Sony: Emotion-Reading Games Possible In Ten Years
Calidreth writes "At Gamescom last week two of Sony's executives stated their belief that in merely ten years' time, video games will have the ability to read more than just movement on the part of the player. Reading player emotions will be a key feature that is possible now and might be implemented into games in the future."
...they will have to install special self-preservation chips inside consoles, so they don't commit seppuku when an Emo kid plays on them.
Tubal-Cain smokes the white owl.
I'd think this could be done now. Game systems with motion controllers should be able to tell when the controller hits the wall when thrown out of frustration. I suppose the Xbox 360 is at a disadvantage though since there's nothing to throw with Kinnect, unless you throw some random nearby item and the Kinnect sees it. That or you start swearing at it and the voice recognition kicks in.
Want to make a bold claim with no evidence whatsoever? No problem! Just prophecy that it will happen "in ten years" and you'll enjoy the infinite respect and worship afforded to the likes of St. Peter, Confucious, and Warren Buffet.
Worried about being wrong? Again, no problem! When the ten year mark rolls around you'll have a wealth of available options for deflecting the shame. Your best hope is that you'll be dead by then. If by some grave misfortune you are still alive, you'll still have an average of 2.7 presidential administrations (adjusting for the assassination coefficient), dozens of religious leaders, and countless others to blame for derailing the path to paradise in the prior decade!
Bet on ten years and no one can ever prove you wrong! Guaranteed or your money back! Thank you for shopping Ronco, operators are standing by!
if you want to do something like that now all you need to do is record and analyze input patterns from regular controllers, redundant or excessively long button presses and repeated attempts at disallowed actions, significantly lower than average (for a particular player) numbers of actions taken in a minute (not counting walking around) and significantly worse than normal reaction times would indicate boredom or lack of attention Redundant button presses along with high actions per minute would indicate excitement.
all without creepy "computer is watching you" shit
Snowden and Manning are heroes.
I believe Hocking is somewhat wrong. Although judgment of visual, external, expressions can give clues to internal states, emotions work a bit like this: 1) a person has various goals. As he progresses through the present world-state, he experiences success or failure, or partial success or partial failure of progress towards goals. 2) this progress is cognitively monitored by parts of the brain, and a metric (an 'emotion type and intensity value') is mapped and matched to the goal progress. 3) other factors are combined with that metric. For example, if you've lost something but someone is attacking you, you do not feel sadness of loss first but instead anger over the attack. The amygdala is kind of an interrupt prioritizer for this. 4) the person internally decides (moderates) an external expression of the metric. (some of this is done at an unconscious level and rapidly). For example, an expression of anger, happiness, sadness.
Now, a camera and perceptual system might be developed that can measure facial metrics and 'decode' them for that person into likely emotion states IF the system can measure and adapt to that person's particular ways of expression (over time and experience (learning)).
But a key problem is that mere external visual cues do not give a reliable backward mapping to the goals or the internal responses to the goal success or blockage, and certainly not be able to deal with the complexities of various simultaneous emotions rising and perhaps inhibiting each other. In other words, the camera-based perceptual system cannot fully recreate the true internal cognitive states that gave rise to the emotion. It would have to be able to guess at and reason about the person's goals and his internal framework of mental state.
I base this on my own work in AI understanding of emotion, which differs a bit from the accepted AI model of emotion used by Ortony, Clore, and Collins.
"It would be a great advance in human computer interaction..."
It would be a great advance in human interaction, if men got such a gimmick to read emotions, like women can do from birth.
There's a Game Developers Conference video for available for free which relates experiments done by Valve with biofeedback integrated into gameplay:
http://www.gdcvault.com/play/1014734/Biofeedback-in-Gameplay-How-Valve
You are awash in a sea of fiercely stated opinions. Obvious exits are: 'File->Quit', 'Reply', and 'Page Down'.