problem solving

Question 1

Define Problem Solving

Answer 1

• it’s purposeful (goal-directed)
• involves controlled processes & isn’t totally reliant on automatic processes
• a problem exists when someone lacks relevant knowledge to produce immediate solution

Question 2

Well-defined vs Ill-defined problems

Answer 2

Well-defined:
* problems where the initial state, goal & the methods available for solving problem are clearly laid out
* e.g. goal of winning chess

Ill-defined:
* problems that are imprecisely specified
* e.g. initial state, goal state & methods available to solve problem may be unclear
* e.g. goal of becoming happier

Question 3

Gestalt: Reproductive vs Productive Thinking

Answer 3

Reproductive:
* involves re-use of pervious experience & knowledge

Productive:
* involves insight after restructuring the problem

Question 4

Knowledge-rich vs Knowledge-lean Problems

Answer 4

Knowledge-rich problems:
* can be solved by those with relevant knowledge

Knowledge-lean problems:
* don’t require such knowledge as most of the info needed is contained in the initial problem statement
* most research focuses on these as such problems minimise individual differences in relevant knowledge

Question 5

6 Theoretical Approaches to PS:

Answer 5

1. associationist & behaviourist
2. gestalt
3. learning set
4. representational change theory
5. information processing
6. adaptive control of thought-rational (ACT-R) model

Question 6

Associationist & Behaviourist Approach

Answer 6

Associationist Approach:
Aristotle: Laws of Association:
- law of contiguity (after events occur together (in spatio-temporal proximity), the reoccurrence of only one event evokes the ‘memory’ of the others)
- law of similarity (experience or recall of one object will elicit the recall of things similar to that object)
- law of contrast (experience or recall of one object will elicit the recall of opposite things)

Behaviourist Approach:
* Trial & error learning (Thorndike, 1898)

Trial & error learning:
- learning curve

Question 7

Associationist & Behaviourist Approach: Evaluation

Answer 7

+trial & error learning can be applied to real-life PS

-the associationist laws & stimulus-response associations are limited to simple problems
-application is to reproductive problems, not productive problems

Question 8

Gestalt Approach

Answer 8

Gestalt Theory (Wertheimer, 1954):
* PS is focused on productive or novel problems

Productive problems involve:
- cognitive restructuring: the representation of a problem is changed allowing a solution
- insight: the sudden realisation of a problem solution; accompanied by the “ah-ha” experience

Restructuring & Insight:
* problem is restructured & solution found
* experience the ‘ah-ha’ phenomenon (insight)
* Kohler’s (1925) apes

Functional Fixedness:
* the function of objects is fixed
* solution derived only if fixedness is overcome
* object used in untraditional way for problem to be solved
* candle problem (Duncker, 1925)

Question 9

Gestalt Approach: Evaluation

Answer 9

+approach has left a legacy of tasks, problem types & data
+offered an alternative approach to the associationists/behaviourists
+Gestalt approach led to the information processing approach

-cognitive constructs ‘under-specified’ (e.g. insight, cognitive restructuring, etc)

Question 10

Learning / Mental Set Approach

Answer 10

Luchins (1942):
* combines Behaviourist (trial & error) & Gestalt (insight)
* the tendency to use a familiar PS strategy that has worked before even when it’s not appropriate
* mental set can impair PS - use past experience knowldege to solve new problems - low level of processing - however is often useful

water jars study (Luchins, 1942):
* 3 water jars, need to make 1 jar 25 quarts
* ppts given familiar problems prior: 95% solved it
* ppts trained on water jar problems: 36% solved it

Bilalić et al (2008):
* chess players used familiar way rather than quickest way to win game
* mental set solves problem but not in the most efficient (quickest) way

Question 11

Information Processing Approach

Answer 11

Newell & Simon (1972):
* people have limited STM capacity
* complex info processing is serial (one process at once)
* e.g. Tower of Hanoi - move one disc at once to solve
* problem space: possible states of problem - initial state, goal state, mental operators (moves)
* use heuristics (rules of thumb) & algorithms (complex procedures) to PS

Question 12

What is the most important heuristic method? (Newell & Simon, 1972)

Answer 12

Means-ends Analysis:
* create sub-goal to reduce difference between current state & goal state
* select mental operator to achieve sub-goal

Question 13

Info Processing Approach & Means-end Analysis: Evaluation

Answer 13

Info Processing Approach:
+comprises of a normative theory of PS (& allows investigation of deviations from the ideal)
+applied to numerous problems, including a re-interpretation of Gestalt theory
+problems solved using general-purpose heuristics

-ecological validity of problems is lacking (unfamiliar, little knowledge, well-defined)

Means-end Analysis:
+generally very useful & assists PS

-people sometimes persist with a heuristic even when it impairs performance (Sweller & Levine, 1982)

Question 14

Representational Change Theory

Answer 14

Ohlsson (1992):
* may encounter a block when PS due to representing problem wrongly
* need to change problem representation for insight to occur
* similar to restructuring in Gestalt, but with more detail about processes leading to insight

Can change this in 3 ways:
* constraint relaxation: inhibitions on what is regarded as permissable are removed
* re-encoding: some aspect of the problem rep is reinterpreted
* elaboration: new problem info is added to the rep

Question 15

Representational Change Theory: Evaluation

Answer 15

+extended Gestalt theory
-can’t predict when or why a problem’s rep will change
-de-emphasised important individual differences in PS skills

Question 16

Adaptive Control of Thought-Rational (ACT-R) Model Approach

Answer 16

• Anderson et al (2004, 2008): ACT-R is a general cognitive architecture
• Early versions (HAM, ACT & ACT*) developed within cognitive science
• Recent approach (ACT-R) combines cog science with cog neuroscience

Question 17

ACT-R Model: 4 Modules involved in PS

Answer 17

6 modules but 4 important ones:
* Retrieval module: maintains retrieval cues needed to access info
- located in inferior ventrolateral prefrontal cortex (VLPFC)
* Imaginal module: transforms problem representations
- located in posterior parietal cortex
* Goal module: keeps track of an individual’s intentions & controls info processing
- located in anterior cingulate cortex
* Procedural module: uses production rules (if…then) to determine action to be taken
- located in head of caudate nucleus

Question 18

ACT-R Model: Evaluation (evidence)

Answer 18

• Qin et al (2004) tested children solving equations over 6 day period
• VLPFC no activity when retrieval not required
• Practice led to reduced activation in brain areas associated with retrieval, imaginal & procedural modules
• Goal module showed no reduced activity during learning
• however, problem-solving strategy remained the same across the activity

Question 19

ACT-R Model: Evaluation (strengths & weaknesses)

Answer 19

+provides theoretical framework for performance on a number of cognitive tasks
+combines cognitive science & neuroscience providing a theory-driven approach
+provides an explanation of how a number of brain regions may be activated when completing a cognitive task

-gives only a minor role to the prefrontal cortex when studies have found the dorsolateral prefrontal cortex involved in PS
-the proposed modules may not be as distinct as assumed (e.g. Danker & Anderson, 2007)