Why visual form matters for information architecture

Victor uses the example of a Chinese menu to describe how familiarity with the visual form of information helps people recognize the object. Original paper here. Its an interesting point, and I suspect the hands of “implicit memory” (a lesser known aspect of human memory). Let me explain.

Victor was elaborating on some well done experiments by Elaine Toms. Three versions of the same content were presented to people for recognition. (The pictures below are Victor’s examples and stolen from his presentation at the IA Summit). In all cases, people are asked to recognize what type of a document it was.

Full version with content and form intact A content version with all structure deleted

A form version, with content changed to X’s and 9’s

Expectedly, the full version was recognized more than the content only version, recognized more than the form only version. The interesting finding is that when the form version was recognized, it happened twice as fast as for the other versions. Victor speculates that this is because the cognitive processing of visual information happens faster than that for verbal information. While there is some truth to that, let’s play detective and identify what cognitive mechanism is the culprit here. What could be better than a good cognitive mystery for a Friday afternoon?

Lets begin by isolating the broad cognitive process at play here. Like always, sensory processing and working memory are involved, but the crux of the task is about retrieval from long term memory. To put it simply, in order to recognize the menu, we need to reach back into our prior experiences, rifle through our long term memory store and bring back memories that are related.

Having established that, lets look deeper into long-term memory. Broadly speaking, there are two types of long-term memory mechanisms: implicit or unconscious and explicit or conscious. An example of implicit memory is when you meet someone at a party, you do not remember meeting them before, but you have a feeling of familiarity. Implicit memory often operates in an automatic manner and is often not under your conscious control. An example of explicit memory is when you meet someone at a party, you know who they are, you can recall the first time you met them. With explict memory on the other hand, you consciously retrieve something from long term memory. It is a slower, more deliberate process.

Retrieval from explicit memory depends on the depth of processing at encoding, but is not influenced by the match of the perceptual characteristics of the stimuli (e.g. type font of written words) at encoding and retrieval. Implicit memory, on the other hand, is often insensitive to level of encoding, but is sensitive to physical characteristics of the stimuli such as type font. Implicit memory processes are also fast and automatic.

Now let us use the implicit-explicit distinction to explain the Chinese menu example. It is possible that recognition of the form only version was triggered by an automatic, implicit memory mechanism. In contrast, in the full and contentonly version, recognition was based on slower, more deliberate conceptual processing.

My speculation above can explain why the form only condition was less accurate (it was based on perceptual recognition that is very specific. If someone’s prior exposure to a Chinese menu did not match this form, then the implicit memory did not get activated). It also explains why the form only condition was faster: since is was based on faster, automatic processes. However, we are still left with the question: why was form only version faster than the full version.

There are two possibilities. One possibility is that in the form only version, people knew it or they did not. Fast, automatic processing lead to a fast correct answer or an incorrect one. In the case of the full version, fast automatic processing was followed by more belaboured conceptual processing to give a final recognition judgement.*

Above is mostly speculation. I would need to know more details of the paradigm and the data to build a firm theory. However, my point is a more general one – about the role of implicit memory in design. Implicit and explicit memory are based on separate memory systems, located in separate parts of the brain. These two memory systems can act in concern or in contradiction to one another. Designers can use both, in very different ways, to aid in better information retrieval.

Currently, implicit memory is not understood very well, and is therefore not used in a systematic way to aid designs. Implicit memory can be a designer’s friend or her enemy. Know it and befriend it, so that it does not come back to bite you.

And thanks to Victor for pointing out an interesting case of the duality of memory systems.

*Note: There is one more possibility that cannot be ruled out without looking at the data itself. Assuming that the Elaine’s results are referring to mean recognition times, it is possible that the reaction times in the full version comprise two sub-groups. Some people recognized the document on the basis of implicit memory (their reaction times are comparable to that of the form only version), while others got it on the basis of explicit memory (their reaction times are more similar to the content only version). The average ends up being higher than that for the form only version. This explanation does not contradict my above theory in any manner.

4 thoughts on “Why visual form matters for information architecture

  1. Very interesting. One point, though. Can the third example really be said to be content-free (‘form-only’)? Perhaps less time is spent processing that example’s content (x’s and 9’s) because after two or three samples, the entire content can be accurately extrapolated, allowing the viewer can move on to other steps, whereas with the other two examples more time may be spent … well … reading?

  2. The third example is content free to the extent that there is no semantic meaning there. It does use numbers and alphabets, so the parts of the brain dealing with that might get involved.

    Your explanation that people spent time reading for the “full” and “content only” version is a possibility. Add there is no real way to rule that out. One would need to know a lot more about the study, what the specific stimuli were (the chinese menu example is not from the study itself), how long the recognition times were etc.

  3. I am curious as to the effect of specificity on the example, and the way we are trained to recognise certain prototypes. For example, I can quickly recognise a car (general prototype) but it takes longer to identify the type of car. Likewise in the chinese menu example I quickly recognise the form model as being a pricelist, but to be more specific fires up some other form of processing. Interesting stuff!

  4. James, the car example that you describe can probably be explained by the concept of “basic level”. Categories vary along a vertical dimension.

    For example, a spaniel > dog > mammal > animal

    People tend to be more responsive to the basic level. If you show me a picture of a terrier and ask me what it is, I am likely to say dog, rather than the subordinate (terrier) or the superordinate (mammal, animal) category. Our memories work well at this level of differentiation. This effect was first described by the famous psychologist Eleanor Rosch.

    There are several theories regarding why the basic level is faster (and easier). It probably has something to do with the nature of memory representations in our long term store.

Comments are closed.