They Lied to You: Punishment is Better Than Reward. This is Why.

hey 42 here the early 90s gave us the spice girls friends and minesweeper in 1992 microsoft released windows 3.1 and they packaged it with two games that would go on to cost companies all over the world billions of dollars in lost worker productivity solita added to windows in 1990 and minesweeper added in 1992 you may assume that microsoft went to the trouble of coding and releasing these free games to provide windows users with endless hours of procrastination-filled fun but you'd be wrong they actually concealed a secret agenda microsoft added these games to windows to teach the world how to use a mouse as the 80s rolled over into the 90s computer mice were only now beginning to attract widespread consumer adoption even though the computer mass was invented by douglas engelbart in 1964 in the 70s and early 80s most people were accustomed to controlling their home computers solely using the keyboard nowadays we grow up with computer mice so the neural circuitry required to develop the innate motor neuron control for efficient mouse usage is formed at such a young age and so for young people it never feels like they had to learn how to use a mouse the brain is much more plastic meaning moldable in our youth so it's far easier to pick up new skills that's why you were able to frag noobs with your sweet flick shots by the time you were 15. but when you present a middle-aged man with a computer mouse who has never used one before he looks at it like you've just asked him to launch a nuclear missile with a sandwich the older the brain gets the more difficult it becomes to learn what young people assume is really basic stuff moving the mouse around a desk to control a virtual environment that's why your grandma probably takes much better to a tablet than a desktop computer pointing out and pressing buttons is a natural movement that she's probably done many times before but a mouse is completely alien to her microsoft knew this so they designed solitaire to teach people how to use a mouse to drag and drop what do you do when you play solitaire you spend hours dragging and dropping cards dragging and dropping minesweeper on the other hand was designed to teach people how to click but more importantly the difference between left and right mouse clicks get the left and right mouse buttons confused in mindsweeper and you'll be virtual toast so it's a fantastic training tool negative reinforcement worked for minesweeper if you blew yourself up you would adjust your strategy so the next time you didn't but we're often told that the best training tool is actually positive reinforcement rather than negative the carrot is better than the stick praise is more effective than punishment but how true is this at the end of this video i'll reveal how israeli fighter pilots disproved the common adage that praise is better than punishment in a dutch neuroscience study eight to 12 year olds were asked to solve puzzles on a computer half of the children were shown a green tick on screen when they answered correctly but no symbol when they didn't the other 50 of children were shown a red cross when they answered incorrectly but no tick when they answered correctly the tasks were then repeated over and over again to measure how the children improved at the puzzles over time as various tv super nannies keep telling us reward wins out over punishment and this experiment reflected that the group of children that were shown positive reinforcement for correct answers the green tick showed a substantial improvement with the accuracy and speed of their answers over the course of the experiment but the other children who were punished for incorrect answers well they showed absolutely no improvement if anything they got worse at the puzzles the more they repeated them so there we go that definitively answers this question and we can all go to sleep peacefully tonight except there's an enormous elephant-shaped caveat lurking in the corner of the room you see the dutch neuroscientists decided to repeat the exact same experiment with a group of older participants more specifically young adults aged 18 to 25. the results for this group were the polar opposite of the children's the young adults showed much greater improvement when they were punished with crosses and almost no improvement when they were rewarded with ticks something changes drastically in the brain between being a child and an adult and since after the age of 21 our brain becomes less plastic and is thus more set in its ways it could be assumed that if the experiment was repeated with any age group older than 25 then punishment would always win out over reward for improving learning and behavior to further prove the point that this change is the result of a fundamental alteration in the construction of the brain as we age the dutch neuroscientists took brain scans of the children 8 to 12 whilst completing the tasks what they discovered was that the learning centers of the brain specifically the cerebral cortex activated much more strongly in the group that received reward than the punishment group yet with the 18 to 25 year olds it was reversed their brains activated more strongly when punished for their mistakes the experimenters concluded that this is because it's more difficult to process negative reinforcement than positive children respond better to and learn easier from reward because it's a really simple concept to grasp in the words of the author of the paper learning from mistakes is more complex than carrying on in the same way as before you have to ask yourself what precisely went wrong and how it was possible we only learn the ability to effectively learn from our mistakes as we mature into young adults and our brains become fully developed but there's a fundamental rule that explains why punishment is more effective than reward and why neither punishment or reward have very little effect on an outcome at all it's called regression to the mean and it was famously demonstrated in the 1960s when daniel kahneman an israeli psychology professor made a landmark discovery whilst lecturing to a class of air force fighter pilot instructors in tel aviv kahneman was driving home the point to the instructors that rewarding good behavior or performance will improve the capability and learning speed of the students whilst punishing bad behavior will reduce morale and not improve learning he was espousing the virtues of the carrot over the stick before he could finish his speech one of the instructors loudly interjected with respect sir what you're saying is literally for the birds i've screamed at people for badly executed maneuvers and by and large the next time they improve he followed up with i've often praised people warmly for beautifully executed maneuvers and next time they almost always do worse don't tell me that reward works and punishment doesn't my experience contradicts it furthermore the entire audience of instructors resoundingly agreed with him that punishment is a more effective teaching tool than praise kahneman decided to investigate further to see if he could definitively quantify whether perhaps he had been wrong all this time and the instructors were right after observing the flight instructors for some time and collecting further data conneman uncovered a rather surprising truth the flight instructor's shouting always preceded the improvement of the student receiving said bollocking but it didn't actually cause it because the student would always improve after messing up regardless of whether or not he was reprimanded kahneman put his findings down to a phenomenon called regression towards the mean this is the universal axiom that in any series of random events there will always be periods of variation peaks and troughs but over time the results will always regress towards the mean if i roll a six-sided die over and over again i may occasionally get lucky and hit three sixes in a row but over enough iterations my average roll is always going to settle at 3.5 because that is the mean of a six-sided die and the law of large numbers dictates that any series of random events must always regress towards the mean to translate this logic to fighter pilots the students all have their own mean level of performance their mean will increase slowly over time sure with enough practice but increasing that skill takes thousands of hours and so in the very short term the performance of their day-to-day flights are going to vary in a seemingly random manner many factors can increase takeoff flight and landing performance on any given day wind conditions temperatures tire pressure how much sleep the pilot had last night and so on if the pilot happens to do better or worse than usual then that is straying away from their mean performance so statistically the following time out they are almost always going to regress towards their mean that means if they performed worse than usual last time they will most likely do better next time out and if they did better than usual then they will probably do worse next time so when the instructors shouted at the students for poor performance and they improved next time out it was because of regression to the mean and had very little to do with the instructor's course language but from the instructor's point of view it would seem that their bollockings made the students perform better but in reality it actually had very little impact on their future performance the students performed better mostly because of chance and when the instructors praised their students for good performance because the performance that incited that praise was better than their average performance they would because of chance do worse next time out so the instructors walked away with the false assumption that their praise actually reduced the performance of their students i'm not saying that praising good behavior is useless no it does actually have a small impact however it has much much less of an impact on future behavior than what we are usually led to believe having said that there are of course many instances in life when a little praise can go a long way for instance not praising your girlfriend for cooking you a lovely dinner after she spends the day working nine to five is a sure-fire way to receive a world of pain thanks for watching i've recently launched my first book which i'm crowdfunding through unbound publishing it's called sticker flag in it a thousand years of bizarre history from britain and beyond if you'd like to get your hands on a first edition signed copy then please head on over to unbound the links in the description and pledge today thank you

Channel: Thoughty2

Views: 902,331

Rating: 4.8959112 out of 5

Keywords: punishment, teaching, reward, punish, learning, punishments, education, teacher, ted talk, teacher tips, educational, new teacher tips, tips for teachers, psychology, study, israel, fighter pilots, brain, evidence, interesting, students, school, mind, reward and punishment, punishment reward, thoughty2, #thoughty2, thoughty 2, thouthy 2, thoughty, toughty 2, better, forty two

Id: _JfbPq6vHcA

Length: 12min 16sec (736 seconds)

Published: Wed Feb 26 2020