Menu

Facing our Past with Facial Reconstruction

Facing our Past with Facial Reconstruction

You’ve probably seen facial reconstructions using human remains in museums or news stories – but how are they created and why?

What is facial reconstruction?

Facial reconstruction, also known as facial approximation, refers to the process of building up the soft tissue face of an individual from their skull. It’s a creative practice informed by scientific insight from the fields of anatomy, anthropology, osteology, dentistry, and orthodontics. It’s often used in archaeology to depict the faces of people from the past and you may have seen examples in museums or on TV.

Digital image of a human skull built up with muscle, then the facial reconstruction of the individual on the right

Archaeological facial reconstruction of the “Griffin Warrior”, excavated from a Bronze Age tomb in Pylos, Greece. Reconstruction shows: A. the skull; B. placement of eyes and tissue depth markers; C. modelled muscles of facial expression and mastication; D. artistically textured face (Image: own work; part of a University of Cincinnati project directed by Prof. Jack Davis and Dr. Sharon Stocker)

How are facial reconstructions made?

Although you may not realise it at first, everyone’s skull is unique. The shape and size of our skull directly influences our facial appearance. Experts use scientific guidelines that refer to the shape of the skull, and involve various measurements from the skull, to estimate the face.

Typically, reconstructions are performed by manually modelling clay over a skull replica, or by digitally sculpting in 3D over a skull scan using specialist computer modelling software. Sometimes, reconstructions are also performed in 2D and involve sketching over a photograph of a skull using tracing film overlays (below).

Collage of three images: three students working on focial reconstructions, one on a computer, another on a clay model and the last drawing in pencil

Forensic Art students at the Centre for Anatomy and Human Identification, University of Dundee, working on facial reconstructions: A. 3D computerised reconstruction; B. manual clay reconstruction; C. 2D reconstruction (Image: University of Dundee)

To start a reconstruction, the skull is positioned in the Frankfurt horizontal plane, which mimics the natural position of the living head when upright and at rest. Several markers of different lengths are placed across various points of the skull. These represent the average recorded depth of soft tissues at those set landmarks.

Next the superficial muscles of facial expression and mastication (chewing) are applied. Each muscle has its own distinct placement, which will be influenced by the shape of the skull. Finally, the skin is added.

Numerous guides, using various skull measurements and observed shape patterns, are used to model the finer details of the eyes, nose, mouth, and ears. For example, the height of the lips is calculated using measurements taken from the central incisors (front teeth); and the chin shape (and whether it is smooth or dimpled) is indicated by the shape of the lower jawbone.

Why are facial reconstructions performed?

Facial reconstruction has two different applications. It’s used within forensics to support unidentified cold-case investigations. Reconstructions are presented to see if members of the public can recognise them, and in turn generate new leads to help establish the identity of the deceased, which can then be confirmed using DNA or dental record evidence.

In archaeology, they offer today’s world a valuable opportunity to come face to face with those who roamed the earth before us. They have allowed us a glimpse into the lives of many, including (but not limited to) ancient royalty, fallen warriors, those accused of witchcraft, and even early hominids from over four million years ago.

They offer a captivating means of educating the audience about past peoples. The bones might present evidence of healed trauma or disease, which would be depicted on the reconstruction if influencing the face. Referring to historic records reconstructions can also be used to demonstrate trends in fashion and hairstyle, their understood living conditions, and general health.

Photo of a 3D digital model of a human head with the reconstructed digital face of a woman on the right

Example of an archaeological facial reconstruction performed on a medieval woman that was found near Blackness Castle in Scotland (Image: courtesy of Mirja Koponen; skull courtesy of the Anatomical Museum, University of Edinburgh)

Above is an example of a local Scottish reconstruction of a woman found buried at Blackness Castle. Blackness Castle was built sometime in the early 15th century and functioned as an important port for trade and defence for the royal burgh of Linlithgow. It also acted as an essential crossing point to Fife, for those traveling to pilgrimage sites in the north.

The woman would have likely formed a part of this bustling medieval community surrounding the castle. Studying her skeleton, archaeologists estimated that she was approximately 18 years old at death. They also believed she was a mother, because she was found buried with a small child. To complete her facial reconstruction, a basic interpretation of hairstyle and fashion is illustrated using a general understanding of medieval fashion.

How reliable are facial reconstructions?

Facial reconstruction is an evolving specialism, with scientists continually working on new and improved guidelines to increase its accuracy. It’s generally understood that an exact portrait of an individual cannot be produced, but a sufficient likeness that prompts recognition can be achieved.

This has been identified in numerous lab experiments that involve performing reconstructions on skulls where actual face images are available for later comparison (see Facial reconstruction – anatomical art or artistic anatomy? – Wilkinson – 2010 – Journal of Anatomy – Wiley Online Library).

Layout of images of six male faces with a digital reconstruction at the top

Example of how a face matching test might look. Can you recognise the face that matches with the facial reconstruction? (Image: Dr Tobias Houlton; includes open-source portrait photos)

Scientists have measured accuracy using matching tests, which are designed to identify how well people can match the reconstructed face against possible actual face images (above). In another type of study, they have also generated colour deviation maps that compare how well the surface of a 3D facial reconstruction physically aligns with a 3D surface scan of the target face (below).

This has been performed using Computerised Tomography (CT) scans of living persons, which capture 3D images of an individual’s skull (for reconstruction) and face (for comparison). Results have been promising and are further supported by the numerous facial reconstruction success cases within real-world forensic investigations.

Digital scan of different angles of a human face taken using a CT scanner

The accuracy of a 3D facial reconstruction can be measured by aligning it with a 3D scan of the target face and generating colour deviation maps. Green surfaces demonstrate a perfect alignment between the reconstruction and target face. When the reconstruction presents a thicker surface that exceeds the target face, the map is coloured red. Where the reconstruction underestimates the target face, the map is coloured blue (Image: Dr Tobias Houlton)

Facial reconstructions are likely to become more accurate as technology advances – who knows what they’ll look like in 100 years?


Ready to dig deeper into the science behind archaeological investigation? Discover radiocarbon dating, landscape survey and more in our Decoding Archaeology series.

Dr Tobias Houlton is a facial anthropologist, and programme lead for the MSc in Forensic Art and Facial Imaging course at the University of Dundee.

Header Image: University of Dundee


Uncover More