Insight, analogy and expertise Flashcards
what is problem solving
navigation and creation of a search space using available allowable operators to move through intermediate problem states
how do we set up a problem state
the challenge is knowing what operators are allowed and possible and what intermediate states might be good ones
how can we define problem spacees (Kotovsky & Simon, 1990)
participants were given a chinese ring puzzle to complete where they have to release a peg
participants rarely solve this problem, even when they are given an hour to do so
the difficulty lies in making moves through a search space over and above anything else
participants’ move behaviour is often dichotomous featuring many non-progressive, error prone moves followed by rapid error free movement to goal
once participants realise the possible moves, they solve the problem quickly in a smooth error free way
such that problem solving is an issue of figuring out what the right problem space is
making moves explicit using a digital isomorph changes the outcome such that participants succeed in about 15 mins
what does gestalt psychology contribute to how we define problem spaces (Maier et al., 1931)
solving problems requires sudden insight into how to restructure a problem space
: two bits of string hanging from ceiling
Must hold them both → out of arms reach
Table with various objects that can be used to solve problem
Use these as weights: tie to string to create momentum
40% solved baseline, addition of hint: 77% (start with them swinging- participants deny this influenced their decision)
Options available are not obvious → need to figure out possible moves
Importance of gestalt structure
what does Metcalfe & Wiebe’s 1987 study tell us about flashes of insight
Contrast these two types of problems
Arithmetic problem: How to factorise X
Insight problem: how is it possible to do Y
Interested in participants experiences of solving task
Asked to rate hot/cold certainty of answer during
Observed different types of patterns for these two problems
Arithmetic: gradually increasing warmth
Insight: cold ratings while setting up problem space, once this is done instantly hot
what evidence does Duncker (1945) provide for the restructuring problem account
participants were presented with matches, thumbtacks and a candle and had to figure out how to attatch a candle to the wall and light it
people fixate on the typical uses of objects and fail to realise they must tac the match box to the wall and support the candle using that
what evidence does Scheerer (1963) provide for the restructuring problem
Draw 4 straight lines through the dots without ever leaving the page
Initially set up incorrectly - have to move away from how it is initially understood
must go outside the box
what is functional fixedness
tendency to approach a problem in a particular way, we struggle if this does not work (in restructuring)
what is Kaplan & Simon’s 1990 mutilated checkerboard problem and what does it tell us about problem representation
700,000 possible combinations
Can’t realistically get the solution through trial-and-error have realise it’s impossible
Either given blank board, checked board, word board (e.g butter and bread)
Different stimuli on different parts of the board will clue people in about which parts of the board you place a domino (vertically or horizontally)
Always cover one white and black, removing a corner removes two black squares
Particularly obvious when bread and butter words are used
Solution times vary across different conditions
Way in which problem is presented give participants clues about how they might solve it
what does silverias (1971) cheap necklace problem show about incubation effects
Sometimes the best thing to do is to take a break
Remove functional fixedness
Cheap necklace problem
First try: 55% correct
30 min correct: 64%
4 hour break: 85% correct
Take individual starting chain, break up and join all three
how can problems be solved by analogy Duncker, 1945
Correspondence with know problem Requires people to find/see parallels Doctor tumor Gick & Holyoak (1980) Asked to remember story How likely are subjects to see problems as analogous No hint: 20% Explicit hint: 92% Often participants don;t naturally see parallels
how can superficial similarity between problems be helpful (Holyoak & Koh, 1983)
Superficial similarity is helpful
Taught about experiment and discussed
Few days later given problem with lightbulb with xrays
81% give convergence answers compared to 10% controls who weren’t given example
High level needed
Ultrasound version drops to 38% convergence
what did Gentner & Gentner’s intuitive analogies demonstrate
Different analogies for flow of electricity
People understand electricity using analogies
Analogies people spontaneously used influenced answers
Water system: perform better with batteries (store)
Race track: perform between with resistors
how can analogies help in solving problems in a lab (Dunbar, 1995)
Interested in the scientific processes
3/4 labs used analogies to solve problems - and solved more problems
When people struggling to get an experiment to work they would compare with what does work and are able to use
Regional analogies: deeper analogies involved in developing theories and planning experiments e.g compare viruses and retroviruses
Provides strucutre
why is the distinction between production and reception of analogies important
Contradiction between findings in labs and irl
People producing analogies vs being given an analogy
Arguing for different sides of whether or not canada can run a budget deficit
Most analogies reflected deep structures
Spontaneous / natural
Gave participants scenarios and arguments
Asked which of these were similar to a new problem
Recall superficially similar ones not deeper structure
Suggests they can’t use them for problem solving
