Cognitive Psychology Of Discovery Learning Research Paper

More realistically, the phrase ‘discovery learning’ denotes, as much as anything, a style or aesthetic of education. Broadly speaking, advocates of discovery learning tend to invoke visual images of children learning in playgrounds, gardens, laboratories, workshops, and studios; the opposing visual images are those of children seated at rows of desks, listening to a teacher, reciting in unison, or writing in lockstep. These contrasting visual images are important because they encapsulate the concerns, and anticipate the rhetoric, of both the proponents and opponents of discovery learning. Proponents focus on notions such as motivation, depth and authenticity of experience, and respect for individual learning styles; opponents focus on notions such as rapid skill acquisition, efficiency of learning, and quantitative assessment of learning outcomes. The arguments between the two camps, then, are in large part rooted in a strong division between two venerable notions of education—one which focuses on the cultivation of deep individual interests, and one which focuses on mastering fundamental skills as the basis for productive citizenship.

This research paper focuses on discovery learning as an object of study in cognitive science. As such it does not, and could not, resolve the philosophical debate over the value of discovery learning; to do so would be to resolve far deeper questions about what the essential purpose of education might be. Nonetheless, a cognitive science perspective can inform this ongoing debate in several ways: in particular, cognitive theory suggests where the strengths or pitfalls of discovery learning may lie, whether there are settings and domains for which it might best be suited, or what sorts of research and development efforts might improve its practice. The following sections explore these issues. By way of preface, Sect. 2 begins by describing the role of discovery learning in the history of science and mathematics education. Section. 3 then focuses on those particular issues for discovery learning that are highlighted by cognitive science research; and Sect. 4 discusses the role of educational technology in the practice (both current and future) of discovery learning.

2. Discovery Learning In Science Education: From Ancient History To Recent Debates

Images of discovery learning have an ancient pedigree in educational writing. Indeed, one could argue that Socrates’ mathematics lesson in Plato’s Meno— perhaps the quintessential classical portrait of pedagogical success—is an instance of ‘guided discovery learning’ in that the slave boy (student) is brought to understand a special case of the Pythagorean theorem through his own responses rather than direct instruction. (Undoubtedly, this instance of discovery learning is very energetically guided.) Elements of discovery learning can be found in the writings of the seventeenth-century Czech religious educator John Comenius:

Men must, as far as possible, be taught to become wise by studying the heavens, the earth, oaks, and beeches, but not by studying books; that is to say, they must learn to know and investigate the things themselves … (N)o information should be imparted on the grounds of bookish authority, but should be authorised by actual demonstration for the senses and to the intellect. (Rusk 1967, p. 97)

A century later, Jean-Jacques Rousseau’s educational classic Emile includes similar passages: ‘Teach your scholar to observe the phenomena of nature … Put the problems before him and let him solve them himself. Let him know nothing because you have told him, but because he has learnt it for himself. Let him not be taught science, let him discover it’ (Rousseau 1992, p. 131). Rousseau’s ideas were taken up in the nineteenth century by the Swiss educator Johann Pestalozzi, and by Friedrich Froebel, the innovative German educator famous for his development of the idea of a ‘kindergarten’ (Brosterman 1997, pp. 18–22, 30–5).

In the past century, the central images of discovery learning are prominently found in the writings of Dewey, particularly in his descriptions of science education (cf. Dewey 1916, pp. 266–76). Piaget’s work is likewise often cited as providing a theoretical foundation in cognitive psychology for discovery learning: the Piagetian child is one who (in the apt summary of Gardner et al. 1996, p. 28) ‘would first learn about the world directly through spontaneous and natural actions upon the physical world; then, acquiring a set of more complex cognitive structures in a preordained fashion, that child could perform ‘‘mental operations’’ of increasing abstraction and power on the representations that she had constructed.’ Perhaps the most eloquent advocate for the role of discovery in education is Bruner, whose essay ‘The act of discovery’ (Bruner 1979) may be read as a sort of manifesto for the entire discovery learning enterprise. In this essay, Bruner hypothesizes four broad benefits that a student might derive from learning via discovery: an increase in intellectual potency (i.e., development of stronger techniques for problem solving); a shift from extrinsic to intrinsic rewards (i.e., a shift to motivation derived from the act of discovery itself ); learning of heuristics of discovering; and improved memory for learned material (since that material was learned in a rich context developed by the student themselves).

After centuries of such energetic and passionate advocacy, one might expect that discovery learning would be a relatively uncontroversial notion in education; but the history of education shows that uncontroversial notions are few and far between. A representative critique of discovery learning may be found in Hirsch (1996, p. 250), who presents a skeptical one-paragraph definition of the concept, noting that ‘(T)here are two serious drawbacks to preponderant or exclusive reliance on discovery learning. First, students do not always make on their own the discoveries that they are supposed to make … Second, discovery learning has proved to be very inefficient.’ This theme of the pedagogical uncertainty of the method is described as well by Driver (1994, p. 42): ‘(T)he connections that are apparent to a scientist may be far from obvious to a pupil.’ Even among educators sympathetic to discovery learning, pitfalls are noted. Papert (1993, p. 16), for example, sounds the important theme of context: ‘It is simply double-talk to ask children to take charge of their own learning and at the same time order them to ‘discover’ something that can have no role in helping them understand anything they care about or are interested in or curious about.’

3. Discovery Learning And Cognitive Science

The preceding paragraphs leave us with a frustrating problem. On the one hand, learning via discovery would seem to be desirable on affective and intellectual grounds; on the other hand, the difficulties of practical implementation of discovery learning are widely acknowledged. One question that we might begin with, then, would be whether there is evidence that children (as opposed to, say, professional scientists) are indeed capable of effective learning via discovery.

One body of relevant evidence comes from recent work in infant cognition. The tasks with which babies are confronted—to learn the grammar of their natural language, to learn the behaviors of objects in the world, to learn about their own and other people’s minds—seem, upon examination, to be approached by persistent and at least somewhat systematic inquiry. Gopnik et al. (1999, p. 86) summarize this research, writing:

Human children in the first three years of life are consumed by a desire to explore and experiment with objects. In fact, we take this for granted as a sometimes exhausting fact of parenting … (F)or our species the dangers of exploration are offset by the benefits of learning. The rapid and profound changes in children’s understanding of the world seem related to the ways they explore and experiment.

It appears, then, that successful discovery learning does occur, given the right conditions. But note that infant cognition may be special in a number of respects: first, the domains in which discovery takes place for infants (language, object recognition) are most likely domains in which a great deal of inborn cognitive and perhaps neurological structure already exists prior to exploration. Second, as Gopnik et al. point out, the infant’s environment is interestingly poised to reward infant experimentation in the form of adult behavior: adults, for example, ease the learning of language through the use of ‘motherese,’ with its short, simple sentences and affectionate intonations (Gopnik et al. 1999, pp. 128–32).

The problems with discovery learning in science— the most typical classroom domain—are more pointed, and form an interesting contrast to the successful learning scenarios of infant cognition. Children (and adults) come to scientific learning with many misconceptions, some of them arguably derived from the very same natural explorations they performed as infants (e.g., the notion that inanimate moving objects inevitably ‘run down,’ Driver et al. 1985). Our everyday environment is not especially well suited to disclosing phenomena that occur in the absence of friction (such as Newtonian mechanics), or over long periods of time (such as evolution), or at high speeds (such as relativistic effects) or small scales (such as quantum effects), making scientific subjects difficult to teach and understand, even for adults. Moreover, as discussed by Kuhn et al. (1988, p. 220), key aspects of sustained scientific inquiry (e.g., maintaining a conscious distinction between evidence and theory) are difficult for both children and adults.

4. Directions For Discovery Learning

Despite the difficulties involved in getting it right, the benefits that accrue from providing students with opportunities for discovery seem profound, not least on the affective dimension. Csikszentmihalyi, who has done extensive empirical work on motivation and creativity, writes:

When people are asked to choose from a list the best description of how they feel when doing whatever they enjoy doing most—reading, climbing mountains, playing chess, whatever—the answer most frequently chosen is ‘designing or discovering something new.’ … (U)nless enough people are motivated by the enjoyment that comes from confronting challenges, by discovering new ways of being and doing, there is no evolution of culture, no progress in thought or feeling. (Csikszentmihalyi 1996, pp. 108–10)

Csikszentmihalyi’s admonition suggests that one promising direction of research in discovery learning is to make it more motivating and enjoyable for students, perhaps by looking to the sorts of factors that encourage discovery and exploration in infants and toddlers. These factors include increased attention to (a) the nature of students’ preexisting theories and to their skills and predilections for theory building; (b) the arrangement of learning environments, both physical and social, conducive to enjoyable discovery; and (c) the design of novel artifacts whose purpose is to encourage and reward sustained periods of exploration and discovery.

Many provocative recent examples of research in this last direction may be found in the development of new technologies for education. The advent of numerous computer ‘microworlds’ (Papert 1980), in which simulations or experimentation can take place under conditions difficult to achieve in the ‘real’ world of objects, is one notable development in the past two decades. More recent developments in the integration of physical and computational elements to produce (for example) new types of ‘digital manipulatives’ (Resnick et al. 1998, Eisenberg and Eisenberg 1999) suggest new and uncharted vistas for creating ‘discovery-friendly environments.’ Such environments, to be pedagogically successful, will need to be designed with an eye both toward cognitive research in students’ scientific understanding, and toward supportive social structures (e.g., for adult mentoring, for apprenticeship, for collaboration). Nonetheless, despite the practical obstacles, there is still life in the centuries-old dream of learning via discovery.

Bibliography:

1. Brosterman N 1997 Inventing Kindergarten. Harry Abrams, New York
2. Bruner J 1979 On Knowing. Belknap Press, Cambridge, MA
3. Csikszentmihalyi M 1996 Creativity. Harper Collins, New York
4. Dewey J 1916 Democracy and Education. Free Press, New York
5. Driver R 1994 The fallacy of induction in science teaching. In: Levinson R (ed.) Teaching Science. Routledge, London, pp. 41–8
6. Driver R, Guesne E, Tiberghien A (eds.) 1985 Children’s Ideas in Science. Open University Press, Philadelphia
7. Eisenberg M, Eisenberg A 1999 Middle tech: Blurring the division between high and low tech in education. In: Druin A (ed.) The Design of Children’s Technology. Morgan Kaufmann, San Francisco, pp. 244–73
8. Gardner H, Torff B, Hatch T 1996 The age of innocence reconsidered: Preserving the best of the progressive traditions in psychology and education. In: Olson D, Torrance N (eds.) The Handbook of Education and Human Development. Blackwell, Malden, MA, pp. 28–55
9. Gopnik A, Meltzoff A, Kuhl P 1999 The Scientist in the Crib. William Morrow, New York
10. Hirsch E D 1996 The Schools We Need. Doubleday, New York
11. Kuhn D, Amsel E, O’Loughlin M 1988 The Development of Scientific Thinking Skills. Academic Press, San Diego, CA
12. Papert S 1980 Mindstorms. Basic Books, New York
13. Papert S 1993 The Children’s Machine. Basic Books, New York
14. Resnick M, Martin F, Berg R, Borovoy R, Colella V, Kramer K, Silverman B 1998 Digital manipulatives: New toys to think with. Proceedings of CHI ‘98, Association for Computing Machinery (ACM) Special Interest group on Computer-Human interaction (SIGCHI), ACM, New York, pp. 281–7
15. Rousseau J-J 1992 Emile. Charles E. Tuttle, Rutland, VT
16. Rusk R 1967 The Doctrines of the Great Educators. St. Martin’s Press, New York