Conformist social learning inoculates against adverse risk aversion

Given the ubiquity of potentially adverse biases incurred by trial-and-error learning, it seems paradoxical that improvements in decision-making performance through conformist social learning, a process widely considered to be bias amplification, still prevail in animal behaviour. Here we show, through model analyses and online experiments with 467 adult human subjects, that conformity can promote favourable risk taking in repeated decision making, even though many individuals are systematically biased towards suboptimal risk aversion owing to the myopia of reinforcement learning. Although positive feedback conferred by conformity could result in suboptimal informational cascades, our dynamic model of behaviour identified a key role for negative feedback that arises when a weak minority influence undermines the inherent behavioural bias. This ‘collective behavioural rescue’, emerging through coordination of positive and negative feedback, highlights a benefit of social learning in a broader range of environmental conditions than previously assumed and resolves the ostensible paradox of adaptive collective flexibility through conformity.

Recursive patterns in online echo chambers

Despite their entertainment oriented purpose, social media changed the way users access information, debate, and form their opinions. Recent studies, indeed, showed that users online tend to promote their favored narratives and thus to form polarized groups around a common system of beliefs. Confirmation bias helps to account for users’ decisions about whether to spread content, thus creating informational cascades within identifiable communities. At the same time, aggregation of favored information within those communities reinforces selective exposure and group polarization. Along this path, through a thorough quantitative analysis we approach connectivity patterns of 1.2 M Facebook users engaged with two very conflicting narratives: scientific and conspiracy news. Analyzing such data, we quantitatively investigate the effect of two mechanisms (namely challenge avoidance and reinforcement seeking) behind confirmation bias, one of the major drivers of human behavior in social media. We find that challenge avoidance mechanism triggers the emergence of two distinct and polarized groups of users (i.e., echo chambers) who also tend to be surrounded by friends having similar systems of beliefs. Through a network based approach, we show how the reinforcement seeking mechanism limits the influence of neighbors and primarily drives the selection and diffusion of contents even among like-minded users, thus fostering the formation of highly polarized sub-clusters within the same echo chamber. Finally, we show that polarized users reinforce their preexisting beliefs by leveraging the activity of their like-minded neighbors, and this trend grows with the user engagement suggesting how peer influence acts as a support for reinforcement seeking.

“Somebody Told Me,” or How Reputations Spread

This chapter deals with the communicative aspect of reputation. It talks about how reputation circulates and through which social and linguistic mechanisms it can be stabilized. A reputation can be ephemeral while at other times it seems set in stone. Gossip, rumors, and informational cascades contribute to the background noise that characterizes the universal human discussion of who did what to whom. The chapter examines the essentially communicative dimension of reputation, such as its existence not only in the eyes of others but within the cascade of communicated words and speeches that others share among themselves. Reputations can occasionally be consciously and successfully manipulated. But this does little to reduce the general anxiety and uncertainty stemming from an ungovernable transmission and propagation of reputations, the risks of defamation, and the difficulty of restoring a reputation once it has been blackened by rumors and gossip.

‘The sun without a permit’: Serbian solar politics, informational risk cascades, and the Great Disappearing Act of August 1999

In the summer of 1999, the Serbian Ministry of Health issued a public health warning about the environmental risks associated with the total solar eclipse to took place on 11 August. The warning contained a list of phantom symptoms unknown to medical profession. Some of these included severe itching, hypertension, cardiac palpitation and frequent urination. Despite the warning’s patent absurdity, the Serbian public widely observed it by seeking indoor and underground shelter from the lunar shadow, participating in what I term a ‘great public disappearing act’. By contrast, the rest of Europe and the Middle East embraced the event as a public spectacle, with millions thronging the streets and observation posts. This paper raises two key questions: Why did the Serbian government issue the odd warning? And why did the Serbian public observe it? In contrast to the conventional readings of the event as a compound effect of a political manipulation and a lack of public scientific education, I argue that the public behavior on the eclipse day was a meaningful response to the social, political and environmental circumstances in the worn-torn Serbia. Using insights from the social amplification of risk framework, I demonstrate that the great disappearing act was a paradigmatic example of herd behavior governed by the media-driven informational cascades. I further argue that the actors involved in the production and reproduction of phantom ecliptic risks – doctors, journalists, government officials, ordinary citizens – jointly enhanced their plausibility in a way that eventually eliminated the possibility of any behavior not mediated by the cascading processes of risk production.

Respecting Tradition

One way to use the ‘wisdom of crowds’ is to look at past judgements and aggregate them. We take as our model courts, which often follow a norm of respecting precedents (‘stare decisis’). The problem, however, with complete deference to past judgements is the emergence of informational cascades. If all judges respect tradition, a situation arises in which everyone sets their own private information to the side and blindly follows the majority judgement of past decisions. Possible solutions are to hide precedents, or to have some judges who are ‘stubborn’ in refusing to follow precedent, or to have some judges who only aggregate informative precedents. But these solutions involve judges not following tradition. Looking beyond courts, we caution against epistemic arguments from traditionalism more generally.

Rumour propagation and the eco-evolutionary dynamics of social information use

Information is a crucial currency for living organisms as it allows them to adjust their behaviour to environmental fluctuations. Thus, natural selection should have favoured the capacity of collecting information from different sources, including social interactions whereby individuals could quickly gain reliable information. However, such conditions may also favour the gathering of potentially detrimental information, such as false or misinterpreted accounts of environmental and social phenomena such as rumours, which may spread via informational cascades. We applied ecological and evolutionary principles to investigate how the propagation of social information at a populational level affects the propensity to assimilate it, here defined as the gullibilty. Our results show that the evolution of an individual's susceptibility to assimilate information strongly depends on eco-evolutionary feedbacks, in particular when both useful and detrimental information circulate. We discuss our results regarding the different information transmission mechanisms involved with particular attention to specific cases of social learning.

Informational cascades in financial markets: review and synthesis

Purpose The purpose of this paper is to review recent contributions to the theoretical and empirical literature on informational cascades. Design/methodology/approach This paper reviews and synthesises the existing literature, methodologies and evidence on informational cascades. Findings Many financial settings foster situations where informational cascades and herding are likely. Cascades remain mainly an area of experimental research, leaving the empirical evidence inconclusive. Existing measures have limitations that do not allow for a direct test of cascading behaviour. More accurate models and methods for empirical testing of informational cascades could provide more conclusive evidence on the matter. Practical implications Outlined findings have implications for designing policies and regulatory requirements, as well as for the design of collective decisions processes. Originality/value The paper reviews and critiques existing theory; it summarises the recent laboratory and empirical evidence and identifies issues for future research. Most of other theoretical work reviews informational cascades as a subsection of herding. This paper focusses on informational cascades specifically. It distinguishes between informational cascade and herding. The paper also reviews most recent empirical evidence on cascades, presents review and synthesis of the theoretical and empirical development on information cascades up to date, and reviews the model of informational cascades with model criticism.