3 ms·

In other words it suggests penalties trump rewards. It seems to work that way with addictions too: http://medcitynews.com/2015/05/to-quit-smoking-turns-out-pen

In other words it suggests penalties trump rewards. It seems to work that way with addictions too: http://medcitynews.com/2015/05/to-quit-smoking-turns-out-penalties-work-better-than-reward-incentives/ http://medcitynews.com/2015/05/to-quit-smoking-turns-out-pen.... That is to say: I suspect the "penalty" of getting "killed" in a humiliating fashion thus thwarting a goal is more of a motivation than the reward of hearing a voice say, "good job for memorizing that numerical pattern."

johnloeber 11y ago
If that turns out to be true, then that's a huge finding. That's especially considering the fact that modern psychology/therapy is generally recommending positive reinforcement (treats) as opposed to negative reinforcement (penalties) for teaching/conditioning behavior. This could lead us to revise commonplace methods for teaching children, training pets, and for managing individuals and groups in general.
- mrob 11y ago
  It's possible that the source of the negative reinforcement matters. In games it comes from either the environment, or from another person who's part of a (temporary) outgroup. If you simulate aggression towards that person you will be rewarded. In schools it comes from another person in a position of authority over you, and if you act aggressively towards them the punishment will escalate to you point where you are almost guaranteed to lose.
tbrownaw 11y ago
No, it could also be the immediate, continuous feedback. Or that they're actually more mentally demanding. Or that they require lots of different parts to work well together, rather than trying to focus on just one thing.
ajuc 11y ago
Simplification. People don't play counterstrike or starcraft to not get killed in it - not playing achieves the same. They play to compete in something they consider "cool". And competing without the possibility of losing is pointless.
pdkl95 11y ago
https://www.youtube.com/watch?v=u6XAPnuFjJc https://www.youtube.com/watch?v=u6XAPnuFjJc This is the RSA Animate of Dan Pink's lecture on motivation. He suggests there is a division between physical and mental tasks. This was tested with several experiments involving different reward schemes. As long as the task involved only mechanical skill, bonuses worked as they would be expected. The higher the pay, the better the performance. [...] Once the task called for even rudimentary cognitive skill, a larger reward led to poorer performance. edit: The idea that penalties trump rewards probably explains why I've always liked roguelike games (and Touhou-style shooters). I've been playing way too much Nuclear Throne recently (and Risk Of Rain). As a roguelite, it strongly embraces the idea of "permadeath". Robotron/SmashTV style extra-fast action, that where you can never make a mistake (a single hit can end the game). Also, Nuclear Throne demonstrates an important implementation detail in something that uses such a strong punishment-based motivator: never put anything in the way of the player restarting after their mistakes. In NT, once the death animation has played (only a few frames, at most 200-300ms), you can hit "R" to restart the game at the first level. No delay whatsoever. Sure, the penalty is harsh, but you can try to overcome that failure immediately.

johnloeber 11y ago

If that turns out to be true, then that's a huge finding. That's especially considering the fact that modern psychology/therapy is generally recommending positive reinforcement (treats) as opposed to negative reinforcement (penalties) for teaching/conditioning behavior. This could lead us to revise commonplace methods for teaching children, training pets, and for managing individuals and groups in general.

mrob 11y ago
It's possible that the source of the negative reinforcement matters. In games it comes from either the environment, or from another person who's part of a (temporary) outgroup. If you simulate aggression towards that person you will be rewarded. In schools it comes from another person in a position of authority over you, and if you act aggressively towards them the punishment will escalate to you point where you are almost guaranteed to lose.

tbrownaw 11y ago

No, it could also be the immediate, continuous feedback. Or that they're actually more mentally demanding. Or that they require lots of different parts to work well together, rather than trying to focus on just one thing.

ajuc 11y ago

Simplification. People don't play counterstrike or starcraft to not get killed in it - not playing achieves the same. They play to compete in something they consider "cool". And competing without the possibility of losing is pointless.

pdkl95 11y ago

https://www.youtube.com/watch?v=u6XAPnuFjJc https://www.youtube.com/watch?v=u6XAPnuFjJc This is the RSA Animate of Dan Pink's lecture on motivation. He suggests there is a division between physical and mental tasks. This was tested with several experiments involving different reward schemes. As long as the task involved only mechanical skill, bonuses worked as they would be expected. The higher the pay, the better the performance. [...] Once the task called for even rudimentary cognitive skill, a larger reward led to poorer performance. edit: The idea that penalties trump rewards probably explains why I've always liked roguelike games (and Touhou-style shooters). I've been playing way too much Nuclear Throne recently (and Risk Of Rain). As a roguelite, it strongly embraces the idea of "permadeath". Robotron/SmashTV style extra-fast action, that where you can never make a mistake (a single hit can end the game). Also, Nuclear Throne demonstrates an important implementation detail in something that uses such a strong punishment-based motivator: never put anything in the way of the player restarting after their mistakes. In NT, once the death animation has played (only a few frames, at most 200-300ms), you can hit "R" to restart the game at the first level. No delay whatsoever. Sure, the penalty is harsh, but you can try to overcome that failure immediately.