Every summer, the calls of thousands of swamp sparrows can be heard across North America's wetlands. These little brown birds know only a few songs, but they know them very well. In fact, their musical set list probably hasn't changed much for centuries.

Like humans, baby swamp sparrows learn to communicate by copying adults. From a young age, they learn to copy, or mimic, songs sung by their elders. "Swamp sparrows very rarely make mistakes when they learn their songs," says biologist Robert Lachlan. In fact, their mimicry is so accurate that the music changes little between generations.

Just like children, the sparrows don't remember every song they hear. Lachlan says. "They don't just learn songs at random; they pick up commoner songs rather than rarer songs." In other words, they learn songs they hear most often. It's an example of a strategy that scientists call conformist bias. Until recently, this learning ability was thought to be special only to humans.

Between 2008 and 2009, Lachlan's research team recorded the calls of 615 male swamp sparrows across the northeastern United States. The researchers used computer software to break each song into a collection of notes, or syllables. They then measured the differences between the tunes.

The research revealed that only 2 percent of male sparrows sang a different song from the standard tune. The combination of accurate mimicry and conformist bias allows the birds to create traditions that last for centuries. "With those two ingredients together, you end up with traditions that are really stable," says Lachlan. "The song-types that you hear in the marshes（湿地）of North America today may well have been there 1,000 years ago." Lachlan's study is also among the first to measure the longevity of song traditions within a bird species.

The findings are really exciting, says scientist Andrew Farnsworth. He hopes that future research will evolve from these studies. For example, scientists may be able to identify how other animals are able to preserve their cultural traditions. "Seeing the potential for it in other organisms is super cool," says Farnsworth.

Waste heat is a worldwide energy problem most people have probably never heard about. Every machine and power station, even renewable energy like wind and solar, creates heat that is simply lost in the atmosphere. Thankfully, scientists and engineers are devoted to capturing heat and transforming it into useful electricity.

Thermophotovoltaic devices (热光电设备) are one promising way to use this waste heat. University of Michigan engineers have created a precisely machined thermophotovoltaic cell that could be used to generate electricity from surplus renewable power. The cells would use heat stored in molten salt and turn it into electricity on demand, in a form that is much cheaper than batteries, say its creators.

One of the factors important to any device's thermoelectric efficiency is its material. If higher electrical energy can be created with smaller temperature differences, then the materials are said to have a high ZT value. And a new material created by scientists in Vienna has broken the ZT value world record.

Any new thermoelectric material has to have a distinct advantage, and research is identifying a number of promising products. Pyroelectric films (热电薄膜) are aimed at waste heat emitted at less than 100 degrees Celsius (212 F) and can generate electricity when a material is either heated or cooled.

Low-temperature generation includes computers and cars, making pyroelectrics particularly useful for squeezing more energy out of electronic systems.

Another relative advantage is inexpensive materials. More abundant elements like tin, selenium, and magnesium are cheap and abundant, with high ZT values when combined in the right way. Their use could lead to greener car engines and industrial facilities.

And when it comes to personal electronics, thermoelectric could eventually see off traditional charging hardware. Chinese researchers have built a wristband that gathers body heat to power a LED and may be able to power smartwatches or other mobile devices in the future.

By the age of seven months, most children have learned that objects still exist even when they are out of sight. But it is something that self-driving cars do not have. And that is a problem. For a self-driving car, a bicycle that is momentarily hidden by a passing truck is a bicycle that has disappeared. How to give AI the reasoning ability for a seven-month-old child is now a matter of active research.

Modern Al is based on the idea of machine learning. If an engineer wants a computer to recognize a stop sign, he does not try to write thousands of lines of code that describe every pattern of pixels(像素) which could possibly indicate such a sign. Instead, he writes a program that can learn for itself, and then shows that program thousands of pictures of stop signs. Over many repetitions, the program gradually works out what features all of these pictures have in common.

Similar techniques are used to train self-driving cars to operate in traffic. But they do not understand many things a human driver takes for granted. In a recent paper in Artificial Intelligence, Mehul Bhatt of Orebro University in Sweden, describes a different approach. He and his colleagues took some existing Al programs which are used by self-driving cars and bolted onto them a piece of software. In tests, if one car momentarily blocked the sight of another, the reasoning-enhanced software could keep track of the blocked car, predict where and when it would reappear, and take steps to avoid it if necessary. The improvement was not huge. On standard tests Dr Bhatt's system scored about 5% better than existing software.

However, the question goes beyond self-driving cars to the future of Al itself." I don't think we're taking the right approach right now,"Dr Marcus, who studies psychology and neural science at New York University, says. "It's not actually the answer to AI. We haven't really solved the intelligence problem. "One way or another, then, it seems seven-month-olds still have a lot to teach machines.