problem solving Flashcards
what is a problem
Lovett 2002
when there is an obstacle between the present state and goal, an it is not immediately obvious how to proceed
what are well defined problems
have a correct answer, you have all the info you need, you know what needs to be done
(most research here)
what are ill defined problems
may have more than 1 correct answer, not obvious when goal has been reached, route to answer not clear
(like most problems in life)
what are knowledge lean problems
can be solved without prior knowledge like most things in life e.g finding a parking space or completing a puzzle
what is a cognitive miser
someone who is economical with their time and effort on a task requiring thinking
what are knowledge rich problems
only solvable with relevant knowledge e.g making a pina colada or solving algebra
describe the monty hall problem
-3 doors, behind 1 is a car, behind 2 are goats
-pp chooses a door and monty opens another door to reveal a goat
-pp asked whether they want to switch doors or stick with original
-better to switch: sticking with original door gives 1/3 chance of getting car but switching means the other 2/3 chance is with the other door
what type of problem is the monty hall problem
well defined: has definite answer
knowledge lean: uses general rather than specific knowledge
what are the 3 main approaches to problem solving
gestalt approach
information processing
using analogies
gestalt approach: who proposed it
kohler 1920s
what does gestalt mean
whole
what is the basis of the gestalt approach
-problems solved by restructuring it in another way (representational restructuring)
-Ohlsson 1992: we encounter mental block when we encounter problem in the wrong way
gestalt approach: the idea of insight and is it real
Insight: the point at which a solution is suddenly seen or problem becomes clearer (aha moment)
-gestalt psychologists believe insight is caused by sudden restructuring of problem
is it real?
-yes, people do experience it
-BUT it may not be separate cog process, we may gradually arrive at solution but only realise once reaching a certain threshold
gestalt approach: triangle and chain problem
-most likely produce aha moment supporting insight
-unclear whether aha moment was separate cog process or built up thinking that tipped over the edge of threshold causing aha
-uncertain what goes on in aha moment
gestalt approach: barriers to problem solving
experience: we have difficulty disengaging our knowledge of how these objects are used correctly
-duncker 1945: use these objects to mount candle on wall
–box of matches and tacs
–tacs should be used to pin match box to wall and put candle on box
-Maier 1931: hold both bits of string at same time
–pliers provided
–pliers should be attached to one piece of string to create pendulum
gestalt approach: what is functional fixedness
when ideas about an objects function can interfere with using an object more usefully e.g matchbox as a platform and pliers as swinging pendulum
adamson 1952: more people solved candle problem correctly if matchbox was empty (80% vs 40%)
gestalt approach: water jug problem
luchins 1942
-jug a = 15l, jug b = 40l, jug c = 3l
-easy task: fill target cup to 18l (fill a then c)
-mid task: fill target cup to 12l (fill a then c from a)
-hard task: fill target cup to 7l (fill b, then a x2 from b, then c)
-pp who did easiest first got 95% correct on mid task
-pp who did hard task first got 35% correct on mid problem
summary of gestalt approach to problem solving
it requires representational restructuring e.g finding other ways to use an object or frame the problem
BUT no explanation given for how the brain is involved
gestalt approach: what helps to increase insight
incubation
-stop thinking about problem for a while
-simon 1966: helps develop new strategy
sleep
-wagner et al 2004: increased performance for identifying a hidden rule amongst letter strings
what is a mental set
continuing to use a strategy that is not the most helpful
relates to luchins water jug task
who proposed the information processing approach
newell and simon 1970s
basis of information processing approach
-problems solved using search process
–there is an initial state, a goal state and a number of intermediate states (subgoals)
–the problem space = all possible intermediate states
–we perform a search of all possible choices in this space
information processing approach: tower of hanoi problem
-move three discs from A to B
-only 1 disc moved at a time
-a disc can only be moved if no discs are on top of it
-larger disc cannot be placed on smaller disc
information processing approach: methods for solving problems
-planning
-heuristic methods: means end analysis and hill climbing
-progress monitoring
information processing approach: tower of hanoi problem and solving problems
planning
-koppenol - gonzalez et al 2010: pp who spent longer planning moves before acting performed better with fewer errors
information processing approach: why is there a limit on the amount of planning
due to STM capacity
information processing approach: what are heuristic methods to solving problems
computationally cheap rule of thumb that can produce reasonably accurate answers
includes: hill climbing and means end analysis
information processing approach: what does hill climbing and means end analysis solve
the problem of limits to STM
information processing approach: hill climbing as a heuristic method to problem solving
1.changes present state to be one step closer to goal
2.mostly used when we have no clear understanding of the way to achieve the goal
3.relatively unsophisticated and little planning needed
4.do whatever it takes to get one step closer
information processing approach: means end analysis as a heuristic method to problem solving
1.set up goal
2.look for difference between current state and goal
3.find an action that would reduce this difference
4.perform the action
5.repeat until goal achieved
(one of the most common methods we use)
information processing approach: what is progress monitoring and research
tracking progress towards a goal and switching strategy if progress is slow
MacGregor et al 2011: performance worse if pp think progress is being made, when pp realised they were likely to switch strategies
neuropsychological evidence for information processing approach
patients with damage to prefrontal cortex perform worse…
-Goel and grafman 1995: on tower of hanoi than controls
-colvin et al 2001: on water jug task when patients used unsophisticated hill climb strategy
information processing approach research by Kleibeuker et al 2013
-used FMRI to study regions involved in planning in adolescents and adults
-pp presented with grid of matchsticks, given 15secs to determine if problem is solvable ‘remove 10 matchsticks to make 2 squares’
-found response in lateral prefrontal cortex (PFC) during problem solving
-activity correlated with performance (better problem solving = more activity)
information processing approach and neural activity for insight
-neural activity before seeing problem predicts whether problem will be solved with or without insight
research evidence for insight and neural activity (information processing approach)
-Kounios et al 2006: problem = what connects the three words, activity seen in posterior middle temporal gyrus and cingulate
–suggests relation to cognitive control and conceptual processing
-Qui et al 2010: neural activity correlated with insight during solving chinese logogriphs
–activity in precuneus, L inferior/ middle frontal gyrus, inferior occipital gyrus
–suggests relation to memory retrieval, changing strategies, visual imagery, attention
evaluation of the studies on neural activity and insight
-no regions in common seen in the 2 studies
-huge differences in the studies (stimuli, task, design etc) so hard to compare
-HOWEVER these studies have attempted to examine the process involved in problem solving unlike gestalt approach
information processing approach and planning research
crescentini et al 2012
-FMRI to look at brain regions used in planning when problem solving
-tower of hanoi: activity in dorso lateral prefrontal cortex (key area involved in initial planning)
strenghts of the information processing approach
+ provides precise account of problem solving by applying rules to reduce complexity
+ general so applies to a variety of problems
+ fits with standard models of memory e.g limited capacity of STM
weaknesses of information processing approach
assumes problem solving is serial
can it be applied to ill defined problems
what are analogies
using solution from one problem to guide solution to similar problem
analogy approach: key example
PROBLEM
duncker’s radiation problem (Gick and Holyoak 1980)
-high intensity lazer: kills tumour and healthy tissue
-low intensity laser: harmless to healthy tissue and too weak to kill tumour
ANALOGY
the fortress
-a general wants to capture a fortress
-roads have mines, small groups of soldiers can pass through safely but large force would detonate the mines
what is analogical/ transfer knowledge for the key example of dunckers radiation problem
multiple low intensity lasers would be best, as would few men on multiple roads
analogy approach: duncker’s radiation problem findings
-10% of pp could solve problems with no analogy help
-30% could solve after hearing fortress story
-75% could solve problem after being told the 2 stories are related
what are the 2 types of reasoning experts use to solve problems
explicit reasoning: slow, deliberate, consciously aware
implicit reasoning: fast, automatic, not always aware
what reasoning do medical experts usually use and evidence
implicit
krupinski 2011: radiologists can fail to detect diseases in about 30% of cases
research evidence for medical experts
krupinski et al 2013: pathologies viewing breast cancer biopsies make less fixations and looked less at diagnostic regions as their expertise increased
Kundel et al 2007: Drs viewing mamograms for cancer are more likely to identify the cancer if they make an eye movement to the cancer quickly
–drs with most experience fixated on cancer almost immediately suggesting they used holistic and global processes rather than serial/bit by bit search
chase and simon 1973 study
task: memorise location of chess pieces arranged randomly or in game positions
-experts remembered game positions better than random
-chess experts do not have better general memory
-they remember better in game positions due to organising the info differently (chunking relevant info)
what is framing the problem
framing: how you frame a problem can influence its difficulty and complexity
e.g problem ‘the lift is too slow’: solution = install new lift, upgrade motor
OR reframe problem to ‘the wait is annoying’: solution = make wait seem shorter by putting mirrors up or playing music
