Skip to main content

Sequential encounters as the evolutionary drivers of choice mechanisms

Final Report Summary - SEEDCM (Sequential encounters as the evolutionary drivers of choice mechanisms)

The overarching goal of our research was to construct a theory of decision-making that applies as broadly as possible across species and takes into account insights from evolutionary biology, economics, psychology and all related sciences. To do this, our experimental work focused on specific decision problems with birds. There is a natural and substantial difference in how decision processes are conceptualised in humans and other species, because in humans introspection creates the perception (often inaccurate) that all decisions are driven by rational evaluation of each option. Since other species do not verbalize their mental processes, hypotheses about how they reach decisions are built from observations of what they actually do.

The specific target of this project was to test in non-humans whether the assumption that decision-makers take decisions by evaluating the options in front of them holds true for birds, as a springboard for an extension to other species, including humans. Previous research has found that European starlings take longer to take a lonely alternative than when they take the same alternative in a choice situation. Further, their choices can be anticipated from the times to accept each option when met alone. To deal with these findings, we proposed a model called the Sequential Choice Model (SCM). The SCM postulates that the mechanisms acting on a choice between options are the same as those present when the subjects face each option alone. The SCM incorporates the idea that these mechanisms evolved as adaptations to environments in which meeting different options simultaneously (simultaneous choice) is rare, but meeting them sequentially (sequential encounters) is common. Thus, there are no special adaptations for simultaneous choices. If more than one option is present then whichever reaches its own threshold first is acted upon. Importantly, the SCM predicts a shortening of the delay to act in simultaneous choice situations.

In the original proposal we described four main areas of interest, each exemplified by a set of particular experiments. Those were: multi-alternative environments, temporally-dependent preferences, risky choice, and comparative valuation. All these research lines have produced interesting results. The SCM was extremely precise in predicting choice from no-choice situations both when two and three options were simultaneously available. Yet, the predicted shortening of decision times was elusive: we either observed no shortening or a very small tendency in that direction. Importantly, we never observed a lengthening of decision times, as expected by the conventional, evaluation-based hypothesis.

Thus, the SCM remains a strong possibility for widespread decision processes for several reasons. First, it fits the data in the only species in which it has been exhaustively tested so far (the starling). Second, it is economical: The same system that drives action in sequential encounters results in choices when options happen to be met simultaneously. Third, it seems ecologically sensible: It is reasonable to assume that in nature foraging animals encounter feeding opportunities sequentially more often than simultaneously. As a result, natural selection has greater opportunity to optimise mechanisms that regulate behaviour in sequential than in simultaneous encounters. If mechanisms that are suitable for sequential encounters are capable of satisfactory performance and good account of data in choice situations, it seems inappropriate to postulate the existence of dedicated mechanisms for choice.

Our continuing aim is to further explore the performance of this model across a larger number of experimental protocols and animal species. For instance, we have recently tested the model in chimpanzees and gorillas in collaboration with the Max Planck Institute for Evolutionary Anthropology. Present empirical results give us reasons for optimism for the predictive power of the SCM.