This post is intended to prompt me, and hopefully you, to think about two things. The first is practical: what constitutes a good style for an academic talk? The second is more abstract: what do the advances in computing speed and power offer to psychology conceived in it’s original form as a science of subjectivity?
I think an excellent starting point for launching both of these discussion is a recent talk at the Institute for Advanced Studies by Christopher Manning of Stanford.
From a purely practical point, many of his slides were solely illustrations. Bullet points and printed repetitions of what he was saying were few and far between. There were plenty of words on his slides, but then his content was computational linguistics - hard to avoid words as illustrations in a talk on language. But when words did appear they were illustrations, not redundant echoes of what he was saying. Manning used a clean white background with little embellishment. Colors, fonts, and changes in arrangements were used sparingly and selectively to group and highlight. In short, I think the formatting of the graphical material in support of the talk was an excellent example of how things should be done. The point is not that we should grab the slide template for use in our own talks, but that we should consider what we are going to show with the same care that we decide what we are going to say, and that just as we use inflection and gesture to enrich the semantics of our speech, we should view the visual opportunities of modern graphics as providing us the same opportunities in the selection and formatting of our visual materials.
Another exemplary feature of this talk is the structure of how information was unfurled. He did not tell us everything interesting he knows or has done, but he edited himself scrupulously to present a single coherent thread from beginning to end. We began at the beginning, and we were led on an apparently seamless trail to the current state of the art. Lots of potential detours were omitted. Not because they were not interesting or important, but because they were not in service of getting us smoothly to our information destination. Thus, this talk is also a good example in plotting a presentation.
Now what about the second goal: telling us how psychological science can benefit from the technical progress in computing speed and power? There things fall short, but that may not be Manning’s fault. There are limits of time, attention, and topic. The goal here was to share some of the accomplishments and potential of computational linguistics, and that is what the talk did. But, given the stylistic and organizational talent, combined with Manning’s position, history, and perspective on the field, one can only hope that he gives the continuation of his talk where he evaluates the contribution of all these practical successes to psychology as a science. It does indeed seem like magic that optimizing a set of vectors (of an arbitrary and comparatively small (100) dimensionality) for the components that probability that they precede and follow other words yields such a high quality result when applied to machine translation. No semantics necessary it seems to say. No knowledge of the world required. Does this empirical result, existentially dependent on computing power, tell us something about the human mind or not? If not, is that just an artifact of the specific application or a more general fault that limits our ability to gain insight into the language of thought by post-hoc reasoning from incremental improvements in Siri, Echo, or Alexa? Hopefully, some day soon Manning will use his rhetorical and graphical talents to shine some light on this knotty problem.
Introductory note: I asked Sherif, now a Research Scientist with Amazon and who I worked with during his undergraduate and graduate years at UWloo to share some advice on what he wished he knew back then. I hope this may help be of help to others considering psychology for undergraduate or graduate study. – Britt Anderson
Professional development advice for my past self
I wanted to write about advice I would give my past self while completing my psychology degrees, but before that, I want to begin with a caveat: each person’s journey is different, and the more you can bring yourself to not feel the pressure to follow any person’s particular advice or prescribed steps, the easier it might be to find your path to what comes next. That was hard for me at first, but you get better at it the more you try.
With that in mind, my story is that I completed my undergraduate studies in psychology and my MA in cognitive psychology at the University of Waterloo. During my MA I worked as a lab manager, a position I enjoyed and continued after graduation. I worked for a few months as a freelance data scientist before taking a position as a Research Scientist at Amazon in Seattle, which is what I’m doing right now. It wasn’t nearly as smooth as it sounds – much of the time I did not know what I was doing, and I had to make guesses and take chances on what would be useful or worthwhile uses of my time. I would apply to positions and not hear back, and have no feedback as to what I could do differently to be more competitive. I would have many ideas for things to learn or projects to do and not know which would give me the best return on the invested time and effort. Your best guesses will probably be different than mine when it comes to decisions like this, and I’ll be the last person to try and convince you that my way is the best.
In hindsight, my trajectory makes perfect sense. During my undergrad I thought I really enjoyed research because I enjoyed reading and discussing papers in seminars and research methods courses. But once I started my MA I quickly realized that, while I found those experiments and articles enjoyable to discuss, I didn’t find them enjoyable to make. I deeply enjoyed the parts that involved writing code and improving a data analysis protocol, and procrastinated on everything else. All the blogs I read and podcasts I listened to were about computers, programming, tech, and statistics. None were about psychology. For me, that was a sign.
But this is not a post aimed exclusively at those who are considering a move from academia to a position outside of it. My goal is to discuss a few things I wish I had known, and think others should know, when thinking about professional development.
Knowing the other side
It’s hard to exaggerate the large range of positions you can go for with a graduate psychology degree. Your skills are much more than just knowing how to analyze data. You have a lot of experience in experimental design and methodology, and great intuition about how to research human behavior. There are few companies who don’t need people who are good at that. Especially when your experience designing, running, and analyzing experiments is nicely complemented by your experience communicating those results in papers, posters, and reports.
The range of the different ways people find those positions is equally wide. I started my work in Amazon after completing my MA; others started after completing their PhDs or mid post-docs. Some, like me, rely on gaining experience through portfolio and project work (more on this below), others join intensive bootcamp programs that prepare them with a mix of training and interview prep.
One way you could learn about what’s on the other side is by pursuing an internship with a company or team you find interesting. This is a low-risk option because you likely don’t have to suspend your academic work to try it, and it has the added benefit of counting as work experience if you decide to pursue that path later. Many companies, especially tech companies, offer summer internships to PhD students nearing the end of their graduate studies.
In addition to exploring internships:
- Go to career workshops in your university’s career center (Centre for Career Action at UW). Those workshops are useful and underutilized.
- Find meetups that match your interests, and go meet people who do what you might be interested in doing, or are trying to learn more about doing it like you are.
- Talk to people who have landed themselves positions like ones you’re pursuing, or made transitions like the one you are considering making. This is one of the best ways to learn about your options. Chances are they got help from someone else and are happy to pass that on and share advice.
Expanding technical and statistical knowledge
It’s a good idea to expand your skills and experience regardless of whether your interest lies in academia or industry. Unfortunately, the technical and statistical tools one learns throughout a psychology degree can be very limited.1 This makes some sense because the primary goal of the training is to teach you to do psychological research, and the tools are a means to that end. Looking back, I think the default tools taught in courses are too limited and I think it’s very beneficial to explore other options.
The technical tools of the trade when I did my program were E-Prime for designing experiments and SPSS for data analysis. The R statistical computing environment is starting to penetrate the undergraduate and graduate courses as an alternative to SPSS, but this depends on where you are. I highly recommend you replace SPSS with R in your work as soon as possible. R is a better tool in every way2 and is quickly becoming – or already is – the preferred tool for statistical analysis and learning in and out of academia. It also has the benefit of easing you into getting some coding experience that will be useful to you no matter what you choose to do in the future.
R might seem intimidating if you’re not familiar with writing code or working in a terminal, but there are so many tutorials and introductory courses that will work for people at all skill levels. DataCamp’s Introduction to R and R Programming by Johns Hopkins University on Coursera are two great starter courses.
Technical tools aside, you may find as I did that you would not be taught a lot more than Analysis of Variance (ANOVA) and simple regression methods in your required statistics courses. Those are powerful tools that I use in my work right now, but there is a much larger world of statistics that psychology degrees might not expose you to. A good way to expose yourself to those methods is to take as many elective statistics courses as you can. When you can spare the time, read some statistics/data analysis blogs, like Variance Explained or newsletters like R-weekly that link to recent blog posts on data analysis. I also recommend podcasts like Linear Digressions and Data Skeptic that introduce a good mix of statistical and machine learning topics to beginner audiences.
What to do to prepare for non-academic positions
There is so much advice on this that it can be, no it is overwhelming. This is where I want to remind the reader of my initial caveat. Most people will not do everything that everyone who has moved from academia to industry advises they do. Usually doing a couple of those things, but doing them well, is enough.
Credentials and degrees matter less today than they ever did in the past. With the exception of professional careers that require certification or licensing, people care little for the title of the degree you hold. I work as a Research Scientist at Amazon with a BA an MA in psychology. My manager completed their undergraduate degree in music, and their manager’s manager – our product’s lead – completed theirs in philosophy. I rest my case.
So what matters? What matters is what you can do for them. People want to be convinced that you will be able to do the things they hope you will do in the position you’re interested in. The most common way an interviewer would judge that is by looking at your job experience. ‘Oh, this person has written R for 3 years in their previous job, it’s reasonable to expect they will be able to write it for me too.’
When you’re just starting and you don’t have experience, you must find a proxy for experience; an operationalization if you will. My proxy was to build a public portfolio of my technical and data analysis work and skills. I created a personal website where I wrote about technical topics, and hosted several projects that I put a lot of effort into on Github. I have enough evidence to know that this work was critical in getting me where I wanted to go, but that was my way. Your way will likely be different than mine, but your goal is the same: accumulate evidence of being good at what you want to be good at.
This post included a lot of dry thoughts and recommendations, so let me end on a different note: the best thing you can do for yourself, no matter your goals or interests, is to realize that you can learn anything and get really good at whatever you set your mind to. It’s never too late, and the lack of formal training is no deal breaker, and might in fact make things easier. The hardest part is to start, and once you do, the second hardest thing is to keep a schedule of learning and practice.
The second best thing is to recognize the skills you gain from your training in psychology. I already mentioned your knowledge in designing and running experiments, but that’s the lowest hanging fruit. You’re good at communicating your approach and findings, you’re good at collaborating with other researchers on projects and papers, and you have experience working independently and taking a project from a design through implementation and delivery.
Finally, while I think “follow your passion” is a bit too simplistic and unhelpfully vague, there is tremendous value in recognizing what you have fun doing and what you find yourself returning to whenever you get some free time. When thinking about your path forward, look for things that will pay you and let you have fun; fun is important!
If you have any questions about academia or industry, or questions about studying at the University of Waterloo in particular, don’t hesitate to get in touch, I’m happy to chat!
Your mileage, as well as the curricula of programs from different universities, may vary. ↩
R is free and cross-platform (works well on Linux, macOS/OSX, and Windows), and has a big, growing community and a massive repository of packages that provide almost any functionality your heart may desire. ↩
While superficially similiar, attending to features (color, orientation, etc.) is associated with distinct behavioural effects, as compared to attending to a location. In Jabar & Anderson (2017), we showed that a similiar distinction exists for probability. While orientation probability affected perceptual precision, spatial probabilty did not.
We wanted to extend that finding to a different task. 2-alternative forced choice (2AFC) tasks are much more common in the literature than estimation tasks, but they typically introduce possible alternate accounts, such as response biases. In a previous post, we outlined a novel method by which response preparation can be measured in a 2AFC task.
That procedure was implemented in a paper that was recently published at Attention, Perception, & Psychophysics. Collaborating our previous work, feature probability was indeed found to be distinct from spatial probability. Chiefly, the two probability types operate under different circumstances, and can affect RT and accuracy in different ways.
View older posts in the Archive.
subscribe via RSS