Clare Walsh L4 Flashcards
What is a problem?
When there is no clear path between a start state and a goal
Well defined problems
Initial goal, goal state and potential moves are known
Ill-defined problems
Start/end-state are unknown and the possible strategies are unknown
Behaviourist approach to learning
Trial and error - Thorndike’s Cat (1898)
- Unsystematic behaviour
- Requires little to no knowledge
- Slow
- Doesn’t work for all problems
- Risky
Gestalt approach to learning
Problem-solving requires insight
Koehler’s (1925) monkey - incubation and Ah ha
Neuroscience and insight
Activation in anterior superior temporal gyrus
Incubation
Walking away from the problem to then come back and be more likely to solve it - rumination? forget misleading information?
Evaluation of the Gestalt approach
- Recognises the role of insight
- Mechanisms underlying insight are unspecified
Representational change theory O
Ohlsson, 1992
changed by altered representation or relaxing constraints on what moves are allowed
Evaluation of representational change theory
Explains some of the mechanisms underlying insight but doesn’t explain what leads to representational change and why incubation helps
Information processing approach N & S
Newell and Simon (1972), computational modelling approach - problem solving is searching through a problem space
Means end analysis
Sub-goal creation
Meet sub-goal and thus closer to overarching goal
Information processing approach evaluation
Led to well-specified computer models but doesn’t work for everyday problems (ill-defined) or insight problems
Analogical problem-solving
Recognise two problems are similar, map them and then apply solution for one to the other
Analogical problem-solving evaluation
Retrieving analogies is hard unless the problems share similar surface features - real-life problems harder since contexts are likely more distant
