Are We Causing Readers to Forget?

Doorway

Copyright (c) 123RF Stock Photos

Could the way we organize content actually be causing readers to forget what they have read, or even why they were reading?

In a post on the Technical Communication Professionals Email List, Mike Tulloch provides a link to a study from Notre Dame that suggest that walking through doors causes people to forget things  (http://newsinfo.nd.edu/news/27476-walking-through-doorways-causes-forgetting-new-research-shows/). The theory is, apparently, that passing through a doorway is a threshold event that triggers the mind to store away information, which then makes it harder to retrieve that information. Mike wonders if there may be similar threshold events in text:

Does whitespace between paragraphs serve as a textual doorway, which causes readers to forget what transpired in the previous paragraph? … Does turning the page causes memory loss? The very concept of “turning the page” implies a forgetting what is behind, so perhaps there is something to all of this.

It is certainly an interesting question. And there are a number of other threshold events that would be worth studying as well. Does going from the text to the index constitute a threshold event? What about going from the index to a page pointed to by an index entry? What about following a reference to a table or an illustration? What about going to the table of contents?

For online content, there are similar questions. Does following a link create a threshold event? Does clicking an up or back button on a topic to get to a table of content page create a threshold event? What about clicking a “next page” link in artificially paginated web content? If there is such an effect, could we or should we change how we design books and websites to lessen the number of threshold events and help readers retain more of what they read?

We won’t know until someone does the necessary studies, so what follows is pure blue-sky speculation, but these are my immediate thoughts:

It seems to me logical to assume that part of what makes passing through a doorway a threshold event is that on the other side of a doorway there is a brand new room full of new sights and sounds to take in. That flood of new information displaces the information that was previously uppermost in the brain. Passing through a door means walking into a new situation full of distractions. What conclusions might we draw from this entirely unsupported hypothesis?

  • It is better to give the reader everything they need to complete one task on one page. In other words, it is better to write Every Page is Page One Topics than to present information in more granular chunks that require the reader to cross multiple thresholds.
  • Sending people to a TOC or an index is sending them to a room full of new possibilities and new distractions that could cause them to forget what they were doing and what they were trying to look up. If you have to send them away from the current page, therefore, it would be better to send them directly to the external information via a link than to send them into a TOC or index.
  • If a reader gets stuck on something on your page, they may Google for more information. (I know I do Highlight > Right Click > Search Google for… all the time when I am reading on the Web.) If whitespace or page turning or checking an index is a threshold event, then doing a Google search is ever so much more so a threshold event. Providing a link within the text that sent the user directly to another page on our own site could avoid this large distraction (besides keeping the reader on our own site).

Of course, if turning a page is a threshold event, then following a link is too. But at least a link would seem to involve exposing the reader to fewer distractions than sending them to an index, TOC, or Google. If there is anything to all this, then putting all the information for one task in a single topic will clearly be the best strategy. But if that is not possible, then it seems reasonable to believe that providing direct links will be better than sending people off to TOCs, indexes, and search.

Of course, there may be nothing to this at all. We would need the studies to be sure. Perhaps someone will do the research. Perhaps someone already has. I’d do a Google search for it, but I’m afraid that before I was finished I would have forgotten I was writing this blog post.

Anyone know of any research that would refute or support these speculations? Anybody got their own blue-sky theories to share?

 

 

, , , , ,

7 Responses to Are We Causing Readers to Forget?

  1. Nuera 2011/11/23 at 07:30 #

    Hi Mark,

    I tried subscribing to the RSS feed to stay updated with your blog, but it shows an error “This XML file does not appear to have any style information associated with it”.

  2. Larry Kunz 2011/11/23 at 10:27 #

    The question you pose in your title is a great one. Given the anecdotal evidence that readers are coming to us with shorter attention spans and with many things to distract them, we need to do everything we can to keep the readers focused and on target.

    For me, clicking a link is definitely a threshold event. Especially if the link opens a new page in the same tab, rather than opening a new tab.

    The IBM Style Guide (a very comprehensive guide for writing software documentation, and now available to the general public) recommends placing links in a list at the bottom of a topic (page) rather than inline. Their research has found that a reader is less likely to become distracted when the links are at the bottom, where they won’t entice the reader to go off-topic but where they still can be used if needed. This is also what I recommend when I teach structured authoring and DITA.

    • Mark Baker 2011/11/29 at 20:10 #

      Hi Larry,

      Thanks for the comment. I certainly agree that if there is any correlation between the physical passage through a door and anything in content, then following a link is going to be a threshold event. You raise an interesting point about where the link opens. I hadn’t considered that. I have always thought of new tab vs. same tab as being largely about the type of link. Forward/back linking seems to call for same tab, sideways/ancillary linking for new tab. If you are right that same-tab linking is more of a threshold event, then that certainly corroborates something I have long felt, which is that forward/back links are just wrong and a topic should be complete and self-sufficient one one page, without linear dependencies on other topics. (Every page is page one!)

      Now, whether merely seeing a link is distracting is a different matter. I certainly used to find links in text distracting 10 or 15 years ago, because my brain saw them as emphasis on the text itself. But they don’t distract me in that way any more, partly because I am just so used to reading text with links in it, and partly because links are much more subtly colored than they were 10 to 15 years ago, and almost never underlined anymore.

      I’m aware of IBMs research on this, but I think time may have passed it by, as it has so much of the early research on web usability that was done when people were still not fully adjusted to the way the web works. And in any case, I think there is a countervailing argument now in favor of in-line linking: anyone can make their own link today by highlighting any phrase and doing Right Click > Search Google for ….

      It is far better, in my view, to put a link on any text that the reader might want more information on (if we have such information, of course), for the following reasons:

      * Encountering a term or a concept you do not understand is, in itself, a threshold event. It is like walking through a doorway into a room filled with smoke. We can’t avoid a threshold event by trying to force people to continue reading a text when they don’t understand the terms or concepts it is using.

      * The characteristic impatience of a modern Internet users means they will not contain their curiosity to wait for links at the end of a topic. And the characteristic P pattern that eye-tracking studies show us, means few people will ever look at a list of links on the bottom. We either serve their curiosity in the moment it occurs, or they will satisfy it themselves.

      * If they choose to satisfy that curiosity through Google, that Google search is a threshold event, and then every page their look at in the list that Google returns is another threshold event. A reader who goes off to Google for themselves will go through multiple threshold events that (if there is anything in this) will greatly increases the chances they will forget what they read on our page.

      * If they choose to satisfy that curiosity through Google, the chances are that they will end up leaving our site and going off to someone else’s content. This is in no way what we want. If they are curious or confused at any point in our text, we want them to turn to some other text of ours to satisfy that curiosity or remedy that confusion. As with customers, it is much more expensive to get a new reader than to keep an existing one.

      So, taking all in all, I think we serve our readers and ourselves much better by placing in-line links in our content.

  3. Kai 2012/01/03 at 05:09 #

    I agree with your idea to “give the reader everything they need to complete one task on one page”. I mainly use links to re-direct users who’ve come almost, but not quite to the page they were looking for. Or to move them across the threshold to the next logical step.

    For example, I link from a concept page to the “how to set it up” page. Or, if there are two separate procedures or user types to set up and operate a function, I link from “how to set it up” to the “how to use it” page.

    Because each such page is logically distinct and can be read alone, it’s fine if following the link causes readers to forget. They often were not interested in the previous page anyway.

    Also, I keep in mind that much of my documentation is meant as external knowledge which I don’t expect readers to retain (though remembering relevant concepts would be helpful). It’s rather training materials often are meant to establish internal knowledge which users remember.

    • Mark Baker 2012/01/06 at 16:55 #

      Hi Kai

      Thanks for the comment. Helping people who are close, but not on the right page, to get where they really should be is something I consider really important. To me, that is a key part of writing a good Every Page is Page One topic.

      You raise a really good point about forgetting. Forgetting is not always bad. Forgetting clears out irrelevant information and allows you to concentrate on what matters in the moment. That one could use page delineation and linking as a means to help users manage what they forget and when is a fascinating idea.

      On reflection, I realize that we do this all the time when we are trying to explain something to someone. They get fixated on some point, and we are eventually forced to say, “Forget about that, focus on this.” Happens all the time, now that I think about it.

      So then I guess the question becomes not “Are we causing readers to forget?” but “Are we causing readers to forget in the right places?”

      • Kai 2012/01/07 at 04:05 #

        I think we write good and helpful documentation if we allow – or even: support – readers to forget the “right” information. Most reference information (such as default settings, which setting maps to which function) should be instantly forgettable, once the reader has achieved the (sub-)goal at hand.

        And readers will feel good about forgetting, if they are confident that they can find the information again. Sort of like in a library or a department store. (And how it really upsets you when “your” place has re-arranged all the shelves again…) Maybe you’ll find an old post of mine helpful: Top 10 things that users want to do in a help system.

Trackbacks/Pingbacks

  1. Re-Thinking In-Line Linking: DITA Devotees Take Note! | The Content Wrangler - 2012/05/03

    […] a comment on my blog post Are We Causing Readers to Forget, Larry Kunz made reference to the IBM style guide (I don’t have a copy myself, so I am […]