At the 2018 International Conference on Social Computing, Behavioral-Cultural Modeling, & Prediction and Behavior Representation in Modeling and Simulation (SBP-BRiMS), I presented a short paper titled “Formal Organizations, Informal Networks, and Work Flow: An Agent-Based Model” (PDF).
The paper presents initial findings of a continuing project (view/follow project on ResearchGate) to develop and refine a generalized organizational agent-based model that includes both formal organization hierarchy (i.e., a so-called “organization chart”) and the informal networks that really matter in a company (i.e., what David Krackhardt and Jeffrey R. Hanson aptly called “the company behind the chart“). Such a generalized model would be useful to create simulations of a variety of individual and organizational processes at multiple levels (e.g., employees, managers, executives, and overall organization) and to precisely quantify processes as the simulations unfold.
Initial findings from early model runs suggest potential decreases in both individual and organizational productivity as supervisory span-of-control increases in organizations with cultures of micromanagement.
Below you can read the paper abstract and find out more information about the model.
Few computational network models contrasting formal organization and informal networks have been published. A generalized organizational agent-based model (ABM) containing both formal organizational hierarchy and informal social networks was developed to simulate organizational processes that occur over both formal network ties and informal networks. Preliminary results from the current effort demonstrate “traffic jams” of work at the problematic middle manager level, which varies with the degree of micromanagement culture and supervisory span of control. Results also indicate that some informal network ties are used reciprocally while others are practically unidirectional.
Keywords: organizations, networks, ABM, boundary spanning
My model will be made available under the Apache 2.0 license from OpenABM for others to use in their own research. Please feel free to use, refine, or extend this model with attribution.
Briggs T.W. (2018) Formal Organizations, Informal Networks, and Work Flow: An Agent-Based Model. In: Thomson R., Dancy C., Hyder A., Bisgin H. (eds) Social, Cultural, and Behavioral Modeling. SBP-BRiMS 2018. Lecture Notes in Computer Science, vol 10899. Springer, Cham https://doi.org/10.1007/978-3-319-93372-6_21 (PDF)
At the Association of Threat Assessment Professionals (ATAP) DC chapter, I had the opportunity to share my perspectives on using a computational social science / complexity science approach for the prevention/mitigation of mass violence.
The ATAP DC group convened in person in Northern Virginia and by videoconference in several other locations along the East Coast. (I’m very grateful to the technical staff at Northern Virginia Community College for all their prework to make sure the technology worked and everything ran smoothly!). The ATAP DC members were a fantastic audience and humored me for what I understand was a slightly different take on mass violence than their usual.
Following a brief overview of computational social science and complexity science, I discussed some of the challenges of researching mass violence: mass violence is rare, complex, difficult to study, lacks agreed-upon theoretical models of causation, and is unfortunately often politicized.
We discussed different types of modeling, from verbal and mental models to mathematical and computational models. I believe that computational models are particularly suited to studying mass violence, and previously constructed one such model – Active Shooter: An Agent-Based Model of Unarmed Resistance. Computational models offer many benefits for mass violence research, education, and training, including the fact that they pose no risk to human subjects, they are infinitely repeatable, they are superlative for studying processes in systems, and can incorporate network features to study the influence of network ties in a particular process or outcome.
I demonstrated several computational models, including my active shooter ABM, as well as Epstein’s civil violence model and an epidemic model showing the spread of a virus between populations. If mass violence is, at least in part, germinated through the spread of the idea of perpetrating mass violence, whether by mass media or the internet, such models are useful in exploring how quickly and broadly such ideas could spread.
Finally, I discussed my view that the cumulative strain model proposed by Levin & Madfis is a verbal model that is ripe for a computational implementation.
I concluded by sharing my view that the threat assessment/threat prevention community could make use of computational modeling for training, for research, and perhaps ultimately for pre-warning. Computational modelers have demonstrated the value of working collaboratively with process stakeholders – for example, key officials in threat planning and response in schools and organizations – to perform “participatory” or “companion” modeling, in which stakeholder input is used to iteratively refine a model such that it is useful for the stakeholders in policy development or in response planning.
In the discussion following my presentation, I received several excellent and thoughtful questions, including whether psychopathy could be represented in agents (yes, through an additional modeling effort), whether my prior model could be extended to include armed responders or law enforcement officers (yes), whether these models can be validated (yes, but validation is challenging for many of the same reasons that mass violence research is difficult), and whether such models can be used for pre-scenario planning (yes).
I thoroughly enjoyed my time with ATAP DC and appreciated the opportunity to contribute.
Mass violence is a rare event. Attempts to empirically study episodes of mass violence can present a variety of challenges. The complex nature of episodes of mass violence, which may have germinated in years prior to actual attacks, make attempts to use conventional statistical techniques problematic.
Complexity science and the relatively new field of computational social science offer new paradigms and computational tools suited to the study of this dynamic problem. This talk reviews some of the challenges of mass violence research, provides an overview of complexity and computational social science, offers a live demonstration of a computational model of an active shooter scenario, and discusses a potential use case to computationally implement the cumulative strain model proposed by Levin and Madfis in 2009.
Why is this research important? Computational approaches enable new and innovative ways of studying, thinking about, and communicating with stakeholders about mass violence, and should become part of the threat assessment community toolkit.
Under conditions of true complexity–where the knowledge required exceeds that of any individual and unpredictability reigns–efforts to dictate every step from the center will fail. People need room to act and adapt. Yet they cannot succeed as isolated individuals, either–that is anarchy. Instead, they require a seemingly contradictory mix of freedom and expectation–expectation to coordinate, for example, and also to measure progress toward common goals.
Brenda Zimmerman and Sholom Glouberman have proposed a distinction among three different kinds of problems in the world: the simple, the complicated, and the complex.
Simple problems, they note, are ones like baking a cake from a mix. There is a recipe. Sometimes there are a few basic techniques to learn. But once these are mastered, following the recipe brings a high likelihood of success.
Complicated problems are ones like sending a rocket to the moon. They can sometimes be broken down into a series of simple problems. But there is no straightforward recipe. Success frequently requires multiple people, often multiple teams, and specialized expertise. Unanticipated difficulties are frequent. Timing and coordination become serious concerns.
Complex problems are ones like raising a child. Once you learn how to send a rocket to the moon, you can repeat the process with other rockets and perfect it. One rocket is like another rocket. But not so with raising a child, Zimmerman and Glouberman point out. Every child is unique. Although raising one child may provide experience, it does not guarantee success with the next child. Expertise is valuable but most certainly not sufficient. Indeed, the next child may require an entirely different approach from the previous one.
And this brings up another feature of complex problems: their outcomes remain highly uncertain.
Yet we all know that it is possible to raise a child well.
It’s complex, that’s all.
The Woman and the Hen
A Woman had a Hen that laid an egg every day. The Fowl was of superior breed, and the eggs were very fine, and sold for a good price. The Woman thought that by giving the Hen twice as much food as she had been in the habit of giving, the bird might be brought to lay two eggs a day instead of one. So the quantity of food was doubled. The Hen thereupon grew very fat, and stopped laying altogether.
It is obvious that effects depend upon causes, but causes also, in a subtle sense, depend upon effects. Every cause itself is an effect of its own causes, which preceded it, and therefore arises in dependence upon its respective causes…effects arise in dependence upon causes. Here cause and effect are in a temporal sequence, an effect occurring after its cause.
Because the designation of something as a “cause” depends upon consideration of its effect, in this sense a cause depends upon its effect. Something is not a cause in and of itself; it is named a “cause” in relation to its effect. Here the effect does not occur before its cause, and its cause does not come into being after its effect; it is in thinking of its future effect that we designate something as a cause.
Agent and action depend upon each other. An action is posited in dependence upon an agent, and an agent is posited in dependence upon an action. An action arises in dependence upon an agent, and an agent arises in dependence upon an action. Nevertheless, they are not related in the same way as cause and effect, since the one is not produced before the other.
How is it that, in general, things are relative?
How is it that a cause is relative to its effect?
It is because it is not established in and of itself. If that were the case, a cause would not need to depend on its effect. But there is no self-sufficient cause, which is why we do not find anything in and of itself when we analytically examine a cause, despite its appearance to our everyday mind that each thing has its own self-contained being.
Because things are under the influence of something other than themselves, the designation of something as a cause necessarily depends upon consideration of its effect.
—Tenzin Gyatso, the 14th Dalai Lama
Linearity is a reductionist’s dream, and nonlinearity can be a reductionist’s nightmare.
Are you risk-literate?
Do you understand how reputable cancer treatment centers in the U.S. lie or mislead you by confounding statistics in their marketing?
Which preventive cancer screenings cause more harm than good? Which preventive cancer screenings are worth getting?
Gerd Gigerenzer’s Risk Savvy (find in a library) is a crash course in risk literacy and a fun romp through several areas in which understanding risk and uncertainty matters enormously: your health and medical care (including defensive medicine and preventive screenings), bank finance and your money, leadership, romance, terrorism, and various runaway panics – Do you remember mad cow disease and how many people died from it?*
I first encountered Gigerenzer while studying cognitive models that could be implemented in computational agents, and specifically, his highly-cited papers on “fast and frugal” heuristics for us boundedly-rational mortals. [Note to economists: this includes you.] While his academic papers are quite accessible, Risk Savvy (find in a library) feels approachable to an even wider audience.
What’s particularly fun and enjoyable about the book is that Gigerenzer doesn’t just pick on laypeople for not understanding risk, but he also picks on experts for not only communicating risk so poorly, but for often lacking risk literacy, themselves. He backs this up with experimental data collected from physicians, bankers, and executives, and uses as a foil some of the best experts at developing simple solutions to complex problems: children.
Gigerenzer includes practical tools that could revolutionize the way we communicate and think about risk – for example, discontinuing the use of relative risks with unspecified or poorly specified reference classes (e.g., a 20 percent risk reduction!) and instead using absolute risks (e.g., a reduction in risk from 5 in 1000 to 4 in 1000).
Gigerenzer also advocates the use of icon boxes and fact boxes (examples available from the Harding Center for Risk Literacy) when health professionals and health-related organizations communicate with individuals.
The concluding chapter suggests ways that we might revolutionize school by teaching risk literacy using simple tools from a very early age. Gigerenzer specifically focuses on applied statistical thinking, rules of thumb, and psychology of risk and suggests focusing these in three areas: health literacy, financial literacy, and digital risk competence. He backs his suggestions with experimental data demonstrating that children as young as the second grade can, when presented with statistical information in the proper format, learn to accurately calculate risk.
Overall, I enjoyed Gerd Gigerenzer’s Risk Savvy (find in a library) and recommend it to just about anyone, and especially to those working in health-related professions and anyone interested in making better decisions about their health, finances, and other areas of life.
*Over 10 years, about 150 people in all of Europe died of mad cow disease. In the same ten years, the other cause that led to an equivalent number dying was drinking scented lamp oil.
The gross national product does not allow for the health of our children, the quality of their education or the joy of their play. It does not include the beauty of our poetry or the strength of our marriages, the intelligence of our public debate or the integrity of our public officials.
It measures neither our wit nor our courage, neither our wisdom nor our learning, neither our compassion nor our devotion to our country, it measures everything in short, except that which makes life worthwhile.
—Robert F. Kennedy
1. Everything we think we know about the world is a model. Every word and every language is a model. All maps and statistics, books and databases, equations and computer programs are models. So are the ways I picture the world in my head–my mental models. None of these is or ever will be the real world.
2. Our models usually have a strong congruence with the world. That is why we are such a successful species in the biosphere. Especially complex and sophisticated are the mental models we develop from direct, intimate experience of nature, people, and organizations immediately around us.
3. However, and conversely, our models fall far short of representing the world fully. That is why we make mistakes and why we are regularly surprised. In our heads, we can keep track of only a few variables at one time. We often draw illogical conclusions from accurate assumptions, or logical conclusions from inaccurate assumptions. Most of us, for instance, are surprised by the amount of growth an exponential process can generate. Few of us can intuit how to damp oscillations in a complex system.
You can’t navigate well in an interconnected, feedback-dominated world unless you take your eyes off short-term events and look for long-term behavior and structure; unless you are aware of false boundaries and bounded rationality; unless you take into account limiting factors, nonlinearities, and delays. You are likely to mistreat, misdesign, or misread systems if you don’t respect their properties of resilience, self-organization, and hierarchy.