Designers, systems thinking, and simulations

How did we get here, and where can we go?

2016 in the news: Climate change and displacement, depleted resources and increased consumerism due to extractive neoliberal ideology, a mounting refugee crisis, increasing xenophobia and public alarm, and the intensifying of U.S. tension due to increased awareness of bias-linked violence. The interlinked causes and influences on these cyclical issues are complex and involve countless international stakeholders. These systems are complex and their problems are “wicked”.

Wicked problems and complex systems

The kind of issues that plague social impact designers and cultural activists have been labeled “wicked problems.” Wicked problems involve a rat’s nest of complexity and immense scale, and are always caused by issues within a system. They are nearly impossible to solve due to their incomplete and contradictory requirements, which are in turn difficult to recognize and make sense of.

Donella Meadows (sustainability scientist, and complex thinking expert) describes wicked problems as being characteristic of the system structures that produce them, and urges us to think about the ancient Sufi story of “The Blind Men and the Matter of the Elephant” as a lesson in tackling social for social impact: the behavior of a system cannot be known just by knowing the elements of which the system is made.

Because these problems arose from interwoven human interactions, tensions between individual and collective interests, and reinforcing feedback loops, designing for them necessarily requires a more holistic approach that accounts for and examines these outcomes as phenomena that arise from aggregate interactions and interconnections.

Below is an example of systems thinking applied to the much misunderstood connection between the Muslim community and terrorism. The left-hand side presents a reductionist’s view of terrorism and is missing a few factors. Breaking down the problem and asking more questions sheds light on its true complexity, articulates the full effect that individual interactions contribute to the overall picture, and ultimately can jumpstart new perspectives towards design action. We must confront complexity, instead of becoming overwhelmed by it.

This is a causal loop diagram. Arrows with (S)’s mean the variables move in the same direction (positive feedback), and (O)’s implying the opposite (negative feedback). How to read a causal loop diagram.

Designing alternate futures with simulations

How do our individual biases, motivations, and actions aggregate into systemic forces over time that change the structures of the world? Which of our local interactions ultimately affect the global system? This thought experiment quickly becomes cyclical — but therein lies the nature of wicked problems and complex systems. Can we, as designers, contribute positively to this discussion, help to steer conversation in a critical way, and create processes to envision a better world?

One way to explore this space and work through future-facing problems is with computer simulations — virtual replicas modeled off our understanding of the world — in order to reproduce the behaviors of a system too complex to be studied with traditional mathematical techniques, and test out ideas too harmful and drastic to try on real people. Designing, building, experimenting with, and analyzing these models can help us develop and improve upon our mental frameworks of social behavior and phenomena, and ultimately expand the way we approach service design and social innovation.

Agent-based modeling (ABM) is a simulation technique with a wide array of applications in biology, network theory, cybernetics, economics, and social sciences. An often cited example is the influential Schelling’s Model of Segregation, which offers a feasible explanation as to why racial segregation persists despite efforts to eradicate it. Tests run with this model seemed to suggest that when racially “tolerant” households cluster over time, patterns of segregation can still emerge. Models like this one become even more interesting when surveying recent news articles about how racism effects U.S. neighborhoods — The Racist History of Portland, the Whitest City in America from The Atlantic, for example. Our personal racial preferences are part of the feedback loop that affects society in ways we cannot expect; the more we understand the invisible forces that create negative impact, the better chance we’ll have at reversing its effects.

(Left) A running of Schelling model shows households segregating into clusters over time source. (Right) Parable of Polygons by Vi Hart and Nicky Case makes the Schelling model more accessible through an interactive web experience source.

ABM simulations are able to articulate larger behavioral phenomena, like the one described in Schelling’s model, by giving researchers the ability to analyze the interconnections and interactions that occur between the following key components:

  • Agents: programmed representations of individuals, organizations, or nation-states
  • Environment: virtual world in which the agents act in — think buildings, cities, states, etc
  • Time (expressed as steps in the model): gives its designer the ability to view, at each stage, the incremental and individual changes that led to a larger shift. This way the path each agent took in the simulation, as well as the solution can be captured: when the model produces a result, an equation describing it can be established, making a system’s dynamic history comprehensible.

Society and SimFutures

SimCity, RollerCoaster Tycoon, Animal Crossing, and similar games are quintessential examples of agent-based systems (watch this video if you need convincing), and are good examples of interactive experiences that have their players thinking about designing alternative worlds. These built environments and simulations provide opportunities for us to question, challenge, and form critical perspectives about our own value systems as we interact with their representations of the world. Urbanists competing in an urban design competition using the 2013 version of SimCity realized that the game encouraged them to “make short-sighted decisions” as it made capitalistic and positive feedback loops addictive, while trivializing “the importance of certain public policies (e.g. investing in education, renewable energy, and pedestrianism).” Revelations experienced from human interaction with AI and computational dynamics, and the concept of a “society in the loop,” articulating “the judgment of society, as a whole, in the algorithmic governance of societal outcomes,” become crucial as systems that we rely on every day begin to face the challenge of moving from automated machines into moral machines.

Building a toolkit

Up until more recent attempts to make these simulations and the concepts communicated in them more accessible to the general public outside of their current academic environment, the use cases of simulations to fabricate a system or world have been limited to scientists and game developers with slow progress being made to liberate it into the planning and public activism community.

System Designer (SYD) is the year-long project Francis Tseng and I will be undertaking as Researchers in Residence at NEW INC’s incubator to tackle that. It is an open-source toolkit and conceptual framework for interaction and game designers, narrative fiction and documentary filmmakers, activists, urban planners, and technologists to experiment with simulating alternative future worlds.

(Left) A diagram from our previous project, The Humans of Simulated New York, completed at DBRS Innovation Lab in March 2016. (Right) Prototypes of SYD’s node-based system designer. Documentation on SYD’s possible features and technical exploration can be found on Francis’ blog, space and times.

We are inspired by Cybersyn, the massive Chilean co-design and computation initiative to coordinate the economy with participation from factory workers, and the speculative exploration of re-building Earth from one woman’s genetic data in Octavia Butler’s Lilith’s Brood, and hope to accomplish these three things right now, in the real world:

  • Open-source toolkit: A mostly code-free, visual, and web-accessible agent-based modeling simulation toolkit that can be used in applications such as modeling alternate economies, the effects of urban planning decisions, and resource management
  • Community and collaborations: The ability for the sharing of modules/models and for others to use them. This would encourage more “social” modeling (in the sense of people modeling collaboratively). Partnerships with organizations would occur with a co-design process while emphasizing designing with complex systems thinking
  • Frameworks: Explore contributing to design thinking frameworks and dialogue in the age of increasingly wicked problems and complex systems

“We shape our images of the future, and meanwhile they shape us.” — Stuart Candy

To simulate and speculate is to be preemptive about designing for better and more anticipative visions of the future. To plan is to design. But as we can see with these wicked problems, oftentimes plans are short-sighted, reactionary, and ultimately result in vicious cycles instead of virtuous ones. Design-school training is lacking, few of us leave our prestigious academies with any idea about the kinds of complex issues interlocked between politics, business, technology, and human and social behavior. The industry is just beginning to encourage dialogue between professionals and the communities they are trying to serve. We must task ourselves to create our own tools for introspection and less extractive enterprises, and reinvest resources back into improving the system.

They will yield only as we reclaim our intuition, stop casting blame, see the system as the source of its own problems, and find the courage and wisdom to restructure it. — Donella Meadows


Wicked problems

Systems thinking

Social innovation