Title: Anomalous cumulative inertia in human behaviour

Abstract: Human behaviour is dictated by past experiences via cumulative inertia (CI): the longer a certain behaviour has been going on, the less likely changes becomes. This is a well-known sociological phenomenon observed in employment, residence, addiction, criminal activity, wars, etc. Fundamentally, these all exhibit a growing resistance to change over time. However, quantifying the strength of this inertia is an ongoing challenge. Here we uncover anomalous cumulative inertia (ACI), ubiquitous across human behavioural patterns, with a much stronger memory dependence than previously anticipated. The behaviours undergo substantially stronger inertia, invalidating classical predictions for recovery, reconciliation, or rehabilitation times. We propose alternative models for predictions of continued anomalous behaviour, and provide means of identifying whether such behaviour is present. The result is a paradigm shift in our understanding of human activity from burstiness to inertia. Our results demonstrate how non-equilibrium models using fractional calculus aptly describe resistance to behavioural change, and produce novel predictions for e.g. rehabilitation of convicted individuals. The presence of anomalous cumulative inertia qualitatively affects the predictions which can be made for behavioural change, and thus forecasts which of these are more or less changeable. These findings are critically important for e.g. recidivism studies and public policy making, by determining more successful intervention strategies and populations where these interventions are most likely to succeed.