I'm reading this book on
Games Theory - some interesting points about the way that Games use feedback. One of the Games Theory gurus (Marc LeBlanc) defines two sorts of feedback: Negative feedback (-ve F) and Positive feedback (+ve F).
With -ve F, which intervenes to reduce the difference between the players or the player and the computer, the game tends to stabilize toward a steady state of activity - prolonging the game. With +ve F, where the intervention enhances the difference between the players, the game tends to destabilize - ending the game.
The book goes on to discuss a sophisticated feedback technique called Dynamic Difficulty Adjustment (DDA) that adjusts the difficulty of play ... i.e. the game 'evaluates' how well the player is doing and, if it senses that the a player is not doing very well (i.e. getting killed too often/quickly or falling too far behind), the game adjusts to make it easier for that player - similarly, it will make the game more difficult for a better player.
The aim being to keep the players playing. (Salen and Zimmerman, Rules of Play: Game Playing Fundamentals, 2004, MIT Press, pp 218-224).
I guess in a Formative Assessment (academic) setting, this would be translated into something like a set of questions/tasks ... after the first ones the computer would assess how good the student was and either present more easier questions/task to poor students or more difficult questions/tasks to good students. After the next set of questions the student would be reassessed and the number and level of difficulty of the subsquent questions would be adjusted accordingly ... i.e. Adaptive Assessment (
http://www.ascd.org/publications/educational-leadership/mar14/vol71/num06/The-Potential-of-Adaptive-Assessment.aspx).