Design UI: Don’t Punish The Humans
Remember the limits of Human Memory
The main reason computers exist is because we humans have very limited memories. Computers help not just in terms of storage, but in terms of working memory, memory that is used for comparison and decision making and calculation.
Rick Bolin
Some Working Principles about Human Memory
- memory is imperfect
- visceral memory is super short (about 200 ms)
- there is very limited short term storage (4 meaningful things +/- 1)
- memory filters (through meaning and significance) as it stores and as it recalls
- memories are reconstructed, from associations, each time you think about them
- memory isn’t instantaneous
Is Memory Analytical or Analog?
A lot of us have a model of memory that is a lot like photography. With a sensitive, impressionable medium, the memory, that is exposed to light/stimuli/input, a full in-situ representation is formed, much like the daguerreotype at the top of this post. We have of course the expression a photographic memory.
Through lots of research, experimentation and observation, the prevailing scientific view of memory is that it is a much more analytical: information processing model.
A Simple Look at the Informational Processing Model
Out of a vast amount of raw data stimulus, the attention collects a limited amount into the sensory memory. Those sensory impressions are held for about 1/5th of a second before either being recognized as a meaning unit or disregarded as a new rush of data comes in.
The working memory is the mechanism that holds our awareness, our decision making and our consciousness. The limit to the human working memory seems to be about 4 meaning units, plus or minus one, for somewhere between 2 and 20 seconds. These meaning units can be tiny atoms of meaning (the color red, for example) or compound chunks of data in a larger meaning unit (red stop light).
When those meaning units are put into a context (placement, functionality, success, failure, timing, emotional experience, for example) they are stored as long term memories, which can last for years. I still can remember where within the Windows 95 Start Menu the network settings were.
Some key conclusions for us to make from this model: we can’t remember what we can’t make sense of and having to choose from more than 4 things is hindered by memory limitations.
Icons For Example
Long a favorite tool of UI designers, icons are aids to memory, an instantly recognizable, discreet glyph that directs users where to get what they are looking for. Sadly, they don’t seem to work that way.
Some reasons this failure of icons is attributable to our limited memory:
- If we can’t recognize what the symbol means, we can’t remember it (is that SEARCH? EXPLORE? DISCOVER? NAVIGATION? STAR?)
- If we can understand the icon in several ways, we can misremember it, confident in our faulty memory (Is that a contact card? Is that messaging? I can’t remember)
- Clearly understood icons (home i.e.) are clearly remembered.
Data Display as a Test Of Our Memory
When we were polishing our new closet design on Tradesy, we had quite a few questions about which data to display and how to display it to help users make quick, confident and meaningful comparison.
Conscious of the amount of cognitive workload we are asking users to bear, just in sifting through our inventory, we had a few hypotheses about how to lighten that load.
One was to reduce the amount of data we were displaying, thinking that that simpler approach (a) would be easier for users to remember the items and ultimately make a choice.
The other hypothesis (b) was that by separating and chunking the information we would be making the data clearer and easier to understand, and therefore easier to remember.
We could see the underlying thought process and value of each approach and so decided to test it to see what users could actually process and remember. One of the reasons we like the 5-second test as a means of understanding our information design success is because our human memories filter/process as they store, so things that are hard to remember are probably poorly structured.
The clearer approach (b) won out by a large margin. Not only did people remember the brand and product much more often, but the accurate price and size as well. The lack of detail in the simpler approach (a) actually seemed to disrupt the information processing and memory.
The Experience Principles I and my team hold dear, that our decisions allow for Kind, Human and Stable experiences, are supported importantly by our making design decisions that work with rather than against how their minds work.
