Probability models and human judgement

For: Learners wanting formal probability foundations before evaluating claims about judgement.

What you will work towards

• Specify a sample space and the assumptions of a probability model.
• Compute a conditional probability and recognise an independence assumption.
• Separate an error relative to a specified model from a disagreement about goals or modelling choices.

Known viewing time: 3 hr 36 min across 4 steps; 2 step durations unknown. Lengths are metadata, not evidence of review.

Download the ordered reading list

1. 01

   MIT OpenCourseWare · Source-checked annotation

   Lecture 1: Probability Models and Axioms

   MIT OpenCourseWare · John Tsitsiklis · intermediate · 51 min

   Begin with the rules and objects of a probability model so that later comparisons have an explicit standard.

   Try this

   Make a sample space for a fictional draw from a box. State what the model assumes about the draw and what it leaves out.

   I have completed this step

   Access evidence checkedOriginal source ↗Watch at source ↗
2. 02

   MIT OpenCourseWare · Source-checked annotation

   Lecture 2: Conditioning and Bayes' Rule

   MIT OpenCourseWare · John Tsitsiklis · intermediate · 51 min

   Conditioning adds information to that model rather than replacing the model with an intuitive story.

   Try this

   Make a table of counts for two colours and a noisy label. Calculate the colour probability after observing the label; then change the base rate.

   I have completed this step

   Access evidence checkedOriginal source ↗Watch at source ↗
3. 03

   MIT OpenCourseWare · Source-checked annotation

   Lecture 3: Independence

   MIT OpenCourseWare · John Tsitsiklis · intermediate · 46 min

   Independence is a distinct assumption, often hidden in informal probability arguments.

   Try this

   Write what independence would mean for two events in your example. Identify a plausible change to the draw that would make the assumption fail.

   I have completed this step

   Access evidence checkedOriginal source ↗Watch at source ↗
4. 04

   MIT OpenCourseWare · Source-checked annotation

   Lecture 13: Thinking

   MIT OpenCourseWare · Speaker not established · foundation · 1 hr 7 min

   With a comparison rule available, examine human problem solving, heuristics and framing.

   Try this

   Choose a puzzle and state the intended rule, the information provided and the response. Explain which claimed error actually follows from those premises.

   I have completed this step

   Access evidence checkedOriginal source ↗Watch at source ↗
5. 05

   Open Yale Courses · Source-checked annotation

   Lecture 4 - Parts of the Soul II

   Open Yale Courses · Speaker not established · foundation · Duration unknown

   Gendler adds sampling and philosophical questions, showing why a description of reasoning needs more than a list of bias names.

   Try this

   Distinguish a sampling problem from a disagreement over values. Give one example of each and identify what would resolve it.

   I have completed this step

   Access not confirmedOriginal source ↗Watch at source ↗
6. 06

   Open Yale Courses · Source-checked annotation

   Lecture 24 - Asymmetric Information: Auctions and the Winner's Curse

   Open Yale Courses · Speaker not established · intermediate · Duration unknown

   The auction example applies conditioning to selection: learning that one won can itself be informative.

   Try this

   Describe what the event of winning reveals in the lecture’s model. State one changed assumption under which the inference would differ; do not turn the exercise into bidding advice.

   I have completed this step

   Access not confirmedOriginal source ↗Watch at source ↗

A short starting route

Steps 1, 2, 3. The probability, conditioning and independence sequence establishes the formal foundation before the psychological comparisons.

Continue more deeply

Attempt the early MIT probability assignments and inspect the provided solutions only after a serious attempt. Continue at source ↗

Limits of this route

Publisher overviews and the opening conceptual part of one official transcript were inspected; recordings were not watched and problem sets were not solved in this check. The later auction lecture presupposes earlier course material. A formal answer is conditional on the chosen model, not a guarantee that its assumptions describe the world. This is an agent editorial check, not human peer review or independent scientific assessment.