archive.Joe Lamantia.com

Several very unpleasant experiences I’ve had with the Lotus Notes webmail client during the past few weeks have brought up some questions about mental models; specifically how users respond to challenges to their mental models, and how resilience plays a part in how changes to mental models occur.
The IAWiki defines a mental model as, “a mental model is how the user thinks the product works.” This is a simplified definition, but it’s adequate for the moment. For a deeper exploration, try Martina Angela Sasse’s thesis
Eliciting and Describing Users’ Models of Computer Systems.
In this case, the model and the challenge are straightforward. My mental model of the Notes webmail client includes the understanding that it can send email messages. The challenge: the Lotus webmail client cannot send email messages – at least not as I experience it.
Here’s what happens my mental model and my reality don’t match:

1. I log in to my email client via Firefox – the only browser on the Mac that renders the Notes webmail client vaguely correctly – (I’m using webmail because the full Notes client requires VPN, meaning I’m unable to access anything on my local network, or the internet, which, incidentally, makes it difficult to seem like a credible internet consultant.) again, because it’s frozen and crashed my browser in the past ten minutes.
2. I realize I need to respond to an email
3. I do not remember that the Notes webmail client is incapable of sending out email messages
4. I open a new message window, and compose a chunk of semi-grammatical techno-corporate non-speak to communicate a few simple points in blame-retardant consultantese
5. I attempt to send this email
6. I am confronted with a cryptic error message via javascript prompt, saying something like “We’re really sorry, but Domino sucks, so you can’t send out any messages using your email client.”
7. Over the span of .376 seconds, I move through successive states of surprise, confusion, comprehension, frustration, anger, resentment, resignation, and malaise (actually, mailaise is more accurate.)
8. I swear: silently if clients are within earshot, out loud if not
9. I switch to gmail, create a new message, copy the text of my message from the Notes webmail window to Gmail, and send the message to some eagerly waiting recipient
10. I close the Notes webmail client, and return to business as usual.
11. I forget that the Notes webmail client cannot send email messages.

Despite following this same path three times per day, five days each week, for the past five weeks, (for a total of ~75 clear examples), I am always surprised when I can’t send a message. I’m no expert on Learning theory but neither lack of attention nor stubbornness explain why seventy-five examples aren’t enough to change my model of how Notes works.
Disciplines including systems theory, biology, and sociology use a concept called resilience. In any stable system, “Resilience generally means the ability to recover from some shock, insult, or disturbance.” From an ecological perspective, resilience “is a measure of the amount of change or disruption that is required to transform a system.” The psychological view emphasizes “the ability of people to cope with stress and catastrophe.”
Apparently, the resilience of my model for email clients is high enough to withstand considerable stress, since – in addition to the initial catastrophe of using Notes itself – seventy-five consecutive examples of failure to work as expected do not equal enough shock, insult, and disturbance to my model to lead to a change my in understanding.
Notice that I’m using a work-around – switching to Gmail – to achieve my goal and send email. In
Resilience Management in Social-ecological Systems: a Working Hypothesis for a Participatory Approach , Brian Walker and several others refine the meaning of resilience to include, “The degree to which the system expresses capacity for learning and adaptation.” This accounts nicely for the Gmail work-around.
I also noticed that I’m relying on a series of assumptions – email clients can send messages; Notes is an email client; therefore, Notes can send messages – that make it logical to use a well established model for email clients in general to anticipate the workings of Notes webmail in particular. In new contexts, it’s easier to borrow an existing model than develop a new one. In short order, I expect I’ll change one of the assumptions, or build a model for Notes webmail.
Here’s a few questions that come to mind:

1. What factors determine the resilience of a mental model?
2. How to measure resiliency in mental models?
3. What’s the threshold of recovery for a mental model?
4. Put another way, what’s required to change a mental model?

Based on a quick review of the concept of resilience from several perspectives, I’m comfortable saying it’s a valuable way of looking at mental models, with practical implications for information architects.
Some of those implications are:

1. Understand the relevance of existing mental models when designing new systems
2. Anticipate and plan the ways that users will form a mental model of the system
3. Use design at multiple levels to further the formation of mental models
4. Understand thresholds and resilience factors when challenging existing mental models

From a broader view, I think it’s safe to say the application of systems theory to information architecture constitutes an important area for exploration, one containing challenges and opportunities for user experience practitioners in general, and information architects in particular.
Time to close this post before it gets too long.
Further reading:
Bio of Ludwig Bertalanffy, important contributor to General System Theory.
Doug Cocks Resilience Alliance
Garry Peterson’s blog Resilience Science

Related posts: