That feeling of awe

I'm in a little bit of shock and I don't know how to put this.

It's 4 in the afternoon and I'm in my office.A little over two years ago when I arrived in Hong Kong, I started a research project with someone in my department. This person is a few years senior to me, and has been a close collaborator with my advisor. So much so that as a grad student I'd effectively been his RA at times, since I used to work at night in the US, which is while he was online in HK. He was also instrumental in bringing me here, pulling for me at recruiting meetings and pushing me hard when I was undecided. Plus he's a great fun person, heavily into cricket, etc. etc. etc.

However, his research area was completely different from mine, but at that stage after tenure, he was looking to broaden out a little. So he had this idea which he broached to me. It didn't sound like the most exciting thing in the world but I said what they hey, let's try it. I thought it would be good to have some research projects with people in my new department, and this one was good enough.

Somewhere along the line, during the discussions, a third person got invited on to the project. He's based in the US and was passing through HK at the time, and got talking about this.

The three of us exchanged a few emails. The two of us in Hong Kong did more of the talking. Eventually, we ran an experiment. It was a very careful, methodical, and interesting study involving a "real behavioral dependent variable" -- something that gets a lot of weight in the field. [It was stuff to do with how people's impulsive behaviors depend on other recent behaviors, and we had our participants doing stuff and then being allowed to eat from bowlsful of cheeseballs which, unknown to them, were counted by our RAs. It was all very cool.] However, none of our initial hypotheses was supported. The only spark was one little pattern, which we thought might be interesting to pursue.

So we thought about it, and delayed, and discussed a little more, emailed the guy in the US a couple of updates, and finally several months later ran a second study with a modified agenda. However, there again our hypotheses weren't really supported except for again, one little pattern that may have been argued to be consistent with the previous one. We didn't really know why, though.

At this stage, I'd about given up on the project. A year and a half had passed and I had other projects that were more promising, more interesting, and more fun. I'd been fortunate to get some good publications in quickly, and didn't really need this distraction, this "dog" project. Plus the brand of cheeseballs that we'd been using went out of business. That seemed like a sign.

But then we got invited onto a session that someone in Germany was organizing for the field's main conference. You get three papers together and make up a session proposal, people evaluate the thing and decide whether to accept the session or not. Getting invited onto this session meant we had to have a coherent story to tell. We almost turned the invitation down, but then thought it might help as a last gasp motivator to revive the project and get something done. So we toiled on - more from a sense of duty than adventure. The US guy had woken up meanwhile, and sent in a couple of enquiries about the progress. We emailed back and forth and finally settled on a story that might explain our data. But we needed more support, especially if we were to present at this conference.

So we had to get some more data, and so we ran a third study back in May. Yet again, the results were disappointing. While the basic pattern was again there in the data, the rest of what we saw refuted what we'd written in the proposal. To make things worse, the session proposal got accepted. So now we really had to figure out how to avoid making fools of ourselves (best strategy: go up there, waffle for fifteen minutes by presenting a literature review aka "other people's work", quickly flash a couple of non-controversial results, smile brightly, apologize for running out of time, invite anyone interested to "contact us after the talk", disappear). July-August saw some frantic emailing back and forth. We remembered that the US guy had proposed a different study back in May. Pressure on him to collect some data there. New story constructed on the basis of three disappointing experiments run to date. Conference in the end of September. Self ~ aka presenter ~ psyching up to be at bullshitting best.

Last week, I put together the presentation on the basis of the three experiments we had. It felt like sculpting bullshit -- that's exactly what I said to a colleague one day. But having done it, I felt better. I almost even believed what I was planning to say. Almost.

This morning, the new data from the fourth study came in. It's a drastically different study with a drastically different methodology, involving response times calibrated down to milliseconds. The guy in the US emailed in practically each analysis as he ran it, while the two of us in HK interpreted the numbers and suggested new tests. It is unbelievable - the patterns almost exactly support the new story, the one we arrived on after months of wrangling and hopelessness. Suffice it to say that we are pretty close to speechless.

There's a bit in Surely You're Joking Mr. Feynman where he talks about walking out into the night after stumbling onto some patterns in his data. He talks about the wonder of being the first person in the world to know something -- being for a brief instant the only person to know it.

In some small little way, that's what I have right now. That feeling of awe.

It's that joy of doing research. That's why I do this.