Clippy & the Problem with Learning Aids

How does the design of a learning environment influence the way people use learning aids? Will a more obvious learning aid lead to higher levels of use? Earlier research has explored whether the content, usability, or supplementary cues might encourage people to use learning aids (think the info hover icon, or the ever-popular Office Assistant, Clippy!), and this study focuses on developing a learning aid that people are most likely to use by testing three different types of learning aids: dynamic, static, and collapsed. The researchers started from a process model of help-seeking (i.e., identifying, strategizing, and looking for help), and investigated how different designs affect the way people use—or don’t use—learning aids. The researchers also assessed the participants’ subject knowledge before and after the learning phase, both to establish a convincing learning environment as well as to eliminate prior domain knowledge as a confounding variable.

Usability rated significantly higher when learning aids were presented statically than when collapsed or dynamically; however, participants actually used the learning aid most when it was presented dynamically. Participants were least likely to look at the learning aid when it was presented in a collapsed form. This was particularly true for taking or modifying notes, and alternating between reading the support area and the content area—the two most frequent forms of usage.

This finding offers insight into the question of why people rarely take advantage of learning aids in computerized learning environments, where help is often initially hidden and activated only with a click. In this case, out of sight really is out of mind: people seem generally unlikely to search for learning aids.

The study is not without limitations. Because the goal was not to evaluate the instructional effectiveness of learning aids, this study did not investigate how people decide to use learning aids after successfully identifying them. Other factors (e.g., motivation to learn) likely affect the decision to take advantage of learning aids. So although participants used the learning aids more frequently in the dynamic and static presentation modes, neither mode resulted in enhanced performance on the knowledge posttest.

Usefulness is another important element in help-seeking behavior. In this case, the researchers were primarily interested in the presentation of learning aids—had they instructed participants on optimal use of the learning aid, they almost certainly would have triggered systematic (i.e., not spontaneous) use of the learning aids, regardless of presentation style.

This study demonstrates that an obvious presentation greatly increases the use of learning aids; however, both the learning aids and the learning environment were fairly artificial. It would be interesting to see further research on more complex learning aids, as well as dynamically presented learning aids in more realistic learning environments.

Ruf, T., & Ploetzner, R. (2014). One click away is too far! How the presentation of cognitive learning aids influences their use in multimedia learning environments. Computers in Human Behavior, 38, 229-239.