Authors

Olga Palacios¹ & Juan A. Barceló²

¹ Independent researcher, olgaapalacios@gmail.com

² Laboratory of Quantitative Archaeology, Department of Prehistory, Autonomous University of Barcelona

Abstract

The application of optimization theory has played a relevant role in archaeology, performing as a theoretical framework to model the socioeconomic decisions of prehistoric communities. Optimization theory expresses the idea that prehistoric human populations likely engaged in decision-making processes aimed at achieving the best possible outcomes given the constraints they faced. From this approach, communities would have behaved as Homo economicus, only interested in maximising their needs with the less investment, comparing benefits and expenses, and focused on their self-interest. However, the oversimplified lens of the Optimal Foraging Theory (MacArthur and Pianka, 1966), applying the rationale of modern capitalist economy to prehistoric contexts, has faced criticism for its inherent limitations and oversimplification.

Diversity within the archaeological record, observed even in similar contexts, challenges the optimization theory. Instances where the same principle does not adequately explain variations in decision-making processes are prevalent. Factors such as food preferences, intricate social relationships within the community and with neighbouring groups, as well as territorial disputes, may have led groups to avoid optimal resource management, even when it would have been beneficial for overall survival.

To explore decision-making variability beyond the constraints of optimization theory, alternative models have been proposed. Two prominent theories, Human Behavioural Ecology (Denham, 1971) and Niche Construction Theory (Laland et al., 1996; Odling-Smee et al., 1996), offer different perspectives on this matter. In Human Behavioural Ecology, decisions are primarily explained by the variability of the landscape, while Niche Construction Theory starts with the premise that human actions can actively modify the landscape. A significant contribution of the Niche Construction Theory lies in its acknowledgment that the most effective behaviour is not always optimal, as niche construction can have negative impacts on fitness.

Decision-making processes considered the individual and community's perceptions of their social and environmental landscape to make choices and select activities that would meet their needs. Understanding these processes necessitates moving beyond the traditional linear theory proposed in the Optimal Foraging Theory, represented by the concept of Homo economicus. To comprehend the variability in the archaeological record, it is crucial to consider all the variables that may have influenced decisions in the past, recognising that people would not have always acted optimally.

In game theory there are numerous examples of contexts in which people do not act rationally, such as the Prisoner’s dilemma, where rational decision-making is often challenged, we recognize the underlying principles of cooperation, trust, and strategic decision-making as relevant to understanding aspects of human behaviour in prehistory.

For achieving our aim of exploring decision-making variability, we propose to use the trial-and-error approach, introducing an iterative process where decisions taken influence the desired outcome. This approach proves useful in gathering potential options to meet goals in similar contexts. To identify the most probable decisions in specific contexts, we leverage a comprehensive database of trials-and-errors from cross-cultural ethnoarchaeological communities, filling gaps in our understanding of decision-making systems that remain elusive.

To enhance our analytical capabilities, we propose building a machine learning model, leveraging its capacity for self-learning from the collected database. Among all the available machine learning algorithms, Bayesian networks emerge as a powerful choice. Bayesian networks (Koller and Friedman, 2009), used for classification, allow output nodes to represent final classes, labels, types, or concepts, while input nodes encapsulate features related to the explanation of the outputs. Bayesian reasoning and probabilistic decision-making will be employed to define the most probable settlement patterns.

This interdisciplinary approach seeks to contribute to a more nuanced understanding of decision-making processes in prehistoric societies. By integrating insights from optimization theory, game-theory, ethnoarchaeology, and machine learning, we aim to incorporate a more comprehensive narrative that captures the essence of decision-making in the tapestry of human history. As we delve into the intricacies of ancient decision-making, we recognize that the path forward requires a synthesis of diverse methodologies, a willingness to challenge established paradigms, and an openness to the possibilities that the archaeological record holds.

Keywords: optimization theory, Prehistory, decision-making, Bayesian networks.

References

Denham, W. W. (1971). Energy Relations and Some Basic Properties of Primate Social Organization 1. American Anthropologist, 73(1), 77-95.

Koller, Daphne, Friedman, Nir. (2009). Probabilistic Graphical Models: Principles and Techniques. Cambridge, Massachusetts: MIT Press.

Laland, K. N., Odling‐Smee, F. J., & Feldman, M. W. (1996). The evolutionary consequences of niche construction: a theoretical investigation using two‐locus theory. Journal of evolutionary biology, 9(3), 293-316. https://doi.org/10.1046/j.1420-9101.1996.9030293.x

MacArthur, R. H., & Pianka, E. R. (1966). On optimal use of a patchy environment. The American Naturalist, 100(916), 603-609.

Odling-Smee, F. J., Laland, K. N., & Feldman, M. W. (1996). Niche construction. The American Naturalist, 147(4), 641-648. https://doi.org/10.1086/285870