Lecture 1 - Intro Flashcards
What is cognitive science?
Cognitive science is the
interdisciplinary study of mind and
intelligence, embracing philosophy,
psychology, artificial intelligence,
neuroscience, linguistics, and
anthropology
(NAPPAL)
-
Thagard, 1996,
Stanford Encyclopedia
of Philosophy
What is John McCathy’s definition of intelligence? (idk if this is important)
“Intelligence is the computational
part of the ability to achieve goals in
the world” (1997)
How did the study of cognitive science start?
At first, just introspection
Then, Wilhelm Wundt, 1879, Uni Leipzig
-Quantitative measurement
-RT
-Sensory experiments
Explain Behaviorism and operant conditioning
B.F. Skinner, 1938 ( “The
Behavior of Organisms:
An Experimental
Analysis”)
Mice pushing lever = treat… see image
-Positive reinforcement = encourage behavior by giving positive feedback
-Negative reinforcement = encourage behavior by removing negative feedback
-Positive punishment = discourage behavior by giving negative feedback
-Negative punishment = discourage behavior by removing positive feedback
-Immediacy: 5 seconds after stimulus vs. 30 seconds… sooner=better
-Contingency (aka consistency)… how being consistent affects the learning behavior… better when not exactly perfectly reliable
-Extinction: unlearning behaviour after certain amount of time
1930, Tolman’s Rat Mazes
Put in rats, skinner conditioning…. Count errors…
3 conditions: no reinforcement (just roaming around, no rewards), regularly reinforced (since the beginning… given rewards), reinforced day 11 (only start getting rewards on day 11…. Best performance out of all of them!)
rats memorize maze w/o reward → internal processes
Behaviorist Model vs. Cognitive Model
Behaviorist model: only study the observable/external behavior
-Stimulus: in the environment
-Black box (brain): can’t be studied
-Response: behavior
Cognitive model: can scientifically study internal behavior
-Input: in the environment
-Mediational process: mental event
-Output: behavior
Turing Machines, 1936
Universal Turing machine
Infinite tape: 0s and 1s
State transition diagram (s1 to s2…)
Turning machine description
Back into universal Turing machine
Can model arbitrary processes as programs with memory
Chomsky’s Structures, 1957
In response to Skinner’s 1957 ‘’Verbal Behavior’’
-Information from the environment…. Language -Goes into Acquisition Device (unconscious process inside child’s mind, used only for learning language)
-Results in Language Acquisition
Innate ability, even without reinforcement
“One would naturally expect that prediction of the
behavior of a complex organism (or machine) would
require, in addition to information about external
stimulation, knowledge of the internal structure of the
organism, the ways in which it processes input
information and organizes its own behavior”
Language processing is innate
Winograd’s SHRDLU: ELIZA
Ex: ELIZA chat bot
Ex: Shapes and computer software… first example of a cognitive scientist putting together language and 3D perception model about how computer could think about the world (he said something like that… I don’t remember)
Microworld with NLP interface
1971 – Shepard/Metzler’s Rotations
1971 – Shepard/Metzler’s Rotations
Reaction time correlated to the degree of rotation (more rotation = takes longer (longer RT)
People have some sort of mental 3D model
Means we have internal 3D models of objects
Marr’s Vision 1982
See image, and more details in other lecture
Vision is multi-stage processing pipeline
Summary
● What is CogSci? → Interdisciplinary study of the mind
● What disciplines factor into CogSci? Neuro, AI, Psyc, Phil, Anthr, Ling (NAPPAL)
● Prehistory: Introspection → Behaviorism
● Tolmanʼs rat mazes: rats memorize maze w/o reward → internal processes
● Turing machines: can model arbitrary processes as programs with memory
● Chomsky: Language processing is innate
● Winograd: Microworld with NLP interface
● Shepard/Metzler: we have internal 3D models of objects
● Marr: vision is multi-stage processing pipeline
