IF YOU have ever concluded that intelligence is in short supply in the modern world, perhaps you are looking in the wrong place. Humanity may seem to be suffering from collective stupidity, but there are still plenty of smarts to be found elsewhere.
You will be familiar with the clever antics of whales and dolphins, chimpanzees and orangutans. But what about wasps? . Or crabs? They use stinging anemones to defend themselves against predators. Then there are alligators that place sticks on their snouts to catch egrets looking for nesting material. And mosquitoes can learn to avoid pesticides after a single taste. . A parasitic vine called a dodder sniffs out its prey with remarkable discernment, for example. Blobby yellow things called slime moulds can learn and teach each other. Even biofilms – collectives of bacteria – possess short-term memory and the ability to make decisions.
Such an astonishing array of aptitude is rather unsettling. It also raises some fundamental questions, including what actually is intelligence, how did it evolve and how do the abilities of various organisms compare? These are hard to answer. Evaluating intelligence in nature is tricky, particularly in life forms that are very different from us. But now a group of neuroscientists, AI researchers and philosophers have come up with a radical idea. They want to create a periodic table of intelligence akin to the one used to categorise the chemical elements. If they succeed, it could radically alter the way we see other species – and ourselves.
Advertisement
It is a bold plan. When the Russian chemist Dmitri Mendeleev invented the periodic table in 1869, his contemporaries immediately recognised it as a monumental feat. It has been called “nature’s Rosetta Stone” and is still seen as one of science’s foremost breakthroughs. By ordering the chemical elements into columns and rows, the table transformed the somewhat mystical field of chemistry (think alchemy) into a powerhouse of hard science. Researchers need only note the position of an element in the table to tell whether it might react vigorously with water, for example, or is likely to conduct electricity. The table has even predicted elements that were subsequently discovered.
Researchers working on a project called hope that a can categorise nature’s varied cognitive abilities in a similar way. But where even to begin? The chemical version sorts elements by their atomic number or atomic mass. For intelligence – broadly defined as an organism’s ability to flexibly respond to a changing environment – researchers are still looking for a criterion that is equally straightforward.
It certainly isn’t brain size, as was long thought. Adult human brains weigh between 1.2 and 1.5 kilograms, contain about 86 billion neurons and make up about 2 per cent of our body weight. But a whale’s brain can weigh 9 kilograms, an African elephant has some 257 billion neurons and the brain of a shrew – a small, mouse-like creature not renowned for its smarts – comprises about 10 per cent of its body mass. What’s more, honeybees seem able to count up to 5 and grasp the abstract concept of zero with a brain measuring .
Behaviour might be a better way to categorise cleverness. Take the courting male cuttlefish that positions himself between a female and his rival and displays female colours on the side of his body facing his competitor and male ones on the side that the female can see. Or consider a rook. If this bird is presented with a treat floating in a water-filled vessel, it drops stones into the liquid until the level has risen high enough to bring the tasty morsel into reach.
But testing for intelligence through behaviour is difficult. Recognising oneself in a mirror is seen as a sign of advanced cognition. Dolphins, magpies and manta rays can do it – but dogs typically can’t. Does this reflect a lack of intelligence in canines or perhaps something else, such as their reliance more on smell than vision? Likewise, many organisms live in environments that are starkly different from ours and so might use senses that we don’t even possess. A shark in its watery, buoyant, blue-green world senses nearby animals through their electrical field. How does that shape its cognition?
Nevertheless, the researchers behind the Diverse Intelligences initiative think intelligence might become more transparent through a combination of behavioural and neuroanatomical features. “We’re going to ask, are there kinds of intelligence, and can we identify structural features that are organisational of those kinds of intelligence?” says at Macquarie University in Australia. “If we can, then we are starting to identify things that could be thought about as possible dimensions [of intelligence].”
Smart selection
Evolution will be a guide. Intelligence seems to have emerged many times and brains take a variety of forms. The researchers hope to gather crucial insights by looking at these different architectures. Jellyfish, for instance, possess a decentralised network of neurons, in which all nerve cells tend to be connected to all others. Worms and sea slugs have a series of dispersed neuron clusters. In vertebrates, the neurons work in loops that feed back into each other. And so on. In each of these architectures, information is organised, processed and spread differently.
Consider the cephalopods, a class of animal that includes octopuses and cuttlefish. They diverged from the evolutionary path that resulted in humans about 600 million years ago, and the architecture of our brains is remarkably different. An octopus possesses about 550 million neurons, of which around 160 million sit in its large optic lobes and 42 million inside a brain shaped like a doughnut because the oesophagus runs through it. Even more neurons – about 350 million – are distributed among the animal’s eight arms, which constantly explore the environment and process information, basically “thinking” independently. Yet cephalopods seem highly intelligent. An octopus will easily open a jar – an object that hasn’t been part of its evolution. And cuttlefish can resist the temptation to eat a treat to get rewarded with a better one a while later. In this version of the marshmallow test, they manage to delay gratification for up to 2 minutes, broadly similar to chimpanzees.
Barron and his colleagues suspect that different neuronal architectures have distinct cognitive advantages and limitations. Decentralised information processing might be faster, for example. Conversely, channelling everything through a central brain could create cognitive bottlenecks. By scouring the literature and performing new experiments, the team will probe the capabilities of a range of animals representative of various cognitive architectures to find out whether a particular brain structure translates into identifiable abilities and limits to learning, which can be used to categorise intelligence into groups. “The breakthrough with the chemical periodic table was realising that there are structural properties in the elements that organise the table,” says Barron. “When you have that, suddenly you have explanatory power.” He hopes to even incorporate artificial intelligence into the table, to see how it differs from natural intelligence.
“Looking at how organisms pull together information makes sense,” says , who researches animal cognition at the Max Planck Institute for the Science of Human History in Jena, Germany. She isn’t involved in the project, but welcomes it. As do others. We need to think about “how we can break down our understanding of intelligence into more atomic parts so that we can see how different kinds of intelligence cut across the tree of life”, says at the University of California, San Diego. He hosts a podcast called Many Minds and is working on a somewhat-related project that tries to map out non-human cognition into an .
However, not everyone is convinced that a periodic table of intelligence can succeed. “Is it meritorious? Obviously. Is it realistic? I’m not so sure,” says at the University of Iowa, who has spent five decades studying the cognitive abilities of eight species, including humans. It is extremely difficult even to systematise intelligent behaviour, he says. “There are so many variables that critically determine the results. It’s mind-boggling.”
Blurred lines
Barron admits that the whole concept might be completely wrong. “There could be no clear boundaries between the different dimensions of intelligence – they could be blurred,” he says. If this were the case, there would be no meaningful structure and, hence, no periodic table. “But even if it falls over, we’ll have learned a lot,” he says.
That might include a deeper appreciation and sense of wonder for the multitudes of intelligence around us. “Whenever you ask, is this species cleverer than we thought?, the answer is yes,” says Cooperrider. “It’s like overturning rocks in the stream bed and we’ve only overturned an extremely small percentage of the rocks so far. Every time we turn one over, we find something cool.”
The project could also fundamentally change how we view other species. “Whether or not an animal is intelligent often informs how we think that animal should be practically and ethically treated,” , one of the project’s researchers and a philosopher of cognitive science at the University of Cambridge. “Some argue that the ability to act intentionally and flexibly is connected to autonomy, and autonomy is sufficient for personhood.” Personhood, in turn, might entitle an animal to legal protections. Already, lawyers have argued in court cases that chimps deserve rights usually reserved for humans, including freedom from unlawful incarceration – although unsuccessfully so far.
We might even learn something about ourselves by trying to work out where we sit on a periodic table of intelligence. Humans don’t outperform other species as much as you might think. Take Ayumu, a chimpanzee taught to remember random sequences of nine numbers that a screen flashes at him for only 60 milliseconds – less than the blink of an eye. Humans don’t do very well at this task. But Ayumu easily types out the digits in the right order. Once, he competed against the – and won handily.
Despite such shortcomings, human intelligence does seem to be special, and constructing a periodic table might help highlight how. “There is something different about it because other organisms on this planet are not having conversations about this,” says Barron. “I don’t think it’s a story of human exceptionalism. But there is something interesting about human intelligence and it will be lovely to understand what that is.”
