17.2 A DEFINITIVE FRAMEWORK

Games and simulations are often referred to as experiential exercises because they provide unique opportunities for students to interact with a knowledge domain. Two concepts important in the analysis of the nature of games and simulations are surface structure and deep structure. Briefly defined, surface structure refers to the paraphernalia and observable mechanics of an exercise (van Ments, 1984). Examples in games are drawing cards, moving pieces around a board, and so on. An essential surface structure component in a simulation, in contrast, is a scenario or set of data to be addressed by the participant.

Deep structure, in contrast, may be defined as the psychological mechanisms operating in the exercise (Gredler, 1990, 1992a). Deep structure refers to the nature of the interactions (1) between the learner and the major tasks in the exercise, and (2) between the students in the exercise. Examples include the extent of student control in the exercise, the learner actions that are rewarded in the exercise or which receive positive feedback, and the complexity of the decision sequence in the exercise (e.g., linear or branching).

17.2.1 Deep-Structure Characteristics

A shared feature of games and simulations is that they transport the players (game) or participants (simulation) to another world. For example, children may be searching for vocabulary clues to capture a wicked wizard (game), and medical students may be diagnosing and treating a comatose emergency room patient (simulation).

Another similarity is that, excluding adaptations of simple games like Bingo, games and simulations are environments in which students are in control of the action. Within the constraints established by die, rules, game players plan strategy in order to win, and simulation participants undertake particular roles or tasks in order to manage an evolving situation. Examples of evolving situations are managing a business and designing and managing research projects on generations of genetic traits.

The deep structure of games and simulations, however, varies in three important ways. First games are competitive exercises in which the objective is to excel by winning. Players compete for points or other advances (such as moving forward on a board) that indicate they are outperforming the other players. In a simulation, however, participants take on either (1) demanding, responsible roles such as, concerned. citizens, business managers, interplanetary explorers, or physicians, or (2) professional tasks such as exploring the causes of water pollution or operating a complex equipment system. In other words, instead of attempting to win, participants in a simulation for the classroom are executing serious responsibilities, with the associated privileges and consequences. Jones (1984, 1987) refers to this characteristic of simulations as "reality of function."

A second difference is that the event sequence of a game is typically linear, whereas a simulation sequence is nonlinear. The player or team in a game responds to a stimulus, typically a content-related question, and either advances or does not advance, depending on the answer. This sequence is repeated for each player or team at each turn.

In a simulation, however, participants at each decision point face different problems, issues, or events that result in large measure from their prior decisions. In a computer delivered simulation, this feature is referred to as branching.

A third difference between simulations and games is the mechanisms that determine the consequences to be delivered for different actions taken by the students in the exercise. Games consist of rules that describe allowable player moves, game constraints and privileges (such as ways of earning extra turns), and penalties for illegal (nonpermissable) actions. Further, the rules may be imaginative in that they need not relate to real-world events. In contrast, the basis for a simulation is a dynamic set of relationships among several variables that (1) change over time and (2) reflect authentic causal processes (i.e., the relationships must be verifiable). For example, in diagnostic simulations in which the student is managing the treatment of a patient, the patient's symptoms, general health characteristics, and selected treatment, all interact in predictable ways.

In addition to these three general characteristics, particular games and simulations also differ in the tasks established for students and the actions that are rewarded in the exercise. These specific differences are discussed later in the chapter.

17.2.2 Experiential and Symbolic Simulations

The broad category of instructional simulations consists of two principal types. One type, referred to as experiential simulations, establishes a particular psychological reality and places the participants in defined roles within that reality. The participants, in the context of their roles, execute their responsibilities in an evolving situation. Experiential simulations, in other words, are dynamic case studies with the participants on the inside (see 23.4.2).

Essential components of an experiential simulation are (1) a scenario of a complex task or problem that unfolds in part in response to learner actions, (2) a serious role taken by the learner in which he or she executes the responsibilities of the position, (3) multiple plausible paths through the experience, and (4) learner control of decision making (see Chapter 33).

Experiential simulations originally were developed to provide learner interactions in situations that are too costly or hazardous to provide in a real-world setting. Increasingly, however, they have begun to fulfill a broader function, that of permitting students to execute multidimensional problem-solving strategies as part of a defined role, The need for such exercises is indicated by several studies. For example, Willems (198 1) found that students in law, social geography, science, and sociology often are unable to apply knowledge they had acquired to the task of solving problems. Further, de Mesquita (1992) found that 53% of school psychology students and graduates initially made an incorrect diagnosis in a school-referral problem involving a third-grader.

Experiential simulations are designed to immerse the learner in a complex, evolving situation in which the learner is one of the functional components. The advent of computer technology, however, made possible the design of a different type of interaction exercise: a symbolic simulation. Briefly, a symbolic simulation is a dynamic representation of the functioning or behavior of some universe, system, set of processes, or phenomena by another system (in this case a computer). The, behavior that is being simulated involves the interaction of two or more variables over time.

A key characteristic of symbolic simulations (like experiential simulations) is that they involve the dynamic interactions of two or more variables. An example of a symbolic simulation is a population-ecology simulation with 75 variables that represents global ecological processes for the 200year period after 1900 (Forrester, 1971; Hinze, 1984).Another is a dynamic computer representation of a complex

equipment system. The student, interacting with a symbolic simulation, may be executing any of several tasks, such as troubleshooting equipment or predicting future trends. However, the student remains external to the evolving events. Many computer exercises erroneously labeled as simulations do not meet this criterion, and this shortcoming arises from the misapplication of the term simulated For example, simulated diamonds are imitation diamonds. Extrapolation of this concept to instructional development has led to the erroneous designation of imitations of objects or events as "simulations." An example is a brief Apple 11 computer program that purports to simulate plant growth. However, the program only presents an outline of a two-leafed plant that shoots up faster, slower, or not at all, depending on whether the student selects "full light ... .. half light," or "no light." The motion of the stilted graphic is a highly simplistic imitation of plant growth, but it is not a simulation. In other words, an animated graphic of some event is not necessarily a simulation.

Symbolic simulations differ from experiential simulations in two major ways. First, the learner is not a functional element of the situation. Instead, symbolic simulations are populations of events or interacting processes on which the learner may conduct any of several different operations. In other words, the deep structure of symbolic simulations is that the learner manipulates variables that are elements of a particular population. The purpose is to discover scientific relationships or principles, explain or predict events, confront misconceptions, and others. Potential instructional purposes for symbolic simulations are described by Riegeluth and Schwartz (1989) as explanation, prediction, solution, or procedure. Tennyson et al. (1987) differentiate simulations as task oriented or problem oriented.

The second major difference is the. mechanisms for reinforcing appropriate student behaviors. The learner in an experiential simulation steps into a scenario in which consequences for his or her actions occur in the form of (1) other participants' actions or (2) changes in (or effects on) the complex problem that the learner is attempting to manage. The learner who is executing random strategies often quickly experiences powerful contingencies for such behavior, from the reactions of other participants to being exited from the simulation for inadvertently "killing" the patient.

The symbolic simulation, however, is a population of events or set of processes external to the learner. That is, there is not an assigned role that establishes a vested interest for the learner in the outcome. Although the learner is expected to interact with the symbolic simulation as a researcher or investigator, the exercise, by its very nature, cannot divert the learner from the use of random strategies.

One solution is to ensure, in prior instruction, that students acquire both the relevant domain knowledge and essential research skills. That is, students should be proficient in developing mental models of complex situations, testing variables systematically, and revising one's mental model where necessary. In this way, students can approach the symbolic simulation equipped to address its complexities, and the possibility of executing random strategies holds little appeal.

Table 17-1 summarizes the primary characteristics of games, experiential simulations, and symbolic simulations. Specific design rules and subtypes are discussed in the following sections.