An Introduction to Simulations for Teaching & Learning

Computer-based simulations have a history as long as the development of the computer itself, dating back to the days of the Manhattan Project.

One could argue that the technique of using simulations for teaching and learning (without technology) has been around for as long as students have been solving problems through role play and the use of data, which in the United States could mean at least as far back as John Dewey's Laboratory Schools around the start of the 20th Century.

Taken together, computer simulations built specifically for teaching and learning have evolved over the past two to three decades to allow for the possibility of dynamic discovery-oriented learning. This primer tackles some introductory considerations for the use of simulations in education.

What is a computer-based educational simulation?

From CCNMTL's point-of-view, a simulation is a teaching tool with three components:

1) A data model or set of algorithms that can be manipulated by the learner and provides dynamic feedback based on those manipulations
The core of a simulation is a model of some kind that attempts to replicate some aspect of reality through a system of data, algorithms, and feedback. On its own, a data model (or simulator) can be an effective teaching and learning device (see Heart Simulator) if the learner is prepared to directly manipulate the model's parameters and respond to feedback without any further context.

But in many cases, further context and structure is needed. This comes in two parts:

2) A role for the learner to adopt
The objective of a role is to offer a point-of-view for the learner's actions. For example, perhaps you are the manager of a humanitarian assistance crisis (see ReliefSim) or you are the investigator of a potential environmental contamination event (see Brownfield Action).

3) An objective to achieve or set of tasks to complete using the data model
Since the data model does not teach directly nor does it lead the learner to act in one way or another, one often needs to present a problem or goal up front for the learner to target. Perhaps you are a farmer in a poverty-stricken village and your goal is to simply stay alive for 50 years (see Millennium Village Simulation).

These three components together make up a simulation experience that can be a very engaging way to learn.

When should you consider using a simulation?

Over the years, CCNMTL has elaborated five tenets for considering simulations as a strategy to enhance learning:

1) Integrating seemingly disparate topics that have definable relationships
It is very common in college programs today for there to be introductory courses that attempt to cover a wide range of seemingly disparate topics that nonetheless have some connection between them or that combine into some system of dynamic parts. In environmental science, this might be the relationships between human health, geology, and toxins. In public health, it might be the challenge of prioritizing disease vaccinations versus sanitation facilities versus food and water assistance with limited resources. In both examples, traditional courses might be organized into a series of lectures by different experts each with their own case studies or problem sets tied to his or her own specialty.

A frequent complaint from students and faculty in courses organized in this way is that students fail to see connections between topics and do not get a sense of the big picture. Because the content is siloed into isolated disciplines with the traditional approach, it is quite natural for this problem to occur. Using a simulation creates the possibility for a multi-disciplinary or multi-concept experience where the challenge might be to try different prioritization strategies in a public health emergency or to explore a simulated geological contamination where human health is at risk. In both cases, identifying the system relationships is necessary to solve the problem or to optimize the system's performance.

Simulations are tailor-made to address this integration problem, as they are in their most basic form a system of related dynamic parts that are meant to adjust and adapt to changing conditions.

2) Learning a system or model of concepts through experimentation
Some instructors have clear goals for teaching a system of concepts or a model, but they attempt to convey these concepts through didactic methods where students simply have to reproduce the model without really demonstrating any expertise in it.

As one might expect, most students are quite capable of this regurgitation, but when faced with a problem, they fail to show an ability to apply the model or set of concepts appropriately. This suggests that students will not be able to transfer the concepts to anything analogous they encounter in future courses or in their everyday experience. Furthermore, even if they are able to solve a problem using the model, students of more didactic methods will be unable to cope with new information or subtle changes to the model, suggesting an inability to generalize the skill to new situations.

Using a simulation with a deliberate discovery approach allows a learner to learn the model through experimentation with it. Rather than memorizing the relationships of a model as a given, students can be tasked with deriving the relationships through manipulating parameters in the model, thus creating their own mental model which they can more readily apply and adapt to new problems and situations.

3) Practicing difficult, rare, or dangerous tasks
There are situations where learners need practice at tasks that are especially difficult, rarely available, or dangerous. Imagine flying an airplane without a flight simulator experience or being asked to perform a rare surgical procedure without a trial run on a dummy patient. Simulations allow a learner to practice a string of difficult maneuvers many times over. They allow a person to have a virtual experience that otherwise might not be available to them based on space or time constraints.

4) Decision-making and prioritizing
Many college courses aim to teach good judgment - which ends up being translated into decision-making and prioritizing tasks where a learner is asked to operate with some combination of incomplete information and time or resource constraints. Case studies often are meant to address these same goals, and in certain instances, especially with a good instructor, they can be very effective. What is sometimes missing, though, is the sense of the drama, perhaps in part because traditional case studies have a 20/20 hindsight element to them.

Simulations are a good way to create drama because they place you in situ and ask you to take action in the moment:

You are the Managing Director of the XYZ Mutual Fund. The Fed has just changed regulations that will have a major effect on how you currently operate. Every minute you don't respond, more of your shareholders are dumping their stocks and calling for your ouster. What do you do?

Many simulation participants, both instructors and students, will tell you that acting in the moment feels more authentic and the in situ feeling adds to the learning experience by creating more motivation, heightened awareness, and a sense of immersion that helps block out distractions and stay focused on the learning experience. For many, their experience with simulation-style video games is similar. For example, some are not only willing but enthusiastic about spending many, many hours working at these kinds of games, losing track of time because of the immersion.

5) Working with ambiguity
A final piece often missing in college classrooms is a requirement for students to work with the unknown and find a way to move forward in the face of ambiguity. All too often, a student can simply ask (or wait) for the answer and plod forward in lockstep with their classmates. Then, when faced with a real-world job or a problem with a set of undefined objectives and no one to follow, that person is frozen, unable to move without that spoon that was feeding them everything they needed to move ahead.

By not directing the user, simulations require one to engage and decide what action to take, what parameter to adjust, and what option or strategy to explore. Many students are resistant to having to direct their own learning, perhaps because their prior schooling did not expose them to it very much, if at all. With the proper open-ended guidance from an instructor, students can learn to explore and create meaning for themselves using a simulation and walk away with the benefit of learning how not to be stymied in the face of ambiguity.

Challenges of Simulations

Simulations are not without their limits. Here are some of the most prominent challenges we've encountered to date:

Inherently reductionist - You can't model everything
Because simulations require one to reduce reality down to a set of data and equations, using them as a teaching and learning tool is inherently reductionist. Therefore, every simulation is subject to a critique of omission of a key component of any reality it tries to represent. For example, in the Millennium Village Simulation, the project team did not take cultural issues into account when creating the model. This was done because (1) culture was not part of the system being taught in the course it was designed for and (2) culture is very difficult to quantify into a set of algorithms. For simulations to be effective, designers and faculty have to make choices about what to represent and what to leave out of a model in order to reasonably reduce reality down to something that can be modeled for the purposes of teaching a particular set of concepts.

CCNMTL's approach is to deliberate carefully up front about which elements are of the highest priority and which will make the biggest educational impact for the given course objectives, and to design models that can incorporate changes and additions in future iterations.

Require a commitment to discovery-oriented learning and related support
Simulations will fail to reach their potential if the project team is unable to follow through with a student-centered discovery approach to learning. Occasionally, after building a simulation and implementing it in a course, faculty will fall back on traditional methods of dictating student work or conveying information at the first sign of student resistance (see previous points about learning through experimentation and working with ambiguity), removing two of the major rationales for choosing this kind of teaching strategy. In other cases, faculty will leave students entirely to their own devices, expecting students to learn on their own. This will also fail. Faculty need to take on a guiding role, which requires patience and a bit of a tough skin in the face of students not familiar with learning this way.

CCNMTL's approach here is to devote time to planning the classroom implementation strategy for the use of the simulation and providing consultation on how to assist students in a manner that encourages discovery, which could include working with lab instructors and TAs.

More work to create than most other types of educational technology projects
Because simulations rely on the creation of a model, they typically take a larger time investment up front before one can see the project's potential. Additionally, simulations require more testing than many other projects to ensure a balanced experience for the learner that will not reward mindless trial and error strategies or other "gaming" of the system to achieve the objective.

More difficult to evaluate what students learn than many other types of educational technology projects
Finally, because a simulation's learning objective is often problem solving skills, the capacity for good judgment, or a more affective emotional connection or awareness of course concepts, it can be especially challenging to develop assessments that demonstrate the learning that took place because of the use of the simulation. It is sometimes difficult to isolate these types of learnings down to the simulation experience and, therefore, one is often left having to be satisfied with correlating new skills gained by students to the use of the tool without being able to say that the simulation caused the learning.

CCNMTL continues to evolve its evaluation methods in effort to capture the experiences of students' use of simulations and is committed to working with its partners to demonstrate their effectiveness.