3 ms·
> Anyone who has ever trained a dog or other animal knows how to use animal psychology / punishments / rewards to change the animal's behavior. Use of punishme
by enragedcacti 4y ago
> Anyone who has ever trained a dog or other animal knows how to use animal psychology / punishments / rewards to change the animal's behavior.
Use of punishment in dog training is actually pretty hotly debated. Not that it doesn't work, but that it has consequences beyond what are intended and that substitution, redirection, and rewards result in a happier and more obedient pet.
- lamontcg 4y agoYeah it works way better to establish the fact that you're the alpha dog via some basic obedience involved in feeding and stuff like that, then build up the dog to want your approval, establish boundaries around things like playtime, give the dog adequate "jobs" and stimulation for the breed so its not bored and anxious, etc. One the dogs needs are all met like that within some structure you're going to have a way happier dog compared to not understanding why the dog is acting out and having to tell it 'no' and discipline it all the time. That requires a commitment from the human a well though, it isn't a one-way street. Which is more less how society should treat humans as well. If we're going to have an article called "Why Punish?" there should really be a companion article on "Why Steal?" And given the right conditions literally every one of us will steal to survive. Every single one of us are still just animals at our core anyway, that means you dear reader.
- leobg 4y agoDebated? B. F. Skinner already found out back in the 1950s that punishment does not lead to any learning of the desired behavior. Rather, it leads to avoidance. So I would argue that the fact that punishment is the most inefficient way to change behavior is pretty much established. In fact, you can look at any area - dog training, marriage, marketing, computer games - and you will find that the most efficient and effective methods of shaping behavior of those that employ positive reinforcement. Things like Facebook and social media are pretty much just Skinner’s techniques amplified by the internet.
- dragontamer 4y ago> and you will find that the most efficient and effective methods of shaping behavior of those that employ positive reinforcement. I attack my opponent when he's waking up (Oki position), he uses a Dragon Punch / Reversal and punishes me. I attack my opponent when he's waking up (Oki position), he uses a Dragon Punch / reversal and punishes me. I attack my opponent when he's waking up. He uses a dragon punch/reversal and punishes me. I block while my opponent is waking up. He uses a dragon punch, loses to my block and I punish him. ------------ An entire genre of video games, fighting games, revolves around conditioning your opponent and getting into their head. The rock-paper-scissors battle of attack vs block vs grab takes place so quickly, you often don't have the time to react (though some players do have insane reaction speeds). When you can't leverage reaction, you leverage conditioning. And over the long term, the only way to really learn the game at a high level is by getting your bad habits repeatedly-punished by a stronger player. Perhaps its not "efficient" to learn, but in this case, it is the only way to learn. Expert players are bad at vocalizing what is or isn't working. There are very few coaches, and almost none of them know how to take you to the highest level (even if they themselves have reached peak form). ------------- There's other forms of punishment as learning. Ex: fall into a pit in Mario. You die, so you try again. Or "Boar Lure" in AoE2. If you fail, your villager dies, and/or you don't get the precious early-game food you need to sustain your production. Or in Basketball. If you get crossed up, you fail to block the opponent and they get a free shot at the basket.
- enragedcacti 4y agoFWIW I agree with you, but many people still employ it and tell others it works despite the academic and professional evidence to the contrary. I probably should have phrased it a little better to make that clear.