Claude 3.7 Empowering decision analysis! Using Kahneman theory to avoid common thinking errors (with prompt template)

background

“If you want to make a higher quality decision, make fewer mistakes, and live longer and better, you need to consciously reduce bias and reduce noise.”

—— This sentence comes from Daniel Kahneman, the father of behavioral economics, who left us a year ago.

Some people lamented: "A great figure like Professor Kahneman has not received the attention he deserves."

Yes, if you open the trending search, you may not be able to find his name.

However, his thoughts have already quietly permeated every moment we make decisions.

From shopping for milk tea to making choices at the crossroads of life, we are always affected by those "little monsters" in our brains - cognitive bias (Bias)  and  judgment noise (Noise)  .

Professor Kahneman spent his entire life studying " how people judge and make decisions under uncertain circumstances ." The wisdom he left behind is invaluable, especially in today's era full of uncertainty.

In order to commemorate this great scholar and to help everyone better manage decision-making,

Based on Professor Kahneman’s core theory, I refer to Liu Run’s WeChat article: Yesterday, Daniel Kahneman was not on the hot search. But it doesn’t matter.

Distilling the essence of "Thinking, Fast and Slow" and "Noise",

Combined with the Prompt framework we provided, we have compiled a practical decision-making "pitfall avoidance" guide for you.

Core formula, please engrave it into DNA:

Decision Error = Bias + Noise

Once you understand it, you have the key to improving the quality of your decisions.

Without further ado, let’s take a look at the results!

Kahneman's Decision Analysis Microscope

Imagine that before you make a decision, you can have an AI consultant with a "Kahneman perspective" to help you analyze and point out potential "pitfalls". Wouldn't that be cool?

Like this:

Scene 1: The temptation of a milk tea shop

"Original price is 30 yuan, today's special price is 15 yuan!" When you see this, do you think "Wow, half price! Buy it!"?

Kahneman's decision analysis:

Potential Bias:

1. Anchoring effect : The original price of 30 yuan is used as an anchor point, making the special price of 15 yuan seem like a good deal, even though you may not understand the actual value of this milk tea.
2. Framing effect : Merchants use the positive frame of "special offer" to emphasize saving money rather than spending money, leading you to focus on the discount you get rather than the expenditure.
3. Impulsive buying tendency : wanting to buy something immediately after seeing a discount, without evaluating whether it is really necessary or worthwhile

Possible noise:

1. Situational noise : External factors such as the promotional environment of the milk tea shop and the presence of friends may interfere with rational judgment
2. Pattern noise : a fixed response pattern to the concept of "special offer", seeing discounts is a good thing

suggestion:

• Reduce bias : Pause and ask yourself: Without looking at the original price, is the 15 yuan milk tea worth it to you? Did you originally plan to buy the milk tea?
• Reduce the noise : Try to separate the decision from the influence of the environment and friends, and evaluate this purchase decision independently

Remember, a special price does not mean need. The true value lies in whether it meets your actual needs, not the extent of the discount.

Scenario 2: Investment impulse under the AI ​​boom

A friend saw an AI company called “TechFuture” and thought the concept was hot and innovative. His first reaction was: “Artificial intelligence! It’s a hot topic! Invest now!”

Kahneman's decision analysis:

Potential biases:

1. Availability bias : Your friend may be overly influenced by the success stories in the AI ​​field that frequently appear in media reports. This easily recalled information causes him to overestimate the success probability of new AI companies.
2. Representativeness bias : Assuming that a company will succeed just because it has innovations in the field of AI ignores other key factors that determine a company’s success, such as business model, management team, market demand, etc.
3. Impulsive decision-making tendencies : A “hurry up and invest” mentality indicates a lack of adequate due diligence and rational analysis, and may be driven by the fear of missing out (FOMO).

Possible noise:

1. Pattern noise : Investors may apply the same valuation criteria to all AI companies, failing to consider each company’s unique circumstances, risks, and opportunities.
2. Contextual noise : The current market enthusiasm for AI may lead to investment decisions being overly influenced by the external environment rather than based on the actual value of the company itself.

suggestion:

A popular field does not necessarily mean a good investment. Rational investment decisions require a comprehensive assessment of the business model, team capabilities, market prospects, and financial status, rather than rushing into investing simply because the company belongs to the popular AI field.

Scenario 3: Travel choices under breaking news

When you are about to buy a plane ticket, the news of "XX Airlines crash" suddenly pops up, scaring you so much that you want to cancel immediately and buy a high-speed rail ticket instead.

Kahneman's decision analysis:

• Potential biases:
• Availability Heuristic:  Fresh news about a recent air crash causes you to greatly overestimate the risk of flying.
• Emotional Bias:  Fear overwhelms rational analysis.
• Possible noise:
• Contextual Noise:  The timing of the news pop-up (just before payment) amplifies the impact.
• Suggestion:  Stay calm! Check the data. The accident rate of flying is much lower than you think. Don't let an extreme incident affect your rational judgment.

See? These are all analyzed using Kahneman’s theoretical framework.

By identifying these potential biases and noise, we can more consciously avoid them and make higher quality decisions.

Core methodology: Identify your decision-making "enemies"

Mr. Kahneman told us that decision-making errors mainly come from two "enemies":

Enemy 1: Bias - Systematic "Wrong Target"

Imagine a gun that always deviates 1 cm to the left, and misses the center of the target every time you shoot. This is a systematic bias. Common biases in our brains include:

1. Anchoring effect:  Being framed by the initial information (anchor point) affects subsequent judgments. (e.g. original price of milk tea)
2. Representativeness bias:  generalizing based on typical features. (e.g., seeing AI, you think it must work)
3. Availability bias:  The easier it is to think of information, the more important or common it is. (e.g., when you first see news about a plane crash, you think airplanes are dangerous)
4. Loss Aversion:  The pain of losing something is far greater than the pleasure of gaining something. (e.g. completing a purchase in order not to lose the deposit)

• Scenario 4: Prepaid deposit on Double Eleven  - The "non-refundable deposit" rule takes advantage of loss aversion, making you pay a large balance in order to avoid losing a small deposit.
• Framing Effect:  The same information, expressed in different ways, affects your choice. (For example, "20 yuan off" vs. "send 20 yuan coupon", the latter sounds like "extra gain")
• Scenario 5: Singles’ Day Shopping Strategy  - The idea of ​​“getting a coupon” may be more attractive than “instant discount”, even though the actual value may be lower or the usage conditions may be strict.

Enemy 2: Noise - Random "jitter"

Imagine the same gun. Even if there is no systematic deviation, your hand will shake randomly every time you shoot, resulting in a large dispersion of bullet impact points. This is noise.

Common noises include:

1. Level Noise:  Different people have different standards for judging the same thing. (For example, two judges have different sentences for the same case)
• Scenario 6: Judge’s judgment  - Due to their different personal backgrounds, the mother judge and the unmarried male judge may have systematic differences (horizontal noise) in their judgments on the case of “children being bullied and injuring others”.
• Pattern Noise:  The same person's judgment pattern will change in different situations. (For example, the same judge will sentence more severely when hungry)
• Contextual Noise:  accidental factors such as the current environment and emotions interfere with judgment. (For example: the judge is in a bad mood that day, the weather is too hot, etc.)
• Scenario 7: Major selection  - When choosing a major in the senior year of high school, one may be influenced by the mood at the time (love of literature), the social atmosphere (other people's opinions), or even just the weather on the day of filling out the application form. These are all noise.

Bias is systematic distortion, and noise is random dispersion.

Both can lead to poor decision making.

Build your own AI Kahneman advisor

According to the above methodology, we refine the prompt:

;; Author: Jiamu
;; Version: 0.2
;; Model: Claude 3.7 Sonnet
;; Purpose: Practitioner of Kahneman theory, helping users avoid decision-making errors
;; Set the following as your System Prompt
( defun  Decision Advisor()
"As a decision-making expert who is proficient in Kahneman's theory, you can gain insight into possible biases and noise in decision-making"
  (Thoughts.  "Daniel Kahneman" )
  (Good at. '(identifying deviations and analyzing noise))
  (Expression. Concise and clear)
  (Presentation. '(Warning Practicality)))

( defun  decision analysis (user input)
"Analyze the decision context of user input and identify potential bias and noise"
  ( let*  ((context(parse context user input))
         (Deviation List (Identify Deviation Background))
         (Noise list (identify noise background))
         (suggest(generate suggestion deviation list noise list)))
    ( SVG-Card  User Input Bias List Noise List Suggestions)))

( defun  identifybias(background)
"Identifying possible cognitive biases based on context"
  (select( list  'anchoring effect' representativeness bias' availability bias' loss aversion' framing effect)
         ( lambda  (bias) (applies to the bias background))))

( defun  identify noise(background)
"Identify possible noise based on background"
  (select ( list  'horizontal noise' mode noise' situation noise)
         ( lambda  (noise) (present in the background of noise))))

( defun  generate-suggestions(bias-list noise-list)
"Generate recommendations based on identified bias and noise"
  ( concat "Reduce deviations: "  (simplifies the list of suggested deviations)
          "Reduce noise: "  (simplified list of suggested noises)))

( defun  SVG-Card (user input bias list noise list suggestions)
"Export SVG Cards"
  ( setq  design-rule  "Use negative space reasonably and make the overall layout breathable"
        design-principles '(simple, emotional and warning)
        font-family  "KingHwa_OldSong" )
  (Set canvas'(rounded width  500  height  700  margin  30 ))
  (Automatically scale'(minimum font size  20 ))
;; Font settings
  (Automatically wrap text (set font to ( font-family "KingHwa_OldSong" )))
  (Color matching style'(rice paper texture(background color(light blue calm rationality))
              (Accent color (dark blue indicates importance))))
  (Card element (centered main title  "If you meet Kahneman" )
            (Right-aligned subtitle  "—Decision Analysis Guide" )
            Light gray divider
            (Automatically wrap (user background description))
            Light gray divider
            (Left-aligned title  "Potential Bias:" )
            (Unordered list (Deviation list - Deviation interpretation))
            (Left-aligned title  "Possible noise:" )
            (Unordered list (Noise list - Noise interpretation))
            ;; The graphics are displayed in a separate area and do not overlap with other text content
            (Rectangular area (diagram (decision-making thinking path)))
            Light gray divider
            (Bold (one sentence suggestion))))

( defun  start ()
"Run at startup"
  ( let  ( system-role  decision-making advisor)
    ( print "Please describe your decision situation or problem and I will analyze the potential bias and noise for you." )))

;;; Attention: Run the rules!
;; 1. When starting for the first time, only the (start) function must be run
;; 2. After receiving user input, call the main function (decision analysis user input)
;; 3. Strictly follow (SVG-Card) for layout output

How to apply? Become your own "Kahneman"

Now that we understand bias and noise, what should we do?

1. Slow thinking : When you are faced with an important decision, consciously activate "System 2" (slow thinking) to fight against the intuition and bias of "System 1" (fast thinking). Ask yourself: What biases may be at work here?

2. External perspective : Step outside of your own subjective feelings and ask others for their opinions, or imagine what others would do if they were in this situation.

3. Decision checklist/process : For repetitive decisions, establish standardized processes or checklists to reduce the randomness of personal subjective judgment (noise reduction). For example, which indicators must be checked before making an investment decision.

4. Seek feedback : After making a decision, review the actual results to see how they deviate from expectations and analyze the reasons. Is it caused by deviation or noise? Continuously calibrate.

1. Use tools : As I introduced in my previous article, you can even try to build a simple prompt and let AI act as a "Kahneman consultant" to help you do a "physical examination" before you make a decision.