From continuous to discontinuous transitions in social diffusion
Models of social diffusion reflect processes of how new products, ideas or behaviors are adopted in a population. These models typically lead to a continuous or a discontinuous phase transition of the number of adopters as a function of a control parameter. We explore a simple model of social adoption where the agents can be in two states, either adopters or non-adopters, and can switch between these two states interacting with other agents through a network. The probability of an agent to switch from non-adopter to adopter depends on the number of adopters in her network neighborhood, the adoption threshold and the adoption coefficient , two parameters defining a Hill function. In contrast, the transition from adopter to non-adopter is spontaneous at a certain rate . In a mean-field approach, we derive the governing ordinary differential equations and show that the nature of the transition between the global non-adoption and global adoption regimes depends mostly on the balance between the probability to adopt with one and two adopters. The transition changes from continuous, via a transcritical bifurcation, to discontinuous, via a combination of a saddle-node and a transcritical bifurcation, through a supercritical pitchfork bifurcation. We characterize the full parameter space. Finally, we compare our analytical results with Montecarlo simulations on annealed and quenched degree regular networks, showing a better agreement for the annealed case. Our results show how a simple model is able to capture two seemingly very different types of transitions, i.e., continuous and discontinuous and thus unifies underlying dynamics for different systems. Furthermore the form of the adoption probability used here is based on empirical measurements.
pacs:adoption, phase transition, mean-field, social contagion, spreading
Spreading processes are ubiquitous in nature: from the contagion of diseases (Anderson and May, 1991), herd behaviour in animals (Sumpter2008, ), the diffusion of innovations (Rogers, E. M., 2010), rumour spreading (Daley, D. J. and Kendall), the evolution of social movements (GonzÃ¡lez-BailÃ³n, Sandra and Borge-Holthoefer), the propagation of hashtags in Twitter (R. Alvarez, D. Garcia), etc. All these processes share similar dynamics; in a population of initially neutral (disease-free, ignorants of some information, etc) agents (humans, animals or even bots), some of them start carrying some information, pathogen, or behavior, i.e. they adopt this innovation. Through a transmission process they can pass it on to other agents, starting in this way the process of adoption diffusion.
The diffusion of adoption has been extensively studied and modeled in several fields including Biology, Physics and Social Sciences (Goel et al., 2012; Lopez-Pintado and Watts, 2008; RevModPhys.81.591, ; RevModPhys.87.925, ). In general, new adopters have been in contact with one or several adopters, with two main mechanisms: in disease-like models (Kermack and McKendrick, 1927; Weiss and Dishon, 1971), adoption takes place with an adoption probability per contact with an adopter which is constant irrespective of the number of adopters; in threshold-like models (Lopez-Pintado and Watts, 2008; Kermack and McKendrick, 1927; Weiss and Dishon, 1971; Granovetter, 1978), adoption happens only after a critical number of adopters has been reached. There are also models of “generalized contagion” (Dodds and Watts, 2004), where both disease-like and threshold behaviors are special cases.
However, while the models describe individual adoption probabilities, most of the related empirical research was based on aggregated data, typically cumulative adoption curves (Bass, 1969; Young, 2009). Recent studies have focused on individuals’ behavior, where the number of adopters accessed by each individual can be measured (Milgram et al., 1969; Dasgupta2008, ; Romero et al., 2011; Gallup et al., 2012). These measurements have a direct connection with the form of the adoption probability. In this paper we explore the probability function obtained by (Milgram et al., 1969) from a social experiment. They analyzed the correlation between the size of a group looking at the same point in the street and the number of passerbies that joined the behavior of looking at that point. The results of the experiment can be fitted with a Hill function for the probability of adoption (Gallup et al., 2012). We will show that the shape of the adoption probability leads to two different behaviors depending on the parameter values: either a continuous or a discontinuous phase transition. This provides a simple model that describes both regimes within the same framework, depending only on two parameters; with a probability function linked to empirical data.
An agent that has not adopted yet, adopts with some probability when interacting with an adopter, which turns her an adopter-maker too. After adoption, the agent is “recovered” at a certain rate and becomes again a potential adopter. Here, we study the consequences of the probability of adoption. The transition from adopter to non-adopter is assumed to occur at some constant rate .
In the standard SIS (susceptible-infected-susceptible) model (Anderson and May, 1991), the adoption probability (from susceptible to infected, S I) is constant for each interaction with an adopter. In general, the adoption probability can be a general function of the number of adopted neighbors, :
In this contribution we will consider the function proposed by Ref. (Gallup et al., 2012)
where is persuasion capacity (similar to for and ), is the adoption coefficient (or Hill coefficient) and controls how fast/slow this probability increases with and is the adoption threshold and fixes the number of adopters needed to reach half the persuasion limit. , and are real positive numbers. This type of function is known as Hill function and has been used in models of population growth and decline (Basios et al., 2016; Gonze and Abou-Jaoudé, 2013; Santillan, 2008). The evolution of such a system in an annealed degree regular network (a network where all the nodes have the same number of neighbors or degree but where they are chosen randomly in the population at each interaction) is determined by
where is the density of adopters and is the probability of adoption given the density and is given by
The number of infected neighbors is assumed to be binomially distributed with a success probability equal to the global density of infected agents. Without loss of generality we get rid of parameter by changing the timescale and rescaling the persuasion capacity
which is equivalent to setting . The equilibrium solutions for the system are determined by the condition
Given a particular value of and , there are at most three possible solutions for (Figure 1): i) , corresponding to the adoption-free regime, ii) , represented by the upper branch and iii) , the lower branch.
The stability of the fixed points can be easily checked by linear stability analysis. The solution changes stability at
being stable for and unstable otherwise. As can be seen in Figure 1, if the solution intersects the upper branch, then that branch is stable and the solution changes stability via a transcritical bifurcation. Then for and for any initial the system will end up in the fixed point (Figure 1(A)A). If, on the contrary, the solution intersects the lower branch, this one is unstable and there is a region for which two stable solutions ( and ) coexist, separated by an unstable solution (Figure 1(B)B). For the two fixed points of opposite stability annihilate through a saddle-node bifurcation, while at we still have a transcritical bifurcation. Therefore in that region the final state of the system will be the upper branch solution if the initial density and otherwise and we can observe hysteresis. For and for any initial the system will end at . Note that is only the critical point for continuous transitions, while for discontinuous ones would be . The sign of the derivative of the function at the intersection of and the other branches determines the type of transition. If the derivative is positive ( intersects ), the transition is continuous, while if it is negative ( intersects ), the transition is discontinuous ((9a) and (9b) respectively).
For the particular case when both and coincide. For this condition one can show, by approximating Eq. 7 to third order in , that the bifurcation diagram is that one of a supercritical pitchfork bifurcation, i.e., the equation is equivalent to (Figure 1(C)C). In this case, the final fate of the system is similar to the continous case. For there is no global adoption and the system ends at , while for any initial condition will bring the system to .
Simulations using a microscopic model are also included in the plots of Figure 1. This microscopic model simulates an SIS dynamics in a degree regular network of that changes at each timestep. From one step to another, an agent is selected; if it is an adopter it recovers with probability , if not, it adopts with probability , where is the number of adopters among randomly chosen agents. There is an initial seed of infected agents which we fix to of the total population.
In pannels (A)A and (B)B of Figure 1 results of the simulations are shown in blue dots over the analytical solution. For panel (C)C, simulations are shown in panel (D)D. As can be seen, the system exhibits hysteresis in the region , where there is bistability. There system ends at or depending on the initial condition.
Fig. 1 also illustrates the two different kinds of transitions. The density of adopters stays at zero until a critical value of , where the system goes to by either a continuous transition or a discontinuous transition. As can be observed, provided a value for , the size of the jump increases with . For values of the system resembles the epidemic-like models while for values the transition is threshold-like.
Fig. 2 shows this parameters space for . The white region represents the parameters combination for a continuous transition while the light gray region corresponds to a discontinuous transition. The dark gray region is the condition that on Eq. (8), that is, that the value where both curves meet is in the range ,
This constraint implies that the in dark gray region in the plot there is only one possible solution, .
Both conditions together, Eq. 10 and 11, predict the values of the parameters for which the model shows one type of transition or another, or none. For example, in panel (B)B of Figure 2, a continuous transition is allowed for all values of and some values of , while the discontinuous transition is only possible for values of higher than 1.25 and values of higher than 1.5. As can be seen in Figure 2, for small values of , there are only continuous transitions, while for higher values of , also discontinuous transitions are allowed. Besides, the higher the value of , the more paramater space allows for solutions.
Finally, we perform simulations to characterize numerically the behavior of the system using a similar microscopic model on quenched regular random network. Again, at each timestep an agent is selected, if she is an adopter it recovers with probability , if not, she adopts with probability , where now refers to the number of adopters in her network neighborhood, which is now fixed. There is an initial seed of infected agents equal to of the total population. The long term values of the fraction of adopters are shown in Figure 3 for 10 realizations and different values of for . The realizations are not averaged to show the low dispersion (inset of upper panel in Figure 3 and lower panel of Figure 3).
As Fig. 2 indicates for and , the system exhibits always a continuous transition no matter the values of (inset of the upper panel). For and , for values of higher than 1.5 the transition is discontinuous, as shown in Figure 2. The upper panel of Figure 3 zooms in the region of the critical point for the case of . It shows the simulations of the microscopic model on a quenched degree regular random network (pink), on an annealed degree regular random network (blue) and the exact solution of the equation (black). As can be seen, there is a small discrepancy for the model on the quenched version of the network. This is because when the topology is fixed correlations appear and in particular the approximation that the infected agents are binomially distributed among the neighbors with a success probability equal to the global fraction of infected agents breaks down. As in the cases presented above, the simulations won the annealed network and the exact solution agree. For both microscopic models, the type of transition is predicted by the parameters space represented in Figure 2.
We have analyzed a model of social contagion (SIS-like) on degree regular random networks with an adoption probability measured in empirical data in (Gallup et al., 2012) that interpolates between the cases of epidemic-like spreading and threshold-like dynamics. We show that this simple model displays both continuous and discontinuous transitions from a disease-free state to an endemic state. We find the values of the parameters that separate this transitions and the critical persuasion capacities by applying standard linear stability and bifurcation theory tools.
The simplicity of the model studied here allows for relaxing some of the assumptions considered here. For example, the stability condition given by Eq. 8 resembles the structure of the critical point in the SIS model in uncorrelated random networks with arbitrary degree distributions. Following this similarity, we conjecture that the solution of our model in complex networks will be given by . Thus degree heterogeneity will lead to the vanishing threshold unless as . This can be achieved for example by considering that . Alternatively, an interesting variation is to consider that the adoption probability depends not on the absolute number of adopters but on the fraction of them. Besides, heterogeneity can emerge not only at the degree level, but also in the distributions of the adoption threshold and adoption coefficient and furthermore they can be correlated with the degree of the nodes. How heterogeneity affects the nature of the transition needs to be explored in detail. Another possible line of research is adding non-Markovianity to the dynamics, for example by letting the adoption probability depend not only on the state of the neighboring agents, but also on some internal time which takes into account when an agent tries to convince another one for adopting the innovation.
Our results highlight that not only the structure of the interaction network neither the dynamics alone are responsible of the type of transition that the system displays. Furthermore this simplified framework is able to capture this seemingly disparate types of transition, which are usually taken as a signature of different dynamics. Furthermore the choice of the adoption probability curve is based on empirical measurements from (Gallup et al., 2012), which highlights the relevance of our results for realistic modeling of social phenomena.
- Anderson and May (1991) Anderson, Roy M. and May, Robert M. Infectious diseases of humans: dynamics and control. Oxford University Press (1991).
- (2) Sumpter D, Buhl J, Biro D and Couzin I. Information transfer in moving animal groups. Theory in Biosciences 127 (2008) 177–186. doi:10.1007/s12064-008-0040-1.
- Rogers, E. M. (2010) Rogers, E. M. Diffusion of Innovations. Simon and Schuster (1991).
- Daley (D. J. and Kendall) Daley, D. J. and Kendall, D. G. Stochastic rumours. Journal of Applied Mathematics (Institute of Mathematics and Its Applications) 1 (1965) 42–55. doi:10.1093/imamat/1.1.42.
- GonzÃ¡lez-BailÃ³n (Sandra and Borge-Holthoefer) GonzÃ¡lez-BailÃ³n, Sandra and Borge-Holthoefer, Javier and Moreno, Yamir Broadcasters and Hidden Influentials in Online Protest Diffusion. American Behavioral Scientist 57 (2013) 943–965. doi:10.1177/0002764213479371.
- R. Alvarez (D. Garcia) GonzÃ¡lez-BailÃ³n, Sandra and Borge-Holthoefer, Javier and Moreno, Yamir Sentiment cascades in the 15M movement.. EPJ Data Science 4 (2015) 6. doi:10.1140/epjds/s13688-015-0042-4.
- Goel et al. (2012) Goel S, Watts DJ, Goldstein DG. The Structure of Online Diffusion Networks. Proceedings of the 13th ACM Conference on Electronic Commerce (New York, New York, USA: ACM Press) (2012), vol. 1, 623–638. doi:10.1145/2229012.2229058.
- Lopez-Pintado and Watts (2008) Lopez-Pintado D, Watts DJ. Social Influence, Binary Decisions and Collective Dynamics. Rationality and Society 20 (2008) 399–443. doi:10.1177/1043463108096787.
- (9) Castellano C, Fortunato S and Loreto, V. Statistical physics of social dynamics. Rev. Mod. Phys. 81 (2009) 591–646. doi:10.1103/RevModPhys.81.591.
- (10) Pastor-Satorras R, Castellano C, Van Mieghem P and Vespignani, A. Epidemic processes in complex networks. Rev. Mod. Phys. 87 (2015) 925–979. doi:10.1103/RevModPhys.87.925.
- Kermack and McKendrick (1927) Kermack WO, McKendrick aG. A Contribution to the Mathematical Theory of Epidemics. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences 115 (1927) 700–721. doi:10.1098/rspa.1927.0118.
- Weiss and Dishon (1971) Weiss GH, Dishon M. On the asymptotic behavior of the stochastic and deterministic models of an epidemic. Mathematical Biosciences 11 (1971) 261–265. doi:10.1016/0025-5564(71)90087-3.
- Granovetter (1978) Granovetter MS. Threshold models of collective behavior. American Journal of Sociology 83 (1978) 1420–1443. doi:10.1086/226707.
- Dodds and Watts (2004) Dodds PS, Watts DJ. Universal behavior in a generalized model of contagion. Physical Review Letters 92 (2004) 218701. doi:10.1103/PhysRevLett.92.218701.
- Bass (1969) Bass FM. A New Product Growth for Model Consumer Durables. Management Sience 15 (1969) 215–227. doi:10.1287/mnsc.1040.0264.
- Young (2009) Young HP. Innovation Diffusion in Heterogeneous Populations: Contagion, Social Influence, and Social Learning. American Economie Review 99 (2009) 1899–1924. doi:10.1257/aer.99.5.1899.
- Milgram et al. (1969) Milgram S, Bickman L, Berkowitz L. Note on the drawing power of crowds of different size. Journal of Personality and Social Psychology 13 (1969) 79–82. doi:10.1037/h0028070.
- (18) Koustuv Dasgupta, Rahul Singh, Balaji Viswanathan, Dipanjan Chakraborty, Sougata Mukherjea, Amit a Nanavati, and Anupam Joshi. Social Ties and their Relevance to Churn in Mobile Telecom Networks (2008). doi:10.1145/1353343.1353424.
- Romero et al. (2011) Romero DM, Meeder B, Kleinberg J. Differences in the mechanics of information diffusion across topics: idioms, political hashtags, and complex contagion on twitter. Proceedings of the 20th international conference on World wide web (New York, New York, USA: ACM Press) (2011), 695–704. doi:10.1145/1963405.1963503.
- Gallup et al. (2012) Gallup AC, Hale JJ, Sumpter DJT, Garnier S, Kacelnik A, Krebs JR, et al. Visual attention and the acquisition of information in human crowds. Proceedings of the National Academy of Sciences USA 109 (2012) 7245–7250. doi:10.1073/pnas.1116141109.
- Basios et al. (2016) Basios V, Nicolis S, Deneubourg J. Coordinated aggregation in complex systems:. The European Physical Journal Special Topics 225 (2016) 1143–1147. doi:10.1140/epjst/e2016-02660-5.
- Gonze and Abou-Jaoudé (2013) Gonze D, Abou-Jaoudé W. The Goodwin Model: Behind the Hill Function. PLoS ONE 8 (2013) e69573. doi:10.1371/journal.pone.0069573.
- Santillan (2008) Santillán M. On the Use of the Hill Functions in Mathematical Models of Gene Regulatory Networks. Math. Model. Nat. Phenom. 3 (2008) 85–97. doi:10.1051/mmnp:2008056.