Take a law of physics (Newton's Laws of Motion, the Laws of Thermodynamics, etc.) and provide a demonstration. Discuss a well-known chemical reaction (vinegar & baking soda, or Mentos and diet soda).

Vinegar and baking soda is a chemical reaction, but Mentos and diet soda is *nucleation*! Explain that one, Mr Non-Teacher!

1) Visit the Exploratorium some time and see how the Explainers do it (or, occasionally, need to work at it). griffinwolf used to be an Explainer and might have some pointers. If it helps, it appears the next Free Day at the Exploratorium is March 3rd.

2) Lightning Talks (see links) and Noisebridge's Five Minutes of Fame talks might be good to see (unfortunately missed the latter, last night in San Fran).

3) Watch some TED talks. It's an opportunity to goof around on the net while actually learning some cool stuff!

4) Watch the masters at work. James Burke (Connections), Carl Sagan (Cosmos), Tim Hunkin (The Secret Life of Machines). You'll find endless clips on YouTube. Another great time sink!

I'm reminded also of something I recall reading about the "demo or die" principle at MIT's Media Lab: they would practice presentations with their hands behind their backs or in pockets in order to avoid the nearly uncontrollable impulse of "hand waving" when explaining abstract concepts. If you can't get your point across clearly in words alone and need all kinds of flailing body language, you're not communicating clearly. When actually giving the final presentation they were free to hand-wave all they wanted, with the catch of staying close to the verbal plan they'd previously developed and practiced...the body language is then an enhancement rather than a crutch.

Ignite Week starts Monday.

Their format is pretty different - 20 slides that auto-advance every 15 seconds - but watching the videos might provide some insight into how to structure a five minute presentation.

I had to do one of those once.
I started with writing Ophidiophobia on the board, then told them I had brought a "friend" to assist me in discussing the term. Then I took a rubber snake out of my pocket. That led to a discussion of snake biology.

You're talking high school science, right? At least, that's what you'd be applying for?

If you're willing to take yourself down to elementary algebra (say, around a grade seven/eight level) there's this one that I did a while ago:

On one desk, put three paper bags and one block. On a different desk, put ten blocks. Explain to the students that there are the same number of blocks on each desk, but most of the blocks on the first desk are in the bags. The only rule is that each of the bags has to have the same number of blocks in it. Their task: figure out how many blocks are in each bag.

When they've finished, write an equation on the board: 3*(paper bag drawing)+1=10. Point out that they just solved the answer to the question, "What number is in the paper bag?" In math, we call this a variable.

Then have them make more equations with paper bags and blocks, and write them using variables. Obviously, you can make the lesson harder by changing up the equation; you could probably introduce a table of values with, say, two different colours of paper bags, where the values of one bag would change in relation to the other, but ONLY in relation to the other. That would take you firmly into middle-school algebra.

Now, this is a problem-solving lesson, so there's some work for the kids to do in the middle of your five-minute lesson. I don't know if that fits into your assignment, but even if it's not stated explicitly, it's not a bad way to go. If you can make the point that you consider experiential learning to be the single most important element in any classroom environment, and teaching students how to problem-solve the cornerstone of math and science education, you've got a better shot at the job than if you go with explain-and-answer-questions.

Your homework: research constructivist models of education and pick a couple of pet topics of yours, preferably ones that fit into your state's curriculum at the level you'd like to teach. Either dig up or create constructivist lessons for your pet topics, with as much hands-on problem-solving and accountable talk (where the kids solve problems by discussing them with peers, then present their findings to the class) as you can squeeze in. If you go into an interview with a bunch of those lesson plans in a portfolio, you're pure gold to them. It will mean, however, that you're committing yourself to a style of teaching you didn't learn under.

Middle school/elementary school, actually. And I didn't follow ANY of that, particularly "constructivist".

The dominant psycho-educational theory at the moment is called constructivist. It means that humans construct their own mental models of new knowledge based on three main factors: their experiences, their prior knowledge, and their social interactions. A teacher's job is to figure out where their knowledge stops on a certain topic, and present them with an opportunity to experience the knew knowledge he wants them to have. Then the teacher should give the students opportunities to work through the experience in conjunction with their peers. The last step is a debriefing in which teacher and students look at, not just the answer and the new knowledge, but the strategies used to get there. Many ways of doing the same thing can be celebrated as right. In math, the goal is to move students from inefficient strategies towards more efficient ones, but only after they've figured out that the inefficient strategies do in fact work to solve the problem.

Traditional education has focused on the teacher imparting knowledge to the students, who were considered to be empty vessels, ready to soak up the knowledge and regurgitate it onto a test. Experiential or constructivist learning has students doing stuff and talking about it, and then presenting what they learn to the rest of the class so they can build their understanding based on what other students did. The teacher guides the process by asking questions, pointing students in a direction that might work if they're stuck, leading the discussion, and assessing the learning and the necessary next steps.

In science, this means figuring out what they'd expect to happen, and either running through the experiment for themselves or having the teacher run through it. Then you ask, "Okay, why did it happen like that?" and get their rationales for it. Once they've constructed a theory as to why it works, you can present the real reason and ask them to compare and contrast their theory with the official one. It actually fits extremely well with the scientific method.

If you can't do an experiment on a topic, you can do a sort. For example, you might take a bunch of life forms from the same order but different family, genus, and species, and put them each on index cards with some pertinent information about their defining characteristics. Then you hand them the stack of papers and ask them to sort the life forms into groups. They decide on the groups and on the criteria for membership in the groups, but they have to be able to justify it later. You can get several groups going with the same set of cards and see if they do things differently, and how. After the class discussion, you present how biologists classify those animals, discuss the names, and compare and contrast. The result is a lesson that gives them some idea how these decisions were made by biologists in the first place; exercises their ability to sort and reason, a key mathematical and scientific skill; and gives them a reason to be interested in the presentation of the actual classifications. They'll probably argue with them, and a few will go away thinking the biologists who made those classifications were really dumb. That's okay; they'll remember them better that way.

The problem with fitting all this into a five-minute lesson is that the explanation comes after the experiential learning, and is dependent on it. The questions are definitely dependent on it. So you may need to create some pseudo-student work in order to base your discussion off of that.

I think your best bet, honestly, instead of picking something that you think is impressive or easily demonstrated, is to pick something you're excited about, think of a simple way to explain it, and let your natural enthusiasm show.

Seriously, if you can let people see how amped you can get about topics, if you show them that, they're gonna see how you can be a great teacher.

If you want to demonstrate seismic waves you might bring a slinky and a box of marbles. This will show how P waves-- primus = primary = "push" or "pressure"- will go through a liquid or a bag of marbles, while secondus = secondary = "shear" or "shake" won't. But will go through the slinky.

Oh, the reason this is useful? It's how we know that the outer part of the Earth's core is liquid. P waves will go through it but S waves don't.

Just a suggestion from the resident geophysicist, if you do decide to go ahead with the seismic wave demonstration.

you can always fall back on the master:

Years back when at college we had to do a talk to our peers ... my subject was radar and I had a notion to use ping pong balls to bounce from objects to rangefind them....
The funny thing was that I bottled out of using the balls - with the consequence that my friends were paying wrapt attention to me.... waiting for me to get my balls out.....

Good Luck!

I've got 14 years background as a technical instructor. Finding a subject that's only 5 minutes long is a real challenge.

When I was "learning to be a trainer" coming up with subjects for 30 minute lessons were challenges. I can't imagine doing a 5 minute one.

I'd tend to look at a simple skill session for your 5 minute session ... like teaching somebody how to properly strip a wire or set a mousetrap.