Is “bias” an asset?
I’m in the midst of a long article I’m slogging through, that I think is trying to make the point that people evaluate alternatives before making decisions (only it’s taking 20 pages to say that, so I’m pretty sure that there’s more to it than that… but the authors keep using the word teleology and I have to keep stopping to look up what that means.)
But this little gem emerged, which is relevant to the Contextual Interviewing class I teach. First, the authors include this model of how learning may take place:
expectation → observation → discrepancy detection → discrepancy-reducing change in expectation → expectation…
In other words: people start with an expectation of what’s going to happen in any situation. They observe what actually happens, noting if it matches their expectation or not. If what happens is different, they change their expectation and repeat the cycle.
Essentially, people come to a situation with a bias, and learn in part as a response to whether that bias is confirmed or not.
But isn’t bias bad?
So that’s where it gets interesting. They ask whether this approach is defective, since not all possibilities get evaluated independently. They cite Rudolf Carnap who argues that approaching a situation with a bias is actually a more efficient way to learn and that “…other things equal, the influence of initial expectations will tend to diminish as experience grows, so that initial bias tends to wash out…” They go on to assert that the fact that if you expect your expectations to change (because it’s happened to you before) you also learn how to learn from experience.
And maybe we are literally wired this way
The authors then describe a “default mode” system in the brain that remains active when the brain is otherwise at rest. The evidence suggests that this part of the brain is constantly running simulations of possible futures: “This supports the intriguing idea that when people are at rest and not engaged in some externally directed task, their mental lives are largely occupied by imaginative simulations.”
What I take away from this
So, people are simulation machines naturally. Which means that you literally cannot eliminate “bias” from your thinking. But what you can do is embrace bias as a tool for learning (if you expect bias, and look for support or counterexamples, you can learn from your bias.)
In my Contextual Interviewing class, people are often concerned about bringing bias into their interviews. And I’ve previously assured them that it’s inevitable. Now I have some scientific evidence that’s true, as well as some reassurance. As the authors said, it’s a feature, not a bug.
Footnotes?
I probably need to figure out how to properly cite references… give me a few more days to practice with Medium. In the meantime, if you want the references, just email me, or comment?
