The CPU Project: Students in control of inventing physics ideas
Fred Goldberg
The Constructing Physics Understanding in a Computer-Supported Learning
Environment (CPU) project is a National Science Foundation supported
project [1] aimed at creating laboratory and computer based materials
to support a learning environment where students take primary responsibility
for developing valid and robust knowledge in physics. The intended
students are mainly secondary school physics and physical science students
and prospective and practicing elementary teachers (through workshops
and University courses). Rather than depending on the instructor as
the source of knowledge, in the CPU classroom students develop, test
and modify their own ideas through experimentation and discussion with
their peers. This does not mean there is no organized structure to
the classroom. Indeed, there is a carefully designed sequence of activities
and a pedagogy that promotes and values extensive intragroup and whole
class discussion. However, the students' own ideas, supported through
experimental evidence, become the standard of authority.
The CPU project began early in 1995 and during the past two and a
half years learning units have been developed in the topical areas
of force and motion, current electricity, static electricity and magnetism,
and light and color, as well as a generic skills unit called Underpinnings.
Additional model units on wave motion and the small particle model
of matter are currently under development.
The CPU units are divided into cycles, each intended to support students'
construction of a relevant model or component of a model. Each cycle
is divided into three phases: Elicitation, Development and Application.
This approach is an extended modification of the Learning Cycle, developed
by Robert Karplus and others as part of the science Curriculum Improvement
Study (SCIS) of the 1960s.
The elicitation phase engages students in an extensive and
robust discussion centered around some interesting phenomenon. They
are usually asked to make predictions, explain their predictions based
on prior knowledge, observe the outcome of the experiment, and then
to suggest ways of making sense of the outcome, which is often a surprise
to many students. The purpose of this activity is not to make judgements
on which ideas suggested by the class are the most "correct" ones,
but instead to open up important issues and ideas that make sense to
at least some of the students in the class and can serve as focal points
of further inquiry.
In the development phase students work in small groups (more
or less independently), testing the class initial ideas in a wide variety
of experimental (hands-on) contexts. They record their observations,
ideas and explanations on the computer with special software and use
computer simulators to receive model-based feedback. The sequencing
of activities within the units is designed to challenge the common
student ideas that are described in the vast literature on research
in student understanding. As students go through the development phase,
they modify some of their initial ideas, cast some aside as not being
useful, and invent new ideas. The students should come to see that
the powerful ideas in science are inventions of the human mind and
not dictums from authority.
The purpose of the application phase is to provide students
with myriad opportunities to see the fruitfulness (and perhaps limitations)
of the class consensus ideas by applying them in a wide variety of
new and interesting contexts. Whereas the development phase engages
all students in a carefully structured sequence of activities to ensure
they all have a common experiential basis for developing shared (consensus)
ideas, the application phase allows students wide latitude to explore
their own questions, using available apparatus, the simulators, or
other resources such as the World Wide Web.
The entire learning process is supported by powerful software designed
especially for the project: a page-layout program for instructors to
author activities and for students to record their predictions, experimental
observations and explanations; a set of pedagogically-oriented physics
simulators that allow students to represent their predictions, and
then provide both phenomenological and model-based feedback; and idea
containers for students to keep track of their evolving ideas.
An Example from the Light and Color Unit-The first cycle of
the Light and Color unit focuses on phenomena involving illumination,
shadows and pinholes. Two ideas that the development phase activities
are intended to help students construct are that an extended light
source can be thought of as a collection of point sources, and that
light travels outward from each point in all directions. Students often
begin their study of geometrical optics by thinking about an extended
source holistically; that is, they think of the entire source, rather
than a point source, as the fundamental entity involved in optical
phenomena. Also, their initial drawings showing how light goes from
a source to a screen rarely show more than a single line connecting
a point on the source with a point on the screen. To facilitate the
invention of more sophisticated ideas about light and sources, students
work with Mini Maglites® (to approximate point sources) and regular
light bulbs (as extended sources) to investigate simple and complex
shadows, and the reproduction of a light source on a screen using a
pinhole. On the computer they write down their predictions, supporting
ideas and draw ray diagrams on top of pictures of the apparatus. After
performing experiments with the actual apparatus, they can set up similar
arrangements with special computer simulations.
Shown below are computer screen snapshots of various (side view)
set-ups in the light and color simulator program: (a) shadow on screen
formed with three point sources and a rectangular blocking object;
(b) shadow using a linear source (representing a clear, tubular showcase
bulb) and the same blocking object; (c) pinhole pattern on screen with
three point sources; and (d) pinhole pattern on screen with an extended
asymmetric source (representing a standard incandescent light bulb).
At the far right of the figure is the elements palette for this
particular simulator. For each set-up the student can obtain a representation
of the pattern of illumination on the front of the screen. These screen
views are shown to the right of the side views of the screen in each
of the four set-ups. In the last set-up (d) the student has dragged
out sprays of light rays from the top and bottom of the source. The
simulator thus provides both phenomenological and model-based feedback.
An Example from the Static Electricity and Magnetism Unit-In
the first cycle of the Static Electricity and Magnetism unit the students
focus on describing the similarities and differences between magnetic
and static electric phenomena. In the second cycle they build a model
for magnetism. It is not uncommon for students to begin this second
cycle with a separation-type model to explain what happens when a magnet
rubs an iron nail (thereby magnetizing it). For example, in the initial
model they often imagine there are two kinds of entities uniformly
distributed inside the nail (represented either with positive and negative
symbols, or with north and south symbols), and in the act of rubbing
the nail with a magnet the two kinds of entities separate to the opposite
sides of the nail. This, they reason, can account for the two-ended
nature of a magnetized object.
This model is immediately challenged in the first development activity,
where students are asked to predict what would happen if a magnet-rubbed
nail were cut in half. The observation that each half of the cut nail
acts as a two-ended magnetized object encourages students to modify
their model. This process of testing their model with various experiments
continues throughout the cycle. One of the last activities has students
look carefully at what happens when a magnet is rubbed over a test
tube partially filled with iron filings. For students who are not already
thinking about a "domain-like" orientation model to explain
the act of magnetization, this analogy provides a strong impetus to
imagine that two-ended entities inside a magnetic material become reoriented
(and aligned) when influenced by an external magnet. One of the simulators
is also used by students to gather evidence to support an orientation
model.
Another simulator used during the Application phase allows students
to explore more complex situations and to receive both phenomenological
and model-based feedback. This simulator shows two magnets influencing
a piece of ferromagnetic material (temporary magnetic dipole) constrained
to rotate about a pivot. Clicking inside the material with a micro
viewer opens a window alongside the ferromagnetic material, providing
a microscopic model view of the inside of the material.
Project Dissemination-The CPU pedagogy and supporting materials
will be disseminated through a series of workshops offered by 24 geographically
distributed teams of college and precollege teachers over the next
three years. Each team of three or four members has been prepared for
this task by participating in special leadership training sessions
held at San Diego State University during the summers of 1996 and 1997,
and by trial testing the materials in classrooms during the intervening
academic year. The teams will offer workshops for two distinct audiences:
elementary teachers, to help them develop a deeper conceptual understanding
of physics; and secondary school physics and physical science teachers,
to help them adapt the pedagogy and materials for their own classrooms.
Information about the teams, their planned workshops, the CPU project,
and the availability of the software (which has been submitted to Physics
Academic Software for publication) and other materials is available
on the CPU project's Web page (http://cpuproject.sdsu.edu).
[1] NSF Grant No. ESI-9454341
Fred Goldberg is
Professor of Physics at San Diego State University and head of the
physics learning research group in the Center for Research in Mathematics
and Science Education. He and Patricia Heller, of the University
of Minnesota, are co-directors of the CPU project.
|
