PSY370H1 Lecture : LEC 3 - Insight Sept 24 2009
PSY370: Thinking & Reasoning
LEC 3 -
Sept, 24th, 2009
Review from Last Lecture
! search spaces are combinatorially explosive
! how do we do it?
! problem formulation is going to be key
! GPS was proposed and failed for 2 flawed assumptions
! 1. all problems are the same
! 2. formulating a problem is a smaller job than solving the job
! is insight just another search inference process or a different one?
! we can distinguish algorithm and heuristic
! most of our problem solving are heuristic based
Gestalt Tradition
! thinking is like perception
! there is an organizational structure that is somehow greater than the sum of the parts
! perception and thinking work in ways of this structure
! the gestalt/structure is changing in the Necker cube ex.
! in opposition to Thorndike's law of effect
! Thorndike try to empirically test animal intelligence
! understand intelligence as the ability to solve problems
! Thorndike's puzzle box
cat inside a box and the cat can see / smell the food outside the box
mechanism within the box can release the cat
measure how long it takes for the cat to get out vs. number of trials
eventually the cat learns how to get out faster, but the process is gradual and there are
jaggedness within the graph
what is important is that b/c the graph is gradual & jagged then that means the cat never
have “insight' or are intelligent
the graph produced is a result of past experience into present experience and
reinforcement loop
! gestalt would challenge that problem solving is a simple extension of past experience into the
present experience and that all insight graph looks like a S graph
! there should be other cognitive processes at work
! Kohler
! doing work during WWI
! Sultan the chimp
! chimp is analogous to the cat experiment by Thorndike
! chimp can get food outside of box via external mechanisms
! Sultan's learning curve looks like an S curve rather than a jagged graph
! this directly challenges Thorndike's idea
! But Kohler's really a German spy and that his cover was being a scientist
! also chimp is not easy to manipulate like cat
! Kohler: restructuring of the given material is insight
! Dunker 1945
! Wrote the book: Odd Problem Solving
! problem solving is restructuring of how we are formulating the problem
www.notesolution.com
Document Summary
problem formulation is going to be key. gps was proposed and failed for 2 flawed assumptions. most of our problem solving are heuristic based. Gestalt tradition thinking is like perception there is an organizational structure that is somehow greater than the sum of the parts. perception and thinking work in ways of this structure the gestalt/structure is changing in the necker cube ex. in opposition to thorndike"s law of effect. thorndike try to empirically test animal intelligence. understand intelligence as the ability to solve problems. gestalt would challenge that problem solving is a simple extension of past experience into the present experience and that all insight graph looks like a s graph there should be other cognitive processes at work. chimp is analogous to the cat experiment by thorndike. chimp can get food outside of box via external mechanisms.
