Artificial Intelligence and Social Simulation: Studying Group Dynamics on a Massive Scale

[We’re pleased to welcome authors, Jesse Hoey of the University of Waterloo, Tobias Schröder of Potsdam University of Applied Sciences, Jonathan Morgan of Potsdam University of Applied Sciences, Kimberly B. Rogers of Dartmouth College, Deepak Rishi of the University of Waterloo, and Meiyappan Nagappan of the University of Waterloo. They recently published an article in Small Group Research entitled “Spotlight on Methods: Artificial Intelligence and Social Simulation: Studying Group Dynamics on a Massive Scale,” which is currently free to read for a limited time. Below, They discusses some of the findings of this research:]

SGR_72ppiRGB_powerpointTechnological and social innovations are increasingly generated through informal, distributed processes of collaboration, rather than in formal, hierarchical organizations. In this article, we present a novel combination of data-driven and model-based approaches to explore the social and psychological mechanisms motivating these modern self-organized collaborations. We focus on the example of open, collaborative software development in online collaborative networks like GitHub (github.com). The synthesized approach is based in affect control theory (ACT), and a recent framing in Artificial Intelligence known as Bayesian affect control theory (BayesACT). The general assumption of ACT is that humans are motivated in their social interactions by affective alignment: They strive for their social experiences to be coherent at a deep, emotional level with their sense of identity and general worldviews as constructed through culturally shared symbols. This alignment is used in BayesACT as a control mechanism to generate artificially intelligent agents that can learn to be functioning members of a social order (see bayesact.ca for further information).

We show in this article how such a model solves two basic problems in the social scientific study of groups and teams. First, because empirical research on groups relies on manual coding, it is hard to study groups in large numbers (the scaling problem). Second, conventional statistical methods in behavioral science often fail to capture the nonlinear interaction dynamics occurring in small groups (the dynamics problem). The ACT-based models we present allow for sophisticated machine learning techniques to be combined in a parsimonious way with validated social-psychological models of group behaviour such that both of these problems are solved in a single computational model.

The purpose of the present article is to discuss the promises of this cross-disciplinary, computational approach to the study of small group dynamics. We review computational methods for using large amounts of social media data, and connect these methods to theoretically informed models of human behaviour in groups. To use a metaphor, we are digging into digital group dynamics data with a sophisticated, artificially intelligent shovel, and showing how computational social science can be taken to a new level with this unique and novel combination of data-driven and model-based approaches. The work is an international collaboration called THEMIS.COG (themis-cog.ca) between researchers in Canada (University of Waterloo), the USA (Dartmouth College), and Germany (Potsdam University of Applied Sciences).

Stay up-to-date with the latest research from Small Group Research and sign up for email alerts today through the homepage!

Will Intelligent Machines Take Over Decision Making in Organizations?

20445410340_c1a0fe6a6a_z[We’re pleased to welcome Sukanto Bhattacharya of Deakin University. Sukanto recently published an article in Group & Organization Management with co-authors Ken Parry and Michael Cohen, entitled “Rise of the Machines: A Critical Consideration of Automated Leadership Decision Making in Organizations.”]

What if it is a machine that provides an organization’s vision for the future instead of a visionary human? Are you willing to accept a machine as your boss? What might happen if your next promotion is decided by a robot?

Intelligent machines, from automobiles to dishwashers, are increasingly making forays into every conceivable dimension of human life with a promise of making things better but perhaps not always quite delivering on that promise. Machine intelligence has permeated various levels of organizational decision-making ranging from robotic technology on production shop-floors to intelligent decision support systems for top management.

Current Issue Cover

In their recent article published in Group & Organization Management, authors Ken Parry, Michael Cohen and Sukanto Bhattacharya hypothesize a scenario where it is possible for an intelligent machine to assume the role of an organizational leader and carry out the decision-making tasks. Without engaging in a debate as to the likelihood of such a scenario, the authors present an overview of the current state of the art in artificial intelligence research, allowing readers to form their own opinion on the plausibility of such a scenario. Assuming the eventuation of such a scenario, the authors then proceed to critically consider some of the potential outcomes, both positive as well as negative, from automated organizational leadership. They posit a design framework for developing an intelligent leadership decision-making system with the objective of ensuring the positive outcomes while thwarting some of the negative (and in some cases, outright dangerous) ones. Their article aims to open up a new line of intellectual deliberations, involving organizational and management sciences on one hand and artificial intelligence as well as systems development on the other, in addressing a number of important moral/ethical issues that they identified.

The abstract for the paper:

Machines are increasingly becoming a substitute for human skills and intelligence in a number of fields where decisions that are crucial to group performance have to be taken under stringent constraints—for example, when an army contingent has to devise battlefield tactics or when a medical team has to diagnose and treat a life-threatening condition or illness. We hypothesize a scenario where similar machine-based intelligent technology is available to support, and even substitute human decision making in an organizational leadership context. We do not engage in any metaphysical debate on the plausibility of such a scenario. Rather, we contend that given what we observe in several other fields of human decision making, such a scenario may very well eventuate in the near future. We argue a number of “positives” that can be expected to emerge out of automated group and organizational leadership decision making. We also posit several anti-theses—“negatives” that can also potentially emerge from the hypothesized scenario and critically consider their implications. We aim to bring leadership and organization theorists, as well as researchers in machine intelligence, together at the discussion table for the first time and postulate that while leadership decision making in a group/organizational context could be effectively delegated to an artificial-intelligence (AI)-based decision system, this would need to be subject to the devising of crucial safeguarding conditions.

You can read “Rise of the Machines: A Critical Consideration of Automated Leadership Decision Making in Organizations” from Group & Organization Management free for the next two weeks by clicking here.

Want to stay up to date on all of the latest research published by Group & Organization Management? Click here to sign up for e-alerts! You can also follow the journal on Twitter by clicking here.

*Binary code image attributed to Christiaan Colen (CC)

Book Review: Mihnea C. Moldoveanu and Joel A. C. Baum: Epinets: The Epistemic Structure and Dynamics of Social Networks

pid_16730Mihnea C. Moldoveanu, A. C. Joel Baum: Epinets: The Epistemic Structure and Dynamics of Social Networks. Stanford, CA: Stanford Business Books, 2014. 187 pp. $44.96, hardcover.

You can read the review by Matthew S. Bothner of European School of Management and Technology and Henning Piezunka of INSEAD, available now in the OnlineFirst section of Administrative Science Quarterly.

From the review:

Epinets is a demanding and brilliant book. It demands and deserves from its audience a very close read. Its theoretical logic builds “line upon line, precept upon precept,” ASQ_v60n2_Jun2014_cover.inddand so this is not a book to be blithely perused. It also demands much of itself. Moldoveanu and Baum not only engage in an act of intellectual brokerage between epistemic game theory (and related fields) and network analysis to introduce what they refer to as epinets (networks of agents’ beliefs); they also seek—staying with Burt’s (2005) theory—to “seed and catalyze closure” (p. 162) among diverse researchers committed to the epistemic turn in social science that they propose.

Scholars from several fields should find much value in their work. These include, first and foremost, network researchers looking for fresh ideas and new methods but also organization theorists more broadly defined, as well as game theorists, strategy researchers, sociologists of knowledge and of religion, and even students of military intelligence. One of the most interesting discussions we had about Epinets took place with a German intelligence expert whose attention was riveted by the book’s core claim: that what you think others think (and what you think they think you think) matters decisively for strategic behavior. Like a shrewd spy who inserts herself in the learning loop of her country’s enemies’ spies, Moldoveanu and Baum’s ideal social actor is an embedded (though not constrained) actor who has much “level 2” knowledge—she knows what others know—and “level 3” knowledge—she can accurately predict what others think she thinks.

You can read the rest of the review from Administrative Science Quarterly by clicking here. Want to know about all the latest research and reviews from Administrative Science Quarterly? Click here to sign up for e-alerts!