On Universality in Human Correspondence Activity

See allHide authors and affiliations

Science  25 Sep 2009:
Vol. 325, Issue 5948, pp. 1696-1700
DOI: 10.1126/science.1174562

You are currently viewing the abstract.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

Correspondence Communications

Statistical physicists and social scientists have attempted to describe human activities, in terms of physical models, and look for universal principles. Correspondence patterns are thought to be driven primarily by the need to respond to other individuals with both e-mail and letter correspondence showing power-law distributions. Because there are different exponents for the two modes of correspondence, it has been suggested that human correspondence falls into one of two universality classes and that e-mail and letter correspondence are fundamentally different activities. Now Malmgren et al. (p. 1696) tested whether human correspondence patterns are instead driven by mechanisms such as circadian cycles, task repetition, and changing communication needs. Letter correspondence, like e-mail correspondence, was accurately modeled as a cascading nonhomogeneous Poisson process giving rise to non-Gaussian statistics, but not to power-law statistics. Instead, the correspondence patterns of each individual could be uniquely characterized by the parameters of the model; that is, the process was shown to be universal, but the parameters were not. Thus, an individual's affinity toward a particular life-style will affect communication patterns, which can be modeled as a complex system.


The identification and modeling of patterns of human activity have important ramifications for applications ranging from predicting disease spread to optimizing resource allocation. Because of its relevance and availability, written correspondence provides a powerful proxy for studying human activity. One school of thought is that human correspondence is driven by responses to received correspondence, a view that requires a distinct response mechanism to explain e-mail and letter correspondence observations. We demonstrate that, like e-mail correspondence, the letter correspondence patterns of 16 writers, performers, politicians, and scientists are well described by the circadian cycle, task repetition, and changing communication needs. We confirm the universality of these mechanisms by rescaling letter and e-mail correspondence statistics to reveal their underlying similarity.

View Full Text