Cross-Validated Neural and Biophysical Models for Pediatric Fingerprint Ridge Breadth Age Estimation with Archaeological Applications
This study establishes a cross-validated neural network model using a dedicated pediatric fingerprint reference from Rhodes to demonstrate that Neolithic clay figurines from Knossos were likely manufactured by children, significantly improving age estimation accuracy over previous adult-calibrated methods.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
In the quiet corners of ancient homes, small clay figures often sit among the debris of daily life. For decades, archaeologists have debated who made these objects. Were they crafted by adults as part of a serious ritual, or were they the playful, imperfect creations of children? The answer matters because it changes how we understand the lives of children in the Neolithic period. If children made them, it suggests they were active participants in their culture, not just passive observers. The problem is that clay does not preserve bones or DNA, and once a figurine is broken, the person who held it is gone. However, there is one clue that survives the firing process: the tiny ridges of a fingerprint. Just as a person's height grows, the spacing between the ridges on their fingertips expands as they age, starting narrow in childhood and widening until adulthood. This biological fact offers a way to measure the age of the maker without needing to see their face or skeleton, turning a simple fingerprint into a clock that ticks only while a child is growing.
A team of researchers set out to test whether this biological clock could finally settle the debate over the Neolithic figurines found at Knossos on the island of Crete. To do this, they first needed to build a reliable reference guide, because previous methods had been tested mostly on adults and failed when applied to children. The team traveled to Rhodes, Greece, where they worked with 200 children between the ages of six and twelve. Each child pressed their thumb into a small ball of soft clay, creating a fresh fingerprint impression. The researchers then measured the distance between the ridges in these prints with extreme precision. They used these measurements to train three different types of computer models to predict a child's age based solely on the width of those ridges. One model followed a standard mathematical curve, another simulated how skin tissue spreads and changes over time, and the third was a complex neural network, a type of artificial intelligence designed to find patterns in data that humans might miss.
The results showed that the artificial intelligence model was by far the most accurate. When the researchers tested it by hiding one child's data at a time and asking the model to guess the age, the average error was just under one and a half years. This was a massive improvement over older methods, which had been off by more than seven years when applied to children. The study also confirmed that the spacing of the ridges changes differently for boys and girls as they get older, with the gap widening more noticeably in boys between the ages of ten and twelve. This level of detail allowed the team to create a new, highly specific tool for reading the ages of ancient makers.
With this new tool in hand, the researchers turned their attention to eleven small clay figurines from the Neolithic period, held in museums in Heraklion and at the site of Knossos itself. They carefully measured the ridge spacing preserved in the clay of these ancient figures. Because clay shrinks when it is fired in a kiln, the team ran their calculations assuming the clay might have shrunk anywhere from nothing to fifteen percent. Even with the maximum amount of shrinkage, the calculations consistently pointed to the same conclusion: the fingerprints belonged to children. The estimated ages ranged from about four to twelve years old. In every scenario, the measurements fell squarely within the childhood range, never suggesting an adult hand.
This finding provides the first direct, physical evidence that children were indeed making these figurines. It moves the idea of child participation from a guess based on the small size of the objects to a fact supported by the biology of the makers' hands. The study does not claim to know exactly how many children made these figures or what specific rituals they were performing, but it definitively rules out the idea that these objects were exclusively the work of adults. By combining a simple biological fact with modern computing, the researchers have opened a window into the daily lives of Neolithic children, showing that their hands, like ours, were busy shaping the world around them.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.