Janus battle simulation download

The Department of Defense believed that the Janus program would be able to train these and future officers to think more clearly about the costs and benefits of battlefield strategies. One officer who let his position deteriorate reached into his arsenal to employ higher-yield weapons that destroyed not only the enemy troops, but also wiped out his own troops. In , the Lab transferred this early version of Janus to the Army, while it continued to refine it.

In the mids, the program was upgraded and made more flexible, incorporating higher-resolution graphics, distributed processing and real-time play with much larger forces and greater variety of combat systems. In , the Army used Janus as an operational planning tool, prior to the invasion of Panama Operation Just Cause to overthrow dictator Manuel Noriega. To control for learning effects, the two scenarios were alternated for each trial.

Out of the eight trials, four were performed with each communication architecture. To control for order effects, each day and team started with different a communication architecture than the previous day or team.

They were then given 20 minutes to work together to plan their mission. To assist with planning routes and tactics, they were provided with a map of Kamaria, a plastic overlay, and some transparency markers. Interactors were intioduced to participants and team members were physically isolated from each other. A radio check was then performed to ensure that all radios were working correctly. Each participant was positioned in front of a computer screen showing a map of Kamaria, with weapon units ready to fight.

Participants were given minutes before begirming game play for instructing their interactor on how to organise the movement routes of these weapon imits. During game play, the experimenters did three things: 1.

Watched the overall battle picture on the Janus screen and made observational notes; and 3. Listened to the radio transmissions and filled in communication check-sheets.

No limits were put on this discussion. After the final trial, they were given the post-trial questionnaire. Results It should be noted that because this is such a new area of research, the current study is largely exploratory. Rather than supplying definitive conclusions, it should be valuable in providing insights into future research areas and methods. As was mentioned, the design of the experiment was based on eight trials four trials imder each communication condition see Table 2.

However, due to time and resource constraints, an equal number of trials for each group was not possible. All three of the civilian groups participated in eight trials.

One military group was involved in experimental refinement, so their results have been excluded from final analysis.

Due to time constraints, the second military group participated in only four trials. Due to small sample size and unequal groups, the following is a descriptive analysis of results. It should reveal trends that could be pursued in further research, rather than finding significant effects. Table 2. The experimental design Military Civilian Open communication 2 trials 4 trials Restricted communication 2 trials 4 trials 4.

Table 3. Table 4 indicates the amount of simulation exposure participants have had before commencing the trials. This table shows that out of the participants who answered affirmatively to having previously played a computer game, the mean amount of time since they had played was approximately 2 years.

The majority However, the infrequency of their game or simulation usage indicates that their expertise in game play would be minimal. Table 4. This generated a performance score relative to that of the enemy. This also allowed enemy performance over time to be monitored. In line with the hypothesis that open communication architecture would facilitate situation awareness, open communication has produced higher performance scores.

Table 5. This supports the hypothesis that military personnel would outperform civilian subjects. Table 7. Neither graph seems to indicate a straightforward development of expertise. Performance seems to increase over time. However, there is no meaningful linear or quadratic function to illustrate this. Dips in performance are found at trials 2,4, and 7. During Days 1 and 3, participants performed better in their irutial trials each day.

In addition, the second day of participation produced higher performance scores than the first. The development of expertise over trials for military participants 2 04 1 2 Run Order Figure 6.

The performance for each group across the eight trials 4. Three peaks on Figure 7 indicate scenarios where the enemy had few losses. The enemy's performance over time 4. Table 8 shows that the most noticeable differences were a larger number of acknowledgmente and updates made by military personnel, while they made less requests for information. Another notewortihy result was that the provision of information was less than the requests for each group.

The military group seems to have the closest relationship between iiiformation request and information feedback. Table 8. All participants except one felt that their situation awareness increased after the first day.

This was true for both military and civilian participants. However, no significant correlations were found between participants' performance score and beliefs about their development of situation awareness. Perception of situation awareness for civilians 4. Hie responses to these questions ranged from disagree to strongly agree on the five-point scale. The majority of participants tended to agree or strongly agree with these statements.

Of the civilian participants, There also seemed to be a general trend for participants to agree to a higher extent with these statements across trials.

However, no significant correlatioris were found between participants' performance score and beliefs about their development of expertise, 4. No participants preferred restricted communication. This preference for open communication is consistent with the finding that open communication produces significantly higher performance scores.

From the post-trial questionnaire, This included explaining capabilities, acting on command, and clarifying commands. This would suggest that interactors would not have negatively influenced the scores. In addition, out of a mean number of moves of This represents an even number of commands issued across trials. The remaining One interactor was responsible for The same interactor was also responsible for A comment made by the participant regarding the interactor was: My interactor took over when he got bored.

I found this very frustrating and felt that I could have gotten more out of the exercise if he let me make my own mistakes at the start. It took a while to learn what was what. I still don't understand it all. I need to be able to make more mistakes. These percentages were OrJy 5 interactors responded incorrectly to commands.

Out of this group, one was responsible for Discussion This study used the computer simulation wargame Janus to investigate the effect of communication architecture, and previous experience on NDM. The development of expertise was also monitored across trials.

The current study has proven successful in identifying issues that need consideration when designing experiments on naturalistic decision-making. It also revealed trends that can be pursued in future research.

This supports the hypothesis that open communication facilitates situation awareness. This result is also consistent with Chapman , where Networked Fire Chief was the experimental tool.

This provides some validation of the use of Networked Fire Chief expenments to develop the NDM theory in relation to the military. For future research, it is recommended that this comparison be investigated with larger participant numbers.

Forming larger teams could also be beneficial, as it may be the case that with larger hierarchies, open communication may not be as effective, as members would have to filter more information. Because the difference was substantial, this would be an important comparison for future research. However, it is important to treat this data with some caution. The imbalanced design, with one military group compared to three civilian groups, as well as the unbalanced number of trials between civilian and military groups, could have produced this result.

It is also necessary for future research to more clearly define what constitutes expertise. The experts used in this study were all military officers. However, they indicated that they were only involved in combat training once per year.

In addition, their roles in the military were not directly related to weapon imit control. Taking the above defirution of expert into account, it should prove valuable to use military participants who are actively involved in the areas in the wargame scenario.

The civilian sample showed a broad range of variability in performance across their trials. However, isolating the first and last trials reveals an increase in performance. The variability of performance in the other trials results could be partially accounted for by fatigue. The participants commented that three trials in one day left them exhausted. One commented that "by the third scenario on each day I think there tended to be a big loss of interest and it was much more difficult to keep focused.

While this varied from group to group, the second or third scenarios being run each day was generally affected. For future research, it would be beneficial to try to reduce participant fatigue and maintain their interest in the task. This proves difficult when running experiments where large numbers of trials are needed to generate meaningful outcomes. The balance between ideal experimental design and participant willingness and availability has to be taken into consideration.

It is important for an experimental design such as this to have an intelligent enemy, who can react to moves made by the participants.

At the same time enemy performance must be held as constant as possible, so as not to confound participants' results across time. Overall, the current study achieved this result. This type of communication helps to facilitate teamwork and situation awareness. This could partially explain the overall difference in performance scores between military and civilian participants.

Unlike the civilians, military personnel also requested information and provided requested information at close to the same frequency. This is what you would expect if each request were given a response. A related issue was the high level of information requests from some of the civilian participants.

This further demonstrates their lack of expertise in the area. In particular, it is likely that they did not know which information to filter out, so were requesting more than was necessary for optimal performance.

However, it is anticipated that by refining teclmiques of extraction, more reliable self-perception data can be obtained. In addition, larger participant numbers would help to consolidate this issue. However, Group 2 reported higher situation awareness on day 2, compared with day 3.

This may be due to different interactors giving var5dng quality in support. The remaining two groups reported approximately the same levels of situation awareness across day two and three. Two members of the military sample agreed that their situation awareness increased over trials, while the third military participant believed that his situation awareness decreased across days. To clarify this issue, it needs to be tested with larger participant numbers.

Also, to obtain a more accurate measure, a different method could be used to measure participants' self-perception of situation awareness. The current study used a 5-point scale, ranging from strongly disagree to strongly agree. Most participants tended to agree or strongly agree. It may be more useful to measure the degree to which they felt their situation awareness had improved via a short interview after each trial.

However, the data on expertise development showed both increases and decreases in performance. It is possible that participants were trying to please the experimenter by answering in a manner expected. Future studies should refine statistical performance measures of expertise development to consolidate conclusions, 5. One of the civilian participants commented; "being aware of what the commander and other commander sub-unit leader were commurucating about helped my situational awareness".

Another participant stated that, "restricted communication introduced a significant delay in requesting artillery support. This made estimating where to order strikes on moving targets much more difficult," Thus, participants recognised that their performance was improving with open communication architecture, 5.

The majority of participants agreed or strongly agreed that they understood Janus capabilities. However, during the trials, interactors provided a large amount of capability explanations. For future investigations it is recommended that the training regime be reviewed, and tested on mock subjects. Clarity of expression, amount of information, necessity of information, and time allocated for training should be considered. The method of presentation should also be reviewed.

In the current study, participants were asked to read through the training protocol. Following this, a Janus expert took them through a practical demonstration. During this time, questions were answered.

This occurred directly before the first trial. If the training occurred on a separate day, and a short practice session was included, participants may feel more comfortable with the system capabilities.

This may improve their performance in initial trials, and remove Janus system learning effects. A consideration, however, is the required time commitment for participants. This would effectively add another morning or afternoon to the testing period. The results indicate that the majority of interactor time was spent performing actions to facilitate experimental outcomes.

However, a number of issues arose that need to be addressed. Firstly, the relationship between the interactor and the participant clearly varied from day to day as different interactors were scheduled on. Some interactors merely acted on commands. Others would laugh at participant's jokes, share smiles, gossip between game play or make suggestions.

This may have influenced performance across trials. In addition, some interactors would indirectly make suggestions. For example, they would ask "Would you like me to? This implies that an action should be taken, but not which action. This may have prompted participants and given them an advantage compared with those where the interactor did not indirectly suggest actions. A more in-depth training scheme should also be established to familiarise interactors with Janus.

Several interactors stated that they did not feel confident with the system at the beginning of trials. This could be combated with more intense training. Also, a requirement for interactor screening should be put in place before the trials begin. As mentioned in the results, there were interactor differences in terms of using their own initiative instead of following orders, and making tactical suggestions.

A participant affected by this became frustrated by the end of the day and commented that "it took a while to learn what was what. I need to be able to make more mistakes". This might have influenced learning and the development of expertise over time as this interactor was scheduled on the first day of the team's trials, where it would be most important for the participants to try the program and make a few mistakes in order to learn. It provided updates of the friendly force positions at regular intervals.

Commanders were provided with transparencies over the BCSS screen on which to mark anything they required enemy position, artillery zones. There were several problems with this setup. For example, when friendly units were destroyed on the Janus screen, they stopped moving on the BCSS screen but were still present.

Commanders reported that they foimd this confusing. This problem could be combated in future research by having a trained BCSS interactor who could remove weapon vmits reported as lost.

A second problem was that the screen could not show a large zoom and still show all the weapons units at the same time. Land types seemed not to match between maps. It is important that a high correlation between maps is present; otherwise effective communication is more difficult, 5. This created a cooperative situation that could have reproduced the same kinds of communication requirements experienced in Networked Fire Chief.

To a certain extent this seemed successful. However, the simplicity of the scenario was of concern, as was the lack of wide variation in the performance scores. The clear difference between the power of the red force compared with the blue force meant that the enemy had many problems posing an effective attack. This is an internal programming aspect of Janus. It may be that in order to create a fairer battle, the enemy needs a greater number of units.

This would possibly create a greater distribution in performance scores. The scenario could also be made more complex by having multiple sub-unit leaders, each of whom would be in charge of certain weapon imits. This would allow use of larger teams. Another aspect of the scenario relates to the duration of the trial. IM hours was recommended by a Janus expert as an appropriate length of time to draw meaningful results. During the trials, one of three things tended to happen: 1. The team defeated all of the enemy's groimd units, and spent the remaining time searching for artillery usually fruitlessly ; 2.

The team surrendered because they felt they had been defeated; or 3. Both teams had reached a town and they spent the remaining time scanning through this area of extremely low visibility. The third response is possibly an artefact from the extremely low visibility in the urban zones of Janus. This caused difficulty locating enemy units within the urban zone.

Therefore time spent searching was lengthened, whereas actual engagement with enemy units was almost impossible. With a more complex scenario and a larger red force, the battle tactics and time-pressured decision-making may continue until the end of the trial.

Therefore, a scenario with a higher level of complexity amongst the blue force, as well as equal distribution of power across the blue and red forces, is recommended. It should also be pointed out to participants that the intent statement is only a guide, and that it is acceptable if they do not achieve the goal within the allocated time. The scenario would often be played out without many interactions with the commander. Because of the random allocation of roles, the most experienced or most confident person for the role did not always play the commander.

In the military, corrunanders are respected because of their experience. In the current experiment, this element was not present, and respect for the commander may have been compromised. This could explain the lack of attention paid to the commander by many sub-unit leaders.

If the commander had more knowledge, it may also have been less intimidating for him to step in and take charge. Another observation related to the level of respect subordinates had for a commander was that often under the open communication condition, commanders would be bypassed in the communications. Instead, subordinates would communicate directly with each other.

This would exclude the commander from the decision-making process, and break down the C2 structure. In future research, a sample of military participants should be in roles consistent with rank and experience. In addition, the commander should be provided with better situation awareness via an interactor at the BCSS monitor. It also revealed trends, and identified issues that need further examination. In addition, it identified the following methodological details that need to be taken into account in future research: Large participant numbers; Amended interactor training; Interactor screening; Revised training regime; An interactor for the BCSS monitor, who is an expert BCSS user; A larger team structure to see more complexity emerge in the commurucations.

This study also provides support for the concept that Networked Fire Chief is an appropriate tool for investigating some aspects of naturalistic decision-making theory. As is suggested in the literature, this study indicates that expertise is largely context specific. There are significant differences between a fire fighting context and a battle context.

It is recommended that future work should further examine the use of Janus in Human Factors research. Also, the validity of Networked Fire Chief should be further investigated to determine the extent to which results can be generalised between areas that require naturalistic decision-making. References Artman, H.

Cooperation and situation awareness within and between time scales in dynamic decision-making. In Waem Ed.

Artman, H. Situation awareness and cooperation within and between heirarchical units in dynamic decision-making. Ergonomics, 42, Beach, L. Why classical theory is an inappropriate standard for evaluating and aiding most human decision-making. Klein, J. Orasanu, R. Zsambok Eds. Decision Making in Action: Models and Methods pp.

Norwood: NJ, Ablex. Brannick, M. Understanding team performance: a multi method study. Human Performance, 6, Brehmer, B. Strategies in real time dynamic decision-making. In Hogarth, R. Insights in Decision Making pp. Chicago: University of Chicago Press. Distributed Decision Making in Dynamic Environments.

Effects of time pressure in fully connected hierarchical architectures of distributed decision making. Waem Ed. Caimon Bowers, J. Establishing the boundaries of a paradigm for decision making research.

Human Factors, 38, Chaloupka, M. S Advanced Concepts Department, Naval reserve project. Naval reserve paper. Chapman, T. The effect of management structure and communication architecture on naturalistic decision making performance. University of Adelaide: Unpublished Honours Thesis. Chase, W. The mind's eye in chess. Chase Ed.

New York: Academic Press. Cohen, M. S, Freeman, J. Meta-recognition in time stressed decision making: recogiusing critiquing, and correcting. De Groot, A. New York: Mouton Original work published Driskell J. The value of small group research to understanding teams. Salas Eds. Dreyfus, H. Intuitive, deliberative and calculative models of expert performance.

Klein Eds. Drillings, M. Naturalistic decision making in command and control. Edwards, W. The theory of decision making, Psychological Bulletin, 51, Fiedler, F. Leader intelligence, interpersonal stress, and task performance.

Wagner, Mind in Context pp. Gaba, D. Anesthia crisis management and human error in anestheology. Proceedings of the 35"' Human Factors Society meeting, 1, George, L. Decision making in complex naval command-and-control environments. Human Factors, 38, Granlund, R. The C3 microworld. Hopple, G. Huf, S. Klein, G. Recognition primed decisions.

Rouse Ed. An overview of naturalistic decision making applications. Analysis of situation awareness from critical incident reports. Garland Rapid decision making on the fire ground. Proceedings of the 30"' Annual Human Factors Society meeting, 1, Lipshitz, R. Decision Making as Argument Driven Action. McDaniel, W. Naturalistic group decision making overview and summary. Castellan, Jr.

Means B. Cognitive task analysis as a basis for tutor development: Articulating abstract knowledge representations.

Psotka, L. Mutter Eds. Intelligent Tutoring Systems: Lessons Learned pp. Means, B. Training decision makers for the real world. Decision hAaking in Action: Models and Methods, pp. Noble, D. A model to support development of situation assessment aids.

Decision Making in Action: Models and Methods, pp. Serfaty, D. Mein Eds. Perla, F. P, A Guide to navy wargaming. Chaloupka, J. Borges-Dubois Eds. Naval reserve paper, pp. Randel, J. M, Pugh, H.

Differences in expert and novice situation awareness in naturalistic decision making. International Journal of Human-Computer Studies, 45, Rasmussen, J. Skill, rules and knowledge: signals, signs and symbols, and other distinctions in human performance models. Thompson, F. Beyond the war game my-ic. War Gaming Antholog. S Advanced Concepts Department, Naval reserve project.. Naturalistic decision-making: where are we now? Mein Eds, , Naturalistic Decision Making pp, Acknowledgments The authors would like to acknowledge the wonderful support that was received from so many people at DSTO throughout the duration of this study.

Firstly a big thankyou to all of the participants who willingly gave up their time to participate in the study and offer feedback. A big time commitment of three days was asked of them and it was wonderful to see so many respondents willing to do this. Thankyou to Rick Halls for rigging up an effective communication system across portable radios.

This was used throughout the study. Thankyou to Dean Bowley for helping us design a suitable scenario for the present study, which created the same kinds of coordination issues experienced in Networked Fire Chief. We would also like to tharik those people who agreed to act as interactors. These people attended several training sessions, and many also put in a lot of their own time to practice using Janus before the trails began.

During the trials they rotated around on a schedule to act as interactors. The study would not have been possible without them. Thanks to Peter Williams who acted as a project adviser in the initial planning stages of the study.

Information for Participants Thankyou for volunteering to participate in this pilot study. Your results will be strictly confidential. However if you would like to have an indication of the study's outcomes, I can e-mail you a summary of results around the end of February. This is a schedule of what you will be tackling over the three days. Under these war game conditions, participants will be tested under the two different communication architectures, defined in Chapman's experiment For each communication architecture the subjects will undergo 4 trials, and learning effects will be monitored.

This way we can investigate the development of expertise, and contrast whether this development is different for military persoitnel compared to civilian subjects. This study should also help to understand whether expert's NDM decision- making performance results are context specific. You will be acting as part of a team in this study. Each team involved in the study consists of 3 members.

One team member acts as the commander, while the other two act as sub- unit leaders controllers. This training session will follow closely a training protocol written by Chapman , which outlines the capabilities of Janus, and the tactical information you will need to understand.

Intent Statement The intent statement for the exercise will be given to the participants to read. Planning In the next phase, you will be given a lesson in the basics of mission planning. You will consequently have a 20 minute planning period, in which you will work with your fellow team mates to form a plan. During the planning you will be provided with a large map of Kamaria the land in which the battle takes place , a plastic overlay, and some transparency markers.

You can plan routes and tactics to accomplish your mission in this session. This planning phase will take place before each of the 8 trials begins. Tanus Trials Following your planning you will be required to perform one and a half-hour war-gaming sessions. This length of time is needed for Janus to generate some meaningful outcome data. The ordering of these sessions is as is scheduled above. Are you, or have you ever been involved in the military?

If so, what was your main role? How long were you or have you been in the military? How often are you involved in practical combat trairung? Please circle appropriate answer. Are you or have you ever been involved in another organisation which requires you to make emergency decisions? Have you worked with your team members before? Have you ever played a computer game or simulation? Please write number in appropriate time span.

Days ago Weeks ago Months ago Years ago 9. How long did you spend on the game or simulation? Please tick appropriate box Less than 30 minutes 30 minutes to 1 hour 1 to 3 hours more than 3 hours When was the time before that that you played a computer game?

Please tick appropriate box Less than 30 minutes 30 minutes to 1 hour 1 to 3 hours More than 3 hours Questionnaire is completed. Notes to the Commander As commander you are responsible for overseeing the actions of your subordinates, and making higher level decisions based upon your knowledge of the "overall picture".

You will need to coordinate your ground forces tanks and APC's with your support forces artillery and recon. During this exercise you are given aids to help you create a good situation awareness of the battle taking place.

You will have a BCSS monitor available. This is a restricted, situation awareness tool that provides you with a map of Kamaria This correlates with the Janus screen.

The system updates the positions of your weapon units controlled by both sub-vmit leaders on that map periodically. Enemy sightings do not appear on your screen.

Rather, you will need your sub-urut leaders to radio through the coordinates to you, and you can update them on the transparency overlay covering the BCSS screen.

If your friendly units are destroyed, they will cease to move on the BCSS screen. Your sub-unit leaders should notify you of casualties and coordinates. Below the BCSS screen you will have a picture of the map on the Janus screen your subordinates are using.

This may help with communication. You also have a whiteboard if you wish to use it. You may want to write tasks here during the planning phase that you can refer to them later. These may be useful also. It is recommended that you keep a few sheets of paper near the BCSS screen so that you can copy down coordinates when they are radioed through, and update them on the overlay later. If you have any questions about any part of this ask me, and we will try and sort it out.

Rules of Engagement 1. You must not move the vehicles forward of the beginning positions during deployment. You are only required to design manoeuvre routes.

The use of mines is not allowed. It is against Kamarian legislation 3. You should give coordinates to your commander regarding enemy sightings and casualties etc Notes to the Sub-unit Leaders Your role is to make lower level decisions, and to report all relevant information eg. Coordinates of casualties. Coordinates of enemy sightings, and situation reports back to your commander. FARH 8 12 Helicopter mm Hzr Howitzer 42 60 Time of flight for the rotmd before impact can be 5 to 58 seconds depending on range.

Radio Protocol In order to make your radio transmissions clearer, we ask that you use the following format. State the person's call sign you wish to talk to. State your own call sign. Relay your message. Finish with either "over" or "out" "Over" indicates that you expect a response from the person you have just communicated with. An example of a communication sequence: Hicon: " Loconl, loconl, this is hicon. Please provide me with a situation report over. We have lost 16 units in total so far, over.

Continue monitoring the enemy to your south east, and report immediately once you have identified the forces. Hicon out. Training Protocol During this study, you will be participating in a two-sided war game. Janus is an interactive simulation war game portra dng realistic events during multi-sided combat. It uses digitized terrain effecting line of sight and movement, depicting contour lines, roads, rivers, vegetation and urban areas.

It has the capability to be networked with other systems, in order to simulate a war game with multiple sides. You will be working with a group of three people in this study, representing a simplistic hierarchy. Each of you will be looking at your view of the battle on a computer screen all physically isolated from one another. Commands will be issued by the commander, and feedback is required from the sub-unit leaders. As well as this, in order to move anything on the screen, carry out any action, or request computer information simply make a request to your "interactor".