Response Sets in Learning and Problem Solving
A set is a tendency to act in a new situation as one acted in a previous situation. Nonspecific transfer, and the learning of principles, are examples of sets.
In this module we will examine the importance of sets in various learning
tasks. As you read the text, try to answer the following questions.
RESPONSE SETS IN ACQUISITION AND PROBLEM SOLVING
The term set refers to a tendency to respond to a new situation in a certain way. The runner on the mark is "set" for a quick start. If the official starting the race calls out, "On your mark, get set, STOP!" the runners would be half way around the track before they even slowed down. A jokester asks his victim how to pronounce T-W-A. The person says "Twah." When asked to pronounce T-W-O,``illustrates a set by saying "Twoh."
Set prepares the subject in some way that facilitates or inhibits some future behavior
Set can be established very quickly, as in many jokes and riddles, or it may be the result of long-established practices. Set is not a behavior itself. It is a tendency which influences behavior. In almost all experiments involving human subjects, the experimenter informs the subject what his task will be during the experiment. The experimenter is thereby establishing a set for the subject. The experimenter may say, "l will show you a figure for three seconds. Then the figure will be shown again, but this time it will be paired with a nonsense word; your task is to learn to anticipate that word when you first see the figure." The set is established and the learning facilitated. The facilitatory and inhibitory effects of set have been investigated in acquisition and problem solving in terms of both immediate and long-term effects.
Many human learning experiments attempt to distinguish incidental from intentional or directed learning. Every day we are confronted with a barrage of stimuli, but we intentionally try to remember only a few of these. We do, however, learn incidentally many other things, such as the color of a friend's clothes or what we had for breakfast. We have somehow remembered these items without being set to do so.
Incidental learning is influenced by repetition and meaningfulness of the learning experience
There is no question that sets, as provided by instructions, promote learning. The data show overwhelmingly that directed learning is superior to undirected, or incidental, learning. The problem remains, however, of explaining the relatively small amount of incidental learning that does take place. Under what conditions can it occur at all?
Mechanic (1962) studied the effects of number of exposures and meaningfulness on incidental learning. Subjects were presented with 12 pairs of nonsense words and asked to judge the similarity of each pair of items on a 1 0-point scale. Then, subjects were told to learn the top member of each pair, thus providing a set to learn that pair only. On a later test, subjects were asked to recall all of the pairs. The experimenter used different groups of subjects, and varied both the number of presentations and the meaningfulness of the items. Some of the data are shown in Figure 16.
As the graph indicates, the amount of incidental learning increases with more frequent exposure and is further facilitated by the higher degree of meaningfulness of the material.
LEARNING HOW TO LEARN
It is often observed that subjects who have learned one list of nonsense syllables are better at learning a second list. This phenomenon has been called "learning how to learn."
Harlow (1949) discovered this phenomenon while studying problem solving in monkeys. The problems used were two-choice discrimination problems. Two objects were presented to a monkey on a tray. One object always covered a bit of food, while the other object covered only an empty food well. Eventually the monkey would promptly move the food-related object, regardless of where it was positioned on the tray.
Solving a problem may create a set for the solution of another problem
Harlow presented a series of 344 such problems in which 344 different pairs of stimuli were used. By the final block of trials the monkeys were choosing the correct object after only one trial. Figure 17 shows the increased efficiency of solving the problems as a function of the number of previous problems solved.
Such experiments, obviously, have far reaching implications for school earning. It is only after many trials on related problems that a child develops facility at executing complex skills. These data argue for a careful analysis of curricula with pre specified long-term objectives, so that appropriate problem- solving practice can be given.
SETS IN PROBLEM SOLVING
Set is an important component of problem solving as Harlow (1949) points out:
The learning of primary importance to the primates, at least, is the formation of learning sets; it is the learning how to learn efficiently in the situations the animal frequently encounters. This learning to learn transforms the organism from a creature that adapts to a changing environment by trial and error to one that adapts by seeming hypothesis and insight.
The phenomenon of insight is in part a function of previously established sets. Certainly, previous experiences influence insight. In a study by Jacks (1942) chimps were given a hoe that could be used to drag in a banana placed beyond arm's length. Those chimps who were allowed to play with sticks for several days prior to the experiment did much better than those who had not had such experience.
Set can inhibit rather than facilitate problem solving. The term functional fixity was first coined by Dunker (1945) to refer to the fact that we tend to react to an object in terms of its usually defined function. That is, a hammer is used to drive nails but is not frequently used as a paperweight, although obviously it could be. In an experiment by Adamson (1954) subjects were presented with a table on which there were a number of items, including three different-sized pasteboard boxes. One box contained candles, another matches, and the third thumbtacks. The subjects were told to mount the candles in a burning position on the wall. The solution would be to empty the boxes, tack them to the wall, and then stick the candles on them. However, the boxes were already perceived as serving the function of containers. Only 12 out of 29 solved the problem, but when the boxes were presented empty, 24 out of 28 attained the solution. The degree to which people employ objects for other purposes (i.e., break the functional-fixity set) has been used as a test of creativity.
During the process of solving, problems may become restructured in such a way that a person sees a resemblance to those problems solved in the past. For example, an equation in mathematics may be solved by adding a constant to both sides. In its new form it may be recognized as an equation for a straight line, or a particular curve, and the final steps to its solution are easy.
Such alterations in the form of the problem are called heuristic solutions. They are further examples of the operation of sets that facilitate problem solving.
Unfortunately, sets can also operate to prevent someone from reaching a solution to a problem, or, at least, from reaching the best solution.
A classical study by Luchin (1942) shows the inhibiting or "blinding" effect that a heuristic set can have. The problem involved obtaining an exact measure of 100 quarts of water from an unlimited supply, when one has 3 jars and each holds 21 quarts,127 quarts, and 3 quarts respectively. (You might try to solve the problem yourself before proceeding.)
Several sample problems were presented (Figure 18). Each could be
solved by the general formula B-A-2C
That is, the solution would be to fill Jar B, pour out some of the water into A until it was full, then fill the C container twice. 127 - 21 - 2(3) = 100.
Problems 6 and 7 can be solved by a simpler and more direct method. However the subjects almost invariably continued to use the more complex solution to the problem.
Jar Capacity Problem No. Jar A Jar B Jar CAmount to be Measured 1 21 127 3 100 quarts 2 14 163 25 99 quarts 3 18 43 10 5 quarts 4 9 42 6 21 quarts 5 20 59 4 31 quarts 6 23 49 3 20 quarts 7 15 39 3 18 quarts
Figure 18. Inhibiting effect of a heuristic set (adapted from 1942)
The facilitating effects of heuristic sets in human problem solving far outweigh the inhibiting effects. Almost always, heuristic sets are more helpful in solving problems than trial-and-error, or, even, learning to learn.
Now test yourself without looking back.
1. Which of these would be an accurate description of set?
a. A condition of preparedness which influences subsequent behavior
b. A response made to a specific stimulus
c. Any incidental learning
2. In Luchin's water jar experiment subjects were less likely to use an easy solution because:
a. of the functional fixity associated with water jars.
b. of the heuristic solution established by previous trials.
3. Learning without being given a formal instructional set is called learning.
4. Which statement is true of sets?
a. Sets generally require previous experience;
b. Sets in human behavior usually inhibit efficient learning.
c. To have insight is to be totally free of set.
d. (none of these)
5. In Harlow's "learning to learn" experiments, increased efficiency in learning was due to:
a. degree of meaningfulness.
b. similarity of the stimuli used in the task.
c. similarity of the rewards used in the task.
d. (none of these)
6. Which of these factors facilitate incidental learning?
a. Meaningfulness of material
b. Frequency of exposure
NOW DO THE EXERCISES
Set is a tendency to respond in a predetermined manner that has
been built up by repeated exposures to tasks of a certain type.
Which of these are examples of set?
a. The first solution to a problem in mathematics.
b. The tendency to attack a problem in mathematics by
using a solution that was successful in previous problems
c. (neither) ___________________________________________________________________2
In many learning tasks, subjects are given the appropriate set by
means of instructions. For which experiment below is a set
a. An experimenter wishes to study incidental learning.
b. An experimenter in a paired-associate task is using the method of anticipation to establish association between pairs of nonsense words._____________________________5
When no formal sets are established, any learning that takes place
would be considered incidental. Below are examples of both
incidental and intentional learning. Label them with the type of
learning most probably involved.
a. Recall a method for obtaining square roots.
b. Recall what time you finished dining last night.
c. Recall the color of your father's hair.
d. Recall the alphabet.
A city child playing in the woods for the first time becomes lost. He eventually finds his way out by trial and error. Was this an example of set?_____________________________________________ 3
In human behavior, sets are important because they increase the
efficiency of learning and problem solving. Which is true?
a. The tendency to respond to new situations in predeter- mined ways usually leads to better performance.
b. Sets which influence behavior in problem situations usually inhibit solutions.
Set can sometimes inhibit problem solving if it causes us to overlook
some simple solution. In which case does set interfere with the solu-
a. A trained carpenter doesn't have the hammer he needs, and ignores a heavy wrench lying nearby.
b. Someone who has never used a wrench picks one up and drives nails with it.
If you overheard a telephone number and a line of poetry, you might remember both. It is more likely, however, that the poetry would be remembered than the telephone number. This indicates that:
a. incidental learning is facilitated by meaningfulness.
b. telephone numbers lead to motivated forgetting.
1 a. intentional
3 No (since no previous set could have been established)
Learning to learn is considered to be due to nonspecific factors.
This means that:
a. the experimental situation and type of problems used are not determiners of transfer in learning-to-learn situations.
b. stimulus and response similarity is not an explanation for learning to learn.
Luchin, in his famous water jar experiment, gave some students
the command, "Don't be blind," between the earlier trials (when the
heuristic solution was appropriate), and the last two trials (which
had simpler solutions). The purpose of this command was to:
a. break the set.
b. establish the set. ________________________________________________________3
Functional fixity refers to the set to view an object only in terms of
its usual function. Birch and Rabinowitz (1951) gave students the
problem of tying two strings together that were hanging from the
ceiling. The difficulty was that the strings were too far apart to
reach. The solution required that one of the strings be swung as a
pendulum, then caught and tied. An electric switch, which could
be used as a pendulum weight, was lying on a nearby table. Some
of the students had used the switch in a previous experiment on
electricity. What would you predict?
a. The subjects who had used the switch in a previous experiment would be more likely to use the switch in this experiment.
b. The subjects who had not used the switch in a previous experiment would be more likely to use it in this experiment._______________________________________1
NOW TAKE PROGRESS CHECK 2
1. Incidental learning is defined as learning without____________________________
2. Continued practice with tasks of a similar type leads to a nonspecific kind of transfer that results in the subject's learning the later tasks more efficiently. What is this phenomenon called? ________________________________
3. What is true of set?
a. Set is any transferred learning.
b. Set is a method of learning used in paired-associate tasks.
c. Set is a state of preparedness which influences behavior.
4. The heuristic approach that the subject used in Luchin`s water jar experiment:
a. facilitated the solution in the earlier problems.
b. inhibited the easy solution to the later problems.
5. The meaningfulness of material facilitates:
a. learning to learn.
b. incidental learning.
c. reflex responding.
d. (none of these)
Return to Unit 4 Contents
4 OR MORE CORRECT GO TO UNIT 7
FEWER THAN 4 CORRECT INSTRUCTOR CONFERENCE