Normally, the trick to learning something at a scientific meeting is to listen to the key lectures. But one afternoon last month, in a conference room at CERN, the European particle-physics lab near Geneva, physicist Matt Strassler managed to convince several researchers that they might learn more if they left the lecture room. He wanted them to avoid hearing the solution to a puzzle they had been working on for months.

Welcome to the strange world of the Large Hadron Collider (LHC) Olympics, a workshop held at CERN in which teams of theorists studied fake data in order to explore unproven theories.

Strassler, a theorist from the University of Washington in Seattle, was one of the organizers of the event, which brought together more than 50 theoretical physicists from across Europe and the United States. These ‘olympians’ have devoted their careers to building mathematical models of the Universe and matter. The LHC Olympics was designed to put their ideas to the test; their challenge was to prepare for the real data expected to emerge from one of the biggest experiments in physics. Strassler and his colleagues had accepted that challenge —and they weren't going to give up on their first attempt.

Monster machine: huge detectors will collect and characterize the fall out from proton collisions. Credit: CERN

That experiment is the LHC, a machine under construction in the tunnels beneath CERN. When it is switched on in 2007, the collider will be the world's most powerful particle accelerator; it is expected to churn out data that will test some of the most cherished theories of physics. But if something turns up in the LHC that no one has predicted, theorists will have to work backwards from a set of observations to try to find a model that fits.

“Everyone in this room wants to be prepared so that when the first slide of the first data from the LHC is presented, we will understand what we are seeing,” says Jesse Thaler, a physics PhD student from Harvard University in Cambridge, Massachusetts.

Bright sparks: Nima Arkani-Hamed (second from left) helped organize a collider workshop at CERN. Credit: CERN

Studying the debris from collisions in particle accelerators has already helped physicists piece together many details of the subatomic world. But the LHC is pushing particle physics to a new frontier.

When high-speed protons smash head-on in the LHC, obliterating each other, seven times more energy will be released than from similar collisions in today's most powerful machine, the Tevatron at Fermilab in Batavia, Illinois. This raises the possibility that in the explosion of particles that sprays out, some new, exotic speck of matter might be found.

Full speed ahead

The experimentalists building the mammoth detectors that collect and characterize the debris will comb through the data, searching for new particles. They have many ideas of what they might find at these energies, but it is possible that there may also be some complete surprises. “To go from seeing particles to understanding theories is a big step. This is where the LHC olympians will come in,” says Albert de Roeck, physics analysis coordinator for the LHC detector known as the Compact Muon Solenoid (CMS).

Physicists have not had to contend with unexpected physics since the ‘standard model’, which divides matter into families of particles, was hammered out in the 1970s.

“The 1960s and 1970s were the golden years for particle physics,” says de Roeck. “The energy that was there then hasn't really been there since. The hope of the whole particle-physics world is that the LHC will bring that era back.”

I can't emphasize enough how tense and exciting this time is. Nima Arkani-Hamed

The LHC Olympics was also designed to attract new people into the fray. Herman Verlinde, a string theorist from Princeton University, New Jersey, is one of the newcomers who helped organize the meeting. “String theorists, like any physicists, are attracted to exciting areas of physics,” he explains.

String theory is one contender for a deeper theory of nature, which goes beyond standard-model physics. Despite its unqualified success, the standard model contains at least 19 arbitrary parameters such as particles' masses that have to be measured from experiments. Physicists seek some greater underlying theory to explain these numbers, and have dreamed up numerous formulations that might provide it.

Going for gold

Some of these theories predict that the standard model will be supplemented by families of heavier particles, which the LHC may find. And the LHC could identify dark matter, uncover evidence for extra dimensions, or create miniature black holes. At the very least, physicists hope it will find the Higgs boson — a hypothetical particle, predicted to exist by the standard model, which is thought to endow other particles with mass. Or it might find nothing at all.

“I can't emphasize enough how tense and exciting this time is,” says Nima Arkani-Hamed, a theorist from Harvard University, and one of the driving forces behind the LHC Olympics.

High-impact: when protons collide in the LHC, their energy will be converted into a spray of particles. Credit: CERN

Last time CERN made a big discovery, Arkani-Hamed explains, the theorists had known exactly what the experimentalists should look for. “This time,” he says, “the theorists are saying ‘maybe this will happen, maybe that will happen.’”

Arkani-Hamed first became worried in 2004 that many theorists were not prepared for the flood of data that will soon issue from CERN. In his original vision for the LHC Olympics, experimentalists would create a simulated data set that obeyed some hidden physics from beyond the standard model, analyse it as they would the real data, and then present their plots to an audience of theorists.

“The idea was that we would have a massive data challenge and then a mini conference,” says Maria Spiropulu, who will work on the CMS data analysis at CERN. At the conference, the theorists would be challenged to explain the experimentalists' plot, “Then they would say, ‘Eureka! This is intersecting D-brane string-inspired supersymmetry.’”

But right now, as the start date for the LHC looms, the experimental teams are too busy to get involved with games. “We're working 16-hour days as it is,” says Spiropulu. “But these people don't give up. The theorists said, ‘If you give us a hint, we'll do it ourselves.’”

So, a few months before the February workshop, several theorists put together three simulated data sets. These ‘black boxes’ were then posted on a website for other meeting participants to download. Could the conference-goers work out, from the plots and tables, what physics the data described?

As you might expect, says Spiropulu, the theorists' data analysis was very basic. “But for them, this was training so that they understand the language that we are using.” With real LHC data, one of the big problems will be distilling the bumps that signify new particles from the ‘background’ events produced by standard-model physics. No background was included in the theorists' black boxes.

Ian Hinchliffe at the Lawrence Berkeley National Laboratory in California, a physicist on the ATLAS detector team, worries about this omission. “It might lead some theorists to wrong conclusions about what is possible,” says Hinchliffe, who ran a data challenge for the people working on ATLAS, which included the necessary background events. That experience, says Hinchliffe, taught them how hard it might be to recognize new events.

Runners up

But the LHC Olympics is not a bad starting point, says Strassler. He describes the black-box exercise as “a sort of flight simulator for someone who wants to understand clearly what the pilots are doing, rather than something for the pilot”. And for the next meeting, planned for August, the theorists hope to develop at least one black box with background.

First to present his black-box solution this time was Verlinde. “Clearly we are not going to impress you with our skills in analysing data,” he began. “You can see how far one can get, starting basically from zero, with minimal students and minimum sleep over the past few days.”

Another presentation came from Harvard University students, who had tested 2,500 models, using an algorithm to optimize their first guess of what particles in the black box might be. “Cheaters,” someone called out when the team thanked the Harvard-Smithsonian Center for Astrophysics for allowing them to use its powerful computers; the heckler's own talk was titled “What you can do with a black box, an undergraduate and two weeks”.

Both Verlinde and his co-workers, and the Harvard students had figured out that the box they had studied was based on a version of supersymmetry — among the leading theories for a physics beyond the standard model. And they identified the masses of most of the particles that this theory predicts — from squarks, to winos and zinos. There were some differences between the two groups, but their answers were almost the same. “We're quite proud of our guess,” says Princeton University theorist Leonardo Rastelli who collaborated with Verlinde.

But Strassler hadn't figured out what was going in the black box he had tackled. That's why, when its solution was to be announced, he persuaded several others who had been working on it to leave the room with him. He wanted to keep going until he cracked it. “If the black box hasn't been decoded, then the learning process isn't over.” And of course, when the real results come in, there won't be anyone who can reveal the answer. It will have to be obtained through hard work.