The two-dimensional map by Bentley et al. concerns decision-making and not games. The east–west dimension is interpreted as the level at which individuals identify with some larger group. We think that this should be linked to the concept of social ties. We argue that social ties will lead to different outcomes in the “north” compared to the “south.”

The textual, big-data literature misses Bentley et al.’s message on distributions; it largely examines the first-order effects of how a single, signature distribution can predict population behaviour, neglecting second-order effects involving distributional shifts, either between signature distributions or within a given signature distribution. Indeed, Bentley et al. themselves under-emphasise the potential richness of the latter, within-distribution effects.

Bentley et al. make the deliberate choice to blur the distinction between learning and decision making. This obscures the social influence mechanisms that operate in the various empirical settings that their map aims to categorize. Useful policy prescriptions, however, require an accurate understanding of the social influence mechanisms that underlie the dynamics of popularity.

I compare the collective behavior map proposed by Bentley et al. (“BOB” for short) with a similar “balance of pressures” (BOP) map proposed by MacCoun (2012). The BOB and BOP maps have important points of convergence, but also some differences. The comparison suggests that they are analogous to different map “projections” for maps of Earth – different ways of simplifying a complex reality.

Bentley et al.'s scheme generates distributions characteristic of situations of high and low social influence on decisions and of high and low salience (“transparency”) of rewards. Another element of decisions that may influence the placement of a decision process in their map is the way in which individual decisions interact to determine the payoffs. This commentary discusses the role of Nash equilibria in game theory, focusing especially on coordination and anti-coordination games.

Bentley et al.’s two-dimensional conceptual map is complementary to cultural evolution research that has sought to explain population-level cultural dynamics in terms of individual-level behavioral processes. Here, I qualify their scheme by arguing that different social learning biases should be treated distinctly, and that the transparency of decisions is sometimes conflated with the actual underlying payoff structure of those decisions.

Recent studies provide convincing evidence that data on online information gathering, alongside massive real-world datasets, can give new insights into real-world collective decision making and can even anticipate future actions. We argue that Bentley et al.’s timely account should consider the full breadth, and, above all, the predictive power of big data.

While a timely conceptual innovation for the digital age, the “map” proposed by Bentley et al. would benefit from strengthening through the inclusion of a non–clock-time perspective. In this way, there could be new hypotheses developed which could be applied and tested relevant to more diverse societies, cultures, and individuals.

Bentley et al. argue for the social scientific contextualization of “big data” by proposing a four-quadrant model. We suggest extensions of the east–west (i.e., socially motivated versus independently motivated) decision-making dimension in light of findings from social psychology and neuroscience. We outline a method that leverages linguistic tools to connect insights across fields that address the individuals underlying big-data media streams.

Contrasting independent with socially influenced decision making does not capture crucial differences in decision making. Independence is fictional, and social influences substantially permeate preference construction. A distinction between deliberate and intuitive decision making would be more useful, and the problem in the big-data era is deciding when it is better to rely on deliberation and when to trust one's intuitions.

Outcome transparency and the weight given to social information both play important roles in decision making, but we argue that an overarching influence is the degree to which individuals can and do gather information. Evolution, experience, and development may shape individual specializations in social decision making that carry over across contexts, and these individual differences may influence collective behavior and cultural evolution.

We are sympathetic with Bentley et al.’s attempt to encompass the wisdom of crowds in a generative model, but posit that a successful attempt at using big data will include more sensitive measurements, more varied sources of information, and will also build from the indirect information available through technology, from ancillary technical features to data from brain–computer interfaces.

Bentley et al. say that economics is the science of their map's northwest quadrant, where choice is individual and transparent. This accepts the picture of the discipline common among behavioral economists who aim to drag economics southward but not eastward. In fact, leading economics journals regularly publish models located in all four quadrants, and the prominence of work from the eastern zone is increasing.

The proposed framework is insufficient to categorize and understand current evidence on decision making. There are some ambiguities in the questions asked that require additional distinctions between correctness and accuracy, decision making and learning, accuracy and confidence, and social influence and empowerment. Social learning techniques are not all the same: Behavior copying is quite different from theory passing. Sigmoidal acquisition curves are not unique to social learning and are often mistaken for other accelerating curves.

The presence of overwhelming amounts of information in our big-data era society is growing. Globalisation is increasingly giving these solicitations (regarding information) a more social aspect causing behavioural changes. While restricting my focus on this aspect of the Bentley et al article, I address related medical questions and pin-point the conceptual interest of the roadmap given therein.

Transparency is a fundamental concept in the target article by Bentley et al. But the text gives cryptic, inconclusive and sometimes conflicting suggestions as to what transparency consists of. Consequently, it is insufficiently clear what the north–south axis of the map of collective behaviour actually represents, or how to order behaviours by transparency.

The past few years have shown a major rise in network analysis of “big data” sets in the social sciences, revealing non-obvious patterns of organization and dynamic principles. We speculate that the dependency dimension – individuality versus sociality – might offer important insights into the dynamics of neurons and neuronal ensembles. Connectomic neural analyses, informed by social network theory, may be helpful in understanding underlying fundamental principles of brain organization.

While Bentley et al.’s model is very appealing, in this commentary we argue that researchers interested in big data and collective behavior, including the way humans make decisions, must account for the emotional factor. We investigate how daily choice of activities is influenced by emotions. Results indicate that mood significantly predicts people's decisions about what to do next, stressing the importance of emotional state on decision-making.

We apply Bentley et al.’s theoretical framework to better understand gender discrimination in online labor markets. Although such settings are designed to encourage employer behavior in the northwest corner of Homo economicus, actual online hiring decisions tend to drift southeast into a “confirmation bias plus weak feedback loops” pattern of discrimination based on inaccurate social stereotypes.

Social influence and payoff transparency interact with each other to influence decision making. Social influence masks payoff transparency, and lacking transparency drives people to seek social influence. Moreover, our survey supports our claim by showing that social influence and payoff transparency correlate with each other (r(53) = –.71). Bentley et al.’s model can be revised to accommodate the covariance.