AMBRIDGE, Mass. — The scant mention of education in President Bush's State of the Union address suggests that the administration feels its work on the subject is done, at least for now. Last year's sweeping bill was a significant achievement, but as with most federal initiatives, it dealt primarily with administrative issues like financing and achievement tests. Little attention was given to the actual process of education: how events in the classroom affect the minds of the students.

Yet a bit of White House leadership might encourage educators and scientists to apply a better understanding of thinking and learning to what happens in the classroom. Bill Clinton, in fact, did show some enthusiasm for this approach — particularly research on the brain. But as exciting as neuroscience is, I suspect it will provide little enlightenment about education. All learning changes the brain, but the changes at the level of brain cells are similar in all complex organisms — including mice, which don't learn to read, write or add.

Rather, it is the patterns of changes across billions of neurons that determine the distinctively human forms of learning in the classroom. To understand these patterns, we need to apply insights from cognitive science, behavioral genetics and evolutionary and developmental psychology.

An important place to start might be in working to apply a scientific mindset to education itself — that is, to determine as best we can whether various beliefs about educational effectiveness are true. Classroom practice is often guided by romantic theories, slick packages and political crusades. Few practices have been evaluated using the paraphernalia of social science, such as data collection and control groups. We already know that some methods of reading instruction work better than others, yet many schools still use methods proved ineffective like "whole language" techniques.

The sciences of the mind can also provide a sounder conception of what the mind of a child is inherently good and bad at. Our minds are impressively competent at problems that were challenges to our evolutionary ancestors: speaking and listening, reading emotions and intentions, making friends and influencing people. They are not so good at problems that are far simpler (as gauged by what a computer can do) but which are posed by modern life: reading and writing, calculation, understanding how complex societies work. We should not assume that children can learn to write as easily as they learn to speak, or that children in groups will learn science as readily as they learn to exchange gossip. Educators must figure out how to co-opt the faculties that work effortlessly and to get children to apply them to problems at which they lack natural competence.

Finally, a better understanding of the mind can lead to setting new priorities as to what is taught. The goal of education should be to provide students with new cognitive tools for grasping the world. Observers from our best scientists to Jay Leno are appalled by the scientific illiteracy of typical Americans. This obliviousness leads people to squander their health on medical flimflam and to misunderstand the strengths and weaknesses of a market economy in their political choices.

The obvious solution is instruction at all levels in relatively new fields like economics, evolutionary biology and statistics. Yet most curriculums are set in stone, because no one wants to be the philistine who seems to be saying that it is unimportant to learn a foreign language or the classics. But there are only 24 hours in a day, and a decision to teach one subject is a decision not to teach another. The question is not whether trigonometry is important — it is — but whether it is more important than probability; not whether an educated person should know the classics, but whether it is more important to know the classics than elementary economics.

This is not just a question of "relevance" to everyday life; these fields are as rigorous and fundamental as those in traditional curriculums. Nor is it a question of tradition being bad and innovation good — many fad movements are just as evasive about setting priorities. Even if learning music were shown to enhance math skills, that doesn't mean it is as effective as the same number of hours spent learning math.

In a world with complexities that constantly challenge the abilities nature gave us, serious thinking about trade-offs in education cannot be responsibly avoided — by scientists, educators or policy makers.

Steven Pinker, professor of cognitive science at Massachusetts Institute of Technology, is author of "The Blank Slate: The Modern Denial of Human Nature."