Dr Dominique Makowski: Breaking Barriers to Reproducibility

Our Openness in Research award winner talks riding the wave of open science, his view on the fear of being scooped and overcoming institutional resistance.

Many of us by now will be familiar with the term ‘open research’. Much more than the narrower concept of open access, it’s the broader name for principles and practices which aim to make the research process as transparent, accessible and trustworthy as possible. The goal is to ensure that each stage of the research lifecycle is open and reproducible, enabling others to verify, build upon and contribute to the work, ultimately driving greater integrity and progress. Surely a good thing, no?

It’s hard to believe that little more than 20 years ago, making research open across the board wherever possible, was not the widely accepted priority it has become today. And even when the open movement took off in the 1990s and 2000s, pockets of resistance remained.

Dr Dominique Makowski, now a lecturer in Psychology and open research advocate at Sussex, developed an interest in openness during his time as a PhD student examining the neurocognitive underpinnings of emotion regulation through fiction. (In other words, how does knowing that something we see is `fake’ impact our emotions towards it). Dominique became frustrated by the many barriers he came up against during his studies: papers behind paywalls, lack of sufficient data for meta-analysis, patchy methodology details and lack of practical tools prevailed.

Importantly, he also saw the personal benefits to making his own work open, by compelling him to make his outputs more clean, modular and reusable. This virtuous cycle would lead to more efficiency, productivity, and satisfaction, paving the way for a wider engagement in the community and the start of large and impactful collaborative projects.

The Reproducibility Crisis in Psychology

Dominique’s awareness of these issues grew alongside a wider awakening of reproducibility issues within certain scientific communities, throwing the reliability of some foundational studies into question. The field of Psychology was identified (in studies such as The Reproducibility Project of 2015) as an area for concern, alongside Medicine, Economics, Political Science and Education. Dominique explains, “The ‘reproducibility crisis’, was identified in Psychology about a decade ago – it made the headlines back then. It was discovered that we can’t replicate many scientific results. And it showed that we needed reform and revolution. So, since then, there has been this wave of open science, and all the new ways of doing science have been evolving with it. But the idea that we need to be more transparent about what we do requires

researchers to know how to make their material open and transparent. How do we code things in a way that can be understood and reused, and communicate results in a clear manner?”

Tools for Open Science: Easystats and NeuroKit

Having realised that one major missing piece of the puzzle was the software needed to facilitate open science work in Psychology, statistics in particular, Dominique set about the parallel work of co-creating software and making his own outputs fully open and accessible. He helped to develop two major tools:

easystats – a popular collection of R packages (add-on software tools used for working with data) designed to make statistical reporting simpler and more consistent. This includes tools like the report package, which automates the generation of statistical models and outputs standardised reports. This saves time, reduces errors and improves reproducibility by ensuring that statistics are reported consistently across papers. It also helps students to learn what information they need to include in their work. Students at Sussex now use easystats, and the software recently hit a couple of million downloads worldwide.

NeuroKit – a Python package designed to process bodily signals like heart rate, breathing or skin conductance (i.e., the sweaty palms when we are stressed). It’s for researchers who are interested in how the body influences cognition (spoiler alert, we are finding out more and more that it’s a lot). NeuroKit enables the easy usage of these methods and tools at the service of psychology, medicine and neuroscience.

Dominique’s bigger idea behind this software is indeed to democratize the use of these advanced research methods, lowering the barrier for those who want to explore the relationship between the body and the brain, or run and communicate complex statistical models. As Dom says, “Very often Psychologists are not limited by their ideas, but by the tools. The idea is to lower the barrier of entry for more people to make use of these and help them realize their wild ideas.”

Overcoming resistance

But changing people’s minds about the way they should conduct science has not always been easy. Dominique confesses, “It’s one thing to have ECR (Early Career Researchers) into open science. The problem that we had a couple of years ago was that there was a real resistance from the establishment and some PIs, because they didn’t want the change. I had an argument with a more senior professor who really wasn’t keen on sharing data or making things transparent at all.”

Dominique identifies two main reasons for this, the first being a deeply ingrained fear of being ‘scooped’. “People think, what if someone faster than me uses my data for

themselves, for better means? I don’t think it happens very often. For me, it has been only beneficial. You connect with people that can help you, it helps others, and you help yourself. Open science enables a mindset shift from competition to collaboration.”

The second reason being imposter syndrome. Dominique explains “If a researcher is not fully confident that what they are doing is the right thing, having the world looking at it can be a bit daunting and scary. Fear of people finding mistakes and damage to reputation is a real thing.”

Despite this, Dominique is hopeful that Psychology have made major steps towards the desired reform, saying “It’s obvious there is still work to be done, but we have come a long way in the last 10 years. We have managed one of the biggest hurdles which is awareness. At least at Sussex, from my experience, most of the ECRs have heard about open science, and know it is good. 10 years ago, that didn’t exist.”

The importance of institutional support

Although open research was born a grassroots movement, Dominique recognises that genuine institutional support is key to achieving the systemic change we want to see. He acknowledges “Sussex is now much ahead compared to other universities in their concrete and visible support for engagement with open practices. The Research Culture and Impact awards are concrete proof of this. It means that PIs are now on our side, ECRs have the awareness, and it creates a lot of good stuff.”

“And awards are one thing, but having actual support for open research is very valuable. The software team here in psychology, the Open Research Technologies Hub – these things give the university an edge. I have been really impressed with the people I have met and their collaborative nature. The open research culture at Sussex is not just talk, it is genuine.”

A vision for the future

Dominique concludes, “We’ve put a lot of work into the tools, creating this software that is capable of many things. The next step is to promote them properly and teach people how to use them. We have so many cool features, but people are not always aware of them. Ideally, I would like to develop online tutorials, or maybe even a book on easystats…”

When asked what conditions he thinks allow research to flourish to its fullest potential, Dominique’s answer is, “Most researchers don’t like to be micromanaged or have too many constraints. They like freedom. But at the same time, feeling that there is support when you need it is important for success. Finding that sweet spot between autonomy and support is the key.”

Written by Alice Sambrook