[eng] Most epidemic spreading models assume memoryless systems and statistically independent infections.
Nevertheless, many real-life cases are manifestly time-sensitive and show strong correlations. We
propose an alternative microscopic description of the susceptible-infected-susceptible model that
accounts for cooperation, and perform simulations on degree-regular networks. Our results show
a large variety of features, including hysteresis e↵ects and loss of universality, and call for further
research.