I’ll admit it: the usability geek in me is smitten with the coolest doodad of the political season, CNN’s live dial-testing of a focus group during the debates. There it is in real-time — the collective response of a group of undecided and presumably impressionable voters — swinging up and down with the pitch and inflection of the arguments being made by our would-be leaders on-screen. Powerful stuff. And terrible, too.

Powerful — and meaningful — if you’re a student of debates, or a political flack who wants nothing more than to slice and dice every mood-altering phrase, every cringe-induced twitch of the dial. Dial-testing gives you access to immediate, emotional, visceral feedback, without any of those messy social and psychological filters.

Terrible — awful, even — if you’re an undecided voter who’s watching the debates at home and trying to come to your own conclusions. Those wandering lines on the screen tend to suggest not only what other folks may be thinking, but what you should be thinking, too, even if you have nothing in common with the folks cranking the meters in their hands.

It’s not a question of whether voters should have access to every bit of information possible to make their decisions, but of whether the information they use to make those decisions is reliable and free from bias… especially from bias that they have no way to discern.

By way of example, this video captures CNN’s dial-testers scoring Joe Biden off the charts during a surprisingly emotional moment of the Vice Presidential debate. It’s hard not to be moved by the double whammy of Joe Biden getting a bit choked up, and the metered results of folks’ reactions to it.

Do you agree with the folks who’s feedback is measured in this example? Doesn’t matter… you’re likely to be influenced by it, regardless. Which is… problematic, at best.