Collective Intelligence & information pooling or aggregation

The review article on “Information aggregation and collective intelligence beyond the wisdom of crowds” describes how collective decision-making is a robust behavioural feature of groups for humans and other gregarious animals.
Pooling individual information is also fundamental for modern societies, in which digital technologies have exponentially increased the interdependence of individual group members. Cognitive and behavioural mechanisms can yield collective intelligence beyond the wisdom of crowds.
Two group decision-making situations can be distinghuised:
– consensus decision-making, in which a group consensus is required, and
– combined decision-making, in which a group consensus is not required.
In both group decision-making situations, cognitive and behavioural algorithms that capitalize on individual heterogeneity are the key for collective intelligence to emerge.

These algorithms include accuracy or expertise-weighted aggregation of individual inputs and implicit or explicit coordination of cognition and behaviour towards division of labour. These mechanisms can be implemented either as
– ‘cognitive algebra’, executed mainly within the mind of an individual or
– by some arbitrating system, or as a dynamic behavioural aggregation through social interaction of individual group members.

As one underlying principle, opt-in or opt-out behavioural mechanisms can promote col­lective intelligence further in both consensus and combined decisionmaking through capitalizing on individual heterogeneity in knowledge, skills and ability. However, many large-scale commu­nication environments in the era of ICT and IoT often lead to less ideal situations for collective intelligence to flourish in. Tribal or intergroup situations can lead to biased and inaccurate information use and greater closed-mindedness exacerbated by social media. In addition, there are many situations in which expertise is difficult to assess.

One potentially fruitful direction may be to more thoroughly explore collective intelligence in model-based decision-making. Model-based decision-making involves decision-makers having a mental representa­tion of the decision environment and how the vari­ous aspects of the environment influence outcomes. Animals (especially humans) often rely on model-based learning in which individuals can lev­erage their own belief about the structure of the envi­ronment — the environment’s model — to guide their future search and decision-making.

Model-based decision-making seems to be useful in groups of micro-size to meso-size with repeated inter­ action between relatively fixed members, but it may not be readily applicable to macro-size situations in which many individuals are temporarily and fluidly involved. A potentially useful approach to such situations may be to explore the integration of AI into human collective intelligence. Machine intelligence is increasingly likely to affect human collective behaviour, spanning human– machine cooperation, group coordination and moral judgement. Incorporating machine intelligence into group decision-making might reduce an individual’s material and psychological burden from participation. It may also enable human–machine ensembles that use each other’s advantages — and even capitalize on human behavioural biases that are persistent but are often eco­ logically rational — for better forecasting and for better health-related decisions.

Cross-disciplinary col­laborations on human–machine interaction are needed to further understand how artificial agents can shape, and be shaped by, human social environments.