Brains, Bees, Balls, and Behavior

[Eravial]

I have a huge soft spot for bee, learning, and foraging behavior research, so I'm totally into this new study that does all three!

Researchers in London co-opted innate behavioral traits of bees to make them learn a completely unnatural behavior: rolling a bee-sized ball into a "goal" to get a sugar water reward. Further, bees could be trained by other bees (or a painted "bee" on the end of a researcher-controlled stick) to complete the task! This could be considered tool-use behavior since the bees use the ball in a specific way to obtain food from their environment.

The innate traits include:
- strong spatial awareness (both with regards to knowing where to find food as well as tracking orientation of self and others within a confined space; the latter refers to the honeybee "waggle dance"),
- social learning and communication by movement (the "waggle dance" occurs in the hive, where scouts return to the hive and communicate to foragers the direction and distance to food using a coded "dance")
- exploratory behavior (bees tend to explore their surroundings, which is adaptive for foraging and evaluating the quality of a possible new hive location)
- tendency towards most efficient foraging

Regarding that last trait, a REALLY cool part of the study showed that bees essentially learned the logic/rules of the task and then used that information to make decisions that increased efficiency. The first group of bees were trained in an arena that had multiple balls, but only the farthest ball could be rolled, so they stopped trying to roll the closer balls. These bees then served as the tutors for the next group, so every time they demonstrated how to do the task to a naive bee, they showed that rolling the farthest ball led to a reward.

When those pupil bees returned to the arena alone, they nearly always chose to move the ball closest to the center, which took the least amount of energy and time to obtain the reward! Further, they still chose to do this when the nearest ball was black instead of yellow. They didn't just copy a behavior; they learned the physics of the environment and optimized a process based on that information.

And because all the best behavior studies have awesome associated videos:

[rowan]

That is SO COOL

*expects bee productivity to tank once they learn to play soccer*

[Sonic#]

When those pupil bees returned to the arena alone, they nearly always chose to move the ball closest to the center, which took the least amount of energy and time to obtain the reward! Further, they still chose to do this when the nearest ball was black instead of yellow. They didn't just copy a behavior; they learned the physics of the environment and optimized a process based on that information.

So in other words, they would follow an experienced bee's lead while being trained, but would immediately try to optimize the process when by themselves? Which might mean they learn the same lesson, but would also mean they'd be open to any changes in the situation's logic?

That's really awesome. And it provides a good illustration countering the assumption I had in high school, that behavioralism necessarily involved being programmed to do particular tasks. If the "programming" is in more general traits, why wouldn't other tasks be teachable and learnable?

[Aum]

Animals are smart and sentient. WHO KNEW

[iamthegate]

Epic! So cool! Can I be a bee when I grow up?

[Nachos]

This is really cool!

[Galen]

I can just imagine the bees filling out their course evaluations. "Professor B is cool and I loved the field trips but he's a bit set in his ways--kept going after the farthest balls instead of the TOTALLY OBVIOUS REALLY CLOSE balls. Duh!"

[Eravial]

When those pupil bees returned to the arena alone, they nearly always chose to move the ball closest to the center, which took the least amount of energy and time to obtain the reward! Further, they still chose to do this when the nearest ball was black instead of yellow. They didn't just copy a behavior; they learned the physics of the environment and optimized a process based on that information.

So in other words, they would follow an experienced bee's lead while being trained, but would immediately try to optimize the process when by themselves? Which might mean they learn the same lesson, but would also mean they'd be open to any changes in the situation's logic?

This touches on something I did find puzzling regarding the black ball experiment. It seems the bees prioritized proximity over color by going for the closer black ball when there were yellow balls in the arena too. Since the black ball was a novel environmental factor, I wonder how they "decided" that the rule was based solely on shape, rather than focusing on maintaining color consistency in completing the task. Color is very important to bees (hence the vibrancy and diversity of flowers), so this was a surprising outcome, to me. I'm sure they would have learned very quickly if that was the incorrect interpretation if they moved the black ball and did not receive food. Maybe most individuals found through their basal "calculation" of foraging cost that trying the closest ball to test the importance of color, finding it unrewarding, and then going to the closest yellow ball would be less costly than only going to the yellow ball, potentially losing out on the efficiency of using a rewarding black ball.

This is why I love foraging behavior. It's so important to fitness that the resources that go into making it flawless (and surprisingly complex) are huge.

[rowan]

Wait was there a different shape to choose from? I think it was distance they decided on, not shape? Or did I miss something. Either way, choosing a closer thing/different shape rather than the usual color/shape thing doesn't diminish your argument.