The Dream of Berners-Lee, Generative Design, and Operant Conditioning

The Competitiveness Spectrum in Designing Social Interfaces interests me because I am fascinated by cognition/behavior and operant conditioning. I have used a few of the methods mentioned in the handbook to train animals (a fish, a rat, a cat, and a dog). Although the methods discussed in the handbook apply to encouraging reputation patterns that fit the culture of a community, both applications pertain to psychology. I've tried a few of the methods the handbook encourages, such as providing "mini-motivations" by rewarding achievements (Crumlish and Malone 156).

In my experience, using positive reinforcement to modify behavior works better than punishment, and this applies to both people and animals. Punishment might deter a behavior, but it can cause the subject to associate the subsequent discomfort, embarrassment, and resentment with the punisher. Several studies also indicate that punishment or violence in childhood can increase aggression in adults. Punishment also does not show the correct behavior, although people can explain the appropriate action to one another. Positive reinforcement, on the other hand, increases the likelihood that the subject will repeat the (correct) behavior, as well as increasing the subject's trust and strengthening bonds.

My Border Collie illustrated this concept. I preferred positive reinforcement while my then-boyfriend used punishment more often. As a result, I often heard my boyfriend yell at the dog, who would then come scrambling down the stairs and hide behind my legs. He also showed much more loyalty to me, and when my boyfriend and I rough-housed, the dog would often "guard" me.

These benefits of positive reinforcement show up in Designing Social Interfaces as effective and motivating leadership methods. Most of the methods also suggest against ranking systems that display everyone rather than targetting only the best (e.g., leaderboards), because displaying low-scoring employees might actually embarrass them and make them resentful. But offering "first-time" achievements encourages people to try new activities (Crumlish and Malone 167), and the English Department even allows peer-to-peer recognition by opening up nominations/recommendations for awards like the Beagle Award and the Rebecca Mason Perry Award. Those potential rewards encourage professors and GTAs to work harder in order to achieve those rewards.

Gathering and processing this data, privately sending out each individual's information, and then rewarding the best performers seems like the best strategy to me. It privately informs individuals of their progress, which allows them to avoid public humiliation and makes it easier for them to recognize both their strengths and weaknesses. Adding a graph can boost motivation by helping individuals visualize their progress. But administrators would have to make sure they "encourage quality participation over mere repetitious activity," a method called "grinding" in the game world (Crumlish and Malone 166; 173). Overall, however, using positive reinforcement to reward individuals has, in my experience, fueled motivation, strengthened bonds, and shown other individuals what they can do to succeed.

References