Stanford research team using AI to discover better taste for plant-based foods

STANFORD, Calif. (KGO) -- Migrating to plant-based foods, including alternative meat, could have a significant impact on our ability to stay Climate Ready.

And now, a team of researchers at Stanford is working to unlock an important secret to making that happen: taste.

Just watching someone bite into a burger, you can probably imagine how it tastes, but there's more to that memory than flavor. There's also the experience you get biting into it.

And now, a research team at Stanford is unraveling the role that texture plays in how we experience taste and the foods we eat. Their discoveries could ultimately help alternatives, like plant-based meat, seem a lot more like the real thing.

"The chewiness, the way it responds to biting, how you break it, the viscosity how much water comes out of it when you bite. So, all kind of reality, the mechanical properties of it," says Professor Ellen Kuhl, Ph.D., who oversees the project.

MORE: Stanford team identifies fish taxis to bring more transparency to global industry

Kuhl says the team is studying a variety of processed meat and meat alternatives, mixing mechanical measurements with AI to map their characteristic.

"So, the idea is that here we map the product onto some properties, like stiffness and texture and everything. And we do this with, and if we have more data, we're collecting more and more data. We could go the way back and we can say, 'OK, if we want this certain desired properties, can we find AI that gives us the formula for the product?'" Kuhl said.

Team member Skyler St. Pierre sliced up a quick sample to place on a benchtop stretcher. The instrument delicately pulls the slice on all four sides until it finally tears, streaming data on its strength into a software model, monitored by colleagues Jeremy McCulloch, Ethan Darwin and Thibault Vervenne.

"And when you see the force go down, that's how you can tell that the material's failed," McCulloch said.

By using multiple instruments to pull, squish and analyze, the team is developing a kind of three-dimensional database that startups and alternative food producers could leverage to improve their products.

MORE: Stanford Health Care study links some asthma inhalers to greenhouse gas

"And what we try to do is we make it open source so that little startups can jump in, can produce something and they can compare it against a whole library of data that we've collected and make it open source available for people to participate in this product design process," Kuhl said.

For the student team, it's a mission to change the world we live in, by changing the food we eat.

"So, for me, it was really more thinking about the environment and the sustainability, especially since as Americans, we're known for overeating and over consuming meat relative to the rest of the global population," St. Pierre said.

These are eating habits they're hoping to tweak by discovering why we enjoy the foods we do.

"Some of the products people did not like that much. That's an opportunity for us to come in and say, 'Hey, maybe this is the reason that people don't like them,' Maybe they're too soft. Maybe they're not soft enough," McCulloch said.

While the engineering team is focusing on texture, they say other labs could potentially add data on the chemistry of flavor, creating an even more detailed artificial intelligence recipe book for alternative meats.