The College Standards and Accreditation Council mandate to implement generic problem solving skills into Ontario community college course curricula warrants evaluation in view of contradictory viewpoints about the nature of problem solving skills. The thesis of this article is that the premise that generic problem solving skills can be taught is mistaken, as problem solving cannot be generalized from one problem to another.
In the 1990 review of Ontario's community colleges and universities, the Vision 2000 investigators identified Lifelong Learning, of which problem solving strategies are considered a component, as the social and educational trend with the third highest importance: “Colleges must provide all learners with the general education and generic skills that will assist their individual development, enhance their employability and provide an important social return on their education”. Research on teaching clinical problem solving was identified as the second highest priority for nursing education. (Tanner, 1987) There is concern that a large number of students suffer from thinking deficiencies, and to remedy poor student performance they need to be taught generic problem solving skills. However, there is a certain humility in the appreciation of how awesomely complex thinking and human action can be, and of just how much a person needs to know before he or she can do much of anything. (Waldrop, 1984, p. 1279) This article defends the assertion of the centrality of knowledge in problem solving.
What evidence is there that problem solving can be taught? The most notable difficulty in evaluating courses designed to teach problem solving skills is the lack of valid and reliable measurement tools to assess generic problem solving skills, and to assess the extent of transfer of these skills to problem solving in specific content domains. Discussion of evidence of improvement in problem solving from educational tests and evaluation tools is a paper in itself and so, the following is a synopsis of the research on the cognitive aspects of problem solving skills. Understanding the cognitive processes of problem solving is a necessary prerequisite to the development of instructional methods for teaching those processes.
The literature informs us how competent individuals solve problems, and how competence changes from novice to expert. Problem solving skills are content specific and are not generalizable. The most reported example is of the chess master whose logistical expertise does not generalize to other fields. It may just be convenient to think that there are a relatively small set of core abilities that improve an individual's performance whatever his or her endeavour. Imagine Isaac Newton being hit on the head by a brown apple. Would he have munched away, cogitating on why that apple fell? He may well have thought a brown apple was a bruised, rotten apple, unless he had the prior knowledge that Russett apples are brown. While one person can be accomplished at many activities, common sense suggests that a person has several kinds of knowledge; not one skill generically referred to as reasoning. (Johanson, 1987, p. 5)
Explicit instruction aimed at enhancing general thinking and problem solving skills, and application of these skills in specific content domains can substantially increase problem solving effectiveness. Indeed, an individual's degree of scholastic improvement has been found to be related to the amount of time spent practising problem solving skills. Examples of behaviours that accompany problem solving failure and success are outlined below.
|1. Random attention. Unable to sustain focus on the critical variables of the problem.||Selective attention.|
|2. Scans compulsively and haphazardly. Does not probe a complex problem until all its components are identified.||Sustained analysis.|
|3. Fails to test prior knowledge against potential relationships in the problem. Does not make analogies.||Analogizing.|
|4. Guesses chronically. Does not assign priorities to regularities but follows the first lead that occurs.||Suspension of closure.|
|5. Fails to test solution.||Auto-censorship.|
The above rules for analysis do not include the interactive and transformative dynamic of knowledge in problem solving.
|Process||Cognitive Domain||Affective Domain||Psychomotor Domain|
|Problem recognition||Realize that there is a problem.||Be aware of own feelings of unease (constructive discontent).||Perceive sensory stimuli (auditory, visual, taste, tactile, smell, kinaesthesia)|
|Problem definition||Translate and interpret all input received. Establish problem boundary.||Be aware of response sets.
Be willing to defer judgment.
|Perceive sensory stimuli and environmental input.
|Problem analysis||Break down the cognitive, affective, and psychomotor components of the problem.
Determine the relationships between elements. Organize the principles involved. Determine desired outcome.
|Be receptive to new relationships. Be aware of own response sets. Feel satisfaction.||Perform habitual tasks related to thinking (Write, pace, chew, read).|
|Data management data collection methods selection||Identify data needed. Relate data to data-collection methods. Select data-collection methods.||Be receptive to new approaches. Be aware of response sets. Be willing to use creative abilities||Perform habitual tasks related to thinking.|
|Data collection||Apply data collection methods.||Respond appropriately to internal and external cues. Be aware of response sets||Perform appropriate complex overt responses.|
|Solution development data analysis||Organize data collected. Classify data collected.||Be aware of own response set. Be willing to defer judgment.||Perform habitual tasks related to thinking.|
This above example of generic problem solving is from physical therapy. (May and Newman, 1980)
The authors of this model tested the hypothesis that problem solving skills would remain the same in a student group using a May and Newman analytic questioning technique treatment and a comparative control group was tested using the Watson and Glaser Critical Thinking Analysis tool. (Slaughter et al., 1989) No statistical improvement of problem solving skills was shown.
Nursing's problem solving heuristic is called the Nursing Process. The five steps of the nursing process are: Assessment, Nursing Diagnosis, Planning, Implementation and Evaluation. The nursing process and nursing diagnosis are nursing education's traditional tools for teaching problem solving. The objective, linear structure of the nursing process actually handicaps the experienced nurse. The knowledge the nurse discovers in interacting with the client leads to logical activities that do not follow the arbitrary structure of the nursing process.
Nursing Diagnosis is an international system of classification that organizes information. It initiates students in the typical problems of clients requiring nursing. It provides a framework for logical, systematic thinking and problem solving. Focusing on a client's nursing diagnosis, however, could lead the nurse to miss important details about the client that are not included in a single diagnosis or group of diagnoses. The number of health problems identified and the accuracy of diagnoses varies as a function of the amount and relevance of the information the nurse possesses. (Tanner, 1987)
Tanner reports studies which used new strategies for teaching clinical problem solving in nursing. Conceptual approaches to teaching problem solving skills include binary decision trees to enable students to use information systematically and to determine characteristics of each condition. This approach showed statistical improvement in students' problem solving. It is uncertain how generalizable this would be to clinical problem solving because in practice, nurses utilize diagnostic hypotheses and systematic review of the client. A structured assessment guide applied to improve abilities to identify pertinent nursing care problems showed that the experimental group had no difference in problem solving behaviours but could produce significantly greater numbers of reasons for problems selected. Tanner explains the failure to find improvement in clinical judgment following generic problem solving teaching interventions. She notes that the experimental treatments have been short, and instrumentation of evaluation is a problem as the search for valid measures of performance continues.
Underlying generic problem solving skill research is the assumption that there is a single process or set of processes of problem solving that can be described, measured and taught. Research suggests that different processes are used by individual nurses depending on their level of expertise, and on the demands of the task. That there is not a single problem solving process is evident in the work on nursing process, and in novice to expert differences.
Studies have found a general problem solving process in medicine. The hypothetico-deductive method can be described as 1. Cue acquisition, 2. Hypothesis generation, 3. Cue interpretation and 4. Hypothesis evaluation. Studies on hypothetico-deductive method did not show construct validity; that is, they gave no evidence that clinicians were consistently better or worse problem solvers than others or that the process of problem solving improved from entry to medical school to entry into practice. What was discovered was that the best predictor of successful problem solving (finding the correct diagnosis) was the content of the diagnostic hypothesis. Thus, the search for a generic problem solving process verified the central role of knowledge.
Geoffrey Norman has researched the cognitive clinical reasoning of experts and students, and identified the expert clinician as a better problem solver by virtue of accumulated experience, not as a result of innate or learned problem solving skills. (Norman, 1984) He says that the process of problem solving in a particular case “is strongly influenced by prior experience, is based on a pattern matching process, shows little evidence of general problem solving strategies which are independent of experience or knowledge, and may explain why studies of clinical problem solving have had difficulty revealing correlations in performance across problems”. In contrast, a reliable estimate of problem solving ability correlated 0.92 with a separate measure of knowledge assessed by multiple choice questions. Norman's research showed that the case being studied and the knowledge and experience of the student will dictate the problem solving strategy. The strategy selected is a function of the amount and relevance of information available.
Robert Glaser explains: “Problem solving in knowledge-rich tasks show strong interactions between structures of knowledge and cognitive processes. Data and theory in developmental psychology, studies of expert/novice problem solving, and process analysis of high and low scorers on intelligence and aptitude test tasks show that a major component of problem solving is the possession of accessible and usable knowledge.” John McPeck says that critical thinking cannot be taught without linking it to a field of knowledge: “Purporting to teach critical thinking in the abstract, in isolation from specific fields or problem areas, is muddled nonsense; thinking of any kind is always “thinking about `x'”. Nickerson points out that thinking is a knowledge dependent activity. He argues: “Knowledge and thinking ability are interdependent and mutually reinforcing; attempting to develop one without the other is like trying to make cloth with only one side”. (in Holbert and Abraham, 1988) Knowledge directs you to your next thought. Problem solving must be flexible according to the data that presents and its relationship to the expert's body of knowledge. Perkins asserts that “thinking strategies often fall flat because the students lack the subject matter knowledge to apply them.” (in Brandt, 1990) If there is little emphasis on content, students will not learn and remember the domain of knowledge central to a course or discipline. It is ironic to note that Arthur Whimbey (1986), one of the strongest proponents of teaching problem solving skills, states, “poor problem solvers frequently express the opinion that either you know it or you don't know it”. If you don't “know it” a problem solving skill won't help.
As college programs and teachers attempt to improve the thinking abilities of students, the abstract conceptualization that guides their problem solving curricula had better be valid. If it is not, it is logical to infer that the acquisition of knowledge will be retarded rather than accelerated. Instead of deepening a student's appreciation of intellectual complexities, generic problem solving skills might solidify uncritical ideas about the nature of a problem by providing a student with an unwarranted justification. (Walters,1986, p. 240) As long as the road map has been followed, the student without a strong knowledge base is led to believe that the destination has been reached. As John Dewey put it, “No one discovers a new world who exacts guarantee in advance for what it shall be, or who puts the act of discovery under bonds with respect to what the new world shall do to him when it comes into vision”. Dewey's vision was missing in Vision 2000. An overemphasis on problem solving skills unwittingly prevents problem solving. Mary Daly (1973) writes:
The tyranny of methodolatry hinders new discoveries. … The worshippers of Method have an effective way of handling data that does not fit into the Respectable Categories of Questions and Answers. They simply classify it as nondata, thereby rendering it invisible.
Those who would teach generic problem solving skills believing that the skills are generalizable to a broad range of situations are treating education as training, and bear the risk of limiting the development of thinking in learners.
Thinking is a human process, not just cognitive and rational, but also spiritual and emotional. We had best be exceedingly careful if we treat thinking as a product to be exploited as human capital. Educational manipulation and control as though the human mind were a machine (cybernetics) is dangerous. The act of thinking is too complex and too poorly understood to be moulded by reductionist rules. The phrase “reflection in action” describes the application of knowledge and experience to a hypothetical solution. It combines pattern recognition with an experimental problem solving approach.
"The inquirer draws on a familiar repertoire which is relevant to the present problem. This inquiry may reframe the situation or suggest an “experiment”. This experiment is not trial and error because the strategy is based on the professional's repertoire of theoretical principles, technical knowledge, and experience. The practitioner searches a mental data bank of knowledge and experience for the most likely explanation or strategy for this unique clinical problem. A recipe does not exist. He or she carries out an experiment which serves to generate new information about the situation and possibly change the situation. The practitioner remains open to feedback from the situation. Data from the first experiment is used to make a second choice.” (Saylor, 1990)
The sort of creative thinking that leads to novel problem solving is not prescriptive but open ended. People are subjects rather than objects. They possess reflective awareness that is individual and unique.
When I started reading the “cognitive process” and “problem solving” literature, I had the intent of finding an instant intelligence model. Instead, I found that there are no generic problem solving skills. In retrospect, problems with generic problem solving skills may stem from the archetypes of Plato's cave and our ancestral knowledge of the landscape. One could contemplate the neurophysiology that has evolved through these two anthropologies in contrast to the cybernetic demands of today's technological world.
What then can guide us in designing curricula for teaching problem solving? Efforts to obtain a general understanding of a topic by using analogies and examples, relating new information to knowledge already available and critical step-by-step examination of details and evidence are desirable. This “content analysis” was found to enhance the level of understanding and long term retention. (Van-Langenberghe, 1988) Harvard's Project Zero identified the need to teach the content better, to ensure conceptual understanding, reflection and strategizing in order to enrich learning. (Brandt, 1990) Effective instruction, we may conclude, allows the learner to translate information into problem solving efforts and thus the goal of instruction should be to facilitate students to become self-propelled thinkers.
Colleges should rethink the inclusion of generic problem solving education in programs, and enhance practice based knowledge exploration to maximize problem solving abilities in learners.
Bloom, B.J. and Broder, L.J. Problem-solving processes of college students: an exploratory investigation. 1950. University of Chicago Press, Chicago.
Brandt, R. “On knowledge and cognitive skills: a conversation with David Perkins.” Educational Leadership, 47 (5), February, 1990, pp. 50-3.
Glaser, R. “The nature of expertise”. National Center for Research in Vocational Education. Ohio State University, Columbus. Occasional Paper No. 107. ED261190
Holbert, C. M. and Abraham, C. “Reflections of teaching generic thinking and problem solving.” Nurse Educator 13, (2), March/April, 1988, pp 23-7.
Johanson, A. “Can informal logic courses teach critical thinking: reflections on McPeck and Paul.” April 12, 1987. Paper presented at the Conference on Critical Thinking. VA. ED292707
May, B.J. and Newman. J. “Developing competence in problem solving; a behavioral model.” Physical Therapy 60 (9), September, 1980, pp. 1140-45.
Norman, G.R. “The role of knowledge in teaching as assessment of problem solving.” ED252528, 1984. Position paper presented at the annual meeting of the American Educational Association, April, 1984.
Saylor, C.R. “Reflection and professional education: art, science, and competency.” Nurse Educator. 15 (2), April, 1990, pp. 8-11. Slaughter, D.S., Brown, D.S., Gardner, D.L. and Perritt, L.J. “Improving physical therapy students' clinical problem-solving skills: an analytical questioning model.” Physical Therapy 69 (6), June 1989, pp. 441-447.
Tanner, C.A. “Research in nursing education: assumptions and priorities.” Journal of Nursing Education 26 (2), February, 1987, pp.50-59.
Van-Langenberghe, H.V. “Evaluation of students' approaches to studying in a problem-based physical therapy curriculum.” Physical Therapy. 68 (4), April 1988. pp. 522-9.
Waldrop, M.M. “The necessity of knowledge.” Science 223, March, 1984, pp.1279-82.
Walters, K.S. “Critical thinking in liberal education: a case of overkill.” Liberal Education, 72 (3), 1986, pp. 233-244.
Whimbey, A. and Lochhead, J. Problem Solving and Comprehension, 4th ed. 1986. Englewood cliffs, N.J.: Lawrence Erlbaum Assoc.
Woditsch, G.A., et al. “The skillful baccalaureate: doing what liberal education does best.” Change, 19 (6), 1987, pp. 48-57.
Vision 2000. (1990). Vision 2000: Quality and Opportunity - A Review of the Mandate of Ontario's Colleges. Toronto: Ontario Ministry of Colleges and Universities.
Leslie Millson Taylor teaches in the School of Applied Arts and Health Sciences at Georgian College in Barrie, Ontario.