Supervision capacity
Welcoming interest while making realistic commitments
Interest in the lab is welcome. It is encouraging to meet people who want to understand research, develop skills, and contribute to questions that matter to them. At the same time, I cannot turn every expression of interest into a research placement. I want the opportunities I offer to come with the time and support they deserve.
Why research capacity matters
The research model has changed substantially over the past decades. Earlier behavioural research could sometimes be conducted with very small samples, little or no compensation for participants, limited checks on statistical power, and flexible analyses that made it easier to mistake noise for a meaningful trend. We now understand the limitations of that model more clearly, and we do not want to reproduce them.
| Then | Today |
|---|---|
| Studies might rely on samples of 10–30 people and still make broad claims. | We use power analysis to plan studies large enough to detect plausible effects and distinguish signal from noise. |
| Participant time was often treated as freely available. | Ethical, inclusive research requires resources to compensate participants fairly, which makes funding a central constraint. |
| Flexible analyses and selective reporting could produce apparently positive findings. | We aim to preregister hypotheses and analyses where appropriate, report results transparently, and resist p-hacking. |
| A publication in a good journal could be treated as the endpoint. | We prioritize careful, reproducible work that can withstand scrutiny and contribute to strong journals without overstating what the evidence shows. |
This is why a serious project may take longer and require more funding than it once did. Capacity is not only a question of how many people can join the lab; it is also the time, participant access, supervision, analysis, and financial support needed to do research responsibly.
In practical terms, I am often the bottleneck. The most consequential decisions still require my judgment: shaping a study, checking an analysis, interpreting uncertainty, and deciding what the evidence can support. Higher standards and new constraints make that responsibility more demanding, so the number of people I can supervise responsibly is limited. This is a constraint on my capacity, not a judgment of anyone’s potential.
Much of our research involves statistics, programming, and careful scientific judgment. AI can help produce early drafts, code, and other materials. In our workflow, producing a plausible result is often faster than establishing whether it is valid. We still need to check assumptions, examine the evidence, test analyses, and decide what conclusions are justified. A polished output does not remove that responsibility.
This also shapes what it takes to bring someone into a project. A task that appears short can require substantial preparation, training, feedback, and review. Supervision includes helping someone understand why a method is appropriate, recognize uncertainty, and learn from mistakes. Those are valuable parts of education, and they require time even when an initial task is completed quickly. Adding people does not automatically increase the lab’s research capacity.
I distinguish between research support that helps an existing project move forward and an educational investment that primarily supports someone’s learning. Both can be worthwhile. A learning opportunity does not have to pay for itself in publications or labour savings. It does, however, need an honest allocation of supervision time and shared expectations about what is possible.
Before offering a research role, we therefore need a real, appropriately scoped task, a suitable supervising contact, and enough capacity to provide feedback and evaluate the work responsibly. Where those conditions are missing, creating a task simply to welcome someone can lead to frustration and promises we cannot keep. Limited capacity is a constraint on what I can offer, not a judgment of someone’s potential.
The proposed SAGE Community offers another way to connect: discussion, peer learning, and a sense of belonging without an implied research placement. It is valuable in its own right. Projects may sometimes emerge from these connections, but they require their own agreement and supervision arrangements. Joining is not an audition, and it does not replace a clear answer to an existing research application.
My aim is to be warm about people’s interest and precise about commitments, so that community participation and research opportunities can each be worthwhile on the terms actually offered.
— Prof. Rémi Thériault