PROBLEM SOLVING

A person experiences a problem when he wants to attain a goal but that goal is not
readily available. An individual experiences a problem when a gap exists between the
current and the future state and he doesn’t know how to cross that gap.
So, a problem consists of three features:
 The Original state: It describes the situation at the beginning of the problem.
 The Goal state: It’s reached when problem is solved.
 The Rules: These are the restrictions to be followed to proceed from the original
state to goal state.
When people solve problems, they seldom take a random trial-and-error
approach, blindingly trying different options until one option finally provides a
solution. Instead they plan their attacks, typically breaking the problem into its
component steps and devising a plan for solving each component. In addition to
plans, problem solvers also use strategies which are likely to produce a solution
relatively quickly.

Characteristics of problem-solving:
1. Goal-directedness: Problem-solving is a mental activity which involves reaching a
specific goal.
2. Sequence of operations: Problem-solving involves sequence or steps of operation.
3. Cognitive operations: Achieving a solution to the overall goal involves the
application of various cognitive operations, like for a long division problem,
operators like adding, subtracting, multiplying etc would be needed.
4. Sub-goal de3composition: A sub-goal is an intermediate goal along the route to
eventual solution of the problem. Sub-goals represent the decomposition or
breaking apart, of the overall goal into separate components.

 Different steps in problem-solving :

1) Understanding the problem : Greeno (1977) believes that understanding a problem
involves constructing internal representation of the problem. It involves 3 features:
 Coherence: Coherent representation is a pattern that is connected so that
parts make sense.
 Correspondence: According to Greeno (1977), understanding a problem
requires correspondence between internal representation and the material
being understood. Sometimes, internal representation is incomplete and
inaccurate. Important relationships among the parts of the problem may be
left out or mismatched.
 Relationship to background information: It involves understanding the
current problem in the context of knowledge or information that the problem
solver already has.
To understand a problem one must pay attention to crucial information, ignoring the
irrelevant ones. The capacity of working memory, if consumed by off-topic, irrelevant
thoughts would have a detrimental effect on problem-solving. E.g. the anxious thoughts
about doing math. Bransford and Stein presented algebra word problems to a group of
college students like, Train travelling in one direction and bird flying in another
direction.” Initial reactions to problem by the students involved thoughts, “Oh no, this is
the mathematical word problem-I hate those things!” and “Boy, I am going to look
stupid.” It was seen that these negative thoughts occurred frequently throughout 5
minutes time which was allotted to solve the problem, which distracted them from
problem-solving.
2) Representing the problem : Once the problem solver has decided which information is
essential and which can be disregarded, the next step is to find a good way to
represent the problem. Simon and Hayes 91976) argue that people regard problems as
‘cover stories’ for the ‘real problem’. Thus, they must discover the abstract puzzle
underneath all the details and then they must find a good way to represent this
abstract puzzle, because an abstract puzzle is difficult to keep in memory while
performing the operations. So, people typically invent some method of representing
the abstract problem in a particular concrete way that shows only essential
information. They make use of symbols, lists, matrices, hierarchical tree diagrams,
graphs, visual images etc. Some problems can not be represented by some of the
methods whereas some can be represented by several possible ways.

3) Adopt the strategies to solve the problem: Once the problem has been represented
one can use many different strategies to attack it.

 Problem-solving approaches:

Some strategies are very time-consuming. E.g. Algorithm is a method that always
produces solution to the problem, sooner or later. It involves searching systematically
through the entire problem space. Newell and Simon (1972) observed that time taken to
search for an answer to a problem is roughly proportional to the total size of problem
space i.e., all possible solutions that have occurred to the solver. It may vary like small
for anagram like YBO and enormous for one like LSSTNEUIAMYOUL.
In problem solving, heuristic is a rule of thumb involving selective search. It
consists of looking at only the portions of problem space that are most likely to produce a
solution. Like, in above situation the strategy adopted may be searching for most likely
spelling combinations.
Though algorithms always guarantee solutions, heuristics are widely used, as
algorithms do not exist in most everyday life. This is the reason why Psychologists have
conducted more research on problem solvers’ heuristics. Most widely used heuristics are
as follows:
 Means-Ends Analysis: In this approach, the problem is divided into a number of
sub problems. Each sub problem is solved by detecting the differences between
original state and goal state and reducing the difference between them. It involves
figuring out the “ends” and then the “means” to reach the ends.
 Generate and test technique: It consists of generating the possible solutions and
testing them. This technique looses its effectiveness when there are many
possibilities and when there is no particular guidance over the generation process.
Also, if there is no way to keep track of possibilities one has tried, along the ones
one has tried; one might be in real trouble.
  Analogy Approach : It involves referring to previous, familiar problem to solve a
new problem. The set of problems with the same underlying structures and the
solutions, but with different specific details and contexts are Problem Isomorphs.
Novick (1988) suggest that problem solvers, especially the novices focus on
salient surface features (specific terms and objects used in the question) but, fail
to emphasize structural features (underlying core that must be understood in order
to solve the problem correctly.)
Novick asserts that solving by analogy requires 4 features;
i) Retrieval or locating appropriate source problem,
ii) Mapping or constructing orderly correspondence between the parts of source
problem and parts of target problem,
iii) Adaptation or determining how to use the same procedures for target problem
that were successful for source problem.
iv) Learning or figuring about an abstract schema for the entire class of problems
that the source and target represents.

 Factors influencing problem-solving:

Characteristics of the solver

1) Expertise: : Experts differ from novices with respect to their memory for task related
material, their knowledge base, method of problem representation, appreciation of
structural similarity, greater elaboration on initial states, speed and efficiency and
metacognitive self-monitoring.
2) Mental set: When problem-solvers have a mental set, they keep trying the same
solutions of previous problems, though new problem could be approached by other easier
ways. The classic experiment in mental set is Luchen’s Water-Jar problem.

Characteristics of “PROBLEM”

3) Functional Fixedness: It means that functions or uses we assign to objects tend to

remain fixed or stable. To overcome functional fixedness, we need to think flexibly about
other ways that object can be used. Duncker’s candle problem is a classic experiment
demonstrate functional fixedness.
4) Well-defined/ill-defined problem: A well-defined problem is the one in which the
original state, the goal state and , the rules are clearly specified. However, when these are
unclear it is an ill-defined problem.
5) Insight/non-insight problem: When we experience insight, solution to a problem
suddenly enters our minds and we immediately realize that the solution is correct.
Wertheimer suggest that insight involves “Productive thinking”-an insightful mode that
goes beyond the bounds of existing ideas or associations or “Reproductive thinking”-
which is based on what is already known.
 RESEARCH ON 9 DOT PROBLEM

The display of 9-dot problem consist of 3 rows of 3 dots each. The rows and
columns are equidistant from one another, and the pattern of 9-dot is perceived as a
square. The task is to connect the nine dots with four straight lines without lifting the
pencil from the paper. Thus, 9-dot problem is a well defined problem, involving insight.
Though it seems to be a simple task investigators have found that people either take along
time to arrive at the solution or fail to solve the problem.
One interpretation for the difficulty of the problem is based on Gestalt Psychologists
viewpoint. According to this approach, Ss are assumed to focus on square shape of the
dot pattern and confine their lines to square area. Problems becomes difficult because of
S’s assumption that lines must stay within points or dots. To solve the problem, S’s need
to restructure the original problem situation in some other way based on a new direction.
Weisberg & Alba (1981) argued against Gestalt viewpoint and proposed a
hypothesis-testing framework. According to them, when Ss confront the problem, they
will sample the hypothesis that the problem can be solved in the way they usually solve
dot-to-dot puzzles. Ss may fail because they do not search the domain of solution within
the allotted time. Even if the domains are exhausted Ss may still not find the solution
because they fail to notice that other domains of solution exist, due to lack of relevant
experience, by showing specific lines that form the solution. However, according to Lung
and Dominowski 91985), there are many problems with Weisberg and Alba’s (1981)
framework. The notion of ‘domain of solutions’ didn’t get much support in their data.
An alternative viewpoint is based on Information-processing approach. According
to this approach, success in solving these problems require;
• appropriate representations of the task
• environments in S’s problem space and
• applications of strategies which execute effective evaluation functions.
Steps undertaken are:
1. Selection of a dot as a starting point
2. Evaluation of available lines
3. Selection of a line based on outcomes of evaluation
4. Selection of a new starting point.
Appropriate strategy to 9-dot problem not only requires that lines must go
outside the boundaries of square, but also requires extending a line beyond the last dot in
that line to a point which is an intersection of that line and a line drawn next. Although, a
hint to go outside the square imply using an outside point as a new starting point, the
implication is not necessarily strong enough to help Ss to solve the problem. This was
probably why most of Weisberg and Alba’s Ss even after receiving the hint could not
solve the problem.
Square shape of the 9-dots is not the only factor which makes it difficult to solve
the problem. The strategy that the problem-solvers use is another critical factor
influencing whether or not the problem is solved. This, however, doesn’t mean that the
two factors are completely independent. The strategy that subject will use for solving the
problem is highly related to how the problem is represented. Subjects may have a strong
tendency to represent the 9-dot problem as a square simply because the way dots are
arranged. This may make it difficult for the subject to grasp the appropriate strategy of
extending and intersecting lines outside the dots. So it’s expected that subjects learn
proper representation and strategy to solve the problem.

 Original experiment by Lung and Dominowski (1985) :

A study was designed by Lung and Dominowski (1985) to investigate several

questions like:
 What were subject’s usual approaches to solving the problem? (Both Ss’
retrospective reports and lines they drew were provide some information).
 How effective it was to instruct the Ss to use a strategy of extending and
intersecting lines outside the dots?
 Could Ss acquire strategies appropriate for solving the problem by experience in
solving other similar problems?
The design used by them was 2 X 2 randomised design. The two independent
variables were Practice (Yes/No) and Strategy Instructions (Yes/No). The practice group
was given 6 practice problems before 9-dot problem. Strategy Instructions group received
Strategy Instructions and then they attempted the 9-dot problem. The Strategy
Instructions + Practice group received Strategy Instructions, attempted the practice
problems and then solved the 9-dot problem. The control group was only asked to solve
the 9-dot problem.
The results indicated that both Strategy Instructions and Practice improved
performance on 9-dot problem. More Ss solved the problem, extended a line and
intersected lines outside the pattern in 3 experimental conditions as compared to Control
condition. This supported the hypothesis that Strategy instruction os extending and
intersecting are helpful. The practice effect that was found disconfirmed Weisberg and
Alba’s (1981) argument that 9-dot problem was so difficult that significant facilitation
could be brought about only by giving Ss detailed solution information.
Although both Strategy instructions and Practice increased the likelihood that Ss
would reach the solution and the intermediate goals, practice was generally more
effective in reducing the number of trials required for the same.
Main results did stress the importance of appropriate strategy but other findings
indicated that perceptual factors can not be ignored. The square shape of the 9 dots did
matter : Ss whose first two lines most closely to the square (two sides) were least likely
to solve the problem.
The finding that practice problems with centre dots were more difficult than those
without centre dots implied that the presence of centre dot contributed to the difficulty in
solving 9-dot problem.
Past experiences biases initial representation of problem in a manner that
hinders solution, & that to overcome this a change in the problem representation is
required ( Ohlsson,1992; Knoblich et al., 1999). People adopt a locally-rational strategy
of drawing lines that intersect the maximum possible number of adjacent dots. This
enables them to approach the goal by the most ‘seemingly-direct route’. Provided that the
resultant move satisfies a “criterion of progress”, it will be selected. When it does not,
then “criterion failure’ occurs, resulting in an impulse to seek alternative moves, which
may lead to the relaxation of constrants. Macgregor et al. (2001) found that criterion
failure was more effective in facilitating solutions than a visual hint to extend lines
beyond the boundary of 9-dots.
Ormerod, Chronicle & Macgregor(2001) suggest that success at re-solving insight
problems depends upon retrieving conceptual, perceptual & procedural knowledge in a
‘useable & integrated’ form.
Eg: redescribing the solution to 6-coin problem as “pincher” gives ‘perceptual’ &
‘procedural’ knowledge or a hint of “destroy in order to create” to necklace problem
gives a ‘conceptual understanding’.
Neuropsychological Findings:
fMRI results revealed an increased signal in the right anterior superior temporal gyrus for
insight relative to non-insight solutions. A EEG finding found a sudden increase in alpha-
band frequency in right visual cortex. ( Bowden,2005 )
Luo and Knoblich(2007) found that an activation in hippocampus using the ‘riddles’ and
an activation of anterior superior temporal gyrus using ‘compound remote association’.
Insight- solving problems also involves reduction in posterior beta oscillations in parieto-
occipital region & increase in gamma band in the right frontocentral region.