Memory-induced spontaneous symmetry breaking

ORAL

Abstract

From genetic encoding in DNA to memory processing in neural networks, information storage in living systems significantly influences collective behaviors, leading to phenomena from evolution to political polarization. Here we use robotic active matter to investigate how decisions based on individual memories can influence their collective behavior. We show that a collective of active agents with two rotational states can spontaneously break the symmetry with sufficiently large individual memory. Each agent we use is a spinning disk (spinner) on an airbed that either spins counterclockwise or clockwise. Thanks to the feature that a pair of same-direction spinners drop their spins upon interaction while a pair of opposite-direction spinners do not, a gyroscope in the microcontroller we mount on each spinner can infer the other spinners around it to some correctness. When each spinner is programmed to follow the majority's spin by checking a list of past interactions with other spinners they keep in the microcontroller, the collective can fall into an all-same-direction state and lock in that state for a very long time when the list size is sufficiently large. A theory considering the the memory evolution and the resultant states well captures the population dynamics of the two states and shows how memory can induce spontaneous symmetry breaking.

* I would like to thank the support from Center for the Physics of Biological Function at Princeton University.

Presenters

  • Shengkai Li

    Princeton University

Authors

  • Shengkai Li

    Princeton University

  • Trung V Phan

    Johns Hopkins University

  • Gao Wang

    University of Chinese Academy of Sciences

  • Ramzi R Khuri

    Baruch College, CUNY, Baruch College

  • Liyu Liu

    Chongqing University

  • Robert H Austin

    Princeton University