Maintaining Transient Diversity Is a General Principle for Improving Collective Problem Solving

Abstract

Humans regularly solve complex problems in cooperative teams. A wide range of mechanisms have been identified that improve the quality of solutions achieved by those teams on reaching consensus. We argue that many of these mechanisms work via increasing the transient diversity of solutions while the group attempts to reach a consensus. These mechanisms can operate at the level of individual psychology (e.g., behavioral inertia), interpersonal communication (e.g., transmission noise), or group structure (e.g., sparse social networks). Transient diversity can be increased by widening the search space of possible solutions or by slowing the diffusion of information and delaying consensus. All of these mechanisms increase the quality of the solution at the cost of increased time to reach it. We review specific mechanisms that facilitate transient diversity and synthesize evidence from both empirical studies and diverse formal models—including multiarmed bandits, NK landscapes, cumulative-innovation models, and evolutionary-transmission models. Apparent exceptions to this principle occur primarily when problems are sufficiently simple that they can be solved by mere trial and error or when the incentives of team members are insufficiently aligned. This work has implications for our understanding of collective intelligence, problem solving, innovation, and cumulative cultural evolution.

Noise and opinion dynamics: how ambiguity promotes pro-majority consensus in the presence of confirmation bias

“…According to this distribution (and in line with other surveys on climate change opinions [23,72]), the majority of people are somewhat concerned about climate change (with 18% in the 'alarmed' and 33% in the 'concerned' categories), but people appear far from having reached a consensus (with significant fractions of 19% in the 'cautious', 12% in the 'disengaged', 11% in the 'doubtful' and 7% in the 'dismissive' categories in 2008) [43]. In some cases, opinion diversity can be beneficial to collective decision-making, for example, when diversity prevents conformity, groupthink or overconfidence [73,74]. However, as a large body of climate experts judge climate change as a 'threat to human well-being and planetary health' [75] and call for an 'emergency response' [76], continued disagreement among large sectors of the population may undermine the support for policies aiming to produce an adequate response.…”

Minority-group incubators and majority-group reservoirs support the diffusion of climate change adaptations

“…Empirical and theoretical work in cultural evolution has shown that homophilous, heterogeneous social structure, characterized by community structure of the networks on which innovations diffuse, promotes greater cumulative cultural complexity [15][16][17]. Diversity in social structure can support problem-solving [18,19] and prescient ideas often emerge from the peripheries of metapopulations [20]. While homophilous subgroups can promote the development of adaptations and support minority groups, social cohesion can also stifle innovation and lead to conflict [8,21].…”

The Psychology of Collectives

“…An example is the phenomenon of hysteresis (Sznajd-Weron et al, 2024), whereby collective change depends on past collective experiences. Another example are transient states of diversity in collective problem solving (Smaldino et al, 2024).…”

“…Other articles investigate collective performance and factors that foster or hinder it. Smaldino et al (2024) emphasize the importance of the transient diversity of solutions during collective problem-solving, whereby successful collectives often use a variety of mechanisms to maintain a diverse set of solutions sufficiently long to explore them but not too long to prevent reaching a common solution. Broomell and Davis-Stober (2024) show how a wisdom-of-crowds perspective can be used to understand collective reactions when collectives include influential subgroups that have a biased view of the societal problem at hand.…”

“…Other authors argue for integrating different models by identifying a general principle for systems with properties shared among the different models (Smaldino et al, 2024), drawing parallels between models of different collective phenomena such as crowd and opinion dynamics (Warren et al, 2024), or combining models of collective dynamics occurring at different scales. For example, Tump et al (2024) argue for a two-layered computational-modeling framework that combines evidence-accumulation and reinforcement-learning processes to capture the dynamic interplay of individual cognition and social systems.…”