ACM Logo  An ACM Publication  |  CONTRIBUTE  |  FOLLOW    

Paying Attention to Attention

By Lisa Neal / October 2004

Print Email
Comments Instapaper

One of the more significant challenges we face in online learning is climbing the wall that blocks our view of learners responding to a course. In a classroom we can see who is making eye contact, nodding in agreement, or sighing with frustration. Above all, in a classroom, you can (usually) tell if somebody is tuning out. Without these cues, we are really just speaking into the void, hoping somebody hears us.

This problem is made all the more critical by the fact that there are so many more distractions online than there are in a traditional classroom. In addition to the classroom options of staring out the window or daydreaming, distractions for the online learner include emails, phone calls, piles of paper, colleagues, friends, and family. Multitasking is not only common—many practice it with pride. As instructors, we don't know who is paying attention and we have good reason to suspect that many are not. The easiest and most typical solution to this problem is to frequently prod the learner for an overt response: Answer a poll, respond to a question, choose an option. However, this sort of frequent artificial interactivity as an attention pulse-check can become tiresome for learners. Luckily, technology advances and people adapt. And, although there are no clean solutions to this problem yet, instructors are learning to subvert current tools to help them better monitor student attention, even as monitoring technologies improve. It's worth taking a moment to consider where we are and where we are going with all of this.

Sneaky teachers trump clumsy technology
Some online technologies provide a way of indicating when a participant is confused or if the pace is too fast or too slow what—Centra, in early versions of its synchronous technology, called "Interactive Body Language." Although this approach mitigates the annoying and distracting sense of being constantly prodded for a response by allowing students to respond when they want to (rather than when prompted and by more closely imitating natural classroom reactions), there are problems. To begin with, this kind of feedback still requires conscious action: The learner has to choose to click a button (and to know where that button is and what it does in the first place). The result is that only some students use these controls and, when they do, the feedback is delayed by enough seconds so that the presenter may not know what the confusion is about. Contrast this to the immediacy of a smile when a student "gets it," or a frown in a moment of confusion.

An alternative would be to physically monitor students. For example, a student of Roz Picard's at the MIT Media Lab experimented with sensors attached to the muscles between the eye brows. With minimal training, this could indicate surprise or confusion (try it with your own facial muscles). But there were a number of problematic issues: the slight delay, the possibility that some people could play with the sensors, whether data should be for the group or for individuals (and, in the latter case, would those individuals be named or would they remain anonymous), and the fact that most people don't like attaching things to their bodies. However, there are other forms of monitoring: using a Webcam—which students are likely to find even more invasive—or monitoring shifting movements in a student's seat, one of Picard's new areas of research.

WebEx recently announced a new feature, an "attention indicator," to determine whether the WebEx client's window is in the foreground. This is another less invasive form of monitoring, however one with only limited potential. A student may be listening attentively even though solitaire or email is in the foreground; and just because the WebEx window is in the foreground a participant is not necessarily paying attention. The use of desktop sharing by some online instructors carries this approach even further since the instructor can see exactly what students are doing without seeing the student (the exact opposite of what happens in a classroom, where you can see students but not necessarily what they are doing). LearnLinc provides a glimpse feature that allows the instructor to not only see but grab a screen capture of a student's desktop. In a demo, a salesperson showed how the solitaire-playing student could be publicly humiliated, an interesting pedagogical technique.

Yet seeing a student multitasking does not necessarily gauge level of attention. Some students may find it easier to listen if their hands or eyes are fully occupied, and some may more willingly attend a seminar, knowing that they can browse email. Chris Schmandt, at the MIT Media Lab, has found that people are good at distinguishing items of interest in audio streams, perking up when there is useful and relevant information. Remember, in school, the attention level when a remark was prefaced by "this will be on the test"?

Do they know that you know?
Having some ability to monitor learners' reactions is useful, but it's even more powerful when the learners know you can see them. Who can forget those terrifying, yet incredibly motivating, moments in elementary school when the teacher called out, "Eggbert, if that note is so funny, would you care to share it with the class?" Some online instructors have managed to simulate this situation by sharing the student's desktop with the entire class. "Eggbert, if that text message made you LOL, would you mind sharing it with the class?"

But this is only part of the classroom dynamic that we want to be able recapture. Student awareness of the teacher's attention can be positive as well as punitive. There's nothing like a moment of eye contact and a quick smile to make a student feel motivated to please the teacher. At the moment, there is no analogue to this kind of multimodal conversation in online learning. Because all communication is overtly directed by our conscious minds and sent mainly by text or disembodied voice, we lack the parallel channels for the rich communication that comes through body language. The learner is speaking into the void (or listening to it) every bit as much as the instructor.

Do they know what you know?
It may turn out that, as with many things in distance learning, the best way to compensate for a loss of capability on the instructor's side is to increase empowerment on the learner's side. We just might be able to help students to monitor their own attention more effectively. Ironically, this is one area where technology is arguably more effective than traditional face-to-face interventions. Biofeedback helps people take more conscious control over their normally unconscious reactions. It has been around for decades and is becoming increasingly available in affordable consumer applications. For example, Journey to the Wild Divine is an interactive game featuring a biofeedback sensor that hooks up to your computer through the USB port. It teaches yogic-like meditation techniques by allowing you to monitor your physiological responses with graphic feedback. For example, when you achieve the right (measurable) level of relaxation, the feather floats on the screen. It's your very own Yoda in a box. In a more practical realm, Ford and other car manufacturers are putting in systems that help drivers monitor their own attention levels by beeping when the car starts to drift out of its lane or when you fail to slow down even though the car in front of you is rapidly decelerating.

As these sensors improve to become more effective, less intrusive, and less expensive, we may be able to augment our distance learning with tools to help learners become more self-aware, making them better learners who need teachers monitoring their attention a little less. This will never be a substitute for the close contact of a face-to-face class, but it will hopefully have its own unique strengths to compensate for its weaknesses.

In the meantime, engaging and interactive lectures tailored to students' needs and delivered at a good pace will always hold attention better than those that are poorly delivered and include uninteresting or poorly prepared material—so will prompting for questions, holding virtual office hours, and asking learners to let us know if they still want to learn.

Lisa Neal is Editor-in-Chief of eLearn Magazine.
Michael Feldstein is CEO of MindWires, Inc., author of the e-Literate weblog, and a member of eLearn Magazine's Advisory Board.


  • There are no comments at this time.