Lecture 8 Flashcards
Why do we study problem solving
Part of everyday life, goal-directed, e.g. how to get to a lecture - but we can use existing schema’s to problem solve. However, a problem only becomes a problem when you don’t know how to solve it and lack all h relevant knowledge to produce an immediate solution
Is problem-solving just in humans?
No. Kohler’s chimp Sultan, observed complex problem-solving.
Discuss Kohlers chimp Sultan’s problem
The banana’s are too far away and too high to reach. However there are obstacles in the environment which he uses to stack and climb to reach the banana’s.
What features can you apply to any problem
Goal-directedness. Subgoal decomposition (simply reaching for food would not have been problem solving there had to be something else). Operator application (action that transforms the problem state into another state).
Define problem solving
Goal-directed behaviour that often involves setting sub-goals to enable the application of operators
When does problem solving take place
If the person is trying to obtain a goal
What is the main task in problem solving
Finding the right intermediate steps to get to the goal state - counter-intuitive move
How do you find the intermediate steps
Initial state, goal state, intermediate steps required
What is an operator
Legal operations or functions.
What are the two types are operation
Legal - means permissible given the rules of the problem. Illegal operations
What is a goal
The ultimate destination, solution or state of problem.
What are the two types of goals
Well-defined goal - any problem with an explicit goal. Ill-defined goal - several different results could satisfy the goal
What is goal directedness
Problems must have a goal
Describe sequence of steps
Must involved multiple operations, simple recall is not problem solving
Define cognitive operations
Done to get from one step to another, steps are discrete and based on cognitive acts such as recall or maths
What is subgoal decomposition
Each step is also a goal, break apart problem into smaller bits. Gives hierarchical structure to problem solving
How do we study problem solving
Speed/accuracy RT measures. Intermediate products. Verbal protocol - however the act of verbalizing the steps might change how the person solves the problems - introspection? Computer simulation - model hypothesized processes
How do we solve problems
Gestalt - focusing on the whole. People generally do not perceive the individual pieces or features in a scene, but extract their relationship.
What is the key to problem solving
Extract the relationship among items - understanding the relationship between items, especially non-obvious relationships, is what allows us to solve problems
Name and discuss the difficulties in problem solving
Persistence of set - old strategies continue to be used, even if less efficient. Incorrect problem representations. Functional Fixedness - tendency to use objects in standard way. Negative set - tendency to solve problems in a set way, even when a different approach may be more successful. Individual differences. Lack of expertise - experts and novices solve problems differently
Why do we use functional fixedness
Semantic knowledge about something leads to recall of normal properties and uses of that object - ALTERNATIVE uses are not obvious
Explain negative set
Einstellung - a bias to solve a problem in only one way, related to functional fixedness. problem solvers keep trying the same solution used in previous problems, even thought the problem could be solved via other ways
What is a set in problem solving
State of mind - an inappropriate set can stop you representing the problem in the most appropriate way
Discuss whether incubation works
Representational change theory - emphasis on factors that allow people to overcome a block to solving. Incubation - a problem is solved more easily if set aside for some time
What is the sequence of problem solving
Identifying problem. Represent problem. Plan the solution. Execute plan. Evaluate the plan. Evaluate the solution
What is on algorithm
A systematic procedure guaranteed to lead to a solution - generally it would guarantee a solution but often time consuming
What is the difference between algorithms and heuristics
Procedures guaranteed to result in the solution = algorithms. Rule of thumb that often leads to a solution = heuristics.
Name the methods of problem solving
Brute force search. Hill climbing. Working backwards. Means end analysis. Mental imagery
What is brute force search
Test all possible states until solution is found. Guaranteed to work eventually, but impractical with large problems
What is hill climbing
States you should always move towards the goal state - never backwards. Can be problematic - correct route may involve moving away from the goal state before returning to it later.
What is working backwards
Approaching the problem from the goal state. Particularly useful when there are too many possible operators at the initial state. MOST EFFICIENT when only one correct solution - rapid elimination of unhelpful paths
When does working backwards actually work
When the goal is singular and there a a variety of alternative routes to take - identifies optimal route. When end results are given or known in the problem and you’re asked for the initial conditions
What is the means-end strategy
Identifies distance between current state and goal state. Finds operator to reduce the difference. Sometimes needs to set sub-goals to remove obstacles. Extremely general and powerful method of problem solving
What are the 5 steps of means-end strategy
Set up goal or subgoal. Look for difference between current state and goal state. Look for an operator that will reduce or eliminate the difference. Apply the operator. Apply steps 2-4 until all goals are achieved.
Why do people only engage in a modest mount of planning when engaged in problem solving
Limited short-term memory capacity. Costs associated with the time and effort. Often simple heuristics suffice.
What is positive transfer
Previous experience allows problem to be solved faster/more easily
What is negative transfer
Previous experience disrupts ability
Discuss analogical problem solving
Analogies = relationship between 2 similar situations, problems and concepts. Difficulty lies in recognising that problems are similar - many problems on the surface appear to be very dissimilar, but are similar at a core level
Discuss the research on analogies
Gick & Holyoak (1980) - radiation, attack dispersion and parade problems used to study analogy. Reading one story influenced whether or not participants solved another story - participants were able to use the first as an analogy for solving the second
Discuss solving the radiation problem (Gick & Holyoak, 1980)
The solution is to use several lower dosage rays to ht the tumor from different angles - the rays are weak enough that they do not damage tissue - but at their convergence, they have enough power to kill the tumor
Discuss the results from the Gick & Holyoak 1980
Parade condition (parade story preceded radiation story) = 49% gave the correct radiation solution. Attack condition (attack story preceded radiation story) = 76% correct solution. Control condition (No story preceded radiation story) = 8% correct
Discuss the conclusions of Gick & Holyoak, 1980
People used previous solutions from analogous stories. But the likelihood of sing a solution depended on how analogous the story was to the radiation problem - the attack story gave more correct solutions and it was most closely matched to the radiation problem
What is the multiconstraint theory
Predicts how many people use analogies. Predicts what factors govern the analogies people construct.
What are the three factors involved in the multiconstraint theory
Problem similarity. Problem structure. Purpose of the analogy
What conditions do people make successful use of previous problems
Chen (2002) identified 3 factors: Problem similarity - there must be a reasonable degree of similarity between the source domain and the target domain. Problem structure - parallel structure between the source and target problems. Procedural similarity - procedures for turning the solution principle into concrete operations are common to both problems - the goals of the stories in the analogy must match
Discuss the properties of expertise
Domain-specific knowledge. Develop superior domain-specific memory. Expert knowledge i highly procedural - faster and fewer mistakes. Experts representations consist of underlying properties
What do experts do that novices don’t?
Experts = elaborate, plan, detect constraints & have better judgments. Whereas novices work on small sub-problems and are less flexible in attempting solution, experts remember more domain-specific information in meaningful clusters.
What are the characterizations of experts
Excel mainly in their own domains. Perceive large meaningful patterns in their domain. Fast. See and represent a problem in their own domain at a deeper level than novices- they see structural similarities between problems. Work forwards from givens to unknowns. Have strong self-monitoring skills
Discuss puzzles
Unfamiliar, involve no prior knowledge. All necessary info is present. Requirements are unambiguous
Discuss real-world problems
Familiar, require prior knowledge. Necessary info is not present and goal is often sometimes ambiguous
Discuss the ways to improve problem solvin
Increase domain knowledge. Automate components. Have a systematic plan. Draw inferences. Develop subgoals. Work backwards. Search for contradictions and relations among problem. Use different problem representatins
What makes and expert an expert
Study exception feats: memory experts, chess experts, musicians, athletes. Practice for 10 years - 10,000-20,000 hours playing… but not always the case
What about talent?
Maybe exceptional performance in some area that can be explained by talent? Ericsson et al. (1990) disagreed that concept of talent is useful or explains anything. Instead, it is the use of deliberate practice & GRIT - perseverance, motivation, dedication
