Skull Theory: Can a 12-Week Scan Show Boy or Girl?

What is skull theory?

Skull theory says you can tell a boy from a girl by the shape of the head in a side-on picture from the 12-week scan. The version shared online goes like this:

People hold their own scan picture up against these descriptions, or against other people's pictures, and make a call. Some then check it against the 20-week scan.

It is easy to see the appeal. A 12-week picture often shows a recognisable profile, and the theory gives you something to look at while you wait. The question is whether it works.

Is there any evidence for skull theory?

No. A search of PubMed and Europe PMC in October 2026 found no peer-reviewed study that has tested skull theory — no study in which head shape on a first-trimester scan was judged and then checked against the baby's sex. The only scientific papers we found using the phrase "skull theory" are about other subjects entirely, including the history of phrenology.

So any accuracy figure you see quoted for skull theory has no published study behind it. That does not prove it is wrong, but the research on how skulls differ between the sexes gives good reason to doubt it.

What the science of skulls actually shows

The theory borrows from something real. In adults, the skull is one of the parts of the skeleton that forensic scientists use to estimate sex, and the features they look at include the brow, the chin, the bony bump behind the ear (the mastoid process) and a ridge at the back of the skull (the nuchal crest). The brow and chin are exactly what skull theory asks you to look at.

The problem is timing. A forensic study tracked those four features on repeated skull X-rays of 10 boys and 10 girls from the Denver Growth Study. Three of the four reached their full adult form at around the same stage as the wisdom teeth came through; the chin got there earlier, but varied a lot between children. The author concluded that these cranial traits "cannot yet be used reliably for sex estimation prior to third molar eruption" — third molars being the wisdom teeth.

Attempts to tell sex from children's bones have struggled even with the actual bones in hand:

• A method claiming 81% accuracy from the shape of young children's jawbones managed only 64% when tested blind on a skeletal collection of known sex and age (the Spitalfields collection), with low consistency.
• A 2026 German study tested several methods on children's skeletons, including one based on the jaw, and checked the answers with DNA. It found "none of the methodologies proved to be suitable for reliable sex estimation on the skeleton of subadults".

If specialists cannot reliably sex a child's skeleton with the bones in their hands, there is no reason to expect anyone to do it from the outline of a 12-week baby's head on an ultrasound picture.

Do boys have bigger heads?

On average, and later in pregnancy — which is a different thing from shape. An MRI study of 200 babies in Sheffield, scanned between 18 and 37 weeks, found the total volume inside the skull grew at a significantly faster rate in boys than in girls after 24 weeks.

Boys are also heavier at birth on average: in one US study of 655 babies, 3,305 g against 3,127 g. But the spread within each group, a standard deviation of about 570 g, was more than three times the gap between them. Averages like these show up well after the 12-week scan, and as the birth weights show, individual boys and girls overlap a great deal, so they cannot sort one baby from a profile picture.

Why skull theory guesses can seem right

Because any guess at a two-way question is right about half the time, and the hits are the ones people post. A picture is also only one moment of a moving baby. The NHS notes that its 20-week scan can take longer "if your baby is moving around a lot or is not in the right position to get clear pictures", and a printout is a single frame of a baby who may have been moving or lying awkwardly when it was taken.

What can tell you at 12 weeks?

If you want an early clue from the scan, nub theory is the one with studies behind it. It uses the angle of the genital tubercle rather than the skull, and in trained hands it improves sharply over a few weeks: in one Taiwanese study, calls were right 71.9% of the time at 11 weeks, 92% at 12 weeks and 98.3% at 13 weeks, where a call could be made.

For a firm answer, see when you can find out the sex, which covers the mid-pregnancy scan and blood tests, and NIPT and what it can tell you. In England, the NHS says you may be asked at the 20-week scan whether you would like to know the sex of your baby; 20 weeks pregnant covers that appointment. What is happening at this stage is in 12 weeks pregnant, and the 12-week appointment itself is explained in the 12-week scan.

Skull theory sits with Ramzi theory and the old wives' tales: fine for fun, not for painting the nursery. If you want a game that is honest about its odds, try our Chinese gender predictor.

Frequently asked questions

Is skull theory accurate?

There is no evidence that it is. No peer-reviewed study has tested it, and research on children's skeletons shows the brow and chin features it relies on are not reliable for telling sex apart until around the time the wisdom teeth come through. A skull theory guess will be right about half the time, like any two-way guess.

Can you use skull theory at 12 weeks?

You can look, but there is nothing to say it works. The features skull theory uses are not reliable for sexing even the bones of young children, let alone a 12-week outline on ultrasound. If you want an early scan clue, nub theory is the only one with published studies, and it is a different feature.

Does skull theory work at 20 weeks?

There is no study at any week. By the 20-week scan there is a much better option: in England, the NHS says you may be asked whether you would like to know the sex. Skull shape adds nothing to what the sonographer can tell you directly, and it should not override what they say.

What is the difference between skull theory and nub theory?

Nub theory looks at the angle of the genital tubercle and has been tested in several studies, which found it unreliable at 11 weeks and much more accurate by 13 weeks in trained hands. Skull theory looks at head shape and has never been tested at all.