what ants can learn

ants can learn a whole lot of things and behave in adaptive ways, but their abilities are limited. Where do these limits lie?

We are working simultaneously on answering a few questions around the learning abilities of ants. We mainly use standard Y-mazes (sometimes T-mazes) in the lab with subcolonies of Lasius niger or Linepithema humile.

idiothetic cues

Ants can easily learn to go to the left (or right) on a Y-maze to get to food. But here, they can usually see visual landmarks that tell them to move e.g. towards the brighter side. Can ants also learn to turn towards their left (or right), irrespective of which compass direction they are heading?

Idiothetic navigation means that a navigator considers its own movement as frame of reference, in this case making left or right turns, as opposed to allothetic (allocentric) navigation, which is based on an external frame of reference, like the compass directions.

In this project, we trained ants to only use ‘turn left’ (or right) as a cue to choose the sugar-rewarded arm of a Y-maze. It did not work very well under standard conditions used in other learning experiments, but when we restricted access to the landmarks, used shorter Y-maze arms, and first trained ants without giving them a choice, they did learn! This is the first demonstration of idiothetic learning ability in Lasius niger.

T-maze with long arm used in this study. In half of the trials, a box with red foil (so we can look in but the ants not out) was placed over the maze.

patterns of right-left-right-left...

Can ants learn that the sugar is always on the other side to the one they have found the reward in the previous run? That is probably not what they ever encounter in the wild, so learning this would show great cognitive flexibility.

reversal learning

We participate in the ManyManys1 global big-team collaboration with our ants. Here, ants are first trained to go towards one of 3 types of cues (color, shape, direction) and then trained to go towards the opposite cue. How many trials will it take them to relearn the cue?

odd-one-out sequences

Can ants get used to a sequence of tones, and recognize when that sequence is all of a sudden different? We are constructing devices which give the ants mild electric shocks along with sound cues, while a camera records their movements, hopefully picking up on indicators of surprise.