How Not to Give a Research Presentation

August 1st, 2006 | View Comments

I spent the better part of last week in Vancouver, attending the annual meeting of the Cognitive Science Society. It was a whirlwind, listening to talks, connecting with people new and old, and taking some time to see the beautiful city as well.

The bad thing about conferences is that it makes it obvious that most academics have absolutely no formal training in public speaking. It is an underappreciated truth in academia that public speaking is performance art and you must put in the time and effort to develop this skill if you want to do it well.

To me, the lax attitude towards training good public speakers is a great tragedy of academia, as so many good ideas get lost in talks that are disorganized or unintelligible. It frustrates me to no end to see people who must give talks regularly as part of their profession make mistakes that I wouldn’t have tolerated in eighth graders during my stint as a middle school forensics coach and judge.

So here is a list of common mistakes and how to avoid making them in your research talk.

Reading instead of talking. It is bad to read your talk verbatim from the sheets of paper you’re clutching in your hand. It is bad to turn your back on the audience and read your talk off your Powerpoint slides. If you’ve ever carried on a conversation about your research, then you are fully capable of talking normally about it, so stop acting like a customer service representative, ditch the script, and just…talk.

Mumbling. A microphone is not a substitute for good enunciation and vocal projection, and given the reliability of auditorium technology, odds are good that you will have to give a talk sans mic some day, so take the time and learn how to talk so people can hear you. The trick to projection is not to use your throat; you’ll tense up and wind up losing your voice. Instead, use your diaphragm to support your voice (scroll about halfway down). The technique is akin to how singers breathe.
Unreadable slides. Keep your slides simple. Contrast is good. If you don’t have an eye for color, err on the side of readability and go with black and white. Large, simple fonts are good. Busy backgrounds are bad. Label your axes. Do not present two-dimensional data with three-dimensional graphs. Make sure that someone sitting fifty feet away in bright light can distinguish between the blue line and the green line.
Slides that distract from what you’re saying. This includes slides that annoy, like slides that make random noises, slides that have bullets flying in from all directions, and slides that take three seconds to dissolve into the next slide. This also includes slides that contain way too much text, because people cannot read your 500-word essay-on-a-slide and listen to you at the same time. Bullet points are called bullet points for a reason. They should be zippy, and they should be used sparingly. Paragraphs are neither bullet-y or pointy; they are blunt, heavy objects. Don’t bludgeon your audience with them.
An obvious lack of preparation. Talks should be crafted carefully, with the time limit in mind. It is unacceptable to prepare 100 slides and just stop wherever your time happens to run out. It is unacceptable to prepare 100 slides and insist on slogging through all of them, even though your time ran out twenty minutes ago. Set up talk triage—know what topics you must get to, what topics would be nice to get to, and what topics you can use as filler if you have time to kill. And then practice your talk to smooth out the ums and ahs and check the timing. If you have a fifteen-minute time slot and your slide title at the 12-minute mark is “Background”, then you need to go back to the beginning of your talk and start slashing and burning.