By Paul Rincon
BBC News science reporter
The famous skeleton from Indonesia nicknamed the "Hobbit" does not belong to a modern human pygmy with a brain disease, as some scientists argue.
That is one of the main outcomes of a detailed examination of the creature's braincase, published in Science.
The authors say their study of the Hobbit's brain supports the idea it is a new, dwarf species of human.
However, others contend the report does little to quash their theory it was actually a small, diseased person.
The remains of the small hominid from the Indonesian island of Flores were unveiled last year to worldwide acclaim.
The 18,000-year-old bones were unearthed at a site called Liang Bua, one of numerous limestone caves on Flores, and were designated LB1 and assigned to a human species new to science: Homo floresiensis.
But the skeleton of this tiny female with a brain no larger than a chimpanzee's soon became mired in controversy.
Several prominent researchers - including the Indonesian palaeoanthropologist Teuku Jacob - argued that the remains were really those of a modern human (Homo sapiens), probably a pygmy with the brain defect known as microcephaly.
Microcephaly is a pathological condition characterised by an abnormally small head and brain and usually associated with mental difficulties.
The new study did not set out to address the controversy directly - but its findings appear to scotch the microcephaly theory, the Science magazine authors say.
The study shows that H. floresiensis managed to pack a number of features of more advanced brains into its tiny skull.
This may explain signs of advanced intelligence shown by the hominid, including the ability to hunt with sophisticated stone tools.
A male Homo floresiensis may have looked something like this (Image: National Geographic)
"The overall shape of LB1's brain resembles Homo erectus (an earlier ancestor of modern humans) more than anything else. But it's its own thing," said co-author Dean Falk, an anthropologist at Florida State University in Tallahassee.
"It has some very advanced features that harken towards [modern] humans. Those features are at the frontal lobe, the temporal lobes at the sides and at the back of the brain.
"In our opinion, LB1 is not in any way, shape or form, a true microcephalic."
Unless other forms of this condition are characterised by a Homo erectus-shaped brain, says Falk and her colleagues, the theory that LB1 is a microcephalic can be rejected.
Because the brain creates a mirror image of its surface inside the skull, scientists can create a cast by applying several layers of rubber paint to the cavity. When dry, this leaves a hollow rubber model - or endocast - of the brain that can be removed.
Professor Falk and researchers at the Mallinckrodt Institute of Radiology used computed tomographic (CT) scan data on the original skull to create a 3D virtual endocast of LB1's brain.
The dimensions and features of this brain cast were then compared with endocasts from a number of different species: the chimpanzee; a normal modern human including the modern pygmy; an individual with true microcephaly; and Homo erectus.
Stone tools were found near the hobbit remains
Further comparisons were made with the ancient human-like creatures Australopithecus africanus and Paranthropus aethiopicus; and even with modern gorillas.
Despite some advanced features, LB1's brain is in other ways intriguingly primitive. The ratio of brain size to body size in H. floresiensis is more similar to that of the australopithecines than to what one would expect if Homo erectus were miniaturised.
"I thought that it was really a very good description," said hominid brain specialist Ralph Holloway, of Columbia University in New York, US.
But, he added: "I think the real weakness was using one microcephalic [skull] in their comparison. I don't think true microcephaly was the one to go with."
Professor Holloway said the team should have tried to get hold of an individual with a more generalised condition called nanocephaly.
"They've made a convincing case for ruling out pathology to 95% confidence. But I think if they had extended their sample, it would have been more interesting," he explained.
Robert Eckhardt, of Penn State University, US, and a member of the group of researchers that believes LB1 was a microcephalic, said the team's work did not sway the disease theory.
"We have some comprehensive analyses underway which I really think will resolve this question," Professor Eckhardt said.
"What I would say for now is that the specimen has multiple phenomena that I would characterise as very strange oddities and probably pathologies."
But Dr Falk was keen to emphasise that the evidence for LB1's position as a separate species extends beyond a lone skull.
"There are other remains of Homo floresiensis, so it's from a population. And it looks pretty interesting from the neck down. So multiple evidence suggests to us that it's not a microcephalic," argued Dr Falk.
"If others are interested in pursuing that line of enquiry, we welcome that and look forward to reading it in a peer-reviewed journal."
A wrangle over custody of the hominid now seems to be drawing to a close. The bones were temporarily moved from the Centre for Archaeology in Jakarta to Professor Jacob's university so the researcher could study them for himself. This prompted fears that access to the remains might be restricted.
But most of the remains have now been returned to a secure facility at their home institution.