BBC Sounds, Music, Radio, Podcasts.

It's spring in San Diego, California, and PhD student Mark Beshe is feeling nervous.

He's made up the steps to the stage without falling over and fired up his PowerPoint.

In a precise Swiss-German accent, he starts his carefully prepared presentation.

He needs this to go well.

The research he's sharing is nothing fancy,

just a small study conducted with two others on a shoestring budget.

But this audience is full of potential employers

and people who can give out lucrative research contracts.

And he's worried they'll judge him poorly.

Because as he comes to the end of his talk,

one slide is projected onto the huge screen.

It's a simple bar chart. but some of the bars are much,

much higher than the others. In fact, it looks a bit of silly. He's not quite sure how to explain

the results. He knows that they are odd, but that is the data they got.

He tries to justify it to himself.

After all, his group was commissioned to run the tests and record the outcomes, not to explain them.

The presentation goes well.

In the follow-up questions, no one challenges the methods or the results.

Another study done and delivered, and now Mark can relax.

What he doesn't know is that he's just set a very small snowball rolling down a very big hill,

because the data that Mark and his colleagues captured will have seismic implications

...