Viewpoint

The Virtues of Virtual Experiments

From across the pond comes the sad news that the University of Reading is phasing out its physics department, due to “budgetary constraints.” Despite protests from the British physics community, led by the Institute of Physics (the rough UK equivalent of the APS), the university’s ruling council voted decisively last fall to close the department by 2010, after the currently enrolled students have had a chance to graduate.

Philip Diamond, assistant director for Higher Education and Science at the IOP, noted that “the Higher Education Funding Council for England has now announced an additional £75m to support very high-cost subjects, including physics, from 2007-08 over three years. It is sad that this funding was not enough to save Reading’s physics department but the institute hopes that it will prevent more closures in the future.”

Diamond’s statement acknowledges the fact that teaching physics is typically a more expensive enterprise than teaching, say, English literature. In major research universities, the extra expense is due in part to the laboratory space and equipment necessary for faculty members to do their research, but in smaller institutions, the big extra item is the space, equipment and staff time for all the teaching laboratories. Many smaller American colleges and universities simply do not offer physics majors, even though they do have majors in chemistry and biology. Physics is squeezed between the high cost of teaching and the typically lower enrollments of majors compared to its sister sciences. In 2004, about 25% of all bachelors’ degrees were awarded to students in colleges or universities that did not have a physics major program.

A ray of hope in this perilous situation may come from a paper published in the recently established online journal Physical Review Special Topics: Physics Education Research. The abstract of the paper, by N. D. Finkelstein et al. (Phys. Rev. ST Phys. Educ. Res. 1, 010103 (2005)), is worth quoting in full:

This paper examines the effects of substituting a computer simulation for real laboratory equipment in the second semester of a large-scale introductory physics course. The direct current circuit laboratory was modified to compare the effects of using computer simulations with the effects of using real light bulbs, meters, and wires. Two groups of students, those who used real equipment and those who used a computer simulation that explicitly modeled electron flow, were compared in terms of their mastery of physics concepts and skills with real equipment. Students who used the simulated equipment outperformed their counterparts both on a conceptual survey of the domain and in the coordinated tasks of assembling a real circuit and describing how it worked.

In addition to being better pedagogically, the simulated laboratory is, of course, also less expensive to operate. Almost all undergraduates these days have their own computers. Although some computers should be made available in the physics building for their use, most of the students could log in and do the simulated experiments from their dorm rooms, libraries or study halls. There will be no need for staff to make sure the equipment is functioning, that proper safety procedures are being followed, and that all the other time-consuming but non-educational aspects of running an instructional laboratory are taken care of. And the considerable space set aside for laboratory work in introductory courses can be reduced to practically nothing.

Computer simulations as instructional devices have a significant history of success, even in areas where learning the material is of critical importance. Perhaps the best example is the use of flight simulators to teach pilots how to fly particular aircraft. First the pilot masters the basics using the flight simulator (undergraduate work) and only then does he or she get to practice on the real thing (graduate work). It’s hard to imagine that significant damage would be done to the training of future generations of physics majors if most of their undergraduate laboratories were replaced by well-designed simulations on the computer.

One should also recognize that computer simulation is an important mode of research in its own right. Very often theories are tested more rapidly and effectively using simulations instead of actual experiment. And experiments depend on computer simulations for both design and data analysis. Event simulators are crucial in planning particle accelerators, and in comparing what is observed with what is predicted. Familiarity with the techniques and capabilities of computer simulation will add to the educational experience, not diminish it. And since simulations also find wide application in the commercial world, physics majors who pursue career options other than graduate school will benefit from using them in their undergraduate courses.

Undergraduate laboratories, in their current incarnation, serve a variety of purposes. They teach physics (although, apparently, less well than analogous simulations); they hammer home the message that physics is, at bottom, an experimental science; they keep students out of mischief on long afternoons; they provide a fund of anecdotes, typically more amusing to retell than to experience–the time I nearly got electrocuted; the time my cell phone fell in the liquid nitrogen; the time I dropped the (fill in your favorite expensive piece of equipment) on the floor. It is unlikely that the good old-fashioned undergraduate laboratory will ever completely disappear. But at the very least, if a college or university is facing the loss of its physics department, the replacement of real experiments by virtual ones is a very minimal price to pay for keeping the enterprise afloat.

–Alan Chodos