The creation of phenomena in interactive biorobotics

Abstract

In so-called interactive biorobotics, robotic models of living systems interact with animals in controlled experimental settings. By observing how the focal animal reacts to the stimuli delivered by the robot, one tests hypotheses concerning the determinants of animal behaviour in social contexts. Building on previous methodological reconstructions of interactive biorobotics, this article reflects on the claim, made by several authors in the field, that this strategy may enable one to explain social phenomena in animals. The answer offered here will be negative: interactive biorobotics does not contribute to the explanation of social phenomena. However, it may greatly contribute to the study of animal behaviour by creating social phenomena in the sense discussed by Ian Hacking, i.e. by precisely defining new phenomena to be explained. It will be also suggested that interactive biorobotics can be combined with more classical robot-based approaches to the study of living systems, leading to a so-called simulation-interactive strategy for the mechanistic explanation of social behaviour in animals.

Sensory collectives in natural systems

“…However, the power of animal–robot interactions is likely as yet unrealised, in sensory-interactions and information transfer. Only by developing our understanding of the interaction between umwelt and collective behaviour may we advance the use of robotics in collective behaviour studies ( Butail et al, 2014 ; Datteri, 2021 ; Swain et al, 2012 ). A further issue in sensory manipulation experiments is that some multimodal signals are ‘fixed’, in that one component of the signal cannot be expressed without the other (e.g.…”

Going Beyond the “Synthetic Method”: New Paradigms Cross-Fertilizing Robotics and Cognitive Neuroscience

“…Robots can be used to study cognition and behavior in many ways, and this diversity can be appreciated by raising one's sight over the level of the individual studies and adopting a more general, methodological perspective. Reconstructing the methodologies adopted in the field may also end up uncovering potentially fruitful experimental strategies that have rarely or never been attempted so far (an example being the simulationinteractive methodology discussed in Datteri (2021b). It may also reveal methodological issues that distinctively arise in one or the other branch of biorobotics.…”

“…This sort of comparison has been relatively unexplored in ethorobotics, while it is made in many social cognitive neuroscience studies as illustrated in section "Bridging Humanoid Robotics and Social Cognitive Neuroscience, " where human reactions to robots are compared with human reactions to other human beings. Another comparative variant of the interactive strategy, envisaged in Datteri (2021b), would be to compare the dynamics of robot-animal (R-L * ) interaction with the dynamics of interaction between the robot and a robotic model of the focal system. This variant might be of interest for studying the internal mechanisms guiding the behavior of the focal systems.…”

Proactivity of fish and leadership of self-propelled robotic fish during interaction

“…Robots, especially micro-robots and biomimetic robots, are widely employed in the study of animal behavior nowadays, where they are able to observe and track animals without being detected or rejected by them. Due to their bionic shape and locomotion mode, biomimetic robots are capable of mingling with animal groups and even imitating animal behavior to interact [1]. Using biomimetic robots to interact with animals is an effective approach for animal interaction studying, even achieving interspecies interaction [2].…”