July 14th, 2015
Several years ago, scientists published an excellent study about how desert ants find their way home after foraging. The story got a lot of media attention; unfortunately, much of the coverage described the ants “counting steps”, which isn’t what the researchers reported and feeds into existing myths rather than broadening our scope. To explain what I think is wrong with that approach, I’m going to tell you a story about ants on stilts, body swapping and how we perceive space.
The actual study was quite straightforward. Unlike many other ants, Saharan desert ants (Cataglyphis fortis) don’t use chemical trails to navigate between their nest and a food source. Instead, they use a “path integrator” to calculate the direct route home after their meandering outward journey. Scientists knew that the ants used a compass based on sunlight to judge direction, but not how they determined the distance.
Researchers had already shot down several ideas, like using visual cues or estimates of energy usage, so the team decided to test a very old idea: that the ants might be using the number of steps taken to reckon distance. The crucial experiment was to change the length of the returning worker’s steps by gluing pig bristles to their legs to artificially lengthen them. When this was done, the ants took longer strides and ended up overshooting their return trip.
The converse experiment also worked: ants with shortened legs had a shorter stride and so overestimated how far they’d gone, stopping before they had reached the nest entrance. After spending a day in the nest, the ants had adjusted to their new legs and were able to make it out and back to the feeding site without any trouble. It was a really neat experiment and did a great job of showing that these ants calculate distance based on how many steps they’ve taken. The researchers called this a “step integrator” or “loosely speaking, a step counter, although the ants most probably do not literally count”.
What did the media suggest?
So, what’s the problem? Unfortunately, a lot of the coverage of this story gives the impression the ants really are counting, which isn’t what the paper claimed. NPR wrote about it under the headline “Ants That Count!” while Livescience went with “When Ants Go Marching, They Count Their Steps“. To their credit, I think New Scientist actually covered the story quite well. A “step integrator” gives the ants a sense of distance based on how many steps they’ve taken, but that’s not the same as counting steps.
Science writers always face the challenging task of simplifying a complex subject which is often full of nuances. There’s always the risk of oversimplifying, which not only does a disservice to readers but can also feed various social myths and constructs, including how special humans are and the onward march of progress. In this case, I think the simplified version draws from and supports the pervasive and enduring myth of a chain of being, with humans at the top and “lower” animals, little more than automata, at the bottom. In fact, at its core a step integrator might not be so different from the way we build our sense of time and space — which is where the body swapping comes in.
The neuroscientists at the Brain, Body and Self lab of the Karolinska Institute in Stockholm regularly make volunteers swap bodies with different kinds of dolls and mannequins to investigate our sense of self. In 2011, they decided to see what the world would look like from bodies that were much bigger or smaller than usual. Participants wore a headset which gave them a first person view of a doll’s body lying in the same position as them. They saw the leg of the doll being stroked at the same time as someone stroked their leg; the combination of touch and sight tricked their brain into thinking the doll’s body was their own. When the doll was threatened with a knife, participants unconsciously responded — breaking into a sweat and showing a change in skin conductance.
As though body swapping wasn’t enough, the team wanted to see how this would affect the perception of size and distance. The participants consistently over- or under-estimated, depending on how big the doll was. People who had swapped bodies with a little Barbie doll thought they were in a world of giants, vastly overestimating how big or how far away everything was. Meanwhile, those who had swapped into 400 cm tall bodies felt like they were giants, thinking everything was tiny and close by. This was true regardless of whether people estimated the distances verbally or tried to walk to an object while blind-folded. It turns out that our brains use our body size to measure space; when we look through the eyes of giants or Lilliputians, everything gets scaled up or down accordingly.
Not actually counting
As the authors point out, the ants most probably aren’t literally counting their steps. Instead, the number of steps taken gives them a sense of how far they’ve travelled. The body swapping experiments show that humans do something similar, estimating distance based on the size of their body, and make the same kinds of mistakes when their stride length doesn’t match their expectations. Though the particulars of the mechanism might be very different, in both studies distance is estimated through some kind of sensation of a body — whether it’s based on stride length or body size.
Yet we persist in describing many organisms as simple automata, sensing and counting and responding, while reserving “awareness” for ourselves. This viewpoint informs how we approach the world around us, as well as how we conduct and report science. I disagree with it and am glad to see mounting evidence forunexpected faculties in other organisms. I have no idea how self-aware an ant is, but the findings fit with the possibility that they are aware of their body size and stride length, which they use to estimate distance. To me, that’s a far more exciting and interesting take on the story, broadening our conception of other organisms and their own unique perspective, rather than reinforcing our illusions of superiority and dominance.
ant behavior experiments, ants count steps, ants behavior, ants behaviour, foraging behavior of ants