Last updated May 15, 2018 at 3:20 pm
Ants can solve problems as a group, even when individual ants don’t realise what the problem is.
If you’re moving a heavy piece of furniture, say a couch, and it requires two or three people, you know how hard it is to co-ordinate. Someone’s walking backwards, you get to the door and one person tries to go through while the other is insisting you should first twist it sideways, while your other friend insists it will be much easier if you just go around the back and try and push it through the back door.
It’s difficult. It’s uncomfortable. It’s inefficient.
Ants, on the other hand, are masters of co-operative transport. When you’re an ant, even if you’re relatively one of the strongest animals on earth, there’s a lot of potential food items that are going to be too big for you to drag back to the nest on your own.
There are more than 40 different species of ants that co-operate to collect food, known as group retrieval. Longhorn crazy ants are one of these species, who can together pick up and carry heavy loads, guided by pheromone trails back to their nests.
How do ants figure out how to get around obstacles?
But what happens if they come up to an obstacle? In the wild, if ants carrying a juicy worm back to their burrow come across an obstacles like a small gap they’ve got two choices – try and squeeze through the narrow opening, or find a path around it.
But how do they co-ordinate their behaviour as a group?
A group of researchers at the Weizmann Institute in Israel wanted to test group behavior when they come up to a barrier that is too small to fit the load they’re carrying through.
They developed a computer model where the carrying group of ants consists of two different types – the ‘lifters’ are the ones carrying the load, who just want to keep moving, and the ‘informed’ ones who have an idea of where the nest is from the pheromone trail and are trying to direct the movement of the group back home.
If they come up against a small gap that individual ants can fit through, but not when they’re carrying something large, you get a conflict – the ‘informed’ ones who know where the nest is want to keep following the path back through the gap, which slows up the ‘lifter’ ants who want to keep moving, but if the group keeps moving they’ll move further away from the pheromone path that marks the way home.
The model simulated the two types of actions that the ant groups could take – persisting in hanging around the opening with the hope you can squeeze the food through, or performing consistent sideways excursions to try and find a way around and back to the nest.
The model showed that the ants would keep switching between these two modes of action, to prevent them getting stuck in either one. It also predicted that the size of the object predicts the mode of motion that will predominate, groups carrying smaller objects would hang around the opening and try to squeeze through smaller bits of food, while larger objects would mean that more sideways motion would be observed.
When encountering the obstacle each ant continues to behave according to the rules that govern their free, unhindered motion – the ‘lifters’ want to lift, and the ‘informed’ ants want to just follow that pheromone trail back home. When the group encounters an obstacle, these same rules result in two different motions that allow them to overcome it. So the problem is “solved” by the collective group without any of the individuals realising the exact nature of the problem.
The computer model matched the behaviour of real life ants
They tested this with some actual ants, convincing them to carry some large plastic rings by dipping them in cat food – making them irresistible to the ants. They found that their model correctly predicted which type of behavior they’d see, with larger objects resulting in move group movement around the obstacle.
“Many social animals–from bird flocks to schools of fish to troops of baboons–perform cooperative tasks, such as foraging, hunting, and migrating,” says study senior author Nir Gov.
“Our results could provide inspiration for similar understanding of other collective animal behavior systems that exhibit coexisting modes.”
Research from PLOS Computational Biology