The usability of your Web site depends upon your users' ability to encode, store, and remember information. If your users can't remember what they need to remember, they will be disgusted and they will fail. If they fail, then your site fails.
Supposedly, humans are able to hold 7 ± 2 items in memory. This number is based on George Miller's (1956) work on digit span recall tasks. In particular, Miller examined how people encode and recall lists of increasingly difficult digit sequences. He found that people were generally only able to recall between 5 and 9 "chunks" of information (even when given other types of material to study, e.g. words). He concluded that the human memory system has a capacity of 7 ± 2 chunks.
I think that 7 ± 2 is a squishy concept since chunks are so very flexible. On the one hand, you might only be able to use small chunks. On the other hand, you might be able to employ huge chunks. You can potentially pack a lot of information into a "chunk". For example, an entire movie passage could be considered a chunk; a lengthy passage is still a chunk. This can be very good to know because, if properly motivated, you can help users chunk information in memory by grouping information on your site. For example, you could structure your global navigation mechanisms at the top of the page (e.g., 'Contact Us', 'Site Map', 'Search', etc.), and you could group your sub-site navigation to the left of the page (e.g., 'Engineering Department', 'Chemical Engineering', 'Quality Control', etc.). In this way, you are taking these ideas and you are chunking them for your users. While users have limited memory capacities, you can optimize your pages to take advantage of what users do have.
So far, I have made it sound as if memory is not all that bad. However, this is seriously wrong: realistically, users' memory capacity in a dynamic environment (e.g., the Web) is only about 2 or 3 chunks (Yntema & Mueser, 1960; Yntema, 1963; Venturino, 1997). That's right, 2 or 3 chunks. Put another way, as your users navigate and digest your site, they are actually only able to keep track of about 2 or 3 things. Your users might not be able to remember information from just a page or two back.
Several people have advocated that Web sites be built shallow and wide, whereby users are not required to click more than three links to get to what they want. This seems to cohere with the empirical work cited above. Users can't remember what they want and need so it is your job to guide them, and guide them well. These details also help to explain why hyperlinks with greater amounts of explanation (or context) are superior to shallow links. If users must click from page to page looking for what they want, they will forget what they are looking for (or they will get distracted).
Some Web sites have decided that certain information (e.g., articles) should be divided into many parts. I will not argue that information should be all on one page; that is silly. However, I will state that breaking pages up for the sake of convenience or aesthetics does not make sense if users must recall information from previous points in time. That is, if users need to remember information from an earlier point in time, then under most circumstances the information should be on a single Web page. Obviously this is not a rule, it is just a statement that memory limitations will prevent users from remembering information from an earlier point. Further, consider that scrolling is easier for users than moving back through pages. Printing even a very large Web page is easier than navigating back and printing previous (smaller) pages. Many of your users print Web pages, since reading off of a computer screen is a hassle. Finally, users can more easily search or scan the information if it is on one screen.
You ought to spend a lot of time improving the navigation of your Web site. In fact, you will need to have navigation that is great if you want to minimize the weaknesses of human memory. Visible, easy to use, easy to understand navigation is critical.
You can augment your users' performance by having them use recognition rather than recall. That is, give your users the information that they need and allow them to pick what they want. Don't force users to recall information, such as passwords, unless absolutely necessary.
Here is some final food for thought. I claim that search engines have largely been successful because they shift a memory burden away from users. Search engines are basically recalling information for users. Furthermore, this is information that users never needed to store. Search engines are so great because they store and recall information for users and then, when the results are displayed, users simply perform a recognition task. It really is an interesting memory phenomenon when you think about it -- in effect, search engines are able shift recall to recognition. Since recognition is better than recall, and since overall performance improves, the usability is augmented. Pretty slick, eh?
Miller, G. A. (1956). The magical number seven plus or minus two: Some limitations on our capacity for processing information. Psychological Review, 63, 81-97.
Yntema, D. B. (1963). Keeping track of several things at once. Human Factors, 5, 7-17.
Yntema, D. B., & Mueser, G. E. (1960). Remembering the present state of a number of variables. Journal of Experimental Psychology, 60, 18-22.
Venturino, M. (1997). Interference and Information Organization in Keeping Track of Continually Changing Information. Human Factors, 39(4), 532-539.
Home | Moving WebWord | Cool Books |
Hot Web Sites
Interviews | News | About WebWord.com
Subscribe to Webword.com
Contact John S. Rhodes, the WebWord.com Editor and Webmaster
(c)1998 by John S. Rhodes. All rights
Do not reproduce or redistribute any material from this document,
in whole or in part, without explicit written permission from John S. Rhodes.