Don’t look now, but this spineless sea creature may be able to count better than your toddler.
Cuttlefish need to be savvy if they want to eat. They’re always on the lookout for shrimp, fish or crabs. When a cuttlefish spots a potential victim, it shoots out two specialized, sucker-bearing tentacles and nabs it. Since these hunters have to make constant judgments about which prey are worth targeting, it would make sense for them to have advanced cognitive skills—say, the ability to count.
To find out whether this was true, Tsang-I Yang and Chuan-Chin Chiao, of Taiwan’s National Tsing Hua University, brought some Sepia pharaonis cuttlefish into the lab for a math exam.
The researchers hatched cuttlefish eggs and waited until the animals were a month old. Then they started testing the young cephalopods. In each test, a cuttlefish waited on one end of a tank, and researchers lowered a two-chambered box with clear walls into the other end. A partition jutted out from the center of the box toward the cuttlefish. As it swam toward the apparatus, the cuttlefish would have to make a choice between one chamber and the other.
As soon as the cuttlefish made its decision and veered toward one side of the box, the researchers yanked the apparatus back out of the water. (This was likely annoying to the cuttlefish, but they got fed at the end of the testing session.)
To see how well the animals can count, the researchers put different numbers of shrimp into each of the box’s chambers, ranging from 1 to 5. If there were 5 shrimp in one side of the box and just 1 in the other, the choice ought to be easy. But could cuttlefish reliably choose 2 shrimp over 1, or 3 shrimp over 2? What about 5 over 4?
The scientists tested 54 cuttlefish. As expected, the animals had no problem picking a bunch of shrimp over a single shrimp. But they also passed every other test they were given. Cuttlefish were significantly more likely to pick the side of the box with more shrimp, even when choosing between 4 and 5.
Cuttlefish took longer to make these decisions as the ratios between the numbers got smaller (for example, 5 to 4 versus 4 to 3). Chiao says this is evidence that the cephalopods were actually counting the shrimp on each side, rather than judging the quantities at a glance.
Further experiments ruled out other possible explanations: Rather than counting, do cuttlefish just look for a denser batch of shrimp? No, because when researchers crowded small numbers of shrimp into tighter spaces to increase their density, the cuttlefish still picked the bigger number. Do cuttlefish simply seek out the wiggliest pile of prey? No again, as the researchers showed using boxes of dead shrimp.
Math isn’t the only factor influencing a cuttlefish’s decisions, though. Cuttlefish prefer live victims, and when researchers offered their subjects 1 live shrimp or 2 dead ones, they chose the former. When offered one big, fat shrimp or two smaller shrimp, hungry cuttlefish took the big one—but cuttlefish that had already eaten chose the two small shrimp. This might be because the big shrimp is a more tempting prize, but riskier for a young hunter to grapple with. For a cuttlefish who isn’t especially hungry, it might not be worth the risk.
Chiao says he wasn’t surprised by how well the cephalopods performed.
“We know that [the] cuttlefish has a complex brain and a sophisticated neural system,” he says. “Cuttlefish have to search for food constantly, so having number sense is important for their life.” Although the experiments stopped at 5 shrimp, Chiao suspects cuttlefish might even be able to count a little bit higher.
Similar research in year-old humans showed that they could judge the difference between 1, 2 or 3 items, but failed when quantities got any bigger. Rhesus macaques could judge between numbers going up to 4, but no higher. So this study, the authors write, “implies that cuttlefish are at least equivalent to infants and primates in terms of number sense.”
Image: by Stickpen (via Wikimedia Commons)
Yang TI, & Chiao CC (2016). Number sense and state-dependent valuation in cuttlefish. Proceedings. Biological sciences / The Royal Society, 283 (1837) PMID: 27559063