Week 10: Problem Solving Flashcards
What is problem solving?
- refers to a cognitive process that involves discovering, analyzing, and solving problems
- is recursive and cyclical
What are the steps to problem solving?
- Recognize or identify the problem.
- Define and represent the problem mentally.
- Develop a solution strategy.
- Organize knowledge about the problem.
- Allocate mental and physical resources for solving the problem.
- Monitor progress toward the goal.
- Evaluate the solution for accuracy.
Initial state
Describes and defines the type of problem
Goal state
A state ofresolution/over-coming the problem
Well-defined problems
have correct solutions & clearly defined states
Ill-defined problems
may not have just one correct answer & problem state may be unclear
The behaviouralist approach
Solutions are discovered via trial-and-error learning.
The Gestalt approach: Wertheimer 1959
Solutions are discovered by cognitive restructuring
The Law of Effect (Thorndike)
- Problem solving may be viewed as a reproductive process, which involves reusing information gained from trial-and-error experiences to generate responses.
- Any response that produces a
satisfying effect/positive outcome
gradually becomes stronger - Any response that does not produce
a satisfying effect/positive outcome
becomes weaker - Precursor to operant conditioning, and modern reinforcement learning methods (in AI).
Wertheimer (1959): The Gestalt Approach
Problem solving is (sometimes) a productive process.
Insight: Metcalfe & Wiebe (1987)
- when a solution to a problem suddenly occurs to you
- not good at predicting how close to solving problem
Productive Process/ non-insight: Metcalfe & Wiebe (1987)
- Thinking about a problem by restructuring and/or recombining
information in your mind (unconscious) - good at predicting how close to solving problem
The cognitive cycle
collect info > choose action>
Recursive Dynamics
Steps are repeated as many times as necessary.
Intermediate States
All of the possible states between
initial and goal states
operators
Actions that transform current problem states into others
problem spaces: (Newell & Simon, 1972)
- Search through the problem space, creating subgoals as needed
- move through problem space using operators
Use of heuristics more common…
o Information overload
o Time-sensitive tasks
o Unimportant circumstances
o Limited information availability
o Competing heuristics
Working backwards (backward chaining)
Focused on the end result to solve a problem
Means-End Analysis
achieving goals by creating subgoals that move you closer to end result
Barriers to solving problems
- ignoring relevant info
- functional fixedness (Maier 1931)- tie ropes together task & most struggled to solve iwthout a cue
ignoring relevant info
- one must decide what is relevant
- is acquired and not innate
- ## this is more difficult when dealing with ill-versus well-defined problems
functional fixedness
- perceiving an item only in terms of its most common use
- inhibits other potential interpretations of problem scenarios
- people tend to focus on a characteristic of a problem which prevents them from finding a solution
creativity
being able to produce novel ideas that are appropriate and relevant to the situation
convergent thinking: Guilford (1967)
Approaching a problem using well established knowledge, rules and methods
Divergent thinking: Guilford (1967)
Approaching a problem by generating many possible solutions (this includes creativity)
Ideational Fluency
The number of ideas a person can generate about a topic or item
Mental sets
The tendency to use solutions that have worked in the past, or to respond to something in a set way
Conditions of mental sets
- blind to alt sol’ns
- tend to create mental sets while paying attention to similarities or relationships between past and current problems
- once that relationship has been established, people will try keep trying solutions that worked in the past even if the current problem requires a different solution
How experts organise knowledge (Chase & Simon 1973)
- chess: experts thoose advantage when placement is incorrect
spend more analysing a problem - spend less time planning steps to solve it
- have automated processes basded on experience
- spend more time comparing previous experiences to current ones
- use what they already know
- quickly execute once they know what to do
The 9 dot problem (Maier, 1930)
Connect the nine dots with four straight lines drawn without lifting your pencil
