Value-sensitive decision-making is an essential task for organisms at all levels of biological complexity and consists of choosing options among a set of alternatives and being rewarded according to the quality value of the chosen op- tion. Provided that the chosen option has an above-threshold quality value, value- sensitive decisions are particularly relevant in case not all of the possible options are available at decision time. This means that the decision-maker may refrain from deciding until a sufficient-quality option becomes available. Value-sensitive collec- tive decisions are interesting for swarm robotics when the options are dispersed in space (e.g., resources in a foraging problem), and may be discovered at different times. However, current design methodologies for collective decision-making often assume a well-mixed system, and clever design workarounds are suggested to deal with a heterogeneous distribution of opinions within the swarm (e.g., due to spatial constraints on the interaction network). Here, we quantify the effects of spatiality in a value-sensitive decision problem involving a swarm of 150 kilobots. We present a macroscopic model of value-sensitive decision-making inspired by house-hunting honeybees, and implement a solution for both a multiagent system and a kilobot swarm. Notably, no workaround is implemented to deal with the spatial distribu- tion of opinions within the swarm. We show how the dynamics presented by the robotic system match or depart from the model predictions in both a qualitative and quantitative way as a result of spatial constraints.
