Babies are Bayesians?

From the abstract of a recent paper in Science:

When 12-month-old infants view complex displays of multiple moving objects, they form time-varying expectations about future events that are a systematic and rational function of several stimulus variables. Infants’ looking times are consistent with a Bayesian ideal observer embodying abstract principles of object motion. The model explains infants’ statistical expectations and classic qualitative findings about object cognition in younger babies, not originally viewed as probabilistic inferences.

I robbed all of the above including the post title from Chris Pincock at Honest Toil.
I’ll likely look up the whole paper the next time I have an opportunity. I’ve been interested in “the Bayesian mind” for some time. If our minds are Bayesain machines, it is not hard to explain our prejudices, our unsupportable beliefs, our unfounded fears but also much of what is best about us. It also explains why experience can be a good thing and why it can be a bad thing.

One thought on “Babies are Bayesians?”

  1. I’m sure for those holding on to religious beliefs, the notion that babies are nothing but elaborate biological machines does a bad number on any strong beliefs in a soul.
    Beyond my linguistics blog, I’ve been carrying a long-standing secret interest in theories on artificial intelligence and information storage. It’s such a boundless and fascinating topic to ponder. So the baby story doesn’t surprise me. What surprises me more is how DNA can code for something so complex as problem-solving. This is surely what’s termed an “emergent” property of brain structure rather than caused directly by the genes themselves. Coincidently, programmers are trying to figure out how to harness the biological “emergence” in order to code more stable and adaptive applications. Positively mind-boggling.

Comments are closed.