Pertinent Problem-Solving Literature

The literature on problem solving does not appear to be bounded by time or space. Historical figures such as James (1890), Thorndike (1898), and Dewey (1933) were contributers and current areas of inquiry include such diverse topics as creativity training and teaching computers to play chess. In his 1966 review, Davis remarked that research in human problem solving has a well-earned reputation for being the most chaotic of all categories of learning. "It is almost definitional of laboratory problemsolving experiments that virtually any semi-complex learning task which does not clearly fall into a familiar area of learning can safely be called 'problem solving' (p. 36)." Duncan (1959) voiced a similar lament in an earlier review, and the word "chaotic" remains an apt descriptor today.

The decision-making and problem-solving literatures overlap to a considerable extent; indeed, some authors use the terms interchangeably (see Krumboltz & Thoresen, 1976, pp. 368-414). Others view decision making as merely a subset of problem solving; for example, D'Zurilla and Goldfried (1971) reduce decision making to a single component in a five-stage general model of a problem-solving process. Still other authors (for example, Brinkers, 1972; Janis & Mann, 1977; Lindley, 1971) treat decision making as a fully autonomous entity. The reader is invited to adopt whatever conceptual framework seems clear and feels comfortable. My own theoretical bias is to set decision making in bold relief as an academic discipline and as a counseling strategy. However, there are anumber of concepts in the amorphous problem-solving literature that have a direct bearing on the conduct of decision-making counseling. This chapter is a concession to their relevance.

Some Definitional Difficulties

---

Counseling problems might be conceptualized in a similar manner. A client has a problem when he or she feels called upon to act but does not know which action is appropriate. What to do about a troubled marriage or how to cope with an intrusive supervisor are typical of the myriad problems clients bring to counselors. However, problems studied in experimental laboratories, for example, how to build a coat rack out of two sticks and a "C" clamp (Maier, 1945), are usually free of agonizing emotional concomitants.

No one has as yet offered a universally acceptable definition of a problem. This lack of a firm conceptual groundwork, however, has not impeded the erection of theoretical skyscrapers. Mahoney (1974) muses that workers in this field seem to have less difficulty describing their tree than their forest. To be sure, such is the case with many if not most concepts in education and the social sciences. Wechsler and Binet may differ in their definitions of intelligence, but researchers who operationally define intelligence in terms of a particular test score can readily communicate with one another. Unfortunately, the problem-solving literature includes a vast array of operational definitions for problem situations and problem-solving performance, each with a host of shortcomings (see Cagné, 1964).

Behavioral Perspectives of Problem Solving

---

It is important to note that the early problem-solving behavior of Thorndike's cat was not random but rather the result of genetic predisposition or prior learning or both. Such behavior might be termed "blind" in that the correct solution is unknown. Moreover, turning a latch is a response not likely to be in the cat's repertoire; so this problem will be harder to solve than another involving, for example, pulling a string. Similarly, because of a rooting tendency, a pig is more likely to solve a problem that requires raising a small platform than is either a cat or a dog (Maier & Burke, 1967).

Campbell (1960) extended Thorndike's trial and error perspective on animal problem solving to include creative thought and other knowledge processes in human beings. His central tenets are as follows:

1. A blind-variation and selective-retention process is fundamental to all inductive achievements, to all genuine increases in knowledge, to all increases in fit of -;vstern to environment.
2. The many processes which shortcut a more full blind-variation and selective-retention process are in themselves inductive achievements, containing wisdom about the environment achieved originally by blind variation and selective retention.
3. In addition such shortcut processes contain in their own operation a blind-variation and selective-retention process at some level, substituting for overt locomotor exploration or the life and death winnowing of organic evolution. (p. 380)

Campbell's point of view has much in common with those of other behaviorists. Blind variation in problem-solving behavior, for example, might be further explained in terms of Hull's (1952) habit-hierarchy mechanism. A problem situation evokes a sequence of solution responses, the order of which is determined by individual reinforcement histories. In other words, we first try what has worked well for us in the past. Should our initial efforts not prove successful, we may exhibit other behaviors that are lower in our response hierarchies. Maltzman (1955) and Staats (1966) elaborated upon Hull's work by including complex Colnbinations of different habit hierarchies in explaining problem-solving behavior.

Skinner's (1966) operant analysis of problem solving greatly expanded Campbell's concept of selective retention. According to Skinner, when we reinforce a response, we increase its probability; this is simple operant conditioning. "Problems" arise, however, when reinforcement contingencies are complex:

For example, there may be no response available which satisfies a given set of contingencies. Or competing responses may be evoked--among them emotional changes which weaken the very response upon which reinforcement is contingent or destroy the power of the reinforcer, Or the contingencies may be satisfied only by a sequence or chain of responses, early members of which are too remote from the terminal reinforcer to be strongly affected by it until conditioned reinforcers have been set up. (p. 226)

Skinner contended that it would be easy to teach Thorndike's cat how to solve its problem by shaping the latch-turning response, but clearly this is not the thing to do if we are interested in teaching problem solving. The latter can be faciliated, for example, by reinforcing the learning of rules that serve as discriminative stimuli orienting the problem solver toward an effective solution. (Such rules are very much akin to Campbell's, 1960, concept of processes that "shortcut" full blind variation and selective retention.)

Gagné (1970) echoed this theme when he suggested that effective problem solving rests upon previously learned rules. Given a problem situation, the learner tries out a number of such rules and tests their applicability. When he or she finds a particular combination of rules that work, not only is the problem solved but something new is learned--an entity not formally different from a rule. "It may be more complex, and it is surely new (to the learner), but it is a rule with the same properties of broad applicability as other rules" (p. 215).

Davis (1966, 1973) elaborated on the visual and thought processes that Campbell (1960) suggested may be substituted for motor trial and error behavior. In his 1966 review, Davis classified laboratory problemsolving tasks according to whether they called for overt trial and error or "implicit problem-solving activity." Most laboratory tasks such as anagram problems call for the latter (except when paper and pencil or letter blocks are provided). Davis (1973, p. 43) neatly summarized the research in this area in one sentence: "The problem-solver will think if he can, he will manipulate if he must."

Classical conditioning may also play a role in problem solving. Kendler and Kendler (1962) point out that problem solving involves both "horizontal" and "vertical" processes. Horizontal processes are best typified by long Skinnerian chains of behavior that occur over time. Vertical processes acknowledge the fact that such chains exist simultaneously and can interact. Serendipity, the art of finding one thing while looking for another, illustrates this interaction. The Kendlers believe that problem solutions often involve a vertical connection between a cue and a response from different chains. Davis (1973, p. 57) has provided an excellent example of this phenomenon: "A bat or porpoise may navigate by bouncing its voice off unseen objects. This information, viewed as a horizontal idea-chain, might suggest a solution for the parallel (vertical) stimulus problem of inventing a navigational aid for blind people."

At the risk of oversimplification, then, I will say the behavioral perspective is that problem solving is basically a trial and error enterprise. But trial and error should be interpreted in a broad sense to include not only physical activity but also seeing and thinking. Given a problematic situation, we normally try out a number of potential solutions, not randomly, but in accord with our learning histories-strongly reinforced responses are generally displayed first. In the course of our experience we will undoubtedly discover a number of problem-solving shortcuts or rules. Like the boxer who learns never to lead with a right, we can acquire these rules through trial and error (ouch!), or they may be transmitted by society. Sometimes experiences we have had in other areas suggest the solution to a problem (vertical processes).

Behaviorists do not have a monopoly on the problem-solving literature. Gestalt psychologists such as Köhler (1925) and Sheerer (1963) have been equally if not more prolific in contributing their views. Köhler stressed the importance of insight, contended that Thorndike's cats merely stumbled onto the solution through chance alone, and argued that true problem solving could occur only under favorable optical conditions. A chimpanzee, for example, after an hour of "fruitless" activity paused to survey its environment, then insightfully used a stick to rake in a banana placed just out of arm's reach. If Thorndike's cats could have seen the relationship between the latch and the access to food, they might have shown the sudden insightful behavior the chimpanzee displayed.

Since Köhler maintained that insight commonly occurred in animals, the concept hardly lends dignity to human beings (cf. Skinner, 1971). Moreover, the Gestaltists have been soundly criticized for a sentimental and theoretically foggy preoccupation with insight, which Campbell (1960) dismissed as merely the successful completion of a blind-variation cycle. (Recall that visual and cognitive trial and error can be substituted for overt activity. The chimpanzee more than likely tried out a number of potential responses in its "mind's eye" before discovering the stick-as-rake solution.) Furthermore, Campbell (1960, p. 390) asserted that when insight is "publicized as part of an ideology of creativity, it can reduce creativity through giving students a feeling that they lack an important gift possessed by some others, a feeling which inhibits creative effort and increases dependence upon authority."

On the other hand, the behavioral perspective has not yet really demonstrated its practical utility. Behavioral researchers have provided us with relatively consistent theorizing that is long on explanation but short on the generation of effective intervention strategies. Although many of the problem-solving treatments to be discussed below are compatible with the behavioral perspective, they did not originally flow from the pens of behaviorists.

Problem Solving and Counseling

---

1. Client concerns may be conceptualized as problems analogous to the sort of problems studied in laboratory settings.
2. Problem-solving ability need not be defined as a psychometric trait subject to the same criticisms leveled by Mischel (1968, 1973) against other traits. Instead, problem solving might best be viewed as a generic term covering a number of individually identifiable and teachable skills.
3. Problem-solving techniques come from a diverse literature that includes, for example, creativity training. The designation of a particular intervention as pertinent to problem-solving counseling is based on demonstrated utility rather than theoretical fidelity. In other words, if a certain technique has been shown to produce a desired effect, it matters not that the technique originated with psychoanalysts, existentialists, behaviorists, or business executives. If need be, translations can be made at a later date.
4. Problem-solving skills clients acquire in the context of counseling for a particular concern may generalize to other concerns clients encounter in the course of their lives.

Early Interventions

Once the problem solver learns to "stop and think," D'Zurilla and Goldfried suggest the counselor's task is to help the client formulate the problem. Clinical problems, in contrast to those encountered in the laboratory, are usually ambiguous. The client will need to define the problem dearly in operational terms, separate relevant from irrelevant information, and identify the primary goals. Although D'Zurilla and Goldfried suggest that training programs like those developed by Karlins and Schroder (1967) and Crutchfield and his associates (Covington, Crutchfield, & Davies, 1966; Olton & Crutchfield, 1969) can help foster pertinent inquiry and formulation skills, most classroom programs of this sort are not applicable to individual counseling with adolescents and adults in distress. More appropriate perhaps would be a variety of strategies such as empathic listening (to extinguish anxiety), shaping of greater degrees of specificity in describing the problem, and model-reinforcement procedures to produce information gathering (see chapter 9).

A final early intervention strategy counters the well-known inhibiting influence of old habits in solving new problems (see Davis, 1973; Luchins, 1942; Taylor & McNemar, 1955). Such corrosive effects can be curtailed by simple instructions to respond in a creative manner (for example, Christensen, Guilford, & Wilson, 1957; Maltzman, Bogartz, & Breger, 1958; Rosenbaum, Arenson, & Panman, 1964). Similarly, Colgrove (1968, p. 1208) told her experimental subjects, "You have the reputation of being a very original person and of being good at coming up with answers to difficult problems. That is why Gus came to you. Keep this in mind while studying the problem." Her results indicate the mere suggestion that a person has the reputation of being an original thinker creates a mental set that upgrades the problem- performance.

Interventions for Generating Solutions

Maltzman's Originality Training

Maltzman (1960) conceptualized originality as nothing more than an operant that happens to be uncommon but relevant to a given stimulus situation. "Original" operants are subject to the same learning principles as are other operant responses. Moreover, according to Maltzman, the mere occurrence of an original operant may be sufficient for its reinforcement-the amount varying inversely with the initial probability of the response. (One's learning history is undoubtedly punctuated with instances of originality met with primary or secondary reinforcers or both.) Finally, Maltzman suggested that reinforcement of original responses could be expected to increase the general tendency to emit uncommon responses.

To enhance originality, Maltzman developed a standardized wordassociation training procedure in which he presented subjects with twenty-five stimulus words, one at a time, and asked them to respond as quickly as possible with the first word that came to mind. He then repeated the entire list five times with instructions to respond with a new word each time. The Maltzman technique has been subjected to a number of experimental investigations (Caron, Unger, & Parloff, 1963; Maltzman, Bogartz, & Breger, 1958; Maltzman, Simon, Raskin, & Licht, 1960; Penney & McCann, 1962; Rosenbaum, Arenson, & Panman, 1964), most of which are highly supportive, particularly when the training procedure is accompanied by instructions to "be original."

Freedman (1965) developed an interesting variation of the Maltzman technique. Basically, the subjects are given thirty seconds to produce as many free associations as possible to each word on a ten-word list. Although the Freedman procedure has produced higher creativity scores than control conditions, it has not been experimentally compared with the Maltzman technique.

The relevance of either the Maltzman technique or the Freedman variation to clinical problems has yet to be determined. Although both procedures have been experimentally examined, only psychometric outcome devices such as Guilford's (1950) Unusual Uses Test or Mednick's (1962) Remote Associates Test have been employed. Perhaps counselors who ask their clients to free associate (or who free associate with their clients) on what might be done to resolve a presenting problem would increase the likelihood of discovering a creative solution.

Brainstorming

In recent years the term "brainstorming" has become popularized and attention to its theoretical underpinnings and procedural guidelines has eroded. Brainstorming (Clark, 1958; Osborn, 1963) is a problem-solving procedure developed largely for use in industrial settings; but its relevance to counseling- group counseling-is readily apparent. Correct use of the term requires close adherence to four general rules:

1. Criticism is ruled out. Group members are told to defer judgment on any member's ideas or suggestions. Potentially good solutions to a problem are more likely to emerge in an atmosphere of unconditional acceptance than one in which criticism (punishment) is possible or probable.
2. Freewheeling is welcomed. Participants are encouraged to offer wild ideas. The more unusual, bizarre, or offbeat the suggestions, the more likely the discovery of a truly original and useful solution. "It's easier to tame down than to think up" is a rule that evolves from the first rule because freewheeling cannot occur in a critical atmosphere. (The reader might note the similarity between freewheeling and Maltzman's, 1960, procedure for training in originality.)
3. Quantity is welcomed. Group members are urged to think up as many solutions as possible. The greater the number of ideas, the greater the likelihood of useful ideas.
4. Combination and improvement are sought. In addition to offering ideas of their own, participants are encouraged to suggest how others' ideas can be improved upon and how two or more ideas can be integrated into a third idea.

In the business world, where ideas are evaluated in terms of how much money was made or saved, anecdotal evidence in favor of brainstorming is impressive: Sylvania Electric's popular flashcube originated in a brainstorming session (Clark, 1958; Mason, 1960; Osbom, 1963). Evaluation of ideas in the experimental laboratory, however, is more problematic (see Davis & Manske, 1968; Davis & Roweton, 1968; Warren & Davis, 1969). Studies on the efficacy of the various brainstorming rules are generally favorable (for example, Bayless, 1967; Davis & Manske, 1966; Parloff & Handlon, 1964; Parnes, 1967). Unfortunately, much of the applied research on brainstorming suffers from a lack of experimental rigor, but its potential as a specific counseling procedure appears promising.

Metaphorical Thinking

In a television comedy sketch George Gobel once dismissed a minor problem in living as "just another barnacle on the sinking ship of youth." In spite of his mournful affect, the audience roared with laughter in appreciation of the unusual metaphor. Indeed, such metaphors seem to be the hallmark of humor and most other forms of creative writing. At least two problem-solving strategies capitalize on the close relationship between creativity and metaphorical thinking.

The first, bionics, stresses the drawing of analogies between biological mechanisms and technological design (Papanek, 1969). Sonar capabilities in bats, for example, have strong implications for the solution of naval engineering problems such as navigation and submarine detection. The second, synectics, developed by Gordon (1961; see also Prince, 1968), emphasizes three kinds of analogical thinking in problem solving: (1) Direct analogy resembles bionics in that nature is often searched for clues in solving problems. For instance the color-changing mechanism of a flounder suggested a process by which roofs might automatically change color from heat-absorbing black in the winter to heat-reflecting white in the summer. (2) Personal analogy involves imagining oneself to be the problem-object. Given such a set, a question such as "What would make me change?" might lead to creative solutions. (3) Finally, fantasy analogy calls for farfetched solutions much akin to the freewheeling rule in brainstorming. Asking how a problem could solve itself may have led to the self-defrosting refrigerator or the self-cleaning oven.

Although anecdotal reports on the utility of metaphorical thinking in industrial problem solving abound, controlled experimental evidence is nonexistent. Undoubtedly many creative solutions attributed to bionics and synectics might just as well have been discovered by simply telling the problem solver to "be creative." Nevertheless, if properly evaluated, both procedures might prove to be quite useful.

The possibility that counseling procedures based on metaphorical thinking may be relevant to client problem solving has never been adequately articulated. In a discussion of cognitive restructuring Lazarus (1971) does suggest the "2-peach parables metaphor designed to reduce sexual guilt. But here the problem solution, guilt reduction, is already known. I am prone to use another metaphor called "the hamster love story" as a counseling technique designed to stimulate alternative solutions to clients who invariably find themselves on the losing side of their dating relationships: My younger brother once raised three hamsters, two females and a male, in separate cages. When they were old enough to breed, he placed all three on the top of a coffee table. One of the females began actively to explore the new surroundings. She noticed the presence of the male but seemed more concerned about being her own hamster: Was there food and water up here? Where might shelter be found? The other female gave a number of nonverbal hamster cues indicating the desire for an immediate serious relationship. As the male hamster began active pursuit of the independent-minded female, I couldn't help but chuckle, "How human!" This metaphor invariably stimulates client thinking about more adaptive ways to handle present and future relationships. Moreover, it is equally appropriate for either sex and can be relevant to a number of marital problems as well.

Comprehensive Training Programs

ICPS Skill Training

Spivack, Platt, and Shure (1976) and their associates contend that psychological adjustment is essentially related to the development of five "interpersonal cognitive problem-solving (ICPS) skills":

1. Sensitivity to interpersonal problems. This general skill covers a variety of topics such as an awareness of the variety of possible problems that beset human interaction and an ability to examine oneself when relating to others,
2. Ability to generate alternative solutions. This skill parallels the process of brainstorming in that numerous alternatives are sought in the context of deferred judgment.
3. Capacity for means-ends thinking. One displays this kind of thinking by articulating step-by-step procedures for carrying out the solution to an interpersonal problem that includes recognizing obstacles that must be overcome and considering how others might react to the implementation of a particular solution.
4. Capacity for consequential thinking. This kind of thinking is reflected by considering the consequences of one's social acts in terms of their impact both on other people and on oneself. Essentially it implies the question: "What might happen as a result of my doing this?"
5. Awareness of personal motivation and interpersonal continuity.

Although there appears, to be considerable overlap in the general descriptions of the ICPS skills, Spivack and his associates have operationally defined most of these skills in terms of tests appropriate to various age groups, For example, alternative thinking in four- and five-year-olds is measured by the Preschool Interpersonal Problem-Solving (PIPS) Test (Shure & Spivack, 1974), which consists of asking the child to think of as many solutions as possible to a "peer" interpersonal problem (wanting a toy another child has) and an "authority" interpersonal problem (having broken mother's flower pot). With children nine to twelve years of age the measurement of alternative thinking consists of presenting seven problems having a greater degree of sophistication. For example: "Johnny wants his friend to go to the playground with him after school but his friend doesn't want to go. What can Johnny do to get his friend to go with him?" One's score on alternative thinking consists of the total number of relevant solutions offered. Other ICPS skills are measured in like fashion.

In their 1976 book Spivack, Platt, and Shure review a steady stream of studies on the ICPS model, beginning with those of the 1960s. Essentially they found that the absence of ICPS skills is strongly related to a wide variety of indexes of maladjustment, including, for example, delinquency and drug use. Consequently, they and their associates have developed and experimentally evaluated ICPS skill training programs for a diverse array of client populations, including a prevention program for kindergarten children, a treatment program for hyperactive children, a program for mothers of young children (ICPS skills-deficient children have ICSP skills-deficient mothers), a school program for third- and fourthgrade children, a school program for fourth- and fifth-grade children, a treatment program for chronic psychiatric patients, a group therapy program for young adults and adults, and a group therapy program for hospitalized psychiatric patients. A critical review of the individual studies supporting these programs would take us far afield. Suffice it to say that in general the programs have been shown to foster not only the development of ICPS skills but concomitant behavioral improvement as well, To encourage these skills, the programs employ a wide variety of techniques, including affect- (empathy training), behavioral rehearsal, cognitive modeling, cognitive restructuring, decision making, self- instruction training, simulation gaming, social reinforcement, videotape modeling, and vocabulary building on ICPS itself through didactic instruction.

Other Comprehensive Programs

Texts by Davis (1973) and Spivack, Platt, and Shure (1976) provide thorough coverage of a wide variety of problem-solving and creativity training programs. The interested reader is urged to peruse these sources for additional information on several of the strategies reviewed in this chapter and a number of other programs redundant or of marginal relevance to decision-making counseling.

The pertinence of comprehensive problem-solving training programs to clinical and educational concerns is receiving increased attention in the counseling literature. Jacobson (1977) developed a multicomponent program for the treatment of marital discord. Mendonca and Siess (1976) combined problem-solving and anxiety-management training in an attempt to reduce indecisiveness about career plans. Finally, Poitras-Martin and Stone (1977) translated the etherial jargon of "psychological education" into a skills-oriented problem-solving framework appropriate to sixth graders. All three of these recently developed comprehensive programs have produced positive effects.

Summary

---

There is no universally accepted definition of "problem"; moreover, unlike counseling problems, most problems studied in laboratory settings are free of agonizing emotional concomitants. According to behavioral theory, problem solving is essentially a trial and error enterprise that includes not only physical activity but seeing and thinking as well. People confronted with a problematic situation can be expected to engage in a number of responses in accord with their learning histories. Problemsolving rules or shortcuts are also learned through trial and error, but sometimes the solution to a problem is suggested by experiences we have had in other areas.

Three categories of problem-solving interventions can be applied in the context of counseling: early interventions used, for example, to thwart impulsive responding and foster an appropriate problem-solving set; interventions for generating solutions, including originality training, brainstorming, and metaphorical thinking; and comprehensive training programs such as ICPS skill training.

With this chapter part II ends. Part IV uses the preceding material to construct a cognitive-behavioral model of decision-making counseling.