MDA: A Design Methodology for Social Interactions


"How do I want this to feel?" is a question we don't often ask, but it is the central question that informs our design practice. If we understand 'interventions' as the creation of spaces for new behaviours and thinking to emerge, then how those spaces feel becomes of particular import.

The MDA model is pulled from the world of video games and has proven an incredibly useful frame by which to design spaces for social interaction. The methodology focuses attention on the experience of the participants in the process and the dynamic behaviour of the system being constructed. The core assumption is that positive interactions with other participants and learning objects (such as generated hypotheses) encourage better outcomes.

Design is not about manipulation but rather about designing spaces and experiences that are more likely to inspire the application of perspectives and heuristics required to address the adaptive challenges of the system.

The MDA Design Approach

The approach is situated in broader conversations about human-centered design and places primacy on the experience of participants in the spaces we co-create. A design methodology allows for diverse expertise to jointly participate in an act of shared creation. Individuals bring their own cultural and disciplinary assumptions into co-creative processes, and a shared orientation to the work allows for those involved to apply themselves to the situation with a minimum of distraction or uncertainty.

One thing we too easily miss is that all social interactions are products of design. Most of the time we arrange spaces and social interactions habitually. Meetings, performance reviews, customer care, presentations and so on are copies of models that we don’t often think about. However, interesting things happen when we first focus on how spaces feel, rather than on the rules and rewards that we can more easily control.

Although evolving, our design approach is co-creative and colloquially known as “MDA”. MDA as a design framework comes from the world of video game design. MDA is an acronym for Mechanics – Dynamics – Aesthetics and was articulated by Hunicke, LeBlanc, and Zubek in 2004. The application of MDA outside of video games appears to be rare, although we have found the framework a powerful and effective tool for designing social interactions for both small numbers of people and large groups over short and long periods.

Some definitions:

Mechanics: Mechanics are the rules of a space. Rules may appear as explicit instructions, reward or praise, or as modeled behaviour. Rules can be either explicit or implicit and define what appropriate action looks like in an environment. In a social environment, there are many mechanics at play, related to etiquette, status, formality, and so on.

Dynamics: Dynamics are the interactions that derive intentionally or unintentionally from mechanics. Dynamics may be between people or between people and objects or symbols or any combination therein. For example, cultures have expectations (mechanics) about lining up, which creates interactions of turn taking (dynamics) that don’t require explicit description or orientation. Failure to abide by the unspoken rules (mechanics) can lead to conflict or resentment.

Aesthetics: Put simply, an aesthetic is how a designed space feels. Aesthetic experiences take place in the imagination of the people who are experiencing the elements of a space. As designers, our ability to control how people experience a space is very limited. Aesthetic responses are an emergent property of the system and the subjective interpretation of the individual in the system. However, by focusing on aesthetics we can make better choices in terms of the interactions that are encouraged and the ‘rules’ that define participation.

Generally, a line is drawn from mechanics to dynamics to aesthetics. Leaders, managers, or designers have an end in mind and set down the rules and expectations that will lead to that end. A rule or expectation in a space leads to desired interactions. How those interactions are experienced is usually an unintended by-product of the system's operation. So, an internet service provider that has rules around controlling costs will hire fewer service support people. This leads to longer waits when service interruptions occur. Some may experience those longer waits as a price to be paid for more affordable service. Others may experience frustration. Ultimately, the design has been “successful” as costs have been controlled. Different mechanics are applied to address the impacts of the initial decision as customers leave or complain. Action can feel incoherent as decisions respond to the outcomes of previous decisions.

However, a line can also be drawn in the other direction, from aesthetics to dynamics to mechanics. We can select how we want people to feel then cultivate multiple interactions most likely to create that feeling and then institute rules that facilitate those interactions. Rules that act against the desired aesthetic can be modified or eliminated through iteration. As an example, a museum wants guests to feel an aesthetic of ‘curiosity’ about artifacts and exhibitions. Interactions that foster curiosity may involve puzzles, intentionally incomplete answers, and means of interacting with exhibits and exhibitors that are open. Then, rules and modeled behaviours can be introduced that facilitate those interactions, such as questions instead of answers on displays, exhibitors that model playfulness rather than expertise, and displays that can be touched and manipulated. The assumption here is that “curiosity” will improve the guest experience. Should evidence suggest that this is not the case, then different approaches can be applied which are similarly coherent and directed. Starting small and prototyping allows for assumptions about appropriate aesthetics to be tested early and cheaply.

The design process is then centered around the following questions:

  1. What is the intended purpose of the social interaction (meeting, discussion, service) being designed?

  2. What feeling (aesthetic) in the participants is most likely to generate productive participation in this social interaction? How do we want them to feel?

  3. What interactions are most likely to generate that feeling?

  4. What control mechanisms (rules, modeled behaviour, rewards) are most likely to support those interactions?

  5. How can we test this?

The appropriate aesthetic will vary considerably given the nature of the challenge and the people involved. An aesthetic of ‘competition’ may spur innovation in some contexts and shut down thinking in others. Rapid iteration and testing allows us to understand both the effectiveness of the design and the suitability of the aesthetic for the system under consideration.

As Hunicke, LeBlanc and Zubeck describe, “thinking about games as designed artifacts helps frame them as systems that build behavior via interaction”. The primary benefit of MDA is that it ensures systematic coherence in created environments.

An approach that starts with mechanics can too often follow an instrumental approach. Rules and expectations are objectified to create the desired behaviour. However, this approach embodies a linear causation that very often fails to recognize the role of the participant as a part of the system under construction. Imagine an organization that operates with very thin margins and struggles with theft by employees (such as some retail sectors). The organization could reduce theft with frequent reminders of property rules, active security checks, and video cameras. This would likely lead to less theft. However, the experience of the employees would be one of feeling mistrusted. Feeling mistrusted leads to interactions where employees are less likely to share discretionary effort or resources. Reductions in thefts may be completely offset by reduced engagement and new interventions must be introduced to address the emerging issues.

The same issue can be seen differently from an aesthetic-centered approach. One assumption might be that employees that feel ownership over the space are less likely to steal from it. Interactions that promote an aesthetic of ownership might involve shared meals, profit sharing, horizontal reporting structures, and a voice in decisions. Mechanics could then be instituted that facilitate these interactions. The need to control costs would still be present, but iteration supports testing of the impact of new mechanics without having to invest fully in a particular course of action. Unintended consequences (both positive and negative) can also be assessed and adaptations made where appropriate.

Iteration focuses on all three levels of design to assess where inconsistencies are occuring. Improvement questions may include:

  • Why isn’t this control (mechanic) generating the desired interaction (dynamic)?

  • Why isn’t this interaction (dynamic) feeling the way we intended (aesthetic)?

  • What interactions (dynamics) might promote the desired feeling (aesthetic)?

  • What instructions, modeled behaviour or tacit rules (mechanics) might encourage the interactions (dynamics) we are seeking?

As a design approach, the MDA model has proven particularly effective in

  • Opening and supporting alternative views, patterns of thinking, perceptions and interactions

  • Synthesizing different disciplines and approaches to meaning making

  • Dismantling invisible or stubborn behaviour patterns

  • Recognizing and reconciling to paradox and contradiction

  • Augmenting reflection and supporting coherent decision making

At its most basic, the MDA process asks us “How do I want this to feel?” before structuring any space or social interaction. This may seem obvious but remarkably few people are this intentional when preparing to bring people together. We are very good at creating spoken and unspoken rules. We are less good at understanding the unintended consequences of those rules on others and how they affect the experience of our collaborators and friends in the work that we do.