Why is ‘doubt’ an important resource in science? How can we support the kind of science that takes intellectual risks – and takes time? Nana-Marie Lemm, Clinical Research Fellow at the Department of Infectious Disease, gives her reflections on the recent Day of Doubt conference, organised by the Good Science Project.
What are the motivations for becoming a scientist? It might be the search for truth or to understand how things work, or to work towards a greater good. But how does that idea compare to the reality of the day-to day work of science, and the culture of doing science?
The Day of Doubt was the first conference of its kind at Imperial College London. Organised by The Good Science Project, the conference aimed to reflect on research culture at the university level and beyond. It provided a space for discussion, reflection, and the opportunity to ask fundamental questions.
So, what is ‘doubt’ and why might it be important in science? The team behind the Good Science Project have helpfully created a thesaurus of doubt and a memo document on the day. In the latter, Dr Anthea Lacchia writes:
“If we take doubt to be the opposite of complacency, it is associated with pause, with thoughtful hesitation. Yet, this act of questioning, of taking necessary time before making a decision, seems at odds with the push for success, publications, and the need for high impact results. In truth, science is full of conflicts, some internal, some relating to wider culture.”
On the day, there was the opportunity to take time to engage in the act of questioning. Reflective sessions ranged from questioning concepts such as, ‘public engagement,’ ‘interdisciplinarity,’ ‘excellence,’ ‘expertise,’ to delving into philosophical discussions surrounding the nature of scientific truth.
A key question to speakers in the first session of the conference that resonated with me was, “How do you talk about doubt when you don’t come from a position of seniority and confidence?”
Navigating doubt and self-doubt
In my reflections, here, I have decided to focus on doubt and self-doubt in the scientific process and journey of researchers early in their career. I am in my final year of my PhD, studying a new way of administering a COVID-19 vaccine by inhalation in healthy volunteers. What I have found especially challenging in immunology research is getting to grips with the technique of flow cytometry analysis. This is a type of analysis which requires a considerable understanding of different cell populations to distinguish them from one another. It has also been criticised by some immunologists due to allowing a degree of subjectivity. Part of my learning journey involved coming to terms with the reality that this is not a black-and white methodology and that it can be approached from different angles, including both manual and unbiased analysis strategies. Is it possible that one dataset might give different answers depending on how you look at it? How do you minimise or account for variability due to both biological and technological reasons and artefacts when you are doing a longitudinal study? How should I phrase a hypothesis when approaching this kind of data? Those were all questions that plagued me and made me doubt the technique I had adopted and doubt myself in being able to analyse my data. But this is only one way of studying immune responses – not the only one!
I therefore found it especially interesting to listen to Sir Paul Nurse at the “Science and Doubt” session of the conference tell his story about being a PhD student. During his PhD, his experimental science project was not going well, prompting him to seek advice from the philosopher, Sir Karl Popper. Popper advised him that a good scientist actively seeks refutation of their conjectures or hypotheses. While this might not necessarily be the approach in every discipline, I found it to be a helpful reminder. However, one researcher’s question, in response to this, that resonated with me was: how does this traditional approach to the scientific process hold up to the emerging nature and methods of dealing with gigantic datasets? Does our personal experience influence the process of “data mining” the attempt to find patterns and meaning in large datasets, and thus undermine Popper’s traditional view of what makes “good science”? Or do we need to reframe concepts of “good science” in an era where artificial intelligence may be usefully applied to big data? As the complexity and volume of data grows, so does our awareness of our own limitations, as well as the need to understand how knowledge is both derived and understood.
Knowing is about understanding
Overall, I believe there is a balance required when incorporating personal experience in the interpretation of data. It is the scientific skill of balancing doubt with confidence. Confidence may come through experiential learning, and that personal component to knowledge is essential. To paraphrase Professor Ian Walmsley, knowing is not just about objective facts; personal understanding is also key to discovery. Knowing is as much about facts as it is about having a personal understanding. In this context, ‘personal’ doesn’t necessarily imply subjectivity. For me, this provides plenty of food for thought about emerging technologies and the data they generate, and how that data will be interpreted.
So, how do we build a culture to allow “good” science to flourish and build confidence in researchers early in their career, while embracing doubt? It sounds obvious, but I think we need to take more time to allow ourselves to think about fundamental questions and take time to discuss these with our colleagues. We can support one another by asking questions that encourage colleagues to reflect on the fundamental nature of our research questions. For example: “What do you think about your hypotheses? What insights and reflections can you share about the methodologies you’ve chosen? What experiment are you most excited about and why?” Doubt should be a collective process that we humbly embrace to create a positive, open and diverse research culture. It should be weaved into the fabric of the everyday rather than being confined to special occasions.