Streaming video: friend or foe?

Progress can bring with it unexpected problems, and new technologies may harbor hidden demons. Asynchrony is one such monster.

It is said that in life, “timing is everything.” This cliché is arguably true for many human endeavors. Sports, love relationships, and business transactions are just a few that immediately come to mind. That it may also be true for distance learning has important implications for how we design online training.

With human interactions, timing is critical.
A too-quick response in a conversation suggests either enthusiasm or a lack of thoughtful processing by the respondent. A too long delay in a response — even momentarily — can cast doubt on the speaker’s interest, sincerity, or intelligence. A split second either way and the implications change.

So what happens if timing is “off” in a media-delivered message — if the video of a “talking head” is ever so slightly out of synch with the spoken words? Will it matter? And will it matter if, in fact, no one notices?

Those who promote the new technologies argue that it makes little difference if it’s noticeable that sound is slightly “out of synch” with the video. They say that it’s a small price to pay for the power of audio/video on our computers. Streaming-video advocates maintain that users will simply discount the irregularity as an unfortunate but minor problem caused by a still-developing technology. Besides, the technologists argue, users are smart and sophisticated enough to understand the limitations of current bandwidth and can accommodate such slight discrepancies in media delivery. Are they correct?

Stanford Studies Explore Asynchrony
Maybe not. A growing body of research suggests that something subtle and insidious is actually happening on the Web when a speaker’s mouth doesn’t quite match its audio. And it doesn’t seem to matter whether or not users are happy to have the latest in online audio/video streaming or smart enough to understand the “minor” problems that accompany it.

One such series of studies was conducted by Byron Reeves and Clifford Nass, professors at the Institute for Communication Research at Stanford University. Drs. Reeves and Nass had participants in their experiments see and hear two levels of audio-video clips:

• perfect synchrony, with no apparent mismatch between the video of people speaking and the words coming from their mouths
• audio and video mismatched by 5 video frames
  (one-sixth of a second)

For each participant, four of eight “talking-head” (news anchor, commentator, etc.) video clips were randomly selected, with each clip running two to three minutes in length.

What did the researchers learn?

Simply stated: Audio-video asynchrony results in diminished evaluations of the speaker and the content being spoken. Even slight imperfections in synchrony create strong negative reactions in the viewer.

Unnoticeable Asynchrony Is Problematic
The researchers were surprised to discover that only half of the participants subjected to the asynchronous video actually noticed audio-video synch problems. But — and here’s the most worrisome part — participants viewing the non-synchronized audio-video clips evaluated the presentations more negatively regardless of whether or not they could identify the audio-video mismatch. The presenters viewed in the asynchronous videos were judged significantly “less interesting, less pleasant, less influential, more agitated, and less successful in their delivery... and, importantly, this effect held for viewers who could not even identify the [audio-video] problem.”

Why does asynchrony create such pronounced negative evaluations? Professors Reeves and Nass posit that audio-video mismatch is fundamentally unnatural to humans, regardless of whether it occurs via media or elsewhere. They hold that human beings using a computer or any form of media delivery, for that matter, expect an interaction — a “social relationship” they call it — that mirrors what humans experience in real life. Any unnaturalness in this social relationship is perceived as problematic.

Furthermore, the researchers conclude that the perceived unnaturalness of the out-of-synch audio/video is not totally attributed to the technology, as some streaming-video advocates may argue, but also to the source and content of the presenter. And when viewers are not aware of the timing problems (about 50% of the time), their negative evaluations fall totally on the presenter!

Sadly, this phenomenon occurs across a slew of media: video-conferencing, web-based corporate presentations, and distance learning, to name just a few.

Talking Heads’ Troubling Implications
The implications of these findings are clear and troubling. What company CEO would want his video-taped annual address delivered over the corporate website if he knew that viewers’ evaluations of him and his corporate vision will plummet? How many team members would choose to meet via video-conference technology if they realized that others’ perceptions of them and their ideas would be deemed “less influential” than when meeting via teleconference or in person? And how many instructional designers would construct online training interventions that include “talking heads” if they knew that learners would find the content “less interesting” and “less important” than when presented in other modalities?

What’s to be done? Should we live with this problem until bandwidth increases or until the technology is enabled to provide perfectly synchronized audio/video over the Web? Or are there temporary solutions?

IBM’s Solution
At IBM Management Development, designers have purposely avoided using video, not just because of synchrony problems, but also because many of its users are working from remote, phone-connected laptop computers where downloading even a short video piece requires a long wait. Hence, in lieu of video, IBMers are presented with a still picture of the presenter with the audio track running. Any nuance or subtlety (for instance, sarcasm) can easily be detected in the audio. Or additional still pictures with different expressions can be added as appropriate to indicate a change in expression. If timing of the particular still is off even by as much as a half second (much less one-sixth second) it does not matter because no words are synched to lip movement.

In fact, IBM has discovered that because the global nature of its business and its employee base, a still photo accompanied by self-paced text is preferable for those users for whom English is not their first language. Users can read and scroll through the text at their own pace, re-read sentences that may be unclear or warrant additional attention, or simply stop and reflect, and then return to the presentation — all without losing their place or missing any of what the source is “saying.” Sometimes, in our eagerness to adopt all the new “bells and whistles” of a new technology, we forget that self-paced text may be the most powerful and preferable learning tool of all.

Be Aware of Asynchrony’s Issues
Progress can bring with it unexpected problems, and new technologies may harbor hidden demons. Asynchrony is one such monster. Until we are able to effectively eliminate a/v streaming’s problems, we should at least be aware of asynchrony’s effects on the user. If so, we will be better able to make informed decisions about using streaming audio/video, or choosing a different approach.

Source: http://www.brandonhall.com
The author, Peter Orton, Ph.D., is Program Director of Global Curriculum Technology for IBM Management Development. Used by permission.