In the last two decades we have witnessed the overall evolution of information systems, driven by the Internet and its derivative instruments, such as social networks and instant messaging platforms. On the one hand, the vast and new availability of data has introduced a certain empiricism in academic discussions of social dynamics and online information, severely limiting the speculative impulses that characterize it. On the other hand, the modulation of content and information has led to the entertainment logic defined by the business model of the platform providing communication services. Unraveling the currents that dominate today's mare magnum of information is a complex and non-trivial task. Looking back at its origins, the Internet was born as a strategic effort to make the distribution of information safer. The infrastructure architecture, consisting of independent computers, but connected by cables, first telephony then optical, eliminates the centralization of information, which is instead distributed across various network nodes. The network strengthens the storage of information, since the possible loss of one node does not have a destructive effect on the entire structure and also facilitates access to it, thanks to the same connections between different nodes that are not arranged hierarchically.
The agility inherent in distributed architectures allows anyone to enter information online with relative ease. This orientation in multiple sources is made possible by search engines which maintain an index of sources for each topic. Therefore, if the technical aspects related to the navigation of marine information magnums are largely resolved by the power and perfection of search engines, it is the human component of the search itself and the associated economic aspects that introduce deviations from the considered optimal trajectory. use of this technology. In fact, apart from the Internet, research and interpretation of information is a fundamental act of interacting with reality, but expensive from a cognitive point of view. As a result, privileged information is one that resonates with an individual's prior knowledge which, sounding familiar, requires less effort of interpretation than completely new information. This trend, which has fallen into the digital world and combined with the timeliness of results provided by search engines, is triggering a positive feedback loop, potentially limiting the correct evaluation of the information landscape level. From an economic point of view, it should be noted that search engine positions are not neutral.
The picture has become more complicated with the emergence of social networks, entertainment tools involving billions of users, who share opinions, emotions and parts of life online. The platform enables a new way of social aggregation, asynchronous and without space restrictions, through the sharing of multimedia content. The platform structure allows for a highly horizontal and inclusive sociality. The increasing number of subscribers has created a huge demand for content, which is met by content creators, both professional and non-professional, operating using large-scale formats and registers, catering to everyone's taste. This dynamic has split the information ecosystem into a galaxy of sources with more or less loyal users. This context has permanently changed the business model of the information and communication world, revealing fierce competition for users' attention. Traditional media have lost their monopoly on topics and methods of information transmission, to the advantage of new players present in the infoosphere, often able to propose new information schemes that are more effective suited to the online environment, but are not always quality oriented. Content that engages more and catalyzes the user's attention gains greater value from an economic point of view also, because ads are delivered to them at the same time. It is estimated that by 2021 74% of advertising investment is intercepted by Amazon, Google and Facebook.
This process is largely responsible for the spread of so-called Fake News, partly or untruthfully, emotionally appealing content that surfs the web for visibility (and advertising value), potentially contaminating public debate and the democratic process. In response to this phenomenon, fact checking has been used, although it is a buffer solution for part of a much more complex problem. Traditional newspapers have suffered the greatest damage from this wave of innovation and transformation of communication tools, finding themselves playing the game on a field and with entirely new rules. Severely affected by the decline in advertising revenue, they are often forced to enforce the attractiveness of their content (for example by using click bait), sometimes to the detriment of the timeliness of information, contributing to the general impoverishment of the quality of information in circulation. The next element to consider concerns the responsibilities of published content. Traditional journalism has always had editorial responsibility, a basic principle of ensuring correct information to citizens. Platforms, on the other hand, are moving in a gray area: on the one hand they deny the role of publishers and therefore responsibility for the content they deliver; on the other hand, many of them actually apply arbitrary regulation and moderation policies to the posted content. Complex academic, political and legislative debates on this topic are still ongoing. It's difficult, for example, to hold Facebook accountable for content posted by its roughly two billion users. Currently, regulatory instruments are inadequate for this new problem, the process is evolving.
Interactions on online platforms generate huge amounts of data, the analysis of which can help shed light on the problems we are experiencing. Our research activities began in 2014 with the aim of studying the data generated by online interactions to understand how content is used and its impact on society. Initially, we focused on how users interact with content that adheres to their worldview, which may even contain intentionally false information. Second, we analyze how users react to information that conflicts with their belief system. The experiment involved a very large population: three million people in the first case, fifty-five million in the second case. The result of the first analysis is that users tend to congregate with those who express similar media diets, creating what is called an echo chamber (Del Vicario 2016a). Overall, the experiment corroborates the hypothesis of confirmation bias, namely the human tendency to acquire information that is consistent with their own vision of the world (Bessi 2015) and reject information that contradicts (Zollo 2017). Users like information that is consistent with their worldview, ignoring other versions. Indeed, if opposed, they appear to be more active in the consumption of beneficial information. But above all it is clear that information tends to be spread by homophiles, making it very difficult to permeate to harmful content (Del Vicario 2016b). To further investigate this issue, we also analyzed how information from newspapers is consumed through social media and on Facebook in particular.
We then analyzed the behavior of 376 million users who had interacted with more than twenty million news articles over a six-year period (2010-2016). The study (Schmidt 2017) confirms the previously observed dynamic: the more active a user is, the more he specializes and focuses only on his favorite sources, in a kind of loyalty dynamic, which in jargon is more technically defined as selective exposure (Cinelli 2020a). These results show that more than the problem of false information, on which the entire discussion in Fake News relies, there seems to be a problem of deconstructing information systems, because of the Internet. We divide into like-minded groups; in these groups, called echo chambers, their arguments are expressed, used to structure and reinforce shared beliefs. Thus, two or more poles are created which divide the user. This phenomenon, which appears to be the dominant element of social-networked information mechanisms, is termed “polarization”, and has been observed in many contexts of Western social debate from vaccines (Schmidt 2018, Johnson 2020), to Brexit. (Del Vicario 2017) to climate change (Falkenberg 2021). The platform has tried to change the presentation of content, implemented a moderation policy, to reduce the effects of polarization and segregation, but has avoided drastic solutions. The risk, in fact, is to put further emphasis on segregation by encouraging users to migrate to new platforms, claiming freedom of expression, in a spiral leading from echo chambers to echo platforms.
Therefore, in social media, users tend to be divided into many tribes around a narrative totem to maintain. The magmatic flow of information that we often encounter pushes us to speed up the way we use content. Time for reading and understanding is greatly reduced and new forms of language emerge through adaptation. One of the most successful is the language of memes, which uses multimedia content such as videos, photos, and images which in itself are an important part of online communication, especially in social media platforms. Online communication can thus be read through the lens of Dawkins memes (Dawkins, The Selfish Gene, Oxford University Press), a definition that applies to almost any online information vehicle. Cultural memes represent units of cultural information that are transmitted and replicated; writing posts, sharing private videos, expressing “likes” are examples of this concept. Although Dawkins' model of cultural evolution is now considered insufficient to understand the complex cultural phenomenon of information transmission, his evolutionary model is still a valid basis for describing the basic characteristics of meme diffusion. In a recent study (Valensise 2021) we investigated the role and evolution of certain types of cultural memes, namely model images that undergo modification or acquire text overlays, conventionally called memes. According to Dawkins' hypothesis, cultural memes are characterized by three essential elements of the theory of evolution: replication, variation, and selection.
In the case of visual memes, the mechanism of replication is obvious. It consists of modifying an image, for example with text, to represent a particular situation. Furthermore, meme replication is facilitated by its consistency with other cultural memes found in the online environment, such as short videos, images, or short texts. Variety is an inherent feature of visual memes. In fact, new memes are constantly being created to emphasize a humorous situation or draw attention to a political or social event and to attract the attention of users flowing through the online community. Finally, selection occurs when a meme fails to capture the user's attention or adapt to deliver new content and, therefore, disappears as it adapts to the typical speed of the online environment. Among online cultural memes that have undergone relatively strong selection, we find, for example, blogs and discussion forums that have been largely replaced by online social media; Likewise, emoji recognition greatly reduces the use of artistic ascii symbols. As a cultural sign, memes are closely linked to the larger cultural system in which they are embedded. Although its theoretical definition remains elusive and debatable, the analysis shows that memes are emerging as one of the most productive and adaptable areas of digital communication, representing a meta-language of cultural dynamics and evolving into progressive forms of textual and visual complexity.
The pandemic in the new information system
The COVID-19 pandemic has offered a very rich analytical environment, in which all the critical issues we have described have emerged, with potentially harmful implications for global health. In particular, we leveraged the novelty of this event to investigate the hypothesis that false information would have a faster speed of online dissemination than true (Vosoughi 2018). Much doubt surrounds this statement because it connotes a clear and Manichean information space which, on the other hand, is very complex. In another study (Cinelli 2020b) we compared the dissemination of information, classified according to reliability, across various platforms (Gab, Facebook, Instagram, Twitter, Reddit) without detecting substantial differences in propagation speed between different ownership classes. . This aspect was further confirmed by other researchers (Juul 2021). In contrast, we note that the dynamics are mainly dominated by polarization and echo chamber effects (Cinelli 2021). The pandemic has highlighted another important aspect of the complex relationship between information and the digital environment, which has been dubbed the "infodemic". Although its formal definition is still the subject of active research, at the heart of the concept is the wealth of information presented online in contexts related to global health. In a recent work published in CELL (Briand 2021) together with other scientists, members of the World Health Organization, and operators of the Centers for Disease Prevention (CDC) we have tried to provide guidance with respect to this process, the impact it has on society that is comparable to the issue of -Issues closely related to health, in the context of political decisions.