Baby V’s noodley experiments teach her how the world works, a process that’s at work in Bayesian statistics.


Tonight, like almost every other night, I had the distinct pleasure of dining with my daughter. Perched in her hippopotamus highchair, Baby V feasted on beef stew, avocado, sharp cheddar cheese and a blueberry pancake. For dessert, she got a strand of linguine.

As far as I can tell, nothing is dearer to Baby V’s heart than linguine. Tonight when she spotted the noodle, her eyes electrified with laser focus, her face turned serious and she started to emit urgent “mmmm” noises. When she got it in her hands, it was all business. First, she held the strand in both hands and pulled hard enough to create nice tension but not so hard as to break it. Then, she moved the noodle to her left hand and shook it over the side of her high chair. Next, it was over to the right hand for the same shakey-shakey on the other side. After the noodle had orbited around her, it was time to get it in the mouth.

She carefully threaded the linguine so that it nestled behind her top and bottom teeth, affording her the opportunity to pull each side as she pleased, kind of like she was flossing. This went on successfully for quite some time. Finally, her new teeth (four on top and two on bottom) compromised the noodle’s integrity and it broke.

Baby V played that noodle for all it was worth. And it’s not just noodles that capture her fancy. It’s watching the key go into the front door, smelling fresh olive oil, seeing herself in the mirror, bouncing on her granddaddy’s knee, peering inside the sudsy washing machine, feeling crunchy leaves on her bare feet. Ordinary, everyday things delight her 9-month-old senses in a way that adults have long forgotten about.

As I watched her make her linguine discoveries like the little scientist she is, I’m reminded of some research I wrote about pre-Baby V (which is hard to even contemplate) that suggests babies and children are capable of sophisticated statistical reasoning. This particular brand of thinking, called Bayesian statistics, is a powerful way to make predictions about the world. Bayesian statistics work because they rely in part on a person’s prior beliefs.

Like most other adults, I’ve amassed a huge collection of prior beliefs. I am certain that my key will unlock my door tomorrow, for instance, because that’s what always happens. I don’t even entertain the idea that the key wouldn’t work.

Baby V, in contrast, is just starting to collect data about the world. She has some ideas about how things should go, but they’re not as solid as mine. She is in full-on discovery mode, and needs to scrutinize my key going into the lock every time, study the rumbling washing machine and figure out her reflection in the mirror. Every new experience is a way for her to accumulate more evidence about her world. Washing machines are noisy. Dry leaves are prickly. Dogs are sometimes wet and sometimes dry.

Far from being dull fact-finding missions, these explorations are exhilarating for Baby V. And for me. There’s nothing better than watching that girl shake her linguine. 

Laura Sanders is the neuroscience writer. She holds a Ph.D. in molecular biology from the University of Southern California.

More Stories from Science News on Health & Medicine