Blog Post

The brain, games and pedagogy


   In Cathy Davidson’s book “Now you see it” she points out how games are compelling for attention and related learning.  Recent neuroscience research has traced the dynamics of how this works in the brain.

   We are constantly predicting what is to come in our world in order to guide our actions.  Our brains operate on a basic algorithm:  If our prediction is correct we needn’t attend or learn.  Walking is a good example.  We predict the floor will remain flat hence we don’t need to attend to each footstep.  If your prediction is in error (PE) this is the signal to turn on attention and learning.  These signals are mediated by the neurotransmitter dopamine (DA).  As the noted neuroscience researcher Wolfram Shultz says:

 “Learning depends on the extent to which behavioral outcomes are different than predicted, being governed by the discrepancy of ‘error’ between outcome and prediction. Outcomes that affect learning in this way are termed ‘reinforcers.’...Learning proceeds when outcomes occur that are not fully predicted, then slows down as outcomes become increasingly predicted and ends when outcomes are fully predicted”

    Video games put the player in a constant state of predicting where to go next.  Good video games (like slot machines) provide ample opportunity for errors in prediction creating a steady dose of dopamine and attention. The player is constantly learning from the gap between his/her prediction and the real. In the computer and AI world this is call Reinforcement Learning.

    Good speakers understand these dynamics instinctively.  Posing questions they know the audience will get wrong spikes attention. Even the pregnant pause in speaking increases anticipation  of what is to come.  Clickers are a good way to create prediction, error and anticipation along with safety.

    Those of us in science will recognize the core of the scientific method here.  Those in literature will recognize plot twists where the reader is lead to certain potential outcomes and their attention is held by the plot shifting to an unpredicted conclusion.  Those in media studies know that any TV show put in a game format where the winner is not known to until the final minute is on the road to higher ratings.

    There is an upper limit to the pleasures of unpredictability where too much becomes chaos and loses its appeal.  Musicians understand this well when their free form improvisations come back down to a familiar melody at various points in the tune.  Students do best in environments where there is plenty of unpredictability within a stable framework.

    The ability to learn from the errors is based on the capacity to stay present with the errors in order to become conscious of the difference between the predicted and the real long enough to update learning hence making a more accurate prediction the next time.  To do this there must be safety.  Fear and shame over mistakes block the attentional space needed for the updating.

    The implications of this research are far reaching when applied to pedagogy.  Environments where students have ample opportunity to make predictions, where it is safe to be wrong and where they are asked to have the metacognition to be aware of the errors accelerate learning.  Compare this to the classes where the information is force fed in continuous streams such as lectures.  In many testing environments predictions that are wrong are severely punished.  Multiple choice exams that are not reviewed leave the taker with their wrong answers as what they have learned.  This has lead to the situation many of us face of passive students limiting their inquiry to what they can anticipate the professor wanting from them.

    In addition to predicting what is to come we are also predicting the reward value of what we are doing.  The prediction of reward spikes DA and related attention and learning.  We have all had this experience when we are anticipating a coming  great date . The arousal and focus kicks in long before.  I believe this can be generalized in the classroom.  We make our predictions of the potential reward based on past experience assuming the future will repeat it.  Imagine two children taking the same difficult, less than stimulating math test.  A child from an environment where they never saw anyone gain from good grades in less than stimulating classes (less affluent communities) does not get the anticipation of reward and related attention.  They can then act bored and indifferent.  The child from the middle class family who knows they can receive future reward from present day bearing down on task gets that spike of DA and attention.

   In my classes in psychology I have shifted my assignments from researching good questions to researching questions after one has made predictions of what they will find.  The most important work is at the end when they reflect on how their prediction varied what was found.  This difference makes crystal clear the domain of new learning.  It puts the student’s critical thinking at the center of learning.

   Perhaps one other way to interpret Dan Simon's gorilla is that the prediction of reward at getting the number of passes counted correctly so rivited attention that the gorilla was lost in the midst.










1 comment

I am so happy to have others profit from your knowledge and wisdom.  Thank you so much for posting.  This is very, very important.