Computer simulation is a form of learning with computers in which the user may experiment with a simulated situation. The challenge for instructional designers is to create engaging, purposeful, educational, and highly interactive simulations. This paper first discusses the concept and common characteristics of computer simulations. Next advantages and limitations are identified. The final section of the paper explores implications for the design of effective computer simulations, including indications and guidelines for the use of computer simulations.

Rothwell and Kazanas (1999) define a simulation as an "artificial representation of real conditions". They further suggest that a simulation is prepared by first preparing a case study, then creating detailed descriptions of the characters in the case study. Min (1995), on the other hand, states that a "representation is not a simulation" but rather a model which serves as a basis for simulation. According to Min, in order to be considered a simulation, a situation must enable experimentation by a user and either strongly resemble reality or be a deliberate simplification. He notes that students often gain better insight into reality through good simplifications than by examination of all components of a complex situation.

Min (1995) describes four different kinds of computer simulation. One of these is simulation of conversation, such as may occur when a student 'nurse' ask a series of questions, with coaching as needed, of a computer simulated 'patient'. A second type is simulation of the behavior of an expert, in which the simulated 'expert' answers, having accessed a knowledge base, after the student has introduced a set of data. The third kind is simulation with moving pictures, which usually includes 'movies' and multiple-choice elements. Min considers the fourth type, simulation of phenomena (model-driven simulation), the most important because it is based on a mathematical model which can be experimentally manipulated by the learner until the 'rules' become apparent and are fully understood.

Boyle (1997) identifies three types of simulations based on the level of learner activity required. These are passive, exploration, and task-based simulation. In passive simulation, a learner simply observes the operation of a simulation as it unfolds. In exploration simulation, a learner selects from multiple paths to navigate through an unfamiliar simulated environment. In task-based simulations, a learner interacts with objects or characters in realistic situations to achieve an effect or goal. In Boyle's view, task-based simulations include case based learn-by-doing environments (CaBLE) and, because of the active, reasoning, driving role required of the learner, are of greatest educational value. In CaBLE, "the learner retrieves information, makes decisions and gets feedback on the consequences of those actions" (p. 39).

Regardless of conceptual approach, computer simulations share some common characteristics.

Simulations, as learning tools, characteristically provide:

• Practice in representative aspects of real situations
• Practical alternative when the real experience is too dangerous, too expensive, too slow, too rapid, or impossible to experience
• Method for rehearsing what to do in stressful situations
• Method for analyzing problems before taking action
• Discovery through immediate and direct feedback
• Opportunity to reproduce a chain of events that could not be repeatedly observed in a natural setting
• Opportunities for change and development during all sequences of the activity
• Interaction among participants, if group activities are included
• Informative feedback
• A critique or debriefing period

Perhaps the greatest advantage of computer simulations is that while working with a computer simulation program the student is experimenting and playing an active role. This active engagement contrasts with 'face-to-face' teaching in which students often listen passively (Min, 1995). In addition, simulations have been reported (Kennedy, 1998; Schank, 1995; Boyle, 1997) to have the following advantages:

• Requires little effort to use; a short introduction is usually sufficient.
• Simulations are based on experiential learning and are excellent discovery learning techniques.
• Students become active participants in the learning process.
• Learners are able to differentiate among input or variables according to their reasonableness, relevance, limitations, and effects.
• Encourage action, behavior, intuitive, and imaginative thinking and flexible problem-solving rather than rote learning.
• Provide experience in decision-making, and allow students to quickly see the consequences of their decisions.
• Participants can repeat simulations, trying different strategies/responses.
• Learning is student-centered rather than teacher-centered.
• Foster an inquiry process that assumes that knowledge is cumulative and constantly expanding.
• Fit easily into interdisciplinary learning.
• Provide near-immediate feedback.
• Provide information at the time the learner is ready for it.
• Time can be greatly compressed or expanded in simulations, so that situations that might otherwise be impractical because they occur too quickly, too slowly, or too complex, can be experienced.
• Dangerous situations can be tested in a non-threatening environment.
• Have a high rate of transfer to real-life situations.
• Are often highly motivating, intrinsically and extrinsically.
• Can be used to evoke emotions and attitudinal changes.
• Can involve group interaction, communication, and collaborative learning.
• Can be designed to accommodate different levels of learning.

While there are many advantages to the use of computer simulations as a learning method, limitations and disadvantages have also been reported (Kennedy, 1998; Schank, 1995; Boyle, 1997). These include:

• No educational tool is effective for everyone.
• Simulations cannot react to unexpected 'sub-goals' which learners may develop during the process.
• Simulations may function well technically, but be difficult to fit into a curriculum.
• Simulations may be more time consuming than alternative learning activities.
• Learners may become wrapped up in the simulated activity and lose sight of its objectives.
• Learners will have varying experiences and may not complete all components.
• Success of the simulation will depend heavily on the design.
• Negative biases may exist in the design, and undesirable attitude changes may be produced.
• Financial support may be difficult to obtain, since those unfamiliar with these materials may equate them with games and entertainment.
• Published research on simulations is sketchy and in many cases inconclusive.
• No widely accepted criteria have been established for decisive evaluation of simulations.

Schank (1995) notes that the major complaint about simulations is that they can never be as rich as and capture all of the intricacies of the real world. The challenge for instructional designers is to create realistic simulations that enable learners to more easily transfer learning to a real situation. Situations need to be designed so that learners can fail and have opportunities to learn from failure.

Schank (1995) further notes that computers can make learning a great deal of fun if designers develop easy to use, purposeful software which is flexible enough to be adapted to different levels of learners. Simulations need to be designed so they can be paused or slowed to enable reflection, research, and problem solving. Schanks states that experts or teachers should be 'embedded' so they can coach learners and interrupt strategically with stories, guidance, information, and commentary as system-monitored learner activity indicates or when requested. Simulations need to provide feedback to students and help them to understand the results of their actions. Schank suggests that simulations should be designed so that learners fall into well-known traps and make mistakes so that they can learn to reason their way out of them.

Min (1995) suggests that the primary focus of instructional developers should be to support the teacher's ability to individualize learning by designing programs that adapt to different circumstances and learner levels. He states that in addition to creating realistic scenarios, repeatability, stopping, restarting, entering and tweaking data, and flexibility of use are important design elements. An important responsibility of instructional developers is identifying when computer simulations should be used and any guidelines regarding their use.

Min (1995) identified a number of situations in which computer simulations are preferred to other tools for learning. These include:

• The process to be investigated takes place in reality too quickly, too slowly, or is to complex to be observed or experienced.
• Simulations allow the study of processes that cannot be examined in their natural occurrence, because they are too rare, too dangerous, or inaccessible.
• The system to be examined can be dangerous to manipulate, e.g. a nuclear reactor, a ship or a human body.
• It can be irresponsible from an ethical point of view to do research through traditional experiments, as with certain diseases for example.
• An introduction to a new subject or certain parts of it is desirable prior to actual practice.
• Experimental manipulation and visualization of the results are desired.
• Individual approaches to investigation are possible and desired.
• Multiple paths to achieve goals are desired.

Min (1995) suggests a number of criteria for the successful design of computer simulations. These include:

• Sufficient realism to convey the essential truths of the process.
• Consistent responses to meaningful and deliberate decisions.
• Adequate testing to determine how it "flows" and what it actually teaches.
• All rules, directions, and options should be clear and consistent.
• Experiences and skills must be at an appropriate level for the target users.
• Should not be so complex or elaborate that it will confuse users.
• Objectives should be clear.
• Resources such as time, equipment, and facilities must be available.
• Debriefing guidelines should be included.
• Supplemental activities should be included or suggested.

The use of computer simulation can not replace practical experience. However, when practical experience is considered important but is unavailable due to limited access, capacity or some other physical limitation, working with a computer simulation program may be an effective alternative. Well-designed computer simulations are motivational, highly interactive, educational, individualized, flexible, and can be fun. They support discovery and inductive learning. They can be designed to support collaborative learning. They also can result in feelings of accomplishment and achievement.

Although no educational tool is sufficient by itself or effective for everyone, many of the limitations of computer simulations addressed in this paper can be reduced through good instructional design. Other noted limitations, particularly the lack of published research on simulations and lack of widely accepted criteria for evaluation of simulations, need to be addressed through research.

References

Boyle, T. (1997). Design for multimedia learning. Hertfordshire: Prentice Hall Europe.

Min, R. (1995). Simulation technology and parallelism in learning environments. De Lier: Academic Book Center. ISBN 90-5478-036-3.

Rothwell, W. J., & Kazanas, H. C. (1997). Mastering the instructional design process: A systematic approach. San Francisco: Jossey-Bass.

Schank, R. C. (1995). Engines for education. Hillsdale, NJ: Lawerence Erlbaum Associates.

Note: The writer found the book by Rik Min to provide the most complete discussion on this topic. The ISBN is included for your convenience.

I look forward to your feedback.