Modelling opinion dynamics in the age of algorithmic personalisation

Nicola Perra,Luis E C Rocha

doi:10.1038/s41598-019-43830-2

Abstract

Modern technology has drastically changed the way we interact and consume information. For example, online social platforms allow for seamless communication exchanges at an unprecedented scale. However, we are still bounded by cognitive and temporal constraints. Our attention is limited and extremely valuable. Algorithmic personalisation has become a standard approach to tackle the information overload problem. As result, the exposure to our friends’ opinions and our perception about important issues might be distorted. However, the effects of algorithmic gatekeeping on our hyper-connected society are poorly understood. Here, we devise an opinion dynamics model where individuals are connected through a social network and adopt opinions as function of the view points they are exposed to. We apply various filtering algorithms that select the opinions shown to each user (i) at random (ii) considering time ordering or (iii) its current opinion. Furthermore, we investigate the interplay between such mechanisms and crucial features of real networks. We found that algorithmic filtering might influence opinions’ share and distributions, especially in case information is biased towards the current opinion of each user. These effects are reinforced in networks featuring topological and spatial correlations where echo chambers and polarisation emerge. Conversely, heterogeneity in connectivity patterns reduces such tendency. We consider also a scenario where one opinion, through nudging, is centrally pushed to all users. Interestingly, even minimal nudging is able to change the status quo moving it towards the desired view point. Our findings suggest that simple filtering algorithms might be powerful tools to regulate opinion dynamics taking place on social networks.

Highlights

Various disciplines, as for example Sociology, Psychology and Behavioral Genetics, investigate the mechanisms leading to opinion formation in groups of people[1,2,3,4,5]
We study the effects of sorting algorithms on the group and on the individual opinion dynamics by first looking at methods REF, OLD, REC and preference method (PR)
If we start the networks with two opinions A and B, in equal proportions (PA = 0.5 and PB = 0.5) and uniformly distributed among the users, the prevalence of both opinions remains stable around the starting values for all four sorting methods for various network configurations

Summary

Introduction

As for example Sociology, Psychology and Behavioral Genetics, investigate the mechanisms leading to opinion formation in groups of people[1,2,3,4,5]. While the details of the algorithm behind the personalisation are a corporate secret, we know that they are based on a combination of explicit information we provide such as demographics, likes, friendship relations, interests and implicit information derived from previous interactions with the platform, IP addresses, locations, cookies, posts age, page relationships, and content quality etc[28,31]. All this information is fused with three main principles of content curation: popularity, semantic and collaborative filtering[32]. At its roots, it depends on the specific definition of democracy considered[33]

Methods

Results

Conclusion