In this panel presentation, we were asked to respond to the challenge of translating research findings in the science of learning into educational applications. To do this, we were asked to provide you with an overview of the current state of research on learning, to consider how the effective application of relevant principles can improve faculty teaching and student learning, and to examine challenges of application within the research university context. As preamble to my talk, I will provide my disciplinary context: although I had a solid liberal arts and science undergraduate education, I grew up as a psychologist, and I became an educational researcher, in fact was the first PhD graduated from the Ontario Institute for Studies in Education at the University of Toronto. I am thus a hybrid, and although I honor my roots in philosophy and psychological theory, I also have the need to test the principles derived from these disciplines in the field. I embody the skepticism of the engineer, asking, “Will this work?” My field is teaching and learning in higher education, and I have spent the last thirty years examining how professors in different disciplines teach, how students learn, and how we might optimize student learning. In this presentation two questions guide my search: What do we know about student learning? What instructional strategies will help students learn to think?

What do we know about student learning?

Helping students learn would appear to be a straightforward goal, but there are many ways of perceiving postsecondary teaching and learning. From the perspective of faculty, learning is a matter of disciplinary knowledge and methods of inquiry, but the expectations of students differ across disciplines. Most physics professors expect students to enter their programs with a high degree of logical ability, while English professors expect students to learn to argue logically in their courses (Donald, 1988). Law professors expect students to learn to think like a lawyer, to acquire the skills and methods of analysis and procedure (Donald, 2002). Since scholars learn and think within disciplines, an important source for what is to be learned is what our disciplines tell us, particularly the methods of inquiry used and the learning tasks prescribed by these methods. Learning theories have a more general effect, influencing what happens in the classroom and how learning is assessed. The experience of adapting to university may lead students to view learning from a very different perspective, “What do I need to do to survive and succeed?” Recognizing this range of perspectives is a first step in responding to the challenge of translating research on postsecondary learning into educational applications.

Disciplinary differences

The primary source of what is to be learned is the discipline. But disciplines are moving targets, uncertain constructs we can only hope to place within certain parameters. Disciplines are classically defined as domains of knowledge that include specialized vocabularies and accepted theories, systematic research strategies with techniques for replication and validation (Dressel & Mayhew, 1974). Among disciplines, the most prototypical are the physical sciences, which have been described as hard, well structured, or paradigmatic (Frederiksen, 1984; Kuhn, 1970). A paradigm consists of a logical structure and governing truth criteria that provide maximum direction to scholars in the field (Kuhn, 1970). In physics, for example, Newton’s laws of classical mechanics form part of the curriculum around the world. The theories that describe physical phenomena, however, are often incongruent with experience, and to be able to problem solve, the main task in the physical sciences, students must frequently make a radical change in their conceptual framework from Aristotelian to Newtonian.

In the social sciences, phenomena are examined at a broader or more general level than in the physical sciences, and one of the learning tasks is to choose among various theoretical frameworks that could describe the phenomena (Donald, 2002). For example, in psychology, there are several models of learning and of human development. In comparison with the physical sciences, where abstract concepts are proven by concrete experiments, in the social sciences multiple variables and their interaction render theories more difficult to test. Methods of analysis therefore assume greater importance in the curriculum, and the student’s task is to locate, recognize and attempt to relate the varied conceptual frameworks within a discipline.

The humanities specify different tasks again. Often they are described as a training in sensibility, and an aesthetic criterion is applied to learning (Donald, 2002). Humanistic truth involves authenticity or genuineness rather than logical or scientific validity (Broudy, 1977). There is a technical language to be learned, however; for example, trope or genre in English literature. The student’s task is to learn how to interpret text using the specified terminology, and how to present an argument. The learning tasks for students in physical and social sciences and the humanities thus differ considerably, and students must adopt a different approach in order to be successful in each of them. In physics, for example, the student must analyze a problem by breaking it down into its elements, then reconstitute or represent the problem. The student in psychology must wrestle with contrasting perspectives or theoretical frameworks in order to approach intellectual closure, but at the same time, needs to be skeptical and to continually search for consistency to validate findings. In English literature, the processes of argument and judgment provide the structure for learning.

Methods of inquiry

The methods of inquiry espoused by disciplines may be part of their heraldry, but they often cross disciplinary boundaries. The earliest method, hermeneutics, or interpretation, was developed in order to analyze biblical text (Table 1). It is the construction of textual meaning which elucidates the connotations that text explicitly or implicitly represents (Hirsch, 1967). The interpreter of the text begins by assuming that the text is coherent, then develops a framework of explanation which is tested by the facts it generates. The method is thus a process of hypothesizing and then searching for corroborating evidence in the text. Although the hermeneutic approach is espoused most frequently in the humanities, discourse analysis as currently utilized in the social sciences owes much to hermeneutics.

Table 1. Methods of inquiry in different disciplines

A method more generally referred to across disciplines, critical thinking, developed out of the Socratic tradition of disciplined inquiry. Defined as a reasoned or questioning approach in which one examines assumptions and seeks evidence (Donald, 1985), researchers suggest that critical thinking includes components of logic, problem solving and Piagetian formal operations (Meyers, 1986; Sternberg, 1985). Different disciplines focus on different aspects of critical thinking - inferential processes in physics compared with testing assumptions in English (Donald, 1985; Meyers, 1986).

In comparison to critical thinking, problem solving is described more specifically and procedurally as a set of steps consisting of formulating or representing a problem, selecting the relations pertinent to solving the problem, doing the necessary calculations, and verifying the logic used to see if the final answer makes sense (Reif, Larkin & Brackett, 1976). Thus problem solving includes critical thinking processes but, in addition, those of implementation or testing; the difference between critical thinking and problem solving is analogous to understanding versus doing. For example, the critical thinker would examine underlying assumptions and deduce their effects; the problem solver would continue from this action to create a strategy for dealing with the problem. Problem solving is most frequently used to describe inquiry in the physical sciences.

A more recent approach to understanding methods of thinking is to examine expertise, because the expert is one who has acquired not only a solid base of knowledge but the ability to apply it (Ericksen & Smith, 1991). The expert in a given area has well-developed representations of knowledge or schemas in the subject matter and can relate the schemas in order to operate intelligently. Research on the development of expertise provides insight into potential pedagogical practices. For example, studies on expert and novice differences reveal that novices use knowledge of surface structures while experts use action schemas (Chi, Feltovich & Glaser, 1981); novices represent problems literally while experts use a scientific and mathematical representation (McDermott & Larkin, 1978). Novices become experts by passing through a stage of analysis where problem solving time increases until they develop the representations and strategies characteristic of the expert. Experts recognize patterns and solve problems efficiently and effectively. They have a sense of the context or parameters, select appropriate information, recognize organizing principles, and verify their inferences. Their action schemas equip them with representations and thinking strategies for applying these representations to problems. What is particularly important about this approach is that it describes the relationship between knowledge and thinking processes, and contrasts the thinking strategies of novices and experts, thus opening the way to promoting such strategies.

Learning theories and implications for instruction

Compared with the methods of inquiry used in disciplines, the influence of learning theories on classroom practice and the assessment of learning is more pervasive though tacit. The history of learning beginning with the earliest universities provides context for this discussion. Scholastics in the middle ages assumed a fixed body of knowledge; they defined that knowledge and were the authorities (Johnston, 1998). The enlightenment and the scientific revolution that followed it challenged the notion of fixed knowledge; a tenet of the revolution was that knowledge was an expanding and open system. Validity was now based in scientific measurement, and dissent was integral to the process of testing hypotheses. The role of the university changed to that of creator of new knowledge, a major transformation in epistemology that led to the increasingly important role of research in the university. It could be expected that the principle of an expanding universe of knowledge would guide instructional practice. But we are still dealing with the quandary of what is foundational or ‘must be learned first’ in many disciplines versus testing hypotheses as a way of learning.

What theories of learning have guided practice in postsecondary education? The discipline of psychology has assumed primary responsibility for the topic of learning, and asks the question, ‘How does learning occur?’ The generally accepted definition of learning - a relatively permanent change in behavior that occurs as a result of practice - renders learning scientifically testable, that is, measurable, but it has certain limitations. The primary limitation is that in order to be measured, the learning task may be construed in an oversimplified manner. This definition of learning is most frequently interpreted reductively as association, that is, a connection between a stimulus and a response. The focus is on specific connections, and practice or repetition explains the process, consistent with experimental findings.

Early learning theories promoted this atomistic approach. In experimental studies of learning, Ebbinghaus in 1885 conceptualized human learning as a process of memorization, especially by repetition, so that one can repeat or reproduce. The emphasis on scientific measurement led him to reason that because words have many previous associations, to control the learning and recall of material, he would use nonsense syllables like glet or roit to study human learning (Woodworth & Schlosberg, 1954). Absent in his reasoning was comprehension that he was thus rendering learning nonsensical. Ebbinghaus’ conception of learning as memorization was accompanied by a model of measurement that still guides much assessment practice. He postulated that there were four stages of memory: impression, retention (persistence of changed performance), recall (reproduction of once learned items) and recognition (awareness of previous experience). We set examinations to measure our students’ recall and recognition. The limitation of this model is that it does not explain the more complex task of testing our students’ understanding of pattern and relationship and their methods of inquiry.

A second early theory of learning focused on the effect of practice. Thorndike in 1914 applied the law of effect, originally developed to explain animal training, to human learning. The law of effect stated that satisfaction following from an act strengthens the bond and leads to its repetition, while annoyance weakens the bond. Satisfaction and annoyance were conceived in terms of synaptic functions, and were thus coherent with biological theory. His law of exercise, that the use of a given connection between a stimulus and a response strengthens the bond, is consistent with the associationist model, and with the saying that practice makes perfect. It is reflected in more recent biological approaches to pedagogy in which learning is described as a process of burning in mental circuitry (Leamnson, 1999). It too, however, neglects the effects of complexity and higher order learning.

The first breakthrough in terms of paying attention to higher order processing was Shannon and Weaver’s (1949) information theory, which drew on communications theory to explain how messages or signals are sent and received. The prototype of an information channel is a perfect telephone line in which information transmission is complete, but information theory took into account the fact that channels do not deliver total output and the receiver is left with some uncertainty (Berlyne, 1965). The receiver may also select information to reduce the uncertainty, and complexity of form influences information transmission. Thus information theory, in which information is encoded and in the process transformed and actively retrieved, is closer to a model of active or directed learning. Information theory also updated theories of memory: the concepts of immediate or short term memory and long term memory were introduced to discriminate between the limited capacity of an individual to attend to data – the magical number seven plus or minus two (Miller, 1956), and semantic or mediated memory.

A more molar approach, based in gestalt psychology which looks for principles of synthesis or organization, pays attention to a wider array of variables influencing learning. One is the tendency or need to categorize or group information, and another is the tendency to encode new information in terms of extant categories. The articulation of new knowledge with already existing knowledge requires attention to what the learner brings to the classroom. Learning therefore depends upon discovering relationships between the concepts or ideas presented and the learner’s extant experience. Patterns of knowledge exist in schemas or cognitive structures, coherent plans displaying the essential or important relations between concepts which learners actively create. This model is coherent with the notion of expertise. The question of why an individual learns led Tolman (1932, 1949) to postulate that the organism responds purposefully and selectively to its environment. Learning is goal oriented. These more molar approaches to learning were the basis for cognitive theory (Woodworth & Schlosberg, 1954) and, more specifically, constructivism, in which individual learners construct their own understanding of organized public knowledge.

Models of learning provide us with insights into our instructional habits in higher education. Association theory supports the custom of professors repeating important concepts in their lectures and courses of study and giving students a series of problem sets to solve – practice makes perfect (Table 2). Association theory also explains the tendency to give frequent tests, based on the laws of effect and exercise, and why students are asked to recall facts or, in the case of multiple choice tests, recognize the best of several alternative answers. The limitation of associationist models lies in their tendency to promote rote rather than conceptual learning, that is, knowledge is construed as bits of information not necessarily related or contributing to a pattern or theory. The learner therefore adds to a storehouse of knowledge without necessarily linking it to other knowledge. Information theory introduces the processes of encoding, transforming and retrieving, and situates the student as an active participant. Constructivist theory suggests that students need to identify themselves as explorers or inventors who select and organize their own knowledge. This theory is more consistent with the methods of inquiry that different disciplines espouse. How do these theories translate into optimizing student learning?

Table 2. Models of learning and implications for students

What do our students know about learning?

We know that student preparation for learning and student goals have changed over the past 30 years. More entering students report experiencing stress; over the last decade, the percentage of students ‘overwhelmed by everything they have to do’ has risen from 16% to 29% (Astin, 1998). Astin also reports that financial well-being is a more important goal for American postsecondary students than developing a meaningful philosophy of life. Students thus tend to not be oriented to a scholarly life. Their orientations are reflected in the priority they assign to different activities – the way they spend their time. In a sample of over 500 students at my university, they told us that they spent an average of 13 hours per week on studying and homework, but almost as much time socializing with friends and partying (9 + 3 hours) (Donald & Dubuc, 1999). Other extra curricular activities took up less time (four hours in exercising or sports; three-to-four hours watching TV and hobbies). They spent less than one half hour a week talking with teachers outside of class, a pattern that is widespread in North America. Students may complain that they have little chance of encountering their professors, but they do not appear to take advantage of opportunities when they arise. We can infer from these findings that students’ priorities are peer oriented rather than academic.

At the same time, our students tell us that they expect to progress on several fronts during their undergraduate years: In their ability to analyze, synthesize and think critically, in their basic communication skills; in independence in learning; in the ability to interact with others; and in clarity of educational and career goals (Donald & Denison, 2001). Table 3 shows significant increases (*) in the importance of these criteria from entry to graduation. Students consider a commitment to learning quite important at entry, and this does not change. Counterintuitively, they rate academic preparedness equally important on graduation and at entry to university. What is perhaps most encouraging is that they attach extreme importance to the ability to analyze, synthesize and think critically on graduation, although their rating is more modest at entry. They are clearly telling us that they expect to learn to think, and that it is highly important they graduate being able to do so.

Table 3. Students' ratings of the importance of criteria for student quality

Given these findings, how can we help students learn? Attention at three levels is needed: the institution, students, and faculty. At the level of the institution, policies must be reconsidered to establish a supportive learning climate. These may include greater access to professors, a statement of the university’s commitment to learning, and clear expectations of student responsibilities. Students need to become aware of their role and responsibilities as learners, but this must be explained and supported by university policies and practices. As faculty, we first need to consider what our conception of learning is and what consensus there is within our field as to the nature of learning. To do this, we need to discuss with our colleagues what learning should be about in our programs. Then we need to make clear to our students what our conception of learning, and particularly higher order learning or thinking, is in our discipline, and instruct and assess our students according to this conception.

What instructional strategies will help students learn to think?

To optimize student learning, the role of the instructor must evolve from a limited but frequently prescribed model of transmission or presentation of information to that of a facilitator of learning. The general question we pose in the courses and workshops we provide for our faculty and graduate students is: How can we help students to become responsible learners? Our primary goal is for participants to understand useful models of higher order learning that are consistent with the framework of a course they are designing or redesigning, and the kinds of instructional and learning strategies needed to achieve this kind of learning.

We describe a variety of models, one of the most comprehensive being the working model of thinking processes developed at McGill University from the postsecondary literature and tested in different disciplines at research universities such as Stanford, Harvard, Cambridge and Monash (Donald, 2002). This model is a detailed set of examples consisting of 30 thinking processes that apply directly to courses at the postsecondary level. It also delineates inquiry models used in particular disciplines, for example, ‘expertise’ (identify the context, select relevant information, evaluate results), so that references can be made in the terms used by a specific discipline. Table 4 shows those thinking processes most frequently used across domains.

In our study of this model, we found that professors across disciplines considered certain thinking processes or strategies important; this suggests that there are thinking processes a student in any discipline will need to acquire, although the discipline will determine the specific characteristics of the process. Greatest agreement across disciplines was found in the importance professors attached to students’ learning to identify the context and state assumptions, in changing perspective, and in selecting relevant information, recognizing organizing principles and synthesis (Table 4).

Table 4. Thinking processes used across postsecondary domains

Identifying the context may consist of processes as diverse as setting up a general framework for a problem, recognizing what kind of problem one is dealing with, finding where a framework fits the processes being studied, or recognizing the history of the period in which the text was written. Stating assumptions is critical to solving a problem, recognizing bias, perspective or the framework being applied, or considering the steps to be taken or individuals to be taken into account. The general importance of changing perspective is consistent with the need for a constructivist approach to knowledge, where in building one’s own cognitive structure, students must be aware of alternative frameworks and their advantages and disadvantages.

All disciplines acknowledge that because of the abundance of information and phenomena, students must learn to select. Recognizing organizing principles is essential to understand the structure of a discipline. Synthesis results in laws in physics, while engineering professors approach synthesis as a pedagogical goal for their students, training them in design skills in team projects. In education, synthesis is important for bringing together the many components of the classroom situation. In English literature, despite multiplicity and paradox as hallmarks, the search for form is central.

These thinking processes originate in different conceptualizations of thinking, for example, identifying the context is the mark of the expert, while stating assumptions is a defining characteristic of critical thinking. Selection has been used to define intelligence (Sternberg, 1998), while analysis and synthesis are found in the problem solving literature. Changing perspective and evaluating results are found in several approaches to thinking – in expertise, problem solving, and critical thinking. The fact that professors from different disciplines agreed on the importance of these thinking processes suggests that they are foundational to postsecondary learning. What if professors across disciplines advised students that these were strategies they needed to learn whatever course of study they were pursuing? What if these processes were deliberately taught and assessed in each course?

To help our students achieve higher order learning, we need to take a constructivist approach in which learning is goal-oriented and consists of the discovery of relationships between new and extant knowledge, where the student is thought of as an explorer and inventor, selecting and organizing knowledge. This means that we must help our students to learn how to judge knowledge on the basis of evidence, think through problems, and integrate and apply knowledge. In order to do this, we need to examine the disciplinary inquiry strategies we are responsible for developing, and how students develop more general learning strategies. For example, a team of professors may be needed to develop an explanation of the major principles and tenets governing the field of study, to describe how knowledge is validated, and to show the gaps or paradoxes and therefore the areas requiring further research and discussion (Donald, 2000). We also need to consider how we will model the processes of inquiry in our disciplines and explain how theory is developed and tested.

To set the stage, we need to show students what it takes to succeed in the context of a course, for example, giving them a sense of the number of hours of study required and the kind of work required. Small group learning experiences such as seminars, tutorials or undergraduate research allow students to develop their exploratory skills. Learning tasks that improve attitudes to learning, for example, participation in class discussion, projects, or explaining material to another student can be included in any course at any level. What these benchmark practices demand of us as professors is course organization, which is the instructional dimension that has the highest correlation with student learning (Feldman, 1989; 1996). When we talk about prospects for supporting students’ higher order learning, however, we find that professors become animated by the possibility of creating learning situations that are exciting and personally fulfilling. Although much groundwork may be needed to produce a constructivist curriculum, this innovative process can be enriching and rewarding.

Acknowledgments
This article is based on research funded by the Social Science and Humanities Research Council of Canada and by les Fonds pour la formation des chercheurs et aide à la recherche du Québec.

Resources/References:

Publications

Astin, A. W. (1998). The changing American college student: Thirty-year trends, 1966-1996. The Review of Higher    Education, 21 (2), 115-135.
Berlyne, D. E. (1965). Structure and direction in thinking. New York: John Wiley.
Broudy, H. S. (1977). Types of knowledge and purposes of education. In R. C. Anderson, R. J. Spiro, & W. E.    Montagne (Eds.), Schooling and the acquisition of knowledge (pp. 1-17). Hillsdale, NJ: Lawrence Erlbaum.
Chi, M. T., Feltovich, P. J., & Glaser, R. (1981). Categorization and representation of physics problems by experts    and novices. Cognitive Science, 5, 121-152.
Donald, J.G. (1985). Intellectual skills in higher education. Canadian Journal of Higher Education, 1 (1), 53-68.
Donald, J.G. (1988). Professors' expectations of students' ability to think. Higher Education Research and    Development, 17 (1), 19-35.
Donald, J. G. (2000). The pedagogue: Creating designs for teaching. In J. Bess, (Ed.), Teaching alone/teaching    together: Transforming the structure of teams for teaching (pp. 35-61). San Francisco: Jossey-Bass.
Donald, J. G. (2002). Learning to think: Disciplinary perspectives. San Francisco: Jossey-Bass.
Donald, J. G., & Denison, D. B. (2001). Quality assessment of university students: Student perceptions of quality    criteria. Journal of Higher Education, 72, 478-501.
Donald, J. G., & Dubuc, P. (1999). Report on students’ conceptualizations of learning, 1994-1995. Centre for    University Teaching and Learning, McGill University.
Dressel, P., & Mayhew, L. (1974). Higher education as a field of study. San Francisco: Jossey-Bass.
Ericksen, A., & Smith, J. (Eds.). (1991). Study of expertise: Prospects and limits. Cambridge: Cambridge University    Press.
Feldman, K. A. (1989). Instructional effectiveness of college teachers judged by teachers themselves, current and    former students, colleagues, administrators, and external (neutral) observers. Research in Higher Education, 30,    137-194.
Feldman, K. A. (1996). Identifying exemplary teaching: Using data from course and teacher evaluations. In M. D. Svinicki & R. J. Menges, (Eds.), Honoring exemplary teaching. 65 (pp. 41-50). New directions for teaching and    learning. San Francisco: Jossey-Bass.
Frederiksen, N. (1984). Implications of cognitive theory for instruction in problem solving. Review of Educational    Research, 54 (3), 363-407.
Hirsch, E. D. (1967). Validity in interpretation. New Haven CT: Yale University Press.
Johnston, R. (1998). The university of the future: Boyer revisited. Higher Education, 36, 253-272.
Kuhn, T. S. (1970). The structure of scientific revolutions. Chicago: University of Chicago.
Leamnson, R. (1999). Thinking about teaching and learning: Developing habits of learning with first year    college and university students. Sterling, VA: Stylus.
McDermott, J., & Larkin, J. H. (1978). Re-representing textbook physics problems. Proceedings of the 2nd    National Conference of the Canadian Society for Computational Studies of Intelligence. Toronto: University    of Toronto.
Meyers, C. (1986). Teaching students to think critically: A guide for faculty in all disciplines. San Francisco:    Jossey-Bass.
Miller, G. A. (1956). The magical number seven, plus or minus two: some limits on our capacity for processing    information. The Psychological Review, 63 (2), 81-97.
Reif, F., Larkin, J. H., & Brackett, G. C. (1976). Teaching general learning and problem-solving skills. American    Journal of Physics, 44 (3), 212-217.
Shannon, C. E., & Weaver, W. (1949). Mathematical theory of communication. Urbana IL: University of Illinois.
Sternberg, R. J. (1985, November). Teaching critical thinking, Part 1: Are we making critical mistakes? Phi Delta    Kappan, 194-198.
Sternberg, R. J. (1998). Abilities are forms of developing expertise. Educational Researcher, 27 (3), 11-20.
Thorndike, E. L. (1914). Educational psychology. Vol. 3. New York: Teacher’s College, Columbia University.
Tolman, E. C. (1932, 1949). Purposive behavior in animals and men. New York: Appleton Century Crofts.
Woodworth, R. S. & Schlosberg, H. (1954). Experimental psychology. New York: Henry Holt.