A psychophysical analysis of the base rate fallacy.

Slides:



Advertisements
Similar presentations
KeTra.
Advertisements

A Review for Zoology Class
Bayes Rule The product rule gives us two ways to factor a joint probability: Therefore, Why is this useful? –Can get diagnostic probability P(Cavity |
Objectives (BPS chapter 12)
Processing physical evidence discovering, recognizing and examining it; collecting, recording and identifying it; packaging, conveying and storing it;
Cognitive Processes PSY 334 Chapter 11 – Judgment & Decision-Making.
Probabilities Random Number Generators –Actually pseudo-random –Seed Same sequence from same seed Often time is used. Many examples on web. Custom random.
Heuristics and Biases. Normative Model Bayes rule tells you how you should reason with probabilities – it is a normative model But do people reason like.
Cognitive Processes PSY 334 Chapter 10 – Reasoning & Decision-Making August 21, 2003.
Inference in Dynamic Environments Mark Steyvers Scott Brown UC Irvine This work is supported by a grant from the US Air Force Office of Scientific Research.
The Study of Adult Development and Aging:
Robert delMas (Univ. of Minnesota, USA) Ann Ooms (Kingston College, UK) Joan Garfield (Univ. of Minnesota, USA) Beth Chance (Cal Poly State Univ., USA)
Statistics: Unlocking the Power of Data Lock 5 STAT 101 Dr. Kari Lock Morgan Bayesian Inference SECTION 11.1, 11.2 Bayes rule (11.2) Bayesian inference.
Review: Probability Random variables, events Axioms of probability
Statistics: Unlocking the Power of Data Lock 5 STAT 250 Dr. Kari Lock Morgan Probability SECTIONS 11.1 Probability (11.1) Odds, odds ratio (not in book)
Probability and inference General probability rules IPS chapter 4.5 © 2006 W.H. Freeman and Company.
©2003/04 Alessandro Bogliolo Background Information theory Probability theory Algorithms.
Research Design Interactive Presentation Interactive Presentation
An Intuitive Explanation of Bayes' Theorem By Eliezer Yudkowsky.
Math The Multiplication Rule for P(A and B)
1/20 Remco Chang (Computer Science) Paul Han (Tufts Medical / Maine Medical) Holly Taylor (Psychology) Improving Health Risk Communication: Designing Visualizations.
Statistics: Unlocking the Power of Data Lock 5 STAT 101 Dr. Kari Lock Morgan 12/4/12 Bayesian Inference SECTION 11.1, 11.2 More probability rules (11.1)
Chapter 2 Research in Abnormal Psychology. Slide 2 Research in Abnormal Psychology  Clinical researchers face certain challenges that make their investigations.
Math In The Science Classroom Guidelines for Preventing Math Errors Keep track of all information (paper trail). Keep track of all information (paper.
The human 3 of 3 U2Mvo&feature=player_embedded the human 3 of 31.
Bayesian Inference I 4/23/12 Law of total probability Bayes Rule Section 11.2 (pdf)pdf Professor Kari Lock Morgan Duke University.
Demonstration and Verbal Instructions
1.3 Simulations and Experimental Probability (Textbook Section 4.1)
© 2008 McGraw-Hill Higher Education The Statistical Imagination Chapter 1. The Statistical Imagination.
Buying and Selling Prices under Risk, Ambiguity and Conflict Michael Smithson The Australian National University Paul D. Campbell Australian Bureau of.
Problem Solving Riddle.
Bayesian Inference, Review 4/25/12 Frequentist inference Bayesian inference Review The Bayesian Heresy (pdf)pdf Professor Kari Lock Morgan Duke University.
Scientific Method. Scientific method: A logical and orderly way to solve a problem. No one set “THE Scientific Method”
We report an empirical study of buying and selling prices for three kinds of gambles: Risky (with known probabilities), Ambiguous (with lower and upper.
Once again about the science-policy interface. Open risk management: overview QRAQRA.
Lecture: Forensic Evidence and Probability Characteristics of evidence Class characteristics Individual characteristics  features that place the item.
Populations III: evidence, uncertainty, and decisions Bio 415/615.
Review: Probability Random variables, events Axioms of probability Atomic events Joint and marginal probability distributions Conditional probability distributions.
 Science has a standard way to test an idea  Cause and effect  What does that means?  That everything that happens in this world is because of the.
Statistics: Unlocking the Power of Data Lock 5 STAT 250 Dr. Kari Lock Morgan Probability SECTIONS 11.1, 11.2 Probability (11.1, 11.2) Odds, Odds Ratio.
 Aim: purpose  Independent Variable (IV): manipulated variable  Dependent Variable: (DV) measured variable  Operationalized Variable: Written so what.
Scientific Method Review.  The scientific method is used by scientists to solve problems  It is organized and reproducible (can be repeated by other.
Topics and Questions. All quantitative research begins with a question of some kind. The title may suggest the type of question being posed. The research.
The Scientific Method. Scientific Method The scientific method is a systematic approach to problem solving. There are seven steps: 1.State the Problem.
Introduction to Research Concepts using the Card Probability Study Chapter 1 Thomas and Nelson.
BC Cancer Agency CARE & RESEARCH Breast Cancer Mortality After Screening Mammography in British Columbia Women Andrew J. Coldman, Ph.D. Norm Phillips,
 experimental group  The object or group of objects that receive the independent variable. (The objects you experiment on.)
Active Learning in the Third Year Statistical Physics Module at the University of the Witwatersrand Jonathan Keartland School of Physics, University of.
HL2 Math - Santowski Lesson 93 – Bayes’ Theorem. Bayes’ Theorem  Main theorem: Suppose we know We would like to use this information to find if possible.
Probability Probability Day 3 Introduction to Probability Probability of Independent Events.
Contact Info: Improving Decision Making: The use of simple heuristics Dr. Guillermo Campitelli Cognition Research Group Edith.
Psychology 101: General  Chapter 1Part 2 Scientific Method Instructor: Mark Vachon.
LECTURE 13: ONGOING RESEARCH: THE ROLE OF INDIVIDUAL DIFFERENCES April 25, 2016 SDS136: Communicating with Data.
Communicating Risk.
Bayesian inference, Naïve Bayes model
Presented by: Karen Miller
Demonstration and Verbal Instructions
Scientific Method.
Scientific Method.
Make an Organized List and Simulate a Problem
Unit 1: Scientific Method
Medical Diagnosis Problem
Probability Probability is the frequency of a particular outcome occurring across a number of trials
theoretical probability p = number of successful outcomes
Lecture: Forensic Evidence and Probability Characteristics of evidence
Key idea: Science is a process of inquiry.
A Guide to Scientific Problem Solving
3. Formulate a Hypothesis 4. Test / Experiment - Retest
Equations Objectives for today’s lesson :
Counting Methods and Probability Theory
Presentation transcript:

A psychophysical analysis of the base rate fallacy

The base rate frequency bias The mammography problem The probability of breast cancer is 1% for a woman at age forty who participates in routine screening. If a woman has breast cancer, the probability is 80% that she will get a positive mammography. If a woman does not have breast cancer, the probability is 9.6% that she will also get a positive mammography. A woman in this age group had a positive mammography in a routine screening. What is the probability that she actually has breast cancer? (Gigerenzer and Hoffrage, 1995) The information we have The “a priori” probability of a breast cancer: P(B.C.) The conditional probabilities to be positive with and without B.C.: P(Pos|B.C.), P(Pos|¬B.C.) The correct solution of the problem The most frequent subjective estimations The 95% of the subjects give an estimation of the probability for the woman to have a breast cancer between 70% and 80% (Eddy, 1982)

The dual process hypothesis Strategic processing –Mental representations –Examples of theories: Human problem solving (Newell & Simon, 1972) Mental models (Johnson-Laird, 1989) –Features: Effortful processing Goal-directed Automatic processing –Features: No attentional control required Obliged course Stimulus dependent –Examples of phenomena Conditioned reflexes Implicit memory

A bayesian game with cards There is a pack of cards, every card has a symbol on a side and a colour on the other side. The symbol could be an O or an X. The colour could be Red or Blue. The probability that a card has an O on one of its side is 5%. If a card has an O on a side the probability is 80% that it is Red on the other side. If a card has an X on a side the probability is 20% that it is Red on the other side. A card from the pack is Red. Which is the probability it has an O on the other side? The outcomes: O or X The data: Red or Blue The information we have: P( O), P( Red| O), P( Red| X) The information we want: P( O| Red) The solution:

The design of the experiment The independent variables The type of presentation of the information: dynamic or static, varying through the treatments The real probability of the outcome given by the Bayes theorem and varying from trial to trial The difficulty: the incidence of the a priori probability on the real probability, varying through the trials with the same real probability The dependent variable The subjective probability: the choice for an outcome and the amount of money of the bet The hypothesis In the second treatment the subjective probability fits with the real probability better than in the first treatment The goodness of fit measure: The d’ : the difference between the z points of the proportions of choices for the outcome when its probability is higher and lower than 50%

The subjective probability The external validity problem –Should we trust the statements of the subjects? –Do the subjects understand the numbers? The subjective probability –The probability is the relative willingness to invest in the occurrence of a future event (De Finetti, 1931) 50%30%10%70%90% ¬AA UncertaintyCertainty 2€1€50c20c 50c1€2€ 0100% Choice:

The meaning of the d’ 0 10% 30% 50% 70% 90% 100% d’ Wrong bets for O Correct bets for O Wrong bets for X Correct bets for X

Results of a pilot experiment Dynamic Game Most Probable Outcome BetOX O X d’0,5078 Static Game Most Probable Outcome BetOX O31738 X d’1,6900 Dynamic Game Most Probable Outcome BetOX O8816 X d’0 Static Game Most Probable Outcome BetOX O14519 X d’1,1989 In all the trialsIn the difficult trials

Most Probable Outcome BetOX O12113 X d’2,3491 Most Probable Outcome BetOX O9312 X d’1,6034 Most Probable Outcome BetOX O10313 X d’1,3023 Most Probable Outcome BetOX O5712 X d’0 Most Probable Outcome BetOX O11516 X d’0,7151 Most Probable Outcome BetOX O7310 X d’0,8430 Dynamic GameStatic Game Considering the money 3,5£ 3£ 2,5£ 2£ 3£ 2,5£ 3,5£

Conclusions The base rate fallacy in these games is not so pervasive as in the written problems It seems to depend more by the amount of ambiguity of the information than by the kind of cognitive process the subjects use There is a clear correlation between the amount of money the subjects invest and their ability to guess the most probable outcome