Is text plus audio really reinforcement—even when they convey the same thing?

We process information from different channels; we process them differently, too. There are some incorrect conclusions based on this, one being that text plus audio means reinforcement (even when they convey the same thing). That idea has been corrected quite a few times—in this post on Cathy Moore's blog, for example. What about pace... what happens if you want to make a courselet self-paced? What works and what doesn’t? Here's a short roundup.

Audio and text is tricky.

Unless the audio is saying the same thing as the text, audio can work with text on the screen. The narrator can quickly talk about what’s on screen. If narration precedes the text, perhaps it can give an idea of the forthcoming topic. (Music can sometimes be used to set the tone; we don’t seem to use it often enough!)

The problem with audio-plus-text arises, most obviously, because you can read quicker than the narrator speaks.

Audio and images blend well.

With images or short animations, a voice-over can work very well indeed. Factual info (including lists and detailed descriptions) is probably better as text; a voice can often augment a presentation that isn’t text-heavy.

Audio isn’t self-paced.

When it goes along with something more complex than images (like animations and simulations), a narration is very often useful, with the trade-off that the material won’t be self-paced. Is this an important consideration?

Maybe. For many people, being tied down to the speed of the narrator won’t matter. But if your learner is racing ahead—if some of the matter is familiar territory, for example—the “slowness” can be unnerving.

The shorter the speech segments, the less of a problem this is.

In just-audio, the key is length.

Podcasts—as an example of pure audio—are great when you’re putting across point, a topic, an idea. As the message becomes more detail-oriented and less about an idea, they’re more restrictive in terms of pace.

With some effort, an audio-only course can be broken down into appropriately-sized chunks. This would be easier if the narrator or speaker were familiar with the subject.