Designing For Learnability – Is It So Simple?
by Kirk Bridger
My eldest daughter has started kindergarten this week. Part of the preparatory material sent to us over the summer included a pamphlet on “intelligence” – apparently teachers get a lot of questions from parents about if their child is intelligent or not. The pamphlet briefly describes 8 “areas of intelligence” and outlines examples of each area, allowing the parent to begin to identify which areas their child excels in, and those in which they may not be so strong. Here are the 8 areas and their brief descriptions:
- Word Smart (verbal linguistic) – capacity to use words effectively in speaking and writing
- Number Smart (logical-mathematical) – capability to use numbers effectively and reason well
- Picture Smart (spatial) – thinking in pictures and images and the ability to perceive, transform and recreate different aspects of the visual-spatial world
- Body Smart (bodily-kinesthetic) – expertise in controlling one’s body movements and handling objects skillfully
- Music Smart (musical) – capacity to perceive, appreciate and produce rhythms and melodies
- People Smart (interpersonal) – ability to understand and work with other people
- Self Smart (intrapersonal) – self-knowledge and the ability to use that self-understanding to enrich and guide one’s life
- Nature Smart (naturalist) – sensitivity to and understanding of the natural world
There have been additional areas proposed since Howard Gardner, Harvard psychologist, developed this theory of “multiple intelligences” almost 30 years ago. They include spirituality, humour, and creativity. The theory has not been well received and has not been clinically proven to be true.
So why am I writing about it? Well, I think it is an interesting take on intelligence and emphasizes the need to accommodate different ways of learning. And this naturally leads me to discuss learnability – see how I did that? It really does make sense in the end!
A key element of a usable and intuitively-used software product is that it is easily learnable by its target users. Often the target users are roughly defined as generalized roles and other times they are more thoroughly defined via archetypes or personas. Oftentimes a persona will include some description of the persona’s level of comfort and familiarity with technology and computers, to try to outline for the project team just how learnable the product has to be. If the target user is an advanced computer user who likes keyboard shortcuts then the team should design it differently than if the target user is a novice user who is uncomfortable with technology overall.
This is where the multiple intelligences comes in, or more accurately the multiple means of learning. When designing a product to be extremely learnable and easy to use, are we falsely assuming a particular kind of intelligence is used by the user? Are we, for example, assuming that all novice users learn in a manner consistent with a person with a high “word smart” intelligence? Are we imagining that clear, concise, and plain words in the instructions will help the user learn how to use the interface? I’m sure anyone familiar with user testing can relate a story or two where the user simply refused to look at the text on the screen when trying to figure out how to accomplish their task. Are these people then simply not highly “word smart”? Another example may be that we assume that including large pictures and diagrams in the interface will help the user learn how best to get their work done. We’ve all seen those “Quick Start” guides that come with new electronic products nowadays – are they a panacea or are they simply catering to people with a high degree of “picture smarts”?
Another question: are there areas of intelligence that we are ignoring, and if so what would that mean? Would it mean that no matter how pretty our design and how clear our words, some users just won’t understand how to use the system because they learn by doing (i.e. high body smarts)? Should we look at self-guided tutorials then?
It might be interesting to look through the areas of intelligence and try to look at the product’s design from that point of view, to see if it is learnable from numerous points of view.
It might also be interesting to incorporate and expand on this idea when defining personas or users. For example we could focus more on mental models than on comfort levels with technology in order to capture learnability requirements. What if we began creating new types of intelligences that were specific to our product or software development – is there value in that? For example:
- Exploration Smart (curiosity) – eager to explore new ideas without fear of failure
- Domain Smart (subject matter expert) – very knowledgeable about the domain
- Interconnection Smart (integration) – easily grasps how things fit and work together to accomplish a greater goal
In the end this post is about looking at how we measure, design for, and evaluate learnability in our products. I suspect many would agree that a single approach does not seem to be robust enough to rely on 100% of the time. Perhaps there are new ways of looking at learnability and capturing this knowledge in our user definitions. Or perhaps Howard Gardner was a nut and this whole idea is out to lunch. If that’s the case then we should fall back on the universally-accepted theory of learning styles. Oh wait, that article alone lists at least 5 “models” – I don’t think people have figured this stuff out yet. So no harm in trying it out to see how it works – right? I guess I would be highly “Exploration Smart”.
(Originally posted here)