Research related to Common Themes has focused on students' understanding of the notion of system, on the theoretical and tentative nature of models, and on the concept of conservation. Some research has found that student misconceptions about certain subjects can arise from their difficulty in recognizing natural phenomena as groups or systems of interacting objects.

The Science Curriculum Improvement Study (SCIS) curriculum led children to approach observation and analysis of natural phenomena by thinking of them as systems of interacting objects (Karplus & Thier, 1969). Research done in connection with SCIS indicates elementary students may believe that a system of objects must be doing something (interacting) in order to be a system or that a system that loses a part of itself is still the same system (Garigliano, 1975; Hill & Redden, 1985). Studies of student thinking show that, at all ages, they tend to interpret phenomena by noting the qualities of separate objects rather than by seeing the interactions between the parts of a system (Driver et al., 1985). Force, for instance, is considered as a property of bodies (forcefulness) rather than as an interaction between bodies. Similarly, students tend to think that whether a substance burns or not is being solely decided by the substance itself, whereas from a scientist's perspective, the process of burning involves the interaction of the burning substance and oxygen.

When students explain changes, they tend to postulate a cause that produces a chain of effects one after another (Driver et al., 1985). In considering a container being heated, students think of the process in directional terms with a source applying heat to the receptor. From a scientific point of view, of course, the situation is symmetrical, with two systems interacting, one gaining energy and the other losing it (Driver et al., 1985). Concentrating on the inputs and outputs of a system often requires a different, time-independent view, which students may not take to be an explanation. Students often do not seem to appreciate that the idea of energy conservation may help explain phenomena. Studies reporting students' difficulties with energy conservation suggest students should have opportunities to describe systems both as sequences of changes over time and as energy inputs and outputs (a systems approach) (Brook & Driver, 1984).

Student explanations of material change seldom include certain kinds of causes that are central to a scientific understanding of the world (Brosnan, 1990); for instance, that parts interact to produce wholes that have properties the parts do not. For children, wholes are like their parts. Brosnan (1990) summarizes all this by presenting two stereotypical views of the nature of change—the common-sense view and the scientific view (pp. 208-209):

Characteristics of a common-sense view of change:

• Properties belong to objects.

• The properties of an object are the same as those of the bits that make it up—not all of which may be visible at any one time.

• There are many kinds of stuff.

• Changes in macroscopic properties are the result of equivalent changes in the microscopic particles.

• If properties change it is because the bits that cause that property have moved away, come into view, changed from, grown or disappeared. New properties can be caused by the arrival of new bits.

Characteristics of a scientific view of change:

• Properties belong to systems.

• The properties of an object are different in kind from those bits that make it up.

• There are fundamentally only a few kinds of stuff.

• Changes in macroscopic properties are the results of changes in arrangements of unchanging microscopic particles.

• If properties appear or disappear it is because the arrangement of an unchanging set of continuing particles has altered—at a fundamental level substance is always conserved.

There is important research into the use of interactive computer models to teach students certain scientific concepts (e.g., Smith et al., 1987; White, 1990). Most models being developed are qualitative for two reasons. Because the prior knowledge and models students bring to their science instruction are themselves usually qualitative, qualitative reasoning is closely connected to that prior knowledge. Moreover, problem-solving studies have shown that qualitative reasoning is not engaged if students move too quickly into memorizing and applying formal laws. There is still a need to examine student understanding and use of models in general and the characteristic knowledge and misunderstandings they hold about models.

Middle-school and high-school students typically think of models as physical copies of reality, not as conceptual representations (Grosslight et al., 1991). They lack the notion that the usefulness of a model can be tested by comparing its implications to actual observations. Students know models can be changed but changing a model for them means (typical of high-school students) adding new information or (typical of middle-school students) replacing a part that was made wrong.

Many high-school students think models help them understand nature but also believe that models do not duplicate reality. This is chiefly because they think that models have always changed and not because they are aware of the metaphorical status of scientific models (Aikenhead, 1987; Ryan & Aikenhead, 1992). These difficulties continue even for some undergraduate chemistry students (Ingham & Gilbert, 1991).

Students may not accept the explanatory role of models if the model shares only its abstract form with the phenomenon, but will usually accept the explanatory role of models if many of the material features are also the same (Brown & Clement, 1989). Middle-school students may have severe difficulties understanding the hydraulic analogue of an electric circuit and think the two circuits belong to entirely different areas of reality (Kircher, 1985).

Middle-school and high-school students may think everything they learn in science classes is factual and make no distinction between observation and theory (or model) (Brook et al., 1983). If this distinction is to be understood, it should be made explicit when models like the atomic/molecular model are introduced (Brook et al., 1983). Irrelevant aspects of the concrete model can distract students and should be pointed out.

Lower elementary-school students fail to conserve weight and volume of objects that change shape. When an object's appearance changes in several dimensions, they focus on only one. They cannot imagine a reversed or restored condition and focus mostly on the object's present appearance (Gega, 1986). The ability to conserve develops gradually. Students typically understand conservation of number between the ages of 6 and 7, of length and amount (solid and liquid) between 7 and 8, of area between 8 and 10, of weight between 9 and 11, and of displaced volume between 13 and 14. These ages will vary when different children are tested or the same children are tested in different contexts (Donaldson, 1978).

Many students cannot discern weight conservation in some tasks until they are 15 years old. The ability to conserve weight in a task involving transformation from liquid to gas or solid to gas may rise from 5% in 9-year-old children to about 70% in 14- to 15-year-old-children (Stavy, 1990). More complex changes, such as chemical reactions, especially those where gas is absorbed or released, are still more difficult to grasp as instances of weight conservation (Stavy, 1990).

Fourth-graders' representations of changes over time are "data-driven" in the sense that the particular data in the problem are the most important. This contrasts with "system-driven" representations in which the emphasis is on overall patterns. Unfortunately, students are typically introduced to system-driven representations while they still think it is a wrong or meaningless way to convey information (Tierney & Nemirovsky, 1991).

No applicable research findings.