could video games train their players for war?

October 29, 2010

It’s not often that a magazine like Foreign Policy devotes a column to video games, and yet soundbytes about the new Medal of Honor game set in modern Afghanistan prompted Matthew Shaer to ruminate on the role of video games in social commentary and vaguely question the validity of the cries about violent and realistic video games somehow turning their players into emotionally sterile psychopaths, an idea I argued against in a post for Discovery Tech based on the simple premise that human can usually tell reality from fiction. Since this topic has really been done to death, there’s very little new material Shaer could offer here, but some of the commenters on the article did bring up an interesting point that’s not often discussed even though it deserves some coverage. One of the key weapons used in Afghanistan (and Pakistan) today is a robotic drone armed with missiles, steered and fired from a remote location, with mechanics similar to video games. And it seems like this is the future of war as ever more robots are being equipped with guns, tracks, and remote controls.

As mentioned previously on this blog, using robots to fight wars is a politically attractive proposition. Should a robot get blown up or disabled, it’s not a human being. While the public is appalled at hearing that thousands of troops died in combat operations, piles of scrap metal hit by missiles or IEDs don’t exactly cause a stir and to grab another one off the assembly line when you need more guns firing at an enemy position during future battles is relatively easy and straightforward. And with robotic guards, scouts, and plans for heavier and better armed machines and unmanned bombers that could clear a path for a human army, the military seems to be very interested in killer robots on the battlefield. In several decades, a few pundits have argued, wars may be fought by remote control and boots on the ground would be needed to secure an area rather than engage in a direct assault on an enemy position. The actual attack would take place from the air by nimble bombers, and by mechanical tanks which could carry much more ammo without their human occupants. Today’s obsessive gamers, the concept usually proceeds to state, would be tomorrow’s top generals.

This is where a number of books and articles raise the concern about emotional detachment from actual war because a significant portion of the fighting would be done remotely, from a faraway bunker. Would the pilot of an unmanned bomber be able to remember that when he or she pushes the button, a real bomb drops on an enemy position and kills real people? Would the operators of robotic siege machines get a little overzealous, as per their gaming experience, and let loose with indiscriminate bursts that kill combatant and civilian alike, something that already happens in the real world in the middle of a battle with today’s soldiers? Being aware and very precise in your attacks is a very important skill that needs to be taught to those who will be operating remote controlled machines of malice, especially since for the foreseeable future, conflict will overwhelmingly likely involve unconventional warfare, in which enemy troops will blend into civilian populations rather than set up huge bases and arm themselves with tanks, planes, and naval destroyers. But against nation-states with conventional militaries, an army which uses vast fleets of robots for its first strikes would have the advantage, maximizing impact, minimizing its own casualties, and attacking well defined enemy targets.

It’s quite possible that training future robot pilots would take the form of video games, or at least something a lot like it, and hopefully reinforced with war games in which they’ll get to see the damage their commands can cause up close. But it’s not going to be done with games like Medal of Honor, or a first person shooter. I’d bet the training games in question would be more like StarCraft, and a modified version of first person shooters may be used to help hone soldiers’ reflexes, while having them wear gear that simulates pain when they’ve been shot by an enemy. I’d be surprised if casual gamers with a taste for first person shooters would want to actually try that level of immersion themselves while playing on their living room couch because to them, the whole point is to briefly escape reality, not to continuously subject themselves to the real world pains of being shot or stabbed. After all, there is a very fine line between simulation and entertainment, and that’s a line that very few commercial games would really want to cross for legal and financial reasons.

Share on FacebookTweet about this on TwitterShare on RedditShare on LinkedInShare on Google+Share on StumbleUpon
  • badbass9

    Most of what you’re saying is being done already. Flight simulators are nothing more than immersive games. Same with tank simulators. And you can bet the military is developing a Caprica style, neural connected V-world for battlefield simulations. Face it. 99% of consumer technology begins as military applications. The internet began as a link between defense contractors and college/military research labs. I didn’t quite word that right, but, you get the idea. The impersonal aspect has been debated as far back as WWII, when bomber pilots were said to have no connection to their targets. While I would agree fighting an enemy like al-Queda is entirely different than fighting nation states, todays smart weapons give the user a huge advantage. After all, Americans (or Russians for that matter) don’t quite fit in with the native population. Why put them on the ground and at risk. Vietnam taught us that, as Afganistan taught the Russians. So, gather intelligence, pick targets, and send in the drones. At least we try to keep civilian casualties to a minimum.

  • Here’s an interesting thought. A certain number of civilian casualties comes about because, when the rounds start flying, sources can be very hard to determine and “get them before they get us” becomes the only operative procedure. When you believe there’s a bullet out there with your name on it, you feel somewhat obligated to keep it from being fired at whatever costs.

    Without the immediate danger of death, however, the remote-operator does not have the same mentality. The penalty for not getting the shooter is a destroyed robot or remote vehicle – serious, perhaps, but still detached. So the remote-operator has more presence of mind to clear their shot and ensure their target is valid. The use of guided mechanical combatants, as long as they provide good simulation (clear views of targets, little lag, etc.) stands a very good chance of reducing civilian casualties.

    Additionally, with good recording systems, even a destroyed remote vehicle can provide the location of the hostile fire to others, thus able to become a sacrificed pawn.

    It costs a lot to train a combat soldier, and a lot more to sustain them remotely. Mechanical combatants might change this drastically, especially when a “killed” operator simply boots another micro-tank and goes back into the fray. No medical corps either.

    But, I suspect this wouldn’t last long. Mechanical combatants would still require a significant amount of ground support, and a reasonably close base of operations – perhaps not as much support as most combat helos, but probably bases closer to the hostile grounds. Those bases are what would become the prime targets of attack, with little emphasis on gaining ground or securing areas.

    For any engagements that have securing an area as an end-goal, little resistance to mechanical combatants would be offered – a smart commander would wait until humans have to enter the “secured” area. Booby-trapping such areas would be very common, and much harder/longer for remote vehicles to detect and clear.

    Communications jamming would be a prime endeavor, requiring constant updates of equipment, meaning the downtime of mechanical combatants might be anywhere from very significant to crippling.

    And finally, when it comes down to evenly-matched armies sending out their Mechwars remotes against each other, would anybody even bother engaging? It’s not about scoring points, it’s about controlling the populaces and/or securing assets. With no casualties, there’s no psychological advantage, no body counts and devastation to induce surrender. So the targets will be switched to something more “precious.”

    I get the feeling that remote combatants will progress much along the same lines as dogfighting/engagement abilities in modern combat jets. The advances among major powers generally cancel one another out, making them relevant only to their use against lesser-equipped opponents. While at this moment such advances look promising, it’s only a matter of time before they render some forms of combat obsolete, perhaps a very short period of time, and different forms of control are adopted.

  • Jypson

    Current video games are a precursor to virtual combat training. As it has not matured to levels that would be acceptable in the field, it is now of course less effective but would yield basic results.

    I am a video game junky. I’ve spent nearly $10K on systems, computers, controllers, racing wheels, joysticks, guns, and the games themselves. My Xbox Live gamer score is around 40,000 points* and I have just over 100 games on my Steam account. I play 1st / 3rd person shooters, RPG’s, RTS’s, Turn Based, racing, flying, adventure, puzzle, and every hybrid thereof and variation of game that does not fit in one of those categories. Now that I’ve exposed my video game geekery for the ridicule of the WoWT readers…

    I can’t say that I feel any more qualified than any novice to pilot a drone of any kind into combat effectively. If I was placed in the cockpit of an actual F1 race car and expected to perform at the level of my digital avatar, I would certainly not live up to my resume. If I was handed a weapon and equipment and shipped off to Afghanistan, I can assure you that I would not do as well as the least of the trained service members with real training.

    I think the best we could hope for right now is basic familiarization and doctrinal knowledge. I’ve learned many strategies to maximize available resources, the best speed and angle to tackle turn one on the race course Laguna Seca, when to use chaff or flares when being fired upon by various types of missiles, when to deploy a speedling / banneling rush based on Terran defenses, and of course when to use the shotgun vs. the rifle against the zombie apocalypse.

    Could “video games” be used in the future to ASSIST in the training of VR pilots? Sure, if the physics are accurate, the simulator hardware is convincing, and the combatant AI is realistic. Could video games as they are now be effective trainers? Well, pass me the joystick to a multi million dollar drone and a waiver stating I won’t be held accountable for its untimely demise and we’ll find out :)

    * Each game has a potential for 1,000 points if all accomplishments are completed. I think I’ve only gotten 100% on one game thus far.

  • Paul

    “while having them wear gear that simulates pain when they’ve been shot by an enemy. I’d be surprised if casual gamers with a taste for first person shooters would want to actually try that level of immersion themselves while playing on their living room couch because to them, the whole point is to briefly escape reality, not to continuously subject themselves to the real world pains of being shot or stabbed”

    I would note the electric shock “pub games” that were popular a couple of years ago. No real game; you just hold the controller, wait for the buzzer, press a button, loser gets a shock, everyone laughs. No lawsuits as far as I’m aware.

    Re: Gaming-type-controllers dehumanising enemies.

    Just Al mentioned a possible reduction in panicked return fire. I’ll add that there may also be less of a culture of the Sacred Soldier (or Marine). Console operators can be more easily investigated, criticised, even jailed, for breaching the rules of engagement than a soldier or marine in the field. Since they are not “putting their lives on the line”, there’s less of the reflexive “you can’t judge them because you’ve never experienced combat”.

    (Plus you’ve got video evidence.)

  • Loeck

    Having played a number of video games myself (tip o’ the hat to Jypson) i can see the validity of using video games to train, not so much the physical aspect but the strategy of the battlefield. Like don’t stand in the open, look around a corner instead of rushing in to a rifle, throw grenades they are replaceable your not, the thinking and to a lesser extent the mentality of the battle filed.

    I think that video games have a huge potential in lowering fatality rates by teaching soldiers in low risk situations, however as previously stated there’s a way to go before this is a valid option, and the army is already working on using them.

  • Amadan

    Any chance we could get robot civilians to play the part of ‘Oops! Guess it was a wedding party after all!?