Input-output relationship in social communications characterized by spike train analysis

Abstract

We study the dynamical properties of human communication through different channels, i.e., short messages, phone calls, and emails, adopting techniques from neuronal spike train analysis in order to characterize the temporal fluctuations of successive inter-event times. We first measure the so-called local variation (LV) of incoming and outgoing event sequences of users, and find that these in-and out-LV values are positively correlated for short messages, and uncorrelated for phone calls and emails. Second, we analyze the response-time distribution after receiving a message to focus on the input-output relationship in each of these channels. We find that the time scales and amplitudes of response are different between the three channels. To understand the impacts of the response-time distribution on the correlations between the LV values, we develop a point process model whose activity rate is modulated by incoming and outgoing events. Numerical simulations of the model indicate that a quick response to incoming events and a refractory effect after outgoing events are key factors to reproduce the positive LV correlations.

Social physics: uncovering human behaviour from communication

“…Karsai et al (2012a) used the data from a European operator (Onnela et al, 2007b) to show that α ≈ 0.7 for calls and α ≈ 1.0 for SMSs. Aoki et al (2016) also used the data from a European operator (Tabourier et al, 2016) and found α to be 1.2 and 1.4 for calls and SMSs, respectively. Using a Chinese dataset, Jiang et al (2013) found that although the aggregate P (τ ) follows a power law, a majority of individual users show Weibull distributions for interevent times.…”

Temporal dynamics of online petitions

“…[15, 18] to decide whether L V consistently characterizes the time series. This means that the variance of L V across different periods in one time series should be smaller than the variance in the population of all time series.…”

“…This means that the variance of L V across different periods in one time series should be smaller than the variance in the population of all time series. Here we subdivide each time series in 20 slices and calculate the corresponding F-values as the ratios between the variance of L V in the population of all time series and the variances across the 20 slices [15, 18]. For different popularity classes, we show the F-values of L V in Table 1.…”

Onset of anomalous diffusion from local motion rules

“…In that case, adopting a local perspective for the walker dynamics might prove useful to test the notion of anomalous diffusion [30]. Another possible application would the modelling of diffusion on temporal networks [31], especially in the presence of burstiness [32] and the number of events within a time window can be broadly distributed, possibly under the form of trains of events [33].…”