Researchers from Griffith University have started a world-first study. They hope to understand the purpose of ancient finger flutings—marks made by human fingers in a soft mineral film known as moonmilk on cave walls. Dr. Andrea Jalandoni, a digital archaeologist with the Griffith Center for Social and Cultural Research, is leading the study. The team hopes to use contemporary technology to never assume who those mysterious marks might be created by — helping to advance the new field of digital archaeology along the way.
Finger flutings have bedeviled archaeologists for generations. The most ancient of these, at least 300,000 years old, have been attributed to Neanderthals in France. Today, you can see these salt-licking markings in pitch-black caves across Europe as well as Australia. They raise important and fundamental questions about who made them. Griffith University researchers used machine learning techniques to address this problem. Through panoramic renderings, they wanted to shed light on the people who may have been making those ancient impressions.
Utilizing Modern Technology
Collaborative research in digital archaeology lays the foundation for investigating the origin and meaning of finger flutings. This agnostic methodology is a unique blending of machine learning and controlled testing. We ran two such experiments involving a total of 96 adult subjects. In these clayfluting experiments, transdisciplinary participant-subjects produced finger flutings utilizing a moonmilk clay analog and virtually via immersive reality/VR with Meta Quest 3. The goal was to gather the kind of data that might allow us to comprehend these precocious artists.
3D representations of all developed flutings were precisely recorded and arranged. This dataset formed the basis for training two standard image-recognition models to query specific features of the flutings. The results proved encouraging, showing that machine learning could serve as a valuable new tool in archaeological research.
“The goal of this research was to avoid those assumptions and use digital archaeology instead,” – Dr. Jalandoni
The paper uses a novel methodological approach to try to shed light on the demographics of finger fluting creators. In particular, it looks at the differences between the sexes among these grownups. Dr. Jalandoni stressed the significance of determining whether these engravings were created by males or females. This knowledge would be revolutionary for our understanding of the social hierarchies and organization of prehistoric human societies.
Mixed Results with Promising Insights
The experience gained from this research produced inconsistent findings, but did reveal some hopeful findings. Dr. Gervase Tuxworth Senior Lecturer in Information Science, School of Information and Communication Technology is a member of the project team. He shared success stories from the machine learning models.
“Under one training condition, models reached about 84% accuracy, and one model achieved a relatively strong discrimination score,” – Dr. Gervase Tuxworth
The findings suggest that further calibration is needed for the models to improve their predictive power. This extension would expand their ability to recognize distinguishing traits in finger flutings. This exciting development in digital archaeology is just one way that researchers hope to find out more about the cultural practices of ancient humans.
Dr. Robert Haubt, co-author and Information Scientist from the Australian Research Center for Human Evolution (ARCHE), expressed commitment to transparency in their research process. He noted that their code is now public-facing. This lets other people reproduce or challenge their findings.
“We’ve released the code and materials so others can replicate the experiment, critique it, and scale it,” – Dr. Robert Haubt
Future Implications for Archaeology
This study is more than just an exercise in determining the authorship of finger flutings. Finally, it shows how digital archaeology can reshape and broaden the public discourse on the past and present of human civilization. The innovative use of technological tools in archaeological explorations can open up a more detailed understanding of how ancient humans interacted with each other and their environment.
Dr. Haubt emphasized that transparency and reproducibility are crucial for developing credibility in scientific research. He touched on the need for better teamwork in the scientific world. He continued, “And that’s how a proof of concept becomes a trusted and dependable tool.”
As archaeologists develop their techniques and adopt emerging technologies, the future holds great potential for digital archaeology. Research like this is leading to more exciting scientific breakthroughs. Our ancient ancestors—and their creative impulses—hold many more mysteries to reveal, and scholars are ready to discover even more.