It doesn't take a degree in human anatomy to realise that there is something unusual about the Cohuna skull. With its flat, sloping forehead and prominent brow ridge, it looks distinctly primitive.

For decades, the prehistoric Cohuna skull and others like it have occupied a central and contentious role in answering one of the most important questions in human evolutionary studies: where did our species, Homo sapiens, come from?

Most anthropologists now agree that the skulls don't demand a rewrite of the human evolution text books, but this, paradoxically, has made them all the more intriguing. It confirms that they owe their strange appearance not to the blind hand of evolution but to the guiding hand of humanity. Australia's ancient inhabitants were among the first in the world to deliberately transform the shape of their own skulls - and their motives for doing were probably not as strange as they might at first appear.

The skull was a source of controversy from almost the moment it was ploughed up in a field near the town of Cohuna in Victoria, Australia, in 1925. Some researchers enthusiastically proclaimed its strange shape and the generous coating of minerals encrusted around it showed that it predated all then known human skulls, and must therefore have belonged to one of our earliest ancestors. Sceptics, who pointed out it was found in geologically recent soil not far below the surface, were just as certain it did not. They soon won the argument. Genuinely primitive and ancient human skulls were discovered in Africa around this time, and over time anthropologists came to accept that it was there, rather than in Australia, that humanity had evolved long ago. The Cohuna skull was clearly unusual and its anomalous shape was difficult to explain, but it was quietly forgotten.

In the late 1960s, even more of the bizarre skulls were discovered

Or, at least, it was forgotten for a time. In 1948, a second skull, almost identical to the first, was unearthed during excavations at Nacurrie, barely 100 kilometres from the spot where the Cohuna skull one was found.

As if that wasn't strange enough, even more of the bizarre skulls were soon discovered. The most deformed of them all – the Coobool Creek 65 skull – was discovered about 30 kilometres from Nacurrie, and in the late 1960s it was followed by more skulls at another nearby site called Kow Swamp.

By now it was possible to estimate the age of fossils using carbon dating: the Kow Swamp specimens were about 9,000 to 13,000 years old, while the Nacurrie skull was about 11,000 years old. The Coobol Creek remains were about 14,000 years old. This makes the fossils prehistoric, just as the enthusiasts in the 1920s had suspected. But the sceptics had been correct too as human fossils found in Africa were dated at between 3 and 4 million years old. Even our species, Homo sapiens, comfortably predates the strange Australians, as skulls with all the features of modern humans are found in rocks as old as 160,000 years.

The age of the Australian skulls did raise a question, though. Their distinctive sloping foreheads and prominent brow ridges are very like those of an ancient human species called Homo erectus, that lived between 2 million and 140,000 years ago. Why did the Australian skulls look so eerily like this long-dead species?

There was an obvious explanation, according to some researchers. H. erectus must have lived on in Australia until as recently as 10,000 years ago. Others went even further, arguing that H. erectus in southeast Asia and Indonesia had gradually evolved into today's indigenous Australians, with the strange skulls evidence of the process in action.

This idea fits in with what is often called the multiregional theory for the origin of our species, and it directly contradicts the other major idea for the origin of modern humans - the Out of Africa theory. Multiregionalists suggest that our species evolved from ancient H. erectus populations living in Africa, Eurasia and Australia. Out of Africa supporters argue that modern humans evolved in Africa and gradually spread out around the world from there some 60,000 years ago.

Today most researchers favour the Out of Africa model - not least because our genes suggest all living humans can trace their ancestry back to Africa about 60,000 years ago. But some multiregionalists remain, says Darren Curnoe at the University of New South Wales in Sydney, Australia, and the unusual Australian skulls are key specimens to support their theory.

It is far more likely that their flat foreheads were sculpted by human hands

Curnoe himself once advocated a version of the multiregional theory in Australia after some initial work on the Kow Swamp skulls - but no longer. "I had a road to Damascus moment," he says, and he now thinks the skulls belong to our species, despite their strange appearance. "The so-called erectus-like features are actually not erectus features at all."

It’s true that H. erectus and the Australian skulls both had a sloped forehead and prominent brow ridges, says Curnoe, but there the similarities end. H. erectus had a wide skull and a small braincase, while the unusual Australian skulls are narrow and have large braincases, just like today’s humans do. This makes it highly unlikely that their flat foreheads were shaped by ancient H. erectus genes - and far more likely that they were actually sculpted by human hands.

"If you have a forehead that is exaggerated in flatness due to artificial deformation, combined with a naturally narrow skull, then you end up with an exaggerated brow ridge," says Curnoe. "That's exactly what we see in the Australian skulls."

The idea that the Australian skulls were artificially deformed was first suggested in the mid 1970s. Over the last 40 years the idea has gradually become the mainstream view. The deformation would have occurred when the skull's owners were infants under 12 months in age. Our skulls are soft during this period, and so parents or other adults can use boards, bandages or regular head massaging to control its growth trajectory.

Chimp and great ape skulls in general are equally soft after birth and equally prone to deformation

Artificial skull deformation is possible largely because of compromises deep in our evolutionary past. Humans, and some of our great ape relatives including the chimpanzee, have larger brains than we should for animals of our size. Most of that brain growth occurs after humans and other apes are born - and the initial growth spurt, in the first few years of life, is so rapid that young brains can grow faster than young bones. Consequently, not all of the bones in the skull are fused together when we are born, to allow the brain to push them apart as it expands. Our skull does not begin to fuse into a single solid unit until we are at least a year old.

This makes the skulls of human and great ape infants surprisingly soft and amenable to moulding into a new shape that will be locked in place when the skull does fuse together. "Chimp and great ape skulls in general are equally soft after birth and equally prone to deformation," says Christoph Zollikofer at the University of Zurich in Switzerland. "However, great apes do not have the cognitive and techno- cultural abilities to [modify their infants' skulls]."

Skull modification might have been done with the aim of making males look more masculine

Human brains began to grow particularly large, and human infant skulls became particularly pliable, about 2 million years ago. The first evidence for possible skull modification among our ancestors comes in the form of some 45,000-year-old Neanderthal skulls, but their unusual shape could equally be due to flexing of the bones following burial, says Zollikofer. There is also a 20,000-year-old skull found near Beijing in China that may have been modified, but the remains have been lost and there are questions over the true age of the skull.

This adds to the importance of the Australian skulls: they have been studied many times, and most researchers are convinced that they have been artificially modified - making them possibly the oldest good evidence we have of the practice.

Perhaps researchers would have realised that the skulls were deliberately modified when they were first discovered, if not for the fact that the alterations made them look similar to those of H. erectus. It’s unlikely that prehistoric Australia's skull shapers deliberately aimed to make their skulls look like those of the long extinct human species. Curnoe suggests that modification might instead have been done with the aim of making males look more masculine, which happens to have accentuated features also seen in H erectus. "But of course I'm speculating," he stresses.

In truth we will never know precisely why some prehistoric Australians practiced skull modification. However it is known that people from several other cultures in the more recent past also did so, and historical accounts of their motivations might provide clues. Michael Obladen at Charité University Medicine Berlin, in Germany, has compiled some of these accounts. His works reveals the practice has generally been carried out as a means to improve the social prospects of infants.

Head shaping seems to be a human cultural achievement rooting in the belief in an 'unfinished self'

In pre-Columbian America, for instance, head shaping helped elite members of society to define themselves: an elongated skull was seen as more beautiful and a sign of noble birth. The same thing applied in 19th century Nicaragua. And for some of the Native American tribes including the Chinook and the Cowlitz in 19th century North America, a forehead that had been deliberately flattened - like those of the Australian skulls - was a mark of freedom while those with rounded foreheads were looked down on.

To nobility, beauty and freedom we can add one more motive of some skull shapers. Even into the 20th century one tribe in Papua New Guinea modified their infants in the belief that it boosted the child's intelligence.

"Head shaping seems to be a human cultural achievement rooting in the belief in an 'unfinished self' - the belief that something can and must be improved in the newborn baby," says Obladen - and that's a belief that still exists even in today's developed societies.

Twenty years ago, health professionals in North America, Europe, Australia and New Zealand launched campaigns to encourage parents to place babies on their backs to sleep, as a way to reduce deaths from Sudden Infant Death Syndrome, also known as cot death. By 2002, the campaigns had paid off. In the US, the proportion of parents placing infants on their back to sleep had risen from 13% to 72% - and the number of deaths from SIDS had halved, to just 6 babies in every 10,000.

But a new problem emerged in its place. Encouraging parents to routinely putting babies to sleep on their backs before their soft skulls harden led to a dramatic increase in cases of plagiocephaly, also known as flat head syndrome. A study published last year found almost half of a sample of 440 healthy young babies attending two clinics in Calgary, Canada, showed signs of it.

We will probably never know exactly why Australia's ancient inhabitants shaped their skulls

The condition can be counteracted with a bit of intentional skull modification in the same spirit as the procedures used in prehistoric Australia, using headbands or cranial moulding helmets to return a more socially acceptable shape. It is, however, debatable whether these interventions are necessary. One medical trial found that encouraging parents to vary the sleeping position of their infant was just as effective as using a cranial moulding helmet.

In many ways, the strange Australian skulls have been gradually losing their scientific importance, first as belonging to our oldest human ancestors and then as exciting support for the multiregional theory of our species' origins. But arguably they have gained social importance, providing a unique insight into the behaviour of humans living more than 10,000 years ago. We will probably never know exactly why Australia's ancient inhabitants shaped their skulls, but all the available evidence from more recent cultures engaging in the same practice suggests they did so to improve their social prospects – mirroring the rationale for the use of cranial moulding helmets today.

In this light, perhaps the Cohuna skull is not quite so strange after all.