Search

In Search of Clarity: Building a better seminar talk



Learning to effectively communicate science is fundamental to academic success and essential to translating work to clinical applications, policy-making, and interested members of the public. Accordingly, we provide graduate trainees with opportunities to develop in the form of seminars and support for conference presentations. However, in many cases these opportunities exist without any clearly stated objectives, and absent the feedback necessary to improve over time. It is thus unsurprising that many adopt poor practices and hold general misconceptions about what makes a talk effective.



In light of this, I thought it might be useful to pull together a short list of recommendations that trainees (and anyone else really) might consider when preparing a seminar talk. This list is by no means exhaustive, and I am certainly not an expert in communications or knowledge translation. However, these suggestions are informed by issues that have impacted my experience as an audience member over the past few years, and I hope you'll find something useful to reflect on!


  1. Know your audience - Effective communication requires that you establish common ground with your audience. When presenting to your lab group, close collaborators, or at a highly specialized conference, you might safely assume some shared understanding of underlying theory, experimental approaches, or other area-specific terminology. However, in all other cases (e.g., department seminars, conferences with broad scope, public speaking engagements), you should give careful consideration to what prior knowledge the audience is bringing to your talk, and how this will need to be built upon in order for the talk to be well-understood. When preparing your talk, consider who among the potential attendees might know the least about your area of study and start from there.

  2. Mind your language - A seminar is your opportunity to share a new and exciting idea; do away with the jargon and technical language that often comprises academic writing* and focus instead on presenting your methods & results in ways that can be easily understood. Conveying complex ideas in a coherent and engaging way is a skill that will take considerable time to develop; however, and I cannot stress this enough - nobody will be impressed by the size of your vocabulary if your words obscure the core message(s) of your talk. Audience members have a limit to the amount of effort they will expend to follow your presentation; any energy spent deciphering your language is better spent understanding the ideas those words were meant to describe. *To be clear, I think there is a need to bring more clarity and less jargon/technical language to our academic writing as well, but that is a topic for another day…

  3. Check the time - We’ve all been to a seminar where the presenter, somehow shocked by the fact that they have once again failed to cover 120 slides in 45 minutes, advances rapidly through a good chunk of their deck in a race to the acknowledgments slide. This is often detrimental to the flow of the presentation and to the clarity with which core messages are delivered. While advice abounds on how many slides you should produce per minute of speaking time, I think a better starting point is to think about how much information you can hope to clearly convey per unit time. 15 minutes - enough time to describe an experiment. Short talks provide enough time to convince the audience that the area of research is important, provide a brief overview of the critical pieces of work that led to your experiment, and outline the methods, results, and interpretation of a single study. 30 minutes - enough time to describe an idea. Given a little more time, you can outline the conceptual framework that guides your research, how it fits with the existing literature in your area of study, and provide the details of 2-3 experiments that create a cohesive narrative in support of the idea. 60 minutes - enough time to describe a trajectory. On rare occasions (e.g., Doctoral viva, applying for a postdoctoral position), you may be asked to give an extended talk. In these cases, you have the time to really dig into the motivations that brought you to the work, why you decided on the approach that you took, and how the results will inform future study. The ultimate goal of this type of talk may vary considerably, but clarity remains critical regardless.

  4. Triage your content - Developing common ground with your audience and clearly articulating the complexities of your work that are fundamental to the bigger picture will require tough decisions about what should/should not be included in your talk. I understand the challenge here - you have agonized over every detail of your experiment. You may even feel passionately about those details. Yet it is important to understand that passion is unlikely to be shared by your audience. Think carefully about which details are necessary to support a clear and convincing narrative and which are extraneous to the point (this holds true for both the oral content of your talk as well as accompanying visuals). Remember - if specialists in the audience are interested in the minutiae of your experimental design, they’ll ask.


The ultimate goal of a research seminar is to teach the audience something new about your area of study. While you may be tempted to show off every piece of hard work you’ve done, or wow your audience with how deeply you understand the complexities of the field, this is very rarely the time/place. Instead, I encourage you to think carefully about the experience of your audience and deliver an accessible, informative, and inspiring piece of work.



50 views0 comments

Recent Posts

See All

I recently shared some thoughts on how service in support of EDI initiatives disproportionately impacts members of underrepresented groups, and how this imbalance necessitates a fundamental shift in t

There are systemic issues within our institutions that continue to present significant barriers to equity, diversity, and inclusion. This is not meant to be a ‘hot take’ - the data are conclusive; eve