Adversarial Collaboration: An EDGE Lecture by Daniel Kahneman

…

My first experience of an adversarial collaboration was about 40 years ago. My wife, Anne Treisman, and I were studying a new paradigm involving apparent motion and priming. It’s a nice effect. There’s a lot of work on it. And quite a few studies since have followed up on this work. Anne and I had many ideas, and we designed a large number of experiments, most of which succeeded. There was only one trouble. We didn’t agree on the nature of the phenomenon, and we had different stories about the role of attention in the effect. The difference didn’t prevent us from planning and interpreting useful experiments, but we found it difficult to construct a coherent theory.

I observed a phenomenon that I called the “15 IQ point benefit.”

Then came February and the invitation to participate in the meeting of the Psychonomic Society in November. I suggested to Anne that we take two consecutive 25-minute slots, which would give us a whole hour. She said, “but we don’t agree,” which I answered by “don’t you believe in the scientific method? We’ll run experiments to determine who of us is right. There’s plenty of time before November.”

So we started a cycle of critical experiments. I would design a study that I believed Anne could not explain, get her to agree that the result I expected would contradict her theory, and we would run that study. The results would come out as I had predicted, and there, I observed a phenomenon that I called the “15 IQ point benefit.”

…

Read the whole story (subscription may be required): Edge