Dumb and Predictable Bots Make Largescale Communication More Complex and Unpredictable

Abstract

Automated communication bots follow deterministic local rules that either respond to programmed instructions or learned patterns. On the micro-level, their automated and reactive behavior makes certain parts of the communication dynamic more predictable. To the contrary, on the macro-level we find that the overall emergent communication process among humans and bots in Wikipedia becomes both more complex and less predictable. The presence of bots is the main explanatory variable for these seemingly contradictory tendencies. In short, individuals introduce bots to make communication more simple and predictable and end up making the largescale dynamic more complex and more uncertain. We also find evidence that bots make the dynamical signature of communication processes more homogeneous. We explain our results with the information processing nature of complex systems. Measures from dynamical systems theory are used to showcase a methodology that naturally lends itself to assess changes in communication dynamics provoked by algorithms.