Proceedings of the Annual Meeting of the Cognitive Science Society

The Ishango bone—a 10cm, approximately 20,000 year old notched bone—has been frequently claimed as evidence of human numerical competence in the Upper Paleolithic, on the basis of patterns in the collections of notches' quantities. We test whether the notch collections are suitable for university-level numerate adults to perform numerical tasks precisely. We find that performance on a range of simple numerical tasks is significantly below the precision required for exact calculation. Moreover, in an explicit counting task, we find that increasing the size of the bone to about hand-length substantially improves performance. These results are at odds with numerical interpretations of the Ishango bone, and raises the question of why the notches were produced at such a small scale given these performance difficulties and their amelioration after moderate size increases.