem.
Each member of the cohort, choose a complex problem in your life that you'd like to fix. For those of you who attended last week's Level Up Session, you can try using one of the problems you worked on during that time.
Set a timer for five minutes. Spend that time discussing the problem as a cohort, without proposing any solutions. The goal is to get an accurate map of the problem without anchoring yourself on a potentially ineffective solution.
Once the time is up, discuss solutions to that person's problem.