Re: [Phys-l] the role of simulation

["John Clement" <clement@hal-pc.org>]

The building of mental models is indeed a difficult task.  This was shown by
Anton Lawson when he found that students need to be at the theoretical level
to be able to easily understand things that you can not see.  While standard
labs can help students build mental models, simulations can be very useful
if they properly show what is going on at the microscopic level.  But they
need to be surrounded by appropriately designed interactive engagement
pedagogy.

What I am trying to point out is that a bias against simulations is not
warranted, and a bias in favor of always doing physical labs is also not
warranted.  Sometimes one may yield better results than the other, and a
simulation may serve as a better introduction than a physical lab.  There is
no blanket rule that one is automatically better than the other in a given
situation.  Only the research can show when one is better.  This is hardly
an "unquestioning" attitude.  For example the human motion labs are probably
better than simulations because they involve familiar things, and
kinesthetic feedback, but when one is dealing with other objects doing the
motion, simulations may actually sometimes be better.

In either case, you have to surround the lab by the appropriate pedagogy,
and this does not mean verification labs where the student is merely
verifying an equation or law that they have been already told.  The
McDermott cycle needs to have immediate feedback, and this can also be
accomplished by simulations more readily that by physical labs.  Certainly
simulations will not substitute for all physical experience in say learning
how to run a lathe.

John M. Clement
Houston, TX

> I, for one, intend to read the research, but I think this is another
> example of the unquestioned usefulness of simulations in teaching
> techniques. Whether it's flying an instrument approach, wiring a circuit,
> or dissecting a worm, simulations are appropriate. My question relating to
> your example, John, would be something along the lines of whether they are
> building a mental model of the interaction between fields and moving
> charged particles when they do simulations. I'm not saying they aren't,
> any more than I'm saying they will reliably build those concepts with
> actual circuits, rather asking how one is better than the other.
> sk