Data-Driven Inference Reveals Distinct and Conserved Dynamic Pathways of Tool Use Emergence across Animal Taxa

Iain G Johnston; Ellen C Røyrvik

doi:10.1016/j.isci.2020.101245

Data-Driven Inference Reveals Distinct and Conserved Dynamic Pathways of Tool Use Emergence across Animal Taxa

iScience. 2020 Jun 26;23(6):101245. doi: 10.1016/j.isci.2020.101245. Epub 2020 Jun 9.

Authors

Iain G Johnston¹, Ellen C Røyrvik²

Affiliations

¹ Department of Mathematics, Faculty of Mathematics and Natural Sciences, University of Bergen, Bergen, Norway; Alan Turing Institute, London, UK. Electronic address: iain.johnston@uib.no.
² Department of Clinical Sciences, University of Bergen, Bergen, Norway.

Abstract

Tool use is a striking aspect of animal behavior, but it is hard to infer how the capacity for different types of tool use emerged across animal taxa. Here we address this question with HyperTraPS, a statistical approach that uses contemporary observations to infer the likely orderings in which different types of tool use (digging, reaching, and more) were historically acquired. Strikingly, despite differences linked to environment and family, many similarities in these appear across animal taxa, suggesting some universality in the process of tool use acquisition across different animals and environments. Four broad classes of tool use are supported, progressing from simple object manipulations (acquired relatively early) to more complex interactions and abstractions (acquired relatively late or not at all). This data-driven, comparative approach supports existing and suggests new mechanistic hypotheses, predicts future and possible unobserved behaviors, and sheds light on patterns of tool use emergence across animals.

Keywords: Biocomputational Method; Bioinformatics; Ethology; Evolutionary Biology.