Recent flurry over schools paying students to achieve led me back to Edward Deci's work. Below is a brief excerpt from Why We Do What We Do: The Dynamics of Personal Autonomy, Putnam 1995.
I have always believed that the experience of intrinsic motivation is its own justification. Smelling the roses, being enthralled by how the pieces of a puzzle fit together, seeing the sunlight as it dances in the clouds, feeling the thrill of reaching a mountain summit: These are experiences that need yield nothing more to be fully justified. And one might go so far as to argue that a life devoid of such experiences is hardly a life at all.
But modern society is not very concerned about all that. Modern society has what philosopher Charles Taylor recently referred to an the malaise of "instrumental reason." Everything gets evaluated in terms of its bottom-line yield--the cost-benefit ration, so to speak. Sadly, even things that should be evaluated by other criteria, like personal relationships, seem to have come under the dark eye of instrumental reason.
"To feel alive, to be interested and engrossed in an activity, to be in a state of flow, is all well and good," some will say, "but what does it get you?" These people want results. They want "noteworthy pictures," and they don't care whether the painter is in "a high state of functioning" while creating them. They want high test scores, and they are not terribly concerned if the students feel good or are interested in school. They want profits, and they do not pay much attention to the professional or personal development of the employees.
Of course it is important to attend to the outcomes of motivation, and although intrinsic motivation is a desirable end in its own right, [Richard] Ryan and I have devoted considerable attention to exploring the concrete consequences of being intrinsically motivated versus externally controlled. Without verifying that it has concrete advantages, we would be on shaky grounds advocating its promotion in schools, homes, and offices--indeed, in society more generally. So we set out to clarify whether people, when they are intrinsically motivated, also achieve at high levels. Robert Henri hinted at the answer with his powerful intuitive observation that intrinsic motivation is in "back of every true work of art." But what does the research say?
The arena of education seemed like ripe territory for beginning this research because countless people had suggested that motivation is the key to success in education. Certainly, learning seems to be great fun for some people and quite tedious for others, and it was the whole issue of motivation for learning that got me interested in intrinsic motivation in the first place. Fortunately, the outcomes of education--learning, performance, and adjustment--can be reliably measured, which is essential for doing research.
In education, grades (sometimes accompanied by other things like gold stars or dean's lists) are the primary means of extrinsic control. They are considered incentives, and it is assumed that people will be motivated to learn so they can get good grades. In one learning experiment I did with former student Carl Benware, we considered the issue of grades as a motivator. We had two groups of college students spend about three hours learning some complex material on neurophysiology--on the machinery of the brain. Half of these students were told they would be tested and graded on their learning, and the others were told they would have the opportunity to put the material to active use by teaching it to others. We expected that learning in order to be tested would feel very controlling to the students, whereas to put the information to active use would feel like an exciting challenge. After students had learned the material, we assessed their intrinsic motivation with a questionnaire, and we found, as expected, that those who learned in order to be tested were less intrinsically motivated.
Then we took it one step further to get at the main issueĂ˘€”the actual learning that had gone on. We tested both groups, even though the one group had not expected it, and the results showed that the students who learned in order to put the material to active use displayed considerably greater conceptual understanding of the material than did the students who learned in order to be tested. As the research made clear, yet again, well-intentioned peopleĂ˘€”for instance, people employing tests to motivate learningĂ˘€”are unwittingly defeating the desire to learn in those people they are attempting to help.
Ryan, working with Wendy Grodnick (now a faculty member at Clark University), did another learning study, this time with elementary-school children. Two groups of children were asked to read two short passages from grade-level textbooks. Some of the children were told that they would be tested and graded on what they read; others were just asked to read the material without any mention of a test. Those who learned the material without expecting to be tested displayed superior conceptual understanding relative to those who were expecting to be test.
An additional, interesting piece of information was picked up in this study. The children who expected to be tested displayed greater rote memorization than those not expecting the test. It seem that when people learn with the expectation of being evaluated, they focus on memorizing facts, but they donĂ˘€™t process the information as fully, so they donĂ˘€™t grasp the concepts as well. On the face of it, this suggests that the type of learning: context that should be created, depends on which type of learning one hopes to fosterĂ˘€”rote memorization or conceptual understanding. But thereĂ˘€™s a catch hereĂ˘€”and a quite fascinating oneĂ˘€”that was discovered in a final phase of this study.
Another adult visited the classrooms of these elementary-school children a week after they had participated in the experiment. The adult introduced himself and reminded the children of their experience the week before with the woman who had given them the material to read. He then said he would like to ask them some questions about what they had read. On that test, all the children recalled less than they had the week before when they had just read the material, but thatĂ˘€™s to be expected. Stunningly, however, those who had learned expecting to be tested had forgotten much more. Their superior rote memorization was no longer in evidence a few days later. Evidently, they memorized the material for the test, and when the test was over, they pulled the plug and let it drain out. Using computer jargon, Grolnick and Ryan referred to this as a Ă˘€ścore dump.Ă˘€ť
With both college students and elementary-school children, the research indicates quite convincingly that the strategy of giving tests is not necessarily productive if the objective is long-term learning. . . .