Tips for success: Giving an effective research talk

Abstract

For many scientists, their working life can be charted as a series of talks: lab meetings, journal clubs, research seminars, and job talks. Based on the sheer volume of attended and given talks, one might think that nearly any scientist should be an expert speaker. As we are all painfully aware, that's not so. Giving an effective talk is a learned skill, built largely on preparation and practice. Presented here are tips from five panelists, each known for their informative, well-organized, and engaging style: Nicole King, PhD, Michael Levine, PhD, Susan McConnell, PhD, Denise Montell, PhD, and Sean Morrison, PhD. While their advice focuses on improving research seminars (departmental seminars, society meetings, and public lectures), they also offer many helpful tips for enhancing general public speaking skills. Developmental Dynamics 239:3492–3496, 2010. © 2010 Wiley-Liss, Inc.

Before typing that first title slide, it is important to consider what you want to get out of the talk you are about to put together. There are many reasons to give a research talk. Often, the first that come to mind are for the purposes of advertising your work and impressing colleagues. Yet in addition to benefiting the audience, the talk can also benefit the speaker.

A talk can be an effective way to crystallize ideas, get feedback on work in progress, and build relationships with audience members who are interested in your work. Keeping the perspective that the audience is there to help you, not judge you, can be motivation for putting together a more open, honest, and genuine presentation. The approach not only flatters your audience by acknowledging their collective intelligence, but can also give your seminar a fresh and engaging air that can otherwise be hard to achieve.

ASSEMBLE A SIMPLIFIED STORY

The next step is finding a straightforward story to tell. “I believe that the most profound insights into how things work are usually simple ideas that are easily conveyed,” said Mike Levine, PhD, Professor of Cell and Molecular Biology at the University of California, Berkeley (Fig. 1). Ideally, a 12-min talk will tell one story, and a 50-min talk may include a more detailed version of that story, or a series of two or three interrelated stories. Sean Morrison, PhD, Professor of the Life Sciences Institute, University of Michigan and Howard Hughes Medical Institute investigator, describes his approach for focusing his talks. “I think about what question my talk should address and what the major take-home message should be. Then I frame the presentation around that question and that message.”

Figure 1. The panelists. (Top row) Susan McConnell, PhD, Nicole King, PhD; (Bottom row) Michael Levine, PhD, Sean Morrison, PhD, and Denise Montell, PhD.

When deciding what supporting data to include, the “shock and awe” approach of hitting the audience with every piece data is disorienting and counterproductive. “For the purposes of a seminar, I think it is fine to ‘cherry-pick’ and show the key results that capture the points you are trying to make. Of course, it is important to note that the raw data are often more complicated than what is being presented,” said Levine. Simplifying a story can also mean leaving out technical details. “The details of how something was done are often not important but take a long time to explain. For example, how one knocks down or overexpresses a gene or protein may differ for different organisms or cell types, but the concept is all the same,” noted Denise Montell, Professor of Biological Chemistry, Johns Hopkins School of Medicine. A story should contain only the minimum essential information needed to set it up and to support the conclusions.

Telling a streamlined story does not mean that a speaker should be afraid to include solid experiments that point to contradictory mechanisms. “In this case, you will want to discuss how different hypotheses are supported or refuted by each experiment. Then try to formulate a model, or two or three, which may explain your unexpected and nonintuitive results. Also describe what experiments you would do in the future to try to distinguish between your alternative models,” said Nicole King, PhD, Assistant Professor of Cell and Molecular Biology, University of California, Berkeley. “Talking about such data can be a great way to get feedback and ideas for future research.” However, she advised to leave out incomplete or inconclusive experiments for which you are not confident in the results. Ultimately, they could work to undermine the main conclusions.

Knowing who the audience will be, will dictate to what degree the story should be simplified. A talk designed for the general public should keep jargon to a minimum, and any necessary jargon should be defined. In addition, details of experimental techniques are probably not appropriate for a public audience, but may be interesting to the audience at a specialized Gordon conference. King cited an example of how her graduate advisor, Rich Losick, PhD, Professor in Molecular and Cellular Biology at Harvard, simplifies bacterial gene names. “The alphabet soup of spoIIAA, spoIIAB, and spoIIAC is reduced to A, B, and sigF, for reasons that are not worth explaining here. With those handles in place, he can worry more about communicating the interesting science, and less about whether the gene names are going to lose his audience.” It is rarely a mistake to simplify—Losick even uses the approach in seminars given to scientists.

A TALK'S STRUCTURE SUPPORTS THE STORY

Introduction

Most are familiar with the mantra, “always start with the big picture”. But what does it really mean? Susan McConnell, PhD, Professor of Biology at Stanford University said that it is never a mistake to start with the biggest question. “Even for someone who is in my field, I like to hear a broad introduction because it lets me know how they are going to approach the problem—which is often differently than I do.” During the introduction the speaker needs to explain the motivation for the research and why it is important. By relating the research to a commonly understood phenomenon such as a specific disease, or very basic biological mechanisms such as cell division, cell fate, patterning, etc., the speaker has the potential to catch the attention of every audience member.

Levine described another tip for helping the audience to think about the problem the way the speaker does. “I try to bring the audience into ‘my world’ by showing simple and attractive images in the first few slides. I find it boring when a speaker starts his or her talk with a bullet list of conclusions and theories.”

The next step is to give enough background so that the audience can understand the data and rationale for the approach. In a 50-min talk, the speaker can devote 5–10 min to the introduction by expanding on the history of the research question, or significant contributions by others. In a 12-min talk, perhaps one slide is devoted to the big picture and just one or two more on the necessary specific background. Levine noted, “Because in a 12-min talk you have time to make just one main point and conclusion; in the introduction, I include a statement of why I'm selecting this one particular topic for presentation rather than other research areas in the lab.”

Results

This section should include a hypothesis, how it was tested, and the data that supports the conclusions. As alluded to previously, the trick here is to keep the audience focused. “I present data in a logical step by step path toward the conclusion, and try not to distract the audience with anything that would take them off that path,” said Morrison. If possible, each point should be clearly supported by one representative experiment and control, and if appropriate, a summary graph or table demonstrating the statistical significance of the results.

Keeping with a streamlined story, data slides must also be uncluttered. McConnell, who has taught a number of workshops on scientific speaking, described a familiar scenario that illustrates the importance of this rule. “The speaker puts up a slide with 14 graphs on it. Right off the bat the audience is lost because they don't know where to look,” she said. “Even worse is when the speaker says, ‘I know this is a busy slide. All I want you to look at is this.’ What that really says is, ‘I'm too lazy to create a simple slide that delivers the message I want to give you.’” Instead of showing every experimental trial, the speaker can mention verbally that the experiment was repeated X number of times. In preparation for possible questions, additional data can be kept as reserve slides.

Another way to keep the audience from becoming overwhelmed is to explain exactly what is depicted in a figure. “For example, a student might show a tissue section stained with two different antibodies in two different colors but not explain what the tissue is, or that it is a single section, or what the two labels represent,” said Montell. “I have a rule that you should only show what you will fully explain and you have to fully explain everything that you show.”

Transitions

Transitions are a frequently overlooked component of a talk that are integral to keeping the audience on track. They are statements, or better yet slides, shown between chapters of the story that summarize what the speaker has just shown the audience, what it means, and how it sets up the next question. McConnell uses a diving analogy to illustrate their usefulness. “You can think of the ‘results section’ of your talk as a series of dives into the data. After diving into detail, you need to come up for air. That's the point at which you reorient the audience with an explicit transition that summarizes what you've just said, and prepares them for the next data dive.” Transitions not only link one part of the story to the next, but also allow drifting audience members to catch up on what they've missed. McConnell noted, “We need to remember that every member of the audience is going to tune out briefly at some point during the talk because they're human, and therefore, distractable.”

A particularly effective way to compose transitions is as a series of related slides. Find an image, or design a cartoon that represents a theme that is central to the talk. At each transition, slightly change the figure, or add onto the figure in a way that depicts what the data has just shown. For example, add segments onto a genetic pathway, or draw cellular components required for a specific mechanism. When the familiar figure flashes up on the screen, it will signal to the audience that it is time to pay attention. By merely glancing at the sequential diagram, they should be able to obtain a basic summary of what the data have shown up to that point.

Conclusions

The conclusions section should be more than just a “completed” sequential transition slide or a series of bullet points summarizing the conclusions. It should include discussion of the significance of the conclusions and how they have changed the way the speaker thinks about a particular problem. The speaker can also jump-start the question-and-answer session by including forward-thinking ideas. This could include a preview of the direction the work will go next, or speculation about larger implications. Ideally, a talk should come full circle, at some point referencing the big picture idea that led the introduction.

In the closing slide, thank funding agencies, collaborators, and the people who did the work and came up with important ideas. A speaker who fails to acknowledge that research is a team effort runs the risk of coming across as arrogant.

DETAILS THAT KEEP THE AUDIENCE FOCUSED

What's in a Slide

McConnell pointed out that some audience members best absorb information visually, some by reading, and others by listening. Recognizing these different learning styles, a well-crafted talk will deliver the same content in all three media. This means short and simple text that can be read quickly and limited to a few bullet points at most, accompanied by a simple visual that matches the words used to explain it. What's more, the three media should occur simultaneously so that the audience is never confused at what to look at, read, or listen to.

At the top of each slide should be a clear and descriptive title that gives the main message of the slide. “Rather than having a title that says ‘β-catenin pull-down’, you would want it to say, ‘β-catenin binds to a classical cadherin’,” noted King. “If an audience member zones for a moment they can simply read the title to know the point of the slide.” Ideally, the speaker makes one point per slide.

Technical Details

The following nitty gritty details may be tedious to incorporate, but they will help keep the audience from being distracted by busy or unreadable slides. Use sans serif fonts (Arial, Helvetica, Verdana), which are easier to read when enlarged on a screen and when seen from a distance. Font sizes should be fairly large so that anyone in the audience can read the text from a distance: 28 or 32 for titles, 20 or 24 for text. Font color should contrast with the slide background, which should be simple. For example, black or dark blue type on a plain white background, or white or yellow type on a plain black or dark blue background. Omit extraneous decoration, i.e., university logos, textured or photo backgrounds, and decorative slide transitions (wipe, blowout, etc.). Only items that are in service of the talk should be included in a slide.

When giving talks in a large auditorium, a dark background will produce the greatest contrast, making text and figures easier to see. When giving talks in a smaller room, or if the lights will be on, use a white background. The exception to the latter is if the presentation contains several fluorescent images. In this case, a dark background is preferred because it yields better contrast.

Animations and Movies

Animations and movies can be a captivating way to get a message across. “We are primates,” reminded McConnell. “We like to integrate things over space and time.” Like figures, movies should be visually simple and free from extraneous information. If a movie does not contain scientific content that supports a point the speaker is trying to make, then it should not be included.

Keeping Time

When asked how many slides to include in a 12- or 50-min seminar, some panelists cited the rule of dividing the duration of the talk by one to one-and-a-half minutes, the latter time referring to the amount a speaker should spend on each slide. McConnell disagreed, “I think the question is inappropriate.” Widely regarded as an expert speaker, she has been known to show 120 slides in a presentation. She explained that a speaker can easily spend well under 1 minute describing a slide that uses a simple cartoon to set up a hypothesis, or a transition slide. Instead of worrying about numbers, a speaker should do whatever it takes to get her points across in a clear, concise manner. Practice talks will reveal whether a speaker needs to add or subtract slides so that a complete story can be told in the allotted time.

When giving a talk, it is common courtesy to adhere to the allotted speaking time, recognizing that audience members have schedules to keep. Going overtime could actually work against the speaker since audience members who leave before the talk is over may miss the main take-home message. “I think there should be no such thing as a 60-min talk,” said Levine. “You can make everyone happy by giving a 45-min talk, field questions for 10–15 min and really finish the entire presentation in an hour. Remember, it took Lincoln approximately 2 min to deliver his Gettysburg address. No one remembers the preceding speech by Everett, which took 2 hr.”

ANTICIPATING QUESTIONS

A wave of relief may wash over the speaker after he goes through that last slide—and quickly vanish when he sees 20 hands jut up into the air. Although the question-and-answer session is impromptu in nature, a speaker may feel more at ease knowing that it has a certain amount of predictability. McConnell explained that questions tend to fall in one of three categories. The first is curiosity-based questions. These questions are about an interesting phenomenon that may or may not be addressed by the work just presented. In reply, cite work, be it yours or someone else's, that has been done or could be done to address the question. The second type of question is a misunderstanding. Reply by gently reminding the questioner of the work that was done to address this point. The third category is a statement that casts doubt on presented conclusions. In this case, stay calm and defend the work, putting up any relevant slides held in reserve. If it becomes clear that the conclusions are not rock-solid, propose ways in which they could be tested further.

Once a speaker has the chance to think about a question, it may not be as challenging to answer as it initially seemed. Therefore, rephrase a question after it has been asked. In addition to buying time to formulate an answer, it demonstrates that the speaker understands what is being asked. If despite these measures, he is still caught completely off guard, it might be best to say that he'll get back to the questioner at another time. However, when given the chance to think about it, a speaker who does careful science and prepares for obvious questions should be able to answer most questions knowledgably.

NAILING THE PRESENTATION

Be Engaging

A sure-fire way to elicit a positive response from the audience is to be passionate. “Let that love of science come through. Make it really clear to people why this is cool,” advised McConnell. Levine gets an extra charge by keeping his talks fresh. “It really helps to include new findings that I am still thinking about and excited about.” Another way to sustain enthusiasm is to intersperse the talk with occasional reminders of how the work is related to the big picture. A speaker's excitement can incite audience members to care about subjects that they ordinarily would not be interested in.

A speaker who is naturally funny has a distinct advantage. “A few jokes can help engage an audience,” said Levine. “Most of us lead stressful lives and are very eager to laugh when given the opportunity.” However, there is always the danger that a joke can bomb. If joke-telling does not come naturally, then an attempt in front of 25 or 200 people can be downright awkward. Even a natural humorist like Levine has had jokes that fail. “Once I attempted to open my seminar with a ‘eunuch joke’ in Beijing,” he recalled. “It was a total failure that kept me distracted for the entire lecture.”

Even the most passionate speaker may lose the audience if they are distracted by unnatural mannerisms. Make eye contact, controlled gesticulations, modulate one's voice to emphasize key words or phrases, and avoid non-words such as “um”, “so”, and “like”. When using the pointer, hold the laser steady on the graphic or text that the audience should focus on, and when shaky, hold it with two hands. Avoid big sweeping motions, and turn off the laser when the point has been made. If the speaker can pretend she is talking to a friend, a natural performance may follow.

Be Prepared

All five panelists insisted that practice is vital to giving a good seminar. The most effective way to practice is under performance conditions. This means standing up, paying attention to body language, using the laser pointer, and practicing in an auditorium, or at least in a room with a projector. The speaker should also practice in front of colleagues. They will undoubtedly have useful suggestions for how to improve the talk, might spot inconsistencies or faulty data or reasoning, and can help with anticipating questions. Finally, a speaker should practice the talk until she can easily walk through each slide, and anticipate what is coming next. Levine quipped, “Practice it to the point where you could give the talk without a break in the action even during an earthquake or hurricane.”

Another effective way to improve a talk is to have it videotaped. “Few of us are experts at public speaking, but we are all experts at being audience members,” said McConnell. “By watching a video of yourself, you immediately become aware of how you are presenting yourself to the audience. It is amazing how fast you can get rid of those nervous tics once you see them for yourself.”

Being prepared also extends to the technical aspects of seminars. If at all possible, give the talk using the same computer that was used during rehearsal. One can never guess how figures and text will be rearranged on a different operating system. Before the talk, set up the computer to make sure its settings are compatible with the projector. In case of computer failure, be sure to bring a back-up of the talk on a DVD or flash drive. If the auditorium's computer must be used, preview the slides to make sure they are not corrupted. Finally, resist using technology that is unfamiliar, for example a simple remote control may seem easy to use, but could flip through five slides at the lightest touch. Technical failures can cut into the allotted speaking time, and get the speaker off to a bad start.

Settling the Nerves

Despite feeling confident in the story and presentation, a speaker may become paralyzed with fear once 100 pairs of eyes focus their gaze upon her. To get through these moments, King employs lessons learned from her days performing as a flute soloist. First, she dresses nicer than usual to get herself in the right frame of mind. To avoid “choking” once she reaches the podium, she memorizes a script for the first few slides. “Once you get through those slides, you'll be in the groove and your nerves will settle.” During the talk, she focuses on her breathing to calm her nerves. She has even found she can turn her nervous energy to her advantage. “I actually like a little pre-talk jitters because it gives my talks a level of energy and intensity that they might not have otherwise.”

Montell has a different approach. She calms herself by sheer force of will. “I used to get so nervous that I would forget to breathe and then have to take an enormous, audible breath,” she said. “I just decided one day that I would not get nervous because it was going to ruin my presentation, which represented so much work and was so important to me.”

Another source of stress is when something unexpected arises during the talk. Montell cited the importance of maintaining composure. When she forgets what she wants to say during a slide, she simply stays calm and moves onto the next one. As for Levine, he can rely on his sense of humor to diffuse a tense situation. “At a talk I gave 20 years ago, a well-known senior scientist asked about a critical control that was indeed missing. I simply said, ‘You have my word as a Christian gentleman that the control was done and checked-out exactly as expected.’ Of course, I am neither a Christian nor a gentleman. People laughed, and I saved my neck.” Morrison also relayed how he handled his most embarrassing moment. “I once discovered a huge rip in the seat of my pants AFTER my talk.” His attitude is a good one to maintain throughout a presentation—whether in the face of an unexpected circumstance or not. “Ultimately, our best defense is to be genuine.”

Acknowledgements

The author thanks the panelists for sharing their insights and personal experiences.