Ta ta, Goldsmiths; Hallo, Amsterdam

Today is my last day working as a Postdoc in the Department of Psychology at Goldsmiths, University of London. Though the Innovate UK grant that I am working on does not officially end until April, I accepted an offer to work with Henkjan Honing and the Music Cognition Group for the next three years. I’m very excited about it. But I don’t start that job until next week, so I’ll wait a bit before writing about what’s next in that department (or Institute for Logic, Language and Computation, shall I say?).

As is now customary for myself, I find it important to look backwards and reflect on each job as I leave it. There has been a lot of that as of late, I think I have had a new “main gig” pretty much every year for the past six years. I know I often complain about the instability of this way of working, not to mention the project-debt I often find myself in, but there has been a lot of good that has come from all this.

Let me first start by saying that it has really been a honor and privilege to work between both the Department of Computing last year and the Department of Psychology this year. As a card-carrying music theorist and an academic that tries his best to reject the whole humanities versus science thing as much as possible, it has felt quite validating to be able to employed by two different non-music departments. I was told dozens of times during my graduate studies in the United States that because I was getting my Ph.D in music, I was essentially resigned to working in music departments if I wanted to stay in academia.

That probably would have been true were I to have stayed in the United States. Luckily, it feels that this side of the pond has a much more relaxed approach to disciplinary boundaries. It also helps the past few roles I have had have been strictly research, which means that the roles tend to focus on the knowledge and skill sets required to do one very specific job, rather than the host of talents needed to teach, research, commit to service, admin, and whatever people do with their personal life.

What was also special about this job at Goldsmiths was that in addition to catering to the music and science dimension of my professional identity, it also had a nice spread between the academic and industry dimensions. The position involved collaborating with Soundout, a company leading the way forward in sonic branding research. And to top it all off, we were looking at questions of music perception at scale. It really was a wonderful fit.

So what did I actually get my hands dirty doing in this job? AKA if I need to describe this job in an interview, how would I describe what I was doing?

The main task of the project was to develop a series of psychologically informed tasks that would result in insights both for both academic music psychologists and with products that Soundout offers to their clients.

We were mostly interested in three big questions:

1. How is music able to influence the semantic associations of what we see?
2. What makes a some audio logos more memorable than others?
3. Are there any individual or musical features that predict an individual’s propensity to buy a product?

Throughout the project we took a few ideas from the music and science literature and modified them so that they can be run at scale. By at scale, I mean collecting hundreds of thousands of responses over tens of thousands of panelists, paid for their responses in our research. This includes both people on Slice the Pie and using MTurk.

The semantic work looks to essentially extend and refine what has been done with film music, but on a much shorter time scale. The memory work looks to see what features of melodies lead to high recognition rates. This includes not only the structure of the melodies (are they simple, are they complicated?) but also timbral features.

We’ve got some data on pretty much everything at this point, but are still running and will be over the next several weeks while the rest of the team continues to build the analysis pipelines. Hopefully you’ll get to hear about this at SMPC this summer.

As I have noted before, the position here was part-time, which, looking back, wasn’t maybe the best idea. Academia often demands a few more hours than the contract states, so setting boundaries with doing what when was admittedly very difficult. I don’t think I’d say yes to that again and probably would not encourage others to accept similar positions. I had other work to fill in the other half of my paycheck and some months just took the time off and went back to a more, “minimalist” living situation, but I would have rather just focused on one thing.

I had all these thoughts at the start of the project, but I have a hard time saying no to work where I’ll get to work with people I really enjoy working with. This has been my second time working with my old MSc advisor, Daniel Mullensiefen, and always enjoy working together.

Luckily we have someone to step into my role and help get all this work over the finish line (right when the fun stuff happens), so more news on all the interesting audio branding and psychology findings in the near future.

I will also keep up the other plates I have been spinning the past few months such as teaching as part of the faculty at Cambridge’s Center for Music and Science, doing some work in the data science education sector, and the other floating manuscripts I am trying to get off to submission.

I’ll still be hanging out in South London for February, but come March, I will be spending a lot more time on the continent, popping back and forth on the Eurostar frequently, pandemic permitting. Luckily remote work is no longer a strange or peculiar request, but I am looking forward to actually getting to work face-to-face with a whole new group of people in 2022. I’m really going to miss Goldsmiths, but I am sure I will be saying the same things about the University of Amsterdam (as I did for LSU) when that chapter ends.