Serious Game Design Guidelines for the Military Training

TNO AND UNIVERSITY OF AMSTERDAM

Serious Game Design Guidelines for the Military Training

Thesis Master Information Science – Game Studies

Author: F.S.G. van Dijen Student number: 10658181 Supervisor: R.C. Boonekamp Final version: 25th of July 2014

Serious Game Design Guidelines for the Military Training

Abstract

In this paper game design concepts are explored for education and training in the Dutch military. By investigating currently available games, learning procedures and expert interviews within this paper guidelines are created that will aid the game designer in developing a game by aligning game mechanics to the learning goals as well as the intended user. Three concepts are proposed for corresponding military levels and design guidelines are formulated to give insight on why some game mechanics are more useful than others when designing for learning goals.

1.0 Introduction

In this paper, the triangular relation between game mechanics, learning goals and user traits will be explored and recommend three game concepts for specific learning goals and groups. Gaming has been in use for military training and will be in the future, but there are no clear guidelines about how to design a game for learning goals and user types through game mechanics. By looking at commercially developed games, as well as games that are created in a military context. In this paper, it is attempted to explicate the game mechanics that help reaching the set of learning goals needed for training on different levels in the hierarchical structure of the modern military. Aspects of tactical decision making, such as situational awareness, thinking in effects and limited resources are learning goals. A game developer has to translate these goals into game mechanics and has to keep in mind what preferences or reservations the user might have (Desurvire, H., Caplan, M., & Toth, J. A., 2004).

Gaming is a relevant tool that could streamline the training process within the military since it can give the player a feel of agency and presence within a situation (Tamborini, R., & Skalski, P. 2006). When it comes to learning about environment, equipment or behavior, games have the potential to accurately represent these components and give direct feedback about the actions taken by the player or changes in the situation (Barnes, T., Powell, E., Chaffin, A., & Lipford, H., 2008). As more and more games are being used in training, it is time to find out where budget should be spent when creating a game with a specific learning goal. In this paper three types of games will be explored and game design guidelines will provided for different levels of the Dutch military.

2.0 Definition

In this paper, there are several key words that describe game types and aspects of game design. Three types of games are highlighted within this paper because they give a good foundation to build upon for more learning goal focused games. Some important definitions are:

1. First person shooters (FPS) – games where the player looks through the eyes of a character and only has control over that one characters actions or can give orders to a small group.

2. Real time strategy (RTS) – games where the player controls multiple units from a gods eye perspective 3. Grand strategy – games where the player controls an

entire military conflict from a high level.

4. Comprehensive approach - effective crisis management involving political, civilian and military instruments.

3.0 Problem definition

In present-day, military training methods of real life practice require a lot of time and resources. These training sessions, can be in the form of a reenactment of a situation with actors and personnel from multiple divisions, which can last for hours and even days. In other instances the training is done via war games

which can be seen as elaborate board games in which the players are accompanied by their subordinates and play scenarios. These sessions can last for a couple of hours but the results can be unclear or take long to produce. Again, there are many people needed to run these sessions (McHugh, F. J., & Fischer, H. F., 1966). Other types of learning in the military involve reading papers and books about the equipment or having classical lectures about tactics. Gaming in the military can help with training the troops as well as commanders. A game like Go4it (van der Hulst, Boonekamp, van den Homberg, 2014) gives players insight in complex situations via engaging gameplay and reflection. With different types of games that operate and simulate different types of military actions, selected skills and learning goals can be achieved.

Although gaming is a relevant training tool, guidelines are needed that help the game developer better align the game mechanics with the intended audience and the learning objective. In this paper it is examined what game concepts exist and how they can be used to improve this alignment.

4.0 Research questions

For the above mentioned problems, in this paper six problem areas are identified that potentially can be remedied with games as training tools. For each of these problem areas, research questions are formulated that help explore how games can help make training and learning more efficient.

4.1 Homogeny

The skill level of an individual can be different from the rest of the group. This problem can lead to dissonance within the group and learning goals not being met. Games would possibly be a solution if they were made with personalization in mind. The analytic capabilities of computer programs and games might give insight in what aspects of the training curriculum are needed and which are already met with sufficient skill.

1. ‘What learning goals can be identified for game-based learning?’

2. ‘How can learning goals be personalized for a specific user?’

4.2 User profiling

The military rank can have impact on the way a learning method has to be presented. With higher ranking officers the notion of gaming might be interpreted as childish and the outcome could undermine their authority. The competences that need to be trained are varying as well and might need different types of games to achieve the correct focus. There are three main user factors that will be explored within this paper:

1. ‘Which game mechanics influence the design?’ 2. ‘How can they be applied for a certain user group?’ 3. ‘What general game concepts for user groups can be

4.3 User bias

Games come in all shapes and sizes but game designers can benefit from having clear guidelines that lead them to the right game for the learning goal. This means that games might differ in fidelity of the graphics to the physical setting they are played in. Some learning goals might be easier to understand when they are very concrete. Others might fare better when taught in a more abstract setting.

1. ‘What are the biases or reservations of these user groups?’

2. ‘How can we design our games to avoid these biases and reservations?’

4.4 Effects

The effects of actions are hard to visualize when going through a scenario with a physical exercise. Not all the factors can be included or represented with real life actors or are apparent when not in the actual context. Games might have the potential to show these effects in a more comprehensible manner.

1. ‘How can game mechanics be used to give insight in the effects of actions?’

4.5 Learning methods for games

There are many perspectives and theories about learning. Games can use one or more of these theories to teach. Knowing what style of teaching can be used in games and how to implement them, is there for an interesting question.

1. ‘What learning methods can be applied in the game design?’

4.6 Fidelity

Creating games can be a costly and time consuming process. Guidelines might make this process more efficient by understanding what game elements need the most attention. A connection between learning goals, users and mechanics might favor an abstract world representation over a realistic 3d environment.

1. ‘What level of fidelity is suitable for the learning goal?’

5.0 Method

To work toward the final design guidelines and concepts, in this paper a number of methods that help with solving the mentioned problems will be explored. With a state of the art analysis currently promising developments are explored. Since the use of gaming in the military is already initialized their setup can be analyzed to find out what type of learning is used and which learnings goals are addressed.

Interviews will be conducted with several experts in the field of training, within the military, from TNO and the Dutch military. With these interviews, it is attempted to find where the potential for improvement is and which issues the experts encountered with traditional types of training and gaming as a training tool.

Finally, a number of games from different genres, types and developers will be analyzed to see what games currently have to offer. In this part, of the paper both commercial as well as military commissioned games will be explored. With this knowledge both a description will be formulated as well as recommendations towards developing training games.

The combination of these three methods will help formulate guidelines and design concepts for different levels within the Dutch army. The first concept will be a training game that will train soldiers on the lower levels where the training is more focused on situational awareness on a small scale. The second concept will be focusing on the command level. Here the goal is to plan a course of action for a large amount of units. The third concept is focused on the highest form of command, where the goal is to plan an optimal course of action for the entire conflict (Doctrine publication 3.2, Royal Dutch army, 2014).

6.0 Game design triangulation

In this paper a link between three factors that are dependent on each other when designing games is proposed. By defining the three factors, in this paper designs for the different levels of military personnel are proposed.

Fig 1.0 Game Design Triangulation

As illustrated in fig 1.0, there are connections between the three elements. The arrows in this figure illustrate the dependencies of the elements:

The User element is defined, in this paper, by military ranking and has learning goals associated with this ranking. Some learning goals are more relevant for one group than the other.

The connection between user and mechanic is based on the different mechanics that work together to create a game that resembles their real life actions. The mechanics of the game can also be influenced by the users’ biases.

Learning goals can have implications for game mechanics to train different skills. Some skills might have to be trained in different settings or possible actions where others might need different visual representations.

For a game to work, the user, the mechanics and learning goals all have to be defined and work together to create a game concept. A game designer has no influence on the user or the learning goals and has to design a game using appropriate mechanics to ensure the goals and expectations are met according to the proposed

triangulation.

7.0 State of the Art

Gaming has been used by the military for some years now in varying context with varying success. The military has used commercial games such as Combat mission: Shock Force (battlefront.com, 2007) and ARMA (Bohemia Interactive, 2007) to simulate situations that might be encountered in the field. In these games they gave the students the freedom to approach these situations to test their training, with minimal risk and cost. In an issue of “Armor” (Armor, 2012), an American journal about the armor divisions of the American army, there is an article about how the instructors use “Combat mission: Shock Force” to train their

students via gaming. There are a number of games that focus on training kinetic warfare and skills on a personal level, but games that focus on the comprehensive approach and thinking in effects are few and often not specifically focused on military training. Currently, the military uses war gaming as a training tool for high ranking officers and their staff. These games are usually table-top games with small figures to represent the troops and an instructor that leads the game to explain the outcome of the actions taken. These games are played with a large group of people and are focused on the commanding officer that makes the decisions. The staff is there for advice and might not be engaged in the situation when they are not needed. These games are very time consuming mainly because there is a large group of people needed to run the game. These games also have a more traditional “Dungeon Master” type of setup, where the outcome of the actions are fabricated and explained by a field expert that leads the game session. Because of this “analog” way of simulating the outcome of the actions it is time consuming when actions need to be reverted and only a few courses of action can be tested within the time span, since it takes time to reset the game to a previous state (McHugh, F. J., & Fischer, H. F., 1966).

An example of how commercial games address homogeny is the “director” in the “Left 4 dead” (Turtle Rock Studios, 2008) series. Here, the director is a part of the program which monitors the player’s actions and if the player is having trouble with a part of the level or certain type of enemies, the director can adjust the flow of enemies by giving more enemies or specific enemy types. By using this system, the game is on the right difficulty level for every player. This type of game design usually falls in the category procedural generation, which means that the game generates the content during the gaming session. This could make the learning curve for students fit their personal needs as well as give them the content they need to advance to the desired level.

It becomes apparent that games are already in use in military training and games in the commercial market have systems that can help shape training games. However, clear guidelines on how these games should be created seem to be lacking.

8.0 Expert interviews results

From the interviews conducted with experts in the field of military training, there are some clear trends when it comes to learning through games. Using these interviews, in this paper it is tried to define the needs within games for the military and what the user profiles are.

8.1 Homogeny

Mentioned by most experts was the interest in a game that could train relevant skills amongst groups of commanders and other military personnel, to form more homogenized groups. In this case, games could be targeting specific skills that the student might need some extra attention and training in. This targeting of specific learning goals, can help make the process more time efficient by only training those skills the student needs, whilst others in the group might train other skills at the same time. The time aspect is very important, since there are many skills to be trained and very little time to do it. Having games as a more efficient way of training may relief some of the pressure on the students and give them a more relevant training, without overloading them with data. There is however concern about the individualistic aspect of such training methods since there is value in training with different people and the collaboration within teams. For games, this could entail that there should be a multiplayer component which makes

the individual task training more challenging. “Games can have benefit when we need to get everyone on the same level before the collective training begins.” – Rudi Gouweleeuw, Project manager, scientist Military Operations, TNO

8.2 Effects

The idea of using games to simulate situations and events is for most experts an interesting way to teach and show the planning process. By using games to show the results of a plan, the student can get a quick and comprehensible analysis of what the effects are of his/her decisions. A game that focuses on making these effects visible can help with teaching strategic decision making. Also the idea of being able to generate random events to train adaptively in planning has potential. The idea that reading a book is enough to understand how to plan a strategic action is questioned by one interviewed expert, he thinks games have the potential to make learning more practical. Here a focus on planning, executing and reflection is a plus that games can have over the instructions on paper. The focus on culture and religion within the comprehensive approach and their influence on the outcome could be well suited for games since the effects of such variables could be made more visible than the pages of a book are able to.

8.3 User profile

Gaming might also be interesting to give feedback on the plans that are devised by officers in high ranking, which might not have someone that watches over them. They rarely get feedback on their actions because of their ranking. These officers are the best in their field and thus have little chance to get a review from a superior. Gaming can be used to make their plan and its outcome more visible so they can do their own review by analyzing the results of the game. This group, however, can have reservations when it comes to gaming and might find games childish. To overcome this sentiment, the games should have clearly defined learning goals and give upfront information about the learning goals that are trained within them.

The higher officers might also be more reluctant to the idea of games judging them and showing their mistakes to their peers. The culture of the higher officers is based on their credibility and thus making mistakes is seen as a flaw. With games that keep this in mind these officers might still be able to learn from their mistakes when the results are made anonymous and peer reviewed. In this way, there is no fear of failure and the officers can try tactics to learn from the results. Having these officers train separately from each other might be needed to keep the anonymity intact. “Nobody wants to stick their neck out. A failure will stay with you within this

organization” -Major T. de Boer, staff officer. Another concern is

the fear of technology noted by Major T. de Boer, staff officer. The technology that comes with the games might be too advanced for some of the older officers. This is a concern that multiple experts had and can be a real problem when we use digital devices to play games. For these players, a more familiar setting such as board games might have a great benefit.

A plus, games might have within the training curriculum, is that games can evoke an intrinsic motivation to play and thus learn. If these games are made so that students will want to play them there is a good chance the games will be played outside of the allocated learning time. “If there is free time, students might want to train because of intrinsic motivation. Here the student can use his/her time more efficiently.” – Major T. de Boer, staff officer.

From the expert interviews it becomes clear that there is potential in gaming as a learning tool and that the proposed research

questions are indeed interesting. Guidelines that aid the game developer should focus on these aspects.

9.0 Field research

With field research, it will be examined which games have relevant mechanics that aid learning goals. Here, interesting game mechanics will be explored to understand their importance for training games. With understanding of these mechanics, we should be able to formulate clear guidelines for future game development. The games used within this research are: Combat mission: Shock Force (battlefront.com, 2007), VBS2 (Bohemia Interactive Simulations, 2007), Civilizations V (Firaxis Games 2010), Hearts of Iron III (Paradox Development Studio, 2009), Europa universalis IV (Paradox Development Studio, 2013), Frozen synapse (Mode 7, 2011) and Democracy 3 (Positech Games, 2013).

9.1 Game mechanics

Game mechanics can be described as the rule set to which the player has to abide by whilst playing the game (Hunicke, R., LeBlanc, M., & Zubek, R., 2004). These rules can make actions possible or impossible and can force a player into a specific style of play. Using game mechanics to steer a player, game designers can train specific learning goals that are needed in specific military functions.

Below, the game mechanics are defined in topics that concern aspects of games and how they are played. Although there are many ways mechanics can be implemented in games, the list below contains of the most used in games, and deemed important by the research within this paper, when making a game for learning. These mechanics are also present in every game in the research.

9.2 Actions

The player has the possibility to take actions within the game that change the state of the game (Bjork, S., & Holopainen, J., 2005). Sometimes, limited with few possible actions and other times, the player has an extensive array to choose from. Actions can also be limited by prerequisites, such as resources or game states. The actions can also vary in scale, the scale of an action, here indicating on what level of command within the military the action takes place. Small scale actions might be first person shooter style actions, where the player only controls a single soldier. Furthermore there is a midlevel scale, where the player controls a platoon that can be placed individually and lastly a grand scale, where the player is more involved in diplomatic actions than the precise positioning of individual troops. Games that fall into the grand scale category are Civilizations V (Firaxis Games 2010), Hearts of Iron III (Paradox Development Studio, 2009) and Europa universalis IV (Paradox Development Studio, 2013). The scale of the actions is important when game designers are designing games for specific learning goals. Learning goals that are aimed at the soldiers on the battlefield are more direct and focused on learning the tactics, movements and command of the squad or platoon they are affiliated with. They will also have learning goals that concern the layout of the area they are going to be operating in and how certain types of cover and terrain advantages or disadvantages might impact their performance. The actions taken by such troops usually are on an individual or squad level and thus can be represented in a game with a less complex set of actions that have a more direct effect on the situation.

When designing for higher level commanders, the learning goals are much more on a managing level, where an action has impact on

many actors and situations. The scale on which these actions are taking place is much larger, and makes actions of smaller scale less important to be managed by the officer, since the officers below him will be responsible for the proper execution of these actions. For games, this means that we can make the actions more abstract and have the effects of smaller scale actions be simulated in the background.

9.3 Information

Information can be used to give the player insight in the conflict or leave room for the player to investigate and explore. Using the information as a mechanic within a game, can direct a player to assessing a situation or making them cope with limited or incorrect information. It also lets them experience how to make a decision when information isn’t perfect and/or complete (Bjork, S., & Holopainen, J., 2005).

Games employ a variety of ways to let the player discover the information they need to complete their goals. Some games employ a “briefing” method in which the player is given information about the state of the game and/or the events leading up to this state. This briefing style can be used to let the player know what the situation is and what is expected from the player to address it. This information is often incomplete and can be imperfect, this gives players a reason to explore and experiment to gather more information. This can help form learning goals that focus on real life situations, where the information in the briefing is not correct or out dated.

During the game, there is usually a lot of emphasis on gathering information about the situation. For strategic games, the information about other players, the world and its events can be seen as a resource without which a player can have a hard time planning their course of action. Systems such as “Fog of war”, which deliberately obscure sight on the battlefield, can keep information from the player and can be a great way to let the player explore the “fogged” areas of the map. With acting on information that might not be perfect, the player will have to make a weighted decision about the actions he/she needs to take to come to the end goal set by them, the instructor or the game (Bjork, S., & Holopainen, J., 2005).

Since a great deal of decision making is based on the information that is available, teaching officers about the reliability of information and the reality of the information not always being accurate is needed when training complex decision making. In some situations, there is not enough information to predict all the variables it might affect. When these cases arise, the decision maker should be able to weight the reliability and completeness of the information that is available. (Elaboration of this topic in the appendix)

9.4 Narrative

The narrative of a game or situation can give the player a sense of presence within the game. By using real life events or locations, the player can be prepared for the real thing, without the need to physically go to the desired location. This can help with learning goals when they are specific to a location or situation that might vary from the local situation of the player (Bjork, S., & Holopainen, J., 2005).

Within the narrative, the player can expect and anticipate effects of their actions. It gives a sense of presence in the scenario and makes it possible to plan actions. The narrative will also govern the

“non-player characters” actions, making sure that the actions are realistic according to the scenario.

9.5 Fidelity

The fidelity of the game can be affected by the learning goals. When the learning goals are about equipment the fidelity needs to be high and represent the real equipment with great accuracy. When the learning goal is about cultural understanding, the accuracy of the equipment carried by soldiers is less important and can even distract players from the intended learning goal (Alexander, A. L., Brunyé, T., Sidman, J., & Weil, S. A., 2005). When designing for tactical decision making on a large scale such as global economic situations or a national conflict between two or more cultural groups, it is important that the models supporting the behavior simulation are extremely accurate whereas the individual people don’t need visual representation. The addition of detail, to elements that are not in direct connection with the learning goal, can lead to distraction of the player. When the decision is made on a high level, the concern is that it happens and not how it happens, that is the responsibility of the person that coordinates such actions. The investment of time and money is also a factor that needs to be considered. The creation of detailed 3d models of objects and environments is a costly and time consuming process, where it might have little to no impact on the learning goals if this type of detail has no influence on the actions.

However, if the learning goal is focused on teaching about the equipment or about the environment where the player is going to operate, the fidelity of the equipment or landscape is very important. Games that focus on tactical combat or on individual soldier actions, the fidelity of the battlefield and other information have to be representative of the real environment it is simulating. Games such as VBS1 (Bohemia Interactive Australia, 2002), ARMA and Combat mission: Shock force (battlefront.com, 2007); need the fidelity in the equipment and environment, since they are teaching the player how to use their environment and equipment to the best possible outcome. Games such as Europa Universalis IV (Paradox Development Studio, 2013), Democracy 3 (Positech Games, 2013) and Hearts of iron III (Paradox Development Studio, 2009); are on a higher level, where the decisions are more concerned with political relations and cultural differences, than the landscape the people are occupying. These games benefit from a more abstract representation of the world to make the connections between the relevant factors more visible (Gopher, D., Weil, M., Bareket, T., & Caspi, S., 1988).

9.6 Feedback

When we use games, the great benefit of using a computer is to have it give direct feedback when actions are made. Unlike traditional teaching methods in which the problems are reviewed over the course of days or weeks, the computer can give almost instantaneous feedback to an action. Having this direct feedback, will allow players to see the impact of their actions and if the game allows it let them retry the situation.

Feedback is very important to games because it is the way of rewarding or punishing a player and gets them to change or reinforce their behavior (Bjork, S., & Holopainen, J., 2005). When the learning goal of a game is working together the feedback from the game should be more rewarding when the players perform actions together. This feedback should inspire them to work on a common goal and can show them which actions are more beneficial for the group.

The use of games to train means that there is also opportunity for the trainers to assess the progress of the players. This can help with more specific instructions when players are struggling with a problem and can red flag certain actions or results. When these occur, the trainer can instantaneously help the player (Sennersten, C, 2010).

9.7 Big game

The big game can be described as what happens outside of the games’ virtual environment. As in the example given earlier the US military uses Combat mission: shock force (battlefront.com, 2007) to train their soldiers and teach them about cover and terrain. They let the students play the game with two students in front of one computer and after twenty minutes, the instructor pauses the gaming session to analyze the actions the students have taken within the game. This “two-student-team-setting” and the analysis outside of the game are what game design describes as the big game.

The big game is essential when game designers want to use reinforcement learning via games. Reinforcement learning gives the student the freedom to choose the actions and analyzes the results, thus reinforcing the effects of the action. Games have great opportunity to give players the freedom to take an action and see the results. When accompanied by an instructor or data output, the results can be analyzed. As Sennersten, C, (2010) concludes “More generally debriefing reveals different aspects of intended and actual learning processes and has an important function of adjusting the cognitive operational models of trainees to consolidate a collective set of models based upon the joint experience of the simulations.”(p. 112, Blekinge Institute of Technology, Doctoral Dissertation Series No. 2010:05).

Another form of using the physical setting to persuade players into a behavior or towards a learning goal, is to have the players facing each other. Andersen, Kristensen, Nielsen, Grønbæk, (2004) state that their board-game version has “maintained more social interaction than computer games” (p. 8). When a group, that has similar and dissimilar interests, have to work together towards a mutual goal, the physical setting helps with inducing collaboration. When players play online games, in most of them, the players use an avatar and/or alias to represent themselves, and this makes players virtually anonymous. A physical setting, in which players are in the same room, can help with trusting on the other players. The fact that they are confronted with each other without an avatar to represent them makes it easier to connect to a player and will make the step towards collaboration smaller. In these games, the feedback that is shown can benefit greatly if it concerns the common collective, dependent of all the players situations. There is, however, a good use for anonymity when players are afraid of rejection or failure. Mentioned by several interviewed experts, some players don’t want their mistakes to be seen since this can undermine their authority. For these players, a more secluded setting might work better. The act of failing is in itself not a problem and can be seen as a great opportunity to learn, but when the mistakes are highlighted in public and can endanger ones career or standing within a group, players might be reluctant to try new tactics or expand their horizons. When we want these players to learn from their mistakes, it is vital that we can review their mistakes and learn from them all without exposing the players’ identity. This can be achieved by having players play against or with each other, without a real name or identifiable avatar. Reviewing the mistakes or innovative solutions made during the

play sessions can then be viewed without shaming or compromising a player.

10.0 Learning methods for games

There are many ways to teach a subject and some subjects have better results when thought one way instead of the other. Games also have the potential to give learning experiences in different fashions. Games can have the benefit of instant reward to acknowledge the players actions instead of textbooks and traditional assignments that give feedback after a significantly longer time period. It can also reinforce its lessons, make the subject matter tangential and might even teach the player without them ever feeling like it was a chore that had to be forced on them. Reinforcement learning is most applicable to games for its focus on learning from the results of actions. It is best applied when the actions are reviewed with the help of an expert or explanation on why the action resulted in this way. The reinforcement learning method also encourages the student to take innovative actions and try new strategies, without punishing the student for mistakes. Instead of rewarding a bad outcome with a low grade or assessment, the focus is more on learning from these mistakes via reflection.

“Reinforcement learning is the learning of a mapping from situations to actions so as to maximize a scalar reward or reinforcement signal” - R.S. Sutton, (1992).

Tangential learning (Floyd and Portnow, 2008) is also a possible way that students get information. The idea of learning by being introduced to situations, subjects and scenes that have interesting material woven into their design without forcing a student can make learning more intrinsic and satisfying for the student. This can be done by referencing design to the actual subject matter or place. Placing the player in a historic or present event lets them experience the narrative of that event. With intrinsic interest in the subject matter, characters and game, the students are more likely to absorb the information presented and might even try to expand their knowledge about the presented setting. Games can reward explorative players or players that have an excellent knowledge about the narrative they are in and thus give the player the extrinsic motivation to learn about new subjects. Such games might even get the student to be more open to new fields information.

11.0 Learning goals

Within the military, the thinking in effects and learn as you work are seen as important to understand situations and learning them by being involved in them (Smith, 2002). Games can have the player experience a situation, as well as giving feedback on the actions they might take. It is also important that when a decision is made or a course of action is formed, that all the factors that are effected are held into account. Actions that have an expected effect might also bare consequences that are less obvious. Making a decision that is aimed at making a situation more stable when the effect on the local community is the opposite, can be less obvious than eliminating the immediate thread. To train this type of decision- making and making these effects visible is a difficult task where games might help (Kim et al, 2009).

By spreading the guidelines over multiple levels of the military hierarchy, a game designer can focus on the learning goals that are relevant for their respective levels. In all the games it is important that the player can experience the situation and be able to see the effects of his/her actions. On the lower levels the effects might be more obvious because their actions affect few variables, but on

higher levels the number of affected variables to consider grows (Doctrine publication 3.2, Royal Dutch army, 2014).

11.2 The user

What kind of user one designs for is very important because it defines what learning goals and user characteristics the designer has to design for. Within this research, the focus is on military personnel of a multitude of rankings that have to make decisions, plan and execute actions. These user groups have their own specific learning goals, needs and reservations when it comes to games and/or the feedback they produce. Within the Dutch army there are six levels that distinguish their ranking and the operations that they are responsible for. For the purpose of making three game concepts the levels will be defied in to three groups of two rankings per group. These characterizations are devised through the doctrines of the levels and the actions they are able to take.

11.2.1 Level 1-2 “individual – group”

This is the lowest ranking group; it is composed of the lower ranking military forces. This group has little to no commanding authority and is the working force that will execute the orders from the higher ranking officers. This user group is usually the youngest and has the highest chance of having gamed in the past or still actively games. This makes the concept of using a game as training a lot more accessible than if students have no experience with games.

On this level, strategic planning and situational awareness are about small scale actions, performed by a small group of military personnel. An example is making a course of action to inveterate a building, not knowing the amount of hostiles and their positions.

11.2.2 Level 3-4 “platoon – company”

This level can be defined as a command level in which orders from higher ranking officers are orchestrated to be executed by the levels 1 and 2. This level is deploying the troops and equipment and is responsible for the positioning of these assets. For this user group it is very important to have an overview of the battlefield and get information from both the lower and higher ranking personnel and translate between the two.

11.2.3 Level 5-6 “battalion- brigade”

On this level the high level strategic planning is done. Here the situations are assets and courses of action are formed to get the most beneficial outcome. On this level the best of the best are operating and have authority and responsibility for a fast number of assets and interests. To get to this high ranking level most officers spend a number of years training and are thus in most cases older than the personnel in ranking one through four. This means that when it comes to games, a designer might have to consider a different approach to present the game and can’t rely on gamer mannerisms that younger and more experienced gamers have been exposed to. The word game itself might give negative associations and can make the program deemed childish. To remedy such events there are cases where the word game is transformed into war-gaming or might even be replace by words such as simulation.

12.0 Design concepts

When a designer designs to train certain skills, the designer wants to challenge the player with tasks that make use of these skills. This means that, when coping with the resources that are given or dealing the pressure of time, the designer must make sure that these challenges are woven into the game either obviously or more obscured. In the list below, three types of game concepts for the three different levels of command, keeping in mind their group

learning goals, command level and user profiles, will be proposed. In the game industry, there are a number of types of games that have specific styles of play, control and perspectives. Within these genres there are a couple that are especially interesting when looking at games to train specific levels of the military (Rapeepisarn, Wong, Fung, Khine, 2008).

12.1 Levels 1 - 2

The first interesting game type for the levels 1 -2 is the first person shooters. This game type has the player experience the game through the eyes of a character. They can see and hear what the character can see and hear and the actions the player can take are limited to the abilities of the character. In these games the environment, equipment, objects and other characters are in 3d to represent the real world as realistic as possible. The learning goals of these games should focus on exploring the environment, using objects in this environment and tactics within small teams. These games are suited for squad style training in which each player can have a role within the squad. By playing these roles, the player should be able to learn about the responsibilities of the role. In this setting, the fidelity of the world representation should be high because most decisions are based on visual cues.

12.2 Levels 3 - 4

For the levels 3 – 4, strategy games are more interesting. These games are usually referred to as real time strategy (RTS), but turn based games are also a viable solution. The difference between the two types is the flow of time within the game. The real time variant has every action play out immediately and other player can do actions simultaneously, where the turn based gives every player a turn in which they can perform a number of actions. Both these types of game have the gods view perspective in which the payer is viewing the battlefield from above and can select units and give them orders. In this type of game there is a focus on positioning of the troops as well as managing resources. The learning goals for these military levels are more focused on commanding and this is represented in this type of game.

12.3 Levels 5 - 6

The levels 5 – 6 are closely portrayed by the grand strategy genre. This genre focuses on the big picture of war where the actions the player makes affect many variables such as hearts and minds of the local population or the standing with allies and enemies. These games give the player an abstract view of the world in which statistics can represent the different variables of the conflict. With these games planning and predicting outcomes is the main focus. All the actions influence a number of variables and the actions almost never have only negative or only positive results. The decision to supply the local population with fertilizer might help the terrorist make bombs. These games have a great emphasis on thinking in effects and can give insight on the outcome of complex actions. Results in these games should be based on “lessons learned” in real life situations.

13.0 General Guidelines

In this section of this paper, three guidelines will be formulated for three different types of games all catering a different set of levels of the military hierarchy. The games are mostly distinguishable by the scale of the actions taken in them as well as their respective learning goals.

13.1 Thinking in effects

Thinking in effects is the most obvious learning goal for games since they are elaborate state machines that will react to the input

of the player and give a computed output (feedback). The effects can be displayed before an action is taken, so that the player can see what kind of effect is expected from the action, or when the action is executed. The actions’ effect can either reinforce the expected outcome or show the difference between expectations and reality. The effects can also turn up later in the game to make the player look back on a longer period of time to show the outcome with more actions taken. This type of learning is based on reinforcement learning (as mentioned in 10.0 where the player is shown the effects of their actions and will review the plans that were made). Here, a replay function can be very useful to let the player test an action and try again when the action doesn’t turn out the way they anticipated. Recordings of game sessions can also be discussed within a group of students under supervision of an expert and can show the effects of all the actions with relative speed and ease. A thing to consider, when reviewing players sessions, is that some players are afraid to lose and fail or have reservations about being displayed when they make mistakes. For this group, an anonymous replay review might be a safer option to not offend them and the anonymity won’t have much impact on the value of the replay material.

13.2 Information

The skills that come with dealing with the flow of information and the inaccuracy, or lack of this information, can be simulated in games with an arrangement of mechanics. Most games have information as a mechanic and in the realm of strategy games it is vital for planning a course of action. Games often are able to give the player enough information about the environment, resources and enemies, but leave room where the player might have interpreted the signals wrongly or was purposely led to a wrong conclusion. This will force a player to rethink a strategy when the information turns out to be false or different from what was expected. Information can be seen as a vital resource and can give players an edge on their opponent. In real life getting the correct information and dealing with inaccurate and/or incomplete information is of vital importance.

13. 3 Homogeny

Amongst all levels, the members have different skills and can have different skill levels. Games can adapt their content to fit the student’s skill level. As shown in the example in the state of the art section (7.0), commercial games have systems to make changes adaptively as the player gets more skilled. With learning goals, the game should also analyze the player’s skill level through behavior and results. For example, a scenario’s difficulty could be increased if the player fulfils objectives in less time than estimated, or decreased when a player is failing an objective several times. Games can also adapt the content to match the specific skills a player might lack by adding more objectives related to that skill. For all levels of the military this “adaptive content” setup can be used to make the training more streamlined. By focusing on the skills the student needs to improve, without wasting time on already mastered skills, we can make a group more homogenous.

14.0 Guidelines for specific levels

14.1 Level 1-2

First person shooter, High fidelity, Virtual reality

The first guidelines are for a game on individual and group level. This level has to follow orders very closely and has little influence on how the troops are deployed, setup and moved besides him/her –self or a small squad. On this level, the learning goals are more

focused on small scale situational awareness. There are multiple functions on this level, such as medics and other types of personnel, which might need more specific games to teach those function-specific needs. The general game should focus on tactics that require planning in small scale situations, such as breaching buildings or providing cover fire for the group.

Fidelity:

For this type of game the fidelity can be very high. Virtual 3d worlds will help the player understand the environment that the mission is in and show the details of the building or area they need to secure. The goal is to teach about situational awareness and tactical decision making on a small scale. Virtual reality and other immersive tools can be very effective. Having virtual reality glasses can help with the exploration of the environment and can make the experience more lifelike.

Narrative:

The narrative for these games can be quite simple and don’t give all the information about the conflict. As in real life the player doesn’t get all the information about the overarching situation just the information that is deemed necessary for him/her to complete the task at hand. These games can have short mission based sessions where the goal is known from a briefing. Not all information has to be available about enemy position but this can depend on the type of mission the instructors want the player to experience.

Information:

The flow of information in these games mostly comes from the narrative in the form of a pre-mission briefing but can also be given more information during the paying session via in-game radio simulation. The player can also gather more information by actively searching for points of interest in his/her environment. In this way, the player can find the best position to fire from or find weak points in the building to breach through. The players can also relay information to each other in multiplayer games. This will help the player learn about the value of information and can help them relaying information in the correct way so that other players have the best possible understanding of the situation.

Actions:

Actions on level 1-2 have very direct consequences and can be calculated with physics. An accurate model of the world and the physics that tie it together is necessary to give the player an accurate experience. The rule set for such a game should be easy to understand and can be very strict, since the role has very direct impact on the scenario.

Big game:

The big game setting can be very variable. Single player games are an option that could work well for these games. Let players try and get to the objective on their own to simulate a situation where they are split up from their group. More elaborate missions will benefit from having multiplayer setups. These setups can be in one room with multiple computers or via an internet connection.

Feedback:

The feedback and assessment of level 1-2 games, is very direct and with little reflection the outcome should be comprehensible. Having a replay function, that shows how they got to this outcome or how other players failed to get to the objective, will reinforce the experience.

14.2 Level 3-4

Real time strategy, Medium fidelity

This game is focused on the commander level. The responsibility of this level is to orchestrate the attacks, do unit (character or vehicle) placement and formation as well as deciding how to respond to actions from other parties. In this type of game the player is directing the battlefield, not necessarily moving only one unit at a time. The player moves big groups of units with a set of commands. A gods’ eye view of the battlefield can help the player get a sense of the size of the environment and learn about the advantages and disadvantages of the terrain.

Fidelity:

The fidelity can have some abstraction on the lower levels. The learning goals have less to do with the inner workings of the equipment and more about how, where and when to deploy them. A 3d environment isn’t absolutely necessary as an abstracted view of the battlefield would also suffice. There is however benefit in a 3d battlefield, since it can give the player a bit more insight on the line-of-sight that specific units might have from their position.

Narrative:

The narrative can, as in the first game, be mission based with one or more objectives. Here, the player will get a starting amount of troops and equipment and has to lead them through the mission. The player is effectively only sending commands to his/her units. The modelling of information flow can give the player insight in the reality that not all information can be relayed perfectly. Since the player controls other personnel, not all his/her actions will be followed and moral simulation can be implemented to make the player aware of this factor.

Information:

Information has high priority in this type of game. Knowing where the enemy is what type of units they have and how they are positioned, are all factors that weigh in on the tactics necessary to get to the objective. In these games the player usually has a gods’ eye view of the battle field and can scroll through the map. In these games “fog of war” is very important to give an accurate representation of the real life situation. Although in real life we might know what the landscape of the battlefield looks like the positions and configuration of the enemy are mostly unknown. To simulate these conditions the enemy soldiers will not be visible when they are out of sight of the players troops. Also, the way the information is relayed amongst the troops can be simulated, by having commander units that have to be near units to keep a clear line of communication. The incompleteness of the information that the player starts with is an accurate representation but the player has the means to get more information about his enemy through scouting and assessing the situations he/she encounters. The briefing might also not be completely accurate and thus the player might find that they have to adapt their strategy when they encounter a situation not defined in the briefing or was giving wrong information about the situation.

Actions:

Actions are on a midlevel. The player is influencing many units and is trying to keep control over the battlefield. The actions might take a while to come to complete and can have different effects in combination with terrain, moral and enemy movement. The rule set that these actions will work with, can have random elements to simulate the individual soldiers. Not all battles will play out exactly the same way and as in real situations the outcome can only be estimated and not predicted with complete certainty.

Big game:

The big game setting for this type of game could be that the player plays alone against an artificial intelligence (AI) opponent or via a

network with other players. The setup could also be mixed up by having two players using one computer and collectively make decisions. In this setting there could be a classroom of players playing the game and evaluate the results amongst the group.

Feedback:

The feedback in this game setting is not as direct as the level 1-2 game’s feedback. Here it is essential that the results are put in perspective to give insight on what has happened and why the actions produced such results. A group evaluation with a trainer or session leader is essential to come to the insights that were designed for in the scenario.

14.3 Level 5- 6

Grand strategy, abstraction, flexible rules

The levels 5 and 6 within the Dutch army are focusing on the strategic decisions that are made when in conflict or being involved in a conflict. These games should give insight on how a decision will affect many aspects of the situation. On this level an understanding of the other perspectives involved in the situation and understanding of how the other parties might react to certain actions. The officer in one of these levels will have to coordinate a large scale battle plan keeping social and political factors in mind.

Fidelity:

The fidelity of a game on this level has to be in the model to simulate the results of taken actions. There is little to no benefit of having a fully 3d modeled world as in the actual situation these officers would not be on the battlefield but have charts, maps and screens to provide them with relevant information. This means that the representation of the battlefield and situations can be very abstract which in turn will make the results more visible and organized.

Narrative:

For this level, the narrative is more important than the previous two. In this game the narrative heavily influences what actions a player can take and how an action influences the game state. The narrative also needs to be of a bigger scale than the previous two to simulate the situation on the level that their influence is used. In this game type the narrative will have to set the scene for countries and/or interest groups within them. Since the goal is to give insight in complex problems, with multiple actors that might have to work together or against each other to achieve their goals. The narrative should set the scene for these interactions and ideally adapt to the taken actions.

Information:

Here information is everything. Without information, the player can’t make the best decision and the focus of these games might very well be on learning to gather and interpret information. All information helps to get situational awareness which for these games is one of the main learning goals. Knowing what is going on in the conflict area and what the thoughts about the situation, from as many perspectives as possible, will give the player the foundation to make a decision. Gathering information in this type of game is done via exploration of the narrative. The player would have to invest resources to get information out of the narrative which can be represented by a computer in diagrams and/or a small piece of narrative given by the instructor. This information can give the player more insight into the situation but won’t provide him/her with an answer that will have only positive or negative effects. The focus of this information is more about what will be affected by which actions and what are the effects of actions already taken.

Actions:

Actions on this level can be very complex and don’t have a completely right or wrong outcome. In this setting an action can influence many variables within the narrative and can be beneficial to multiple parties within the game. Actions that can be executed collectively, or sponsored by multiple parties, are a way of giving the players a reason to think about negotiation and cooperation. The rules are also less strict than in the first two game concepts. The set of rules in this type of game have to be flexible and might have to bend a little to get the scenario in the desired state. It is possible that the scenario is being steered in a direction where the narrative has no answer to give as a result or misses the intended learning goal. In these situations, an instructor that leads the game session is a must. This session leader should have knowledge about the narrative and lead the events towards the intended situations that are designed for the learning goals.

Big game:

Big game for this type of game is essential and benefits greatly from having a physical setting in which the players can see each other and have a face to face interaction to stimulate communication. This makes a board game type setting ideal. The results can still be displayed digitally and the model can also be calculated via a computer simulation but the interaction is more intense and more realistic when the players have direct contact. A multiplayer setting is in most cases the best option since the learning goal often is to negotiate and collaborate with other parties that might have different goals and perspectives. However, the users targeted for these game types are in high command positions and have subordinates that have to follow their instructions. Having them fail in a game or struggle with a situation, might be perceived as weakness and can undermine their authority. These players are more reluctant to try new and risky actions that might lead to great benefit or significant loss. To accommodate for such specific variables of the user the design must keep in mind that the player has these reservations and when reviewing results might have to make the player anonymous. Online play can make players more anonymous but might also impede the communication.

Feedback:

Feedback on these games is harder to make visible than on the previous two levels. The actions can affect many variables that are in the model and will have to be made visible in an abstract version of the real situation. All actions will have an outcome that is more nuanced than the black and white output from the first two. The actions will be done to get the most beneficial outcome but it is never an action that is only positive or negative. The instructor will have to explain the outcome of the actions alongside the charts and other abstract representation. By having an instructor explain why the action has led to this result, can have the added benefit that it can be presented in a way that the group is more accustomed to.

15.0 Conclusion

15.1 Homogeny

In the entertainment industry there are multiple examples where the game adapts to the skill level of the player and analyzes the behavior, strengths and weaknesses (7.0). For military training this is also relevant. Using games to monitor behavior and progress of the student and have the program adapt to their skill level can be very helpful for efficient training. Having a tool that targets the specific skill that the student might need some extra attention on can make the training more focused and thus less time consuming. This way only the relevant skills are addressed to make the group a homogenized whole.