The serial position effect can also have an impact on a list of links.
The “serial position effect” describes how list items that are first and last in a list are more likely to be remembered than the items in the middle.
Furthermore, the medium of the list, how fast the list items are presented, and the time between exposure and decision, will greatly affect which is better: being first or being last.
The primacy effect describes the likelihood of remembering items that are first on a list – items presented first on a list are stored in long-term memory more easily than items further down the list (“Storage Mechanisms in Recall”, M. Glanzer, 1972).
When presenting list items to your users in video or audio (or any format where they don’t control the pace), beware that the faster you present the less your items will be stored in long-term memory. The primacy effect only occur when list items are presented slowly or at the user’s own pace.
The recency effect describes the likelihood of remembering items that are last on a list – items that are last on a list are still in working memory (“Two storage mechanisms in Free Recall”, M. Glanzer, 1966). There are two interesting characteristics of the recency effect:
The following graph compares the concepts of the primacy and recency effect:
This graphs shows how serial position effect works. The first items on a list are stored in the long-term memory. The last items are in working memory and are almost equally likely to be remembered at first, but only for a short period of time of 10-30 seconds. (Source: Universal Principals of Design, Lidwell et al., 2003 Rockport.)
You should take advantage of this knowledge when presenting information in lists (be it a set of links, your sales pitch, a feature list, client list, etc.).
Here’s some ideas on how to use the primacy and recency effect in practice:
You should also be aware of the primacy and recency effects if you conduct surveys or otherwise have people choose from a long list, as they can represent a margin of error. The “evil” researcher can use it to manipulate the outcome of a survey to his desires, the honest researcher will shuffle the list positions for each vote.
Authored by Christian Holst on October 23, 2010
Join 30,000+ UX professionals and get a new UX article every second week.