Skip to content
Tutorials Center
My learning

Topics

Lesson 42 of 43 Β· Projects and Practice

Cheat Sheet: Cognitive Science in Two Pages

A two-picture recap of models of cognition, language, vision, embodiment, AI and consciousness, plus the key formulas and a table of landmark names and findings.

This is the revision page for the second half of the course. Two summary pictures cover chapters 5 to 8, a short table collects the formulas that appear in the lessons, and a final table lists landmark names with what they found.

Cards recapping models, language and vision, and embodiment.
Part 1: models of cognition and the perceiving, acting mind.
Cards recapping AI and mind, theories of consciousness, and evidence.
Part 2: minds, machines and consciousness.

What you will learn

  • The one-line version of every major idea in chapters 5 to 8.
  • The formulas worth remembering.
  • Which name belongs to which result.

Formulas worth remembering

Idea Formula Plain reading
Bayes' rule P(H given D) = P(D given H) * P(H) / P(D) Posterior from prior and likelihood
Perceptron update w = w + rate * (target - output) * x Nudge weights toward fewer errors
Cue combination weight_i = precision_i / sum of precisions Trust the reliable cue more
Precision precision = 1 / variance Reliability of a signal
TD error error = reward + V(next) - V(now) Reward received versus expected
d-prime d' = z(hit rate) - z(false-alarm rate) Sensitivity apart from bias
Hopfield weights w[i][j] = (1/N) * sum of patterns' x[i] * x[j] Hebbian storage
Hopfield capacity about 0.14 * N random patterns Beyond this recall collapses

Landmark names and findings

Name and year Finding or idea
Newell and Simon, 1976 Physical symbol system hypothesis
Marr, 1982 Three levels of explanation
Rumelhart, Hinton, Williams, 1986 Backpropagation popularized
Fodor and Pylyshyn, 1988 Systematicity objection to connectionism
Ernst and Banks, 2002 Humans combine sight and touch near-optimally
Meyer and Schvaneveldt, 1971 Semantic priming in word recognition
Hubel and Wiesel, 1959 to 1962 Orientation-tuned cells in V1
Milner and Goodale, 1992 Vision for perception versus action (patient DF)
Simons and Chabris, 1999 Inattentional blindness (gorilla)
Braitenberg, 1984; Brooks, 1991 Behavior from sensor-motor coupling and layers
Clark and Chalmers, 1998 Extended mind
Turing, 1950; Searle, 1980 Imitation game; Chinese room
Schultz, Dayan, Montague, 1997 Dopamine as reward prediction error
Nagel, 1974; Chalmers, 1995 What it is like; the hard problem
Baars, 1988; Dehaene Global workspace
Tononi, 2004 onward Integrated information theory
Green and Swets, 1966 Signal detection theory
Owen et al., 2006 Covert awareness in a vegetative-state patient

Five habits of a good cognitive scientist

  1. Ask what is stored, how it is used, how it is learned and where the model fails.
  2. State which level of explanation a claim is at.
  3. Distinguish similar behavior from similar mechanism.
  4. Prefer replicated results and be skeptical of small surprising effects.
  5. Keep first-person reports and third-person data in view together.

Check your understanding

Quick check Β· 3 questions

of 3 answered You scored / 3 Β· perfect!

  1. 1. In cue combination, a cue's weight is proportional to its:

    Reliable cues, those with high precision, get more weight.
  2. 2. Which formula gives the temporal-difference error?

    The error compares reward plus the next state's value to the current estimate.
  3. 3. Who is paired correctly with the hard problem?

    Chalmers introduced the easy versus hard problem distinction in 1995.

Summary

  • Models differ in what they take knowledge to be; perception and action rely on inference and context.
  • AI sharpens questions about understanding, and consciousness science struggles to measure experience.
  • A handful of formulas recur: Bayes, the perceptron rule, precision weighting, TD error, d-prime.
  • Good habits matter as much as facts.

Next: Further study