Indi Young  

Mental Model Building: Step by Step

August 12th, 2005 by Indi Young :: see related comic

A year ago I was casting about for a new example to use in my mental model presentations. I wanted something that was more globally applicable than a kitchen remodel, which was our old example at Adaptive Path. (Imagine someone living in a tiny Tokyo apartment trying to make sense of the task flow supporting the planning around custom kitchen cabinets!) I wanted a topic that was engaging, so that people would be interested. I wanted an example I could present to information architects in India or Thailand, should the time come.

##Mental Model Diagrams
But first, let’s take a step back. What is a mental model? My mental models are diagrams that represent the thought- and action-process used to achieve a set of goals in a narrowly defined scope. For example, you could draw a mental model of what you do in the morning to get ready for work. You could interview other people and add their approach to the mental model diagram as well. Different kinds of people might employ different steps, like a mother might include getting the kids ready for school in her morning routine, whereas a teenage girl might include three trips back to her wardrobe to change clothes to match her mood. You could interview people from other cultures in other countries to see what task differences there are.

With each interview, though, you must remember you have defined the scope as “get ready for *work*.” You want to avoid letting your research get out of hand, interviewing folks who don’t work. Folks who don’t work may be just as important as those who do, but supposedly up front you decided there was more value in studying just those who work. Stick to your focus.

Sometimes the scope of a mental model might be broad enough that you could nest another mental model within it. For example, you could draw up a mental model of how people brush their teeth and nest it inside the “Get Ready for Work” mental model. (I don’t actually nest the diagrams, but reference the “Brush Teeth” diagram from within the “Get Ready for Work” diagram.) For businesses creating mental models, they often start with a broadly defined scope (our website) and discover along the way that they need much more detailed data for a few key sections. These key sections become mental models of their own, and thus can define other standalone offerings, such as,,,, or

My mental model diagrams are horizon charts that look like city skylines.
The buildings (towers) have windows (boxes in the towers) that represent tasks. The buildings themselves represent task groups. There are thick vertical black lines that divide the buildings into city blocks, and these blocks each define a certain mental space. So, in your “Get Ready for Work” mental model, you might have task boxes such as “Pull Up the Sheets,” “Straighten the Blanket,” “Fold the Comforter,” “Fluff the Pillows,” and “Arrange Pillows on the Bed” in a task group tower called “Make the Bed.” “Make the Bed” might belong with other task group towers like “Put Away Clothes” and “Re-Organize Nightstand” in a mental space called “Tidy Up Before Leaving Bedroom.” Maybe certain audience segments don’t do these tasks at all, in which case you mark the tasks with indicators of which audience segments participate. Or maybe some audiences do different tasks, such as “Fold the Futon” and “Put Away the Coverlet” for some folks in Japan.

##Movie Audiences
Speaking of Japan, it finally hit me one afternoon when I was reading an email from a friend about the latest movie from his favorite Japanese anime
studio: Ghibli. Movies! People all over the world see movies. Movies are universally appealing. There are defined audiences to take apart and examine. There are tasks beyond buying tickets. It is a subject that makes people happy. So I set out to create a mental model about movie goers.

The complexity of my choice became evident the first day. I always perform a task-based audience segmentation as my first step in mental model research.
(See While it was fairly obvious that the standard G, PG, NC-17, and R rating system mainly represented a demographic separation of age and of aversion to graphic scenes, it was not obvious what task-oriented approach replaced it. Would segmentation based on genre preference be appropriate, such as “Action Boy,” “The Romantic,” or “The Space Cadet?” Or would there be too much cross-over in the tasks? They certainly all buy tickets to the theater, but those who tend to buy tickets in advance versus those who decide to see a movie on the fly and buy tickets at the counter may not divide strictly along genre preference lines.

So I listed all the tasks I could think of, from “Find a Parking Space” to “Move seats when someone tall sits in front of you.” I included obscure tasks, like “Learn a skill you saw in a movie” and “Sing songs from the movie.” I ended up with a list of 173 tasks. I grouped these tasks to make a more manageable number, figured out who was likely to do the tasks, and ended up with six audience segments:

- Social Movie Goer: tends to see movies to be with friends, be entertained.
- Movie Buff: likes the craft of filmmaking, any story
- Big Fan: follows the work of a certain director, actor/actress, or genre.
- Film Purist: prefers a thought-provoking story or call to action
- Make Believe Artist: enjoys submersing themselves in another world
- Facilitator: seeks to please others by taking them to the movies

This hypothetical set of audiences was what I began to work with. The Movie Buff, the Big Fan, and the Film Purist seemed to me to be quite close in nature, but I decided to let some actual interviews help me decide whether the distinction was needed. Also, some people that I interviewed crossed over between two of these segments, but one segment was always dominant.

##Non-Directed Interviews
My next step was to recruit people to interview who represented the different audience segments. To keep things within the scope of my time and budget, I chose to examine only the first four segments. While interviewing these people, I soon found out that I do not take movies as seriously as many folks do! I, myself, fall into the Social Movie Goer category. Here are the five prompts that I used during my hour-long, non-directed conversations with each participant:

- Tell me about the movies you see.
- What do you do before you get to the theater?
- What is important to you at the theater?
- What about beyond the theater experience?
- What is unique about your interaction with film?

A word about non-directed interviews: you don’t want to bias your data!

Listen to a talk show host, like Terry Gross on NPR radio, and notice how heavily they guide the conversation to certain provoking topics. In your interviews, you want to do just the opposite. Let your participant do all of the talking. Let your participant guide the conversation and direct the order the topics. Your prompts are only a general list of topics that serves as a checklist, nothing more. You will not necessarily hear about each of the topics or hear the same language in each interview. This is not a survey where it’s important to ask the same question to many people and perform statistics on the answers. Each interview is a unique conversation with a single person.

##The Movie Mental Model
The mental model that arose out of the analysis of the interview transcripts appears to hew to most people’s approach to movie-going. It is thrilling to see tangible mental spaces and tasks appear out of all the analysis. Here are the mental spaces and task groups that I found (click to enlarge):

Movie Mental Model

Maybe not all tasks are represented in the model, but certainly all the mental spaces are there. If I interview more and more people, I am confident that new tasks and task groups will find places among the existing mental spaces. This level of confidence is what makes a mental model so powerful. Add that to the fact that mental models tend to shift very slowly, and you have a tool that can guide your design and prioritize your projects clearly.

##Next Steps…
I have yet to take the final step in the process: re-evaluation of the original audience segments. Remember how The Movie Buff, the Big Fan, and the Film Purist seemed to me to be quite close in nature? By tagging the actual tasks in the mental model by audience segment I will be able to see correlations between these segments, and will be able to decide whether some of them need to be grouped together.

Anyway, now I have a mental model diagram that is fun to look at! It will help ease the tedium of the classroom when I teach seminars about this technique. It has already taught me how much more there is to movie-going that meets the eye!

So, what are you doing tonight? Wanna go see a movie?

Indi Young is a co-founder of Adaptive Path where she helps clients create mental models and derive useful tools and clean navigation. Before Adaptive Path, she consulted on her own for 10 years. She was formerly a software engineer. She is the least-public member of Adaptive Path, leaving the limelight to more vocal members.

6 Responses to “Mental Model Building: Step by Step”
Henk Wijnholds wrote:

I think I like this, it’s a great ACD method ;) Though I’m still wondering how I should mould this into a readable, understandable diagram. Maybe something like page 52/53 of ‘Don’t make me think’..

Mental Model wrote:

Re-cognize significant credibility is lost when a long list of labels and stereotypes is presented and then a statement is made like: “You don’t want to bias your data!”

Also many people may not be going to a cubicle for work some people may be working out in the fresh air and so-on.

p.s. Please read Women, Fire, and Dangerous Things
by George Lakoff (especially if y’all are interested in teaching)

betting wrote:

I am halfway through Women, fire and dangerous things and it has been a great read so far. So thanks to Mental model for the recommendation.

Kim wrote:

Thanks for the more layman look at this topic. I was wondering how one might apply this theory to ideas like social issues, such as Crime, Employment, Health Care, Culture, etc.
I’m due to teach this concept for the first time and I’m trying to get an introductory grasp of how exactly to simplify the concept for the student who would know virtually nothing about it, and so may tend to retreat to limited views, which seem, as I research, to be contrary or at least only a fraction of the theory as a whole.

Jonathan wrote:

“Anyway, now I have a mental model diagram that is fun to look at! It will help ease the tedium of the classroom when I teach seminars about this technique.”

I’m sorry you find teaching so boring.

It’s hard to tell what the point of all this listing activity actually is. How do you “re-evaluate the original audience segments” with this exactly?

Also, if I’d rather not draw up my task lists to look like a cityscape, is that OK? As a consultant, I have enough problems dealing with accusations of wasting time as it is without deliberately antagonising clients with eye candy.

And while I’m a-ranting: I really hope people don’t start narrowing the meaning of “mental model” to mean JUST a diagram. Enough damage has been done by Rational’s arrogant hijack of the phrase “use case” in that respect.

Miu Miu Jewelry wrote:

This can be definitely wonderful news. Thank you for discussing it with us!

Beneficial brief and this write-up helped me alot. Say thank you I searching for your facts.

Leave a Reply

OK/Cancel is a comic strip collaboration co-written and co-illustrated by Kevin Cheng and Tom Chi. Our subject matter focuses on interfaces, good and bad and the people behind the industry of building interfaces - usability specialists, interaction designers, human-computer interaction (HCI) experts, industrial designers, etc. (Who Links Here) ?