Lecture 3 - Bayes' Theorem and human rationality Flashcards
(lecture):
Describe what being ‘rational’ is.
(lecture):
Being rational is not the same as being right: it’s about being sensitive to all relevant evidence in a situation and the usefulness/informativeness of that evidence when it comes to dealing with uncertainty about the judgement you’re trying to make.
(see page 1/2 on G doc notes for an example)
(lecture):
Read orange writing on page 2 on G doc notes.
(lecture):
(lecture):
Read purple writing on page 2-4 on G doc notes.
(lecture):
(lecture):
Describe a study that investigates the presentation of data and the way in which we can improve people’s reasonings.
(lecture):
See the red/orange writing on page 3 on G doc notes.
(lecture):
What are the 2 types of probability?
(lecture):
- Frequentist
- Bayesian
(lecture):
Describe the difference between frequentist and bayesian probability.
(lecture):
“Frequentist”:
- “Dice-rolling”-type probability
- Based on well-defined sampling procedures
- Type usually taught in probability/statistics
“What is the probability that two dice rolls sum to 6?”
“Bayesian”:
- “Single-event” probability
- Updates of “personal” probability
- Associated with (potentially informed) subjective beliefs
“What is the probability that it will rain tomorrow?”
This is a personal thing that you’re updating with evidence. This is a single event, not ‘what’s the probability that it will rain on days like today?’. So, this is not sampling from a population.
(lecture):
Describe how Cosmides & Tooby, 1996 investigated the 2 different probabilities (frequentist and bayesian).
(lecture):
See end of page 4-6 on G docs notes.
(lecture):
Describe what (Gigerenzer, 1991, p. 109) and (Kahneman & Tversky, 1996, p. 589) debated about the 2 types of probability (frequentist and bayesian).
(lecture):
See page 6 on G doc notes.
(lecture):
Things to get out of the lecture/reading:
- Some of the sorts of methodologies/experiments researchers use to investigate judgement and decision making. You can change data presentation in ways that seem irrelevant but then gain a different result.
- How the way people perform in those tasks is used to make inferences about psychological processes (e.g., heuristics) underlying that performance.
- Why particular patterns of performance in particular tasks is taken to suggest the existence of particular kinds of underlying psychological processes.
- How those inferences might change if we question the methodological choices and the assumptions implicit in the argument relating performance to process.
(lecture):
