Susan McConnell (Stanford): Designing effective scientific presentations

Susan McConnell (Stanford): Designing effective scientific presentations


You’re probably tuning into this talk because you’re interested in improving your speaking skills and you know, let’s face it, none of us really have had a lot of explicit training in presentation as scientists. So, ok, we’re not experts. We may not be experts at public speaking, but I would actually suggest to you that we are all experts at listening to talks. Think about the number of hours that each of us has sat in a chair, in a lecture room, in a seminar, in an auditorium, listened to journal clubs, to all kinds of talks, we’ve spent hours and hours and hours listening to talks. So we’re experts at knowing what a good talk is because we know what we want delivered as an audience member. So, if you think about it, why do we go to a talk? Well, we’re interested, obviously, in learning some about interesting and novel science. We are going to listen to speakers who we think are credible and knowledgeable and doing novel work. We are responsive to speakers who are enthusiastic and who just keep us awake and we want a talk that is well organized, clear, a talk that we can follow, a talk that’s not laden down with jargon and a talk that gives us enough background to understand what’s going on, we want to be able to see the data, understand it, interpret it and then make our own decision about the science. So, we’re all experts at listening to talks, why can’t we just translate what we know we want as an audience member into how we give a talk as individuals. Let’s think about something; how many times have we each seen a slide like this. We’ve seen it a zillion times, right, we’re sitting there, in a chair, listening to a seminar, a slide like this comes on. And we have no idea of what to look at. Are we supposed to read all of this text? Are we supposed to look at this figure, or that figure or that figure? We’re overwhelmed visually and we’ve all seen this slide a thousand times. Now, I just want to make a quick disclaimer. This was not my work and I’m not suggesting for a second that the authors of this really gorgeous paper would have shown a slide like that. But, if I were giving a journal club on that topic, I might have produced a slide like this for you in trying to tell you about these data. So, when we think about what we know instinctively is just a disaster, why do we keep doing it again and again and again? I think that we need to know some basic rules about power point and how to structure a talk in order to enable these visuals to work effectively as we’re teaching, as we’re speaking about our own science, as we’re presenting journal clubs. So, what I’m going to do is first of all go through some basics about power point and then we’re going to think about how you structure a talk in order to lead a member of the audience through the data in a way that they will understand each piece that you’re presenting and be able to really understand what the take-home message is. First of all, what font should you use? It turns out that some of the default fonts in power point and keynote are exactly wrong. The font you should use is a Sans Serif font. What is a Sans Serif font? San Serif fonts are fonts without little doober-hicky lines and stuff at the bottom of letters. And that’s to be contrasted with Serif fonts, which have all the little doober-hickies at the bottom of each letter. So, why do you use one and not the other? Well, visual psychophysicists have discovered that using a Sans Serif font when you’re projecting text onto a screen like this makes it easier for the audience to read the text quickly. These fonts down here, the Serif fonts, this is the font if you have novel you were going to read it in bed, it’s very easy to read. And that’s why books are printed with Serif fonts, but you should choose, and there are a range of choices that you can make, one of the Sans Serif fonts for your scientific presentations. How big should your font be? Again, be careful of some of the default font sizes. One of the defaults in Power Point is astronomically huge, bigger than this one right here, it’s like 42. It’s way too big! You don’t need that. So, the range of font sizes that you should use are shown here. Anywhere from 18 down to 36, with one exception. We often, in our scientific presentations, will want to insert a reference to a published paper, which isn’t really the focus of the slide, but we want to have it up there for scientific accuracy. So, under those circumstances, you might use a 14 point font, to put your reference at the bottom of the slide. Other than that, just work in this range and you’ll be good. Avoid using capital letters. This makes it clear why. It’s actually really hard to read all capitols. And secondly, in this age of email, I think we all know that capitalizing everything is the visual equivalent of shouting at someone. So it’s just not polite. In addition, when you’re thinking about titles or how you would actually capitalize words, use a sentence format. In other words, don’t capitalize every word of your title or your text down here. Having capitals of every word slows down the eye and makes it difficult to read. So, remember, the text and the slide is really there for the audience to be able to read quickly and easily. Color schemes are really important and thinking about the background of your slides. Now, we’ve all seen a lot of slides that have very fancy backgrounds and in fact, you want the simplest possible background. For example, a plain white background with dark letters, either black or blue. That works very well; it’s high in contrast and the only thing that appears on the slide is the information that your audience needs to understand this slide in the context of your talk. Avoid using the fancy patterned backgrounds. You don’t need your company or university logo on every slide. You do not need a DNA double helix running down every slide. All of that is extraneous. Maybe it might appear on your first slide, but then get rid of it. Streamline it down to as simple as possible and think about the contrast of letters. So dark against light works really well, light against darks works very well. A black background with white letters or bright yellow. Again, high contrast, easy for the audience to read. When should you use which? Well, if you were to give a talk for example, at a very large symposium, with hundreds of people, a very big room, psycophysically, and in terms of the projector, the contrast is greatest if you use a dark background and light letters, under that context. So it’s easier to project at high contrast. If you’re teaching in a small room, or you’re in a circumstance where you’re worried about people staying awake, then I would use a white background, with dark letters. That is actually more effective. However, you do want to think about your science. For example, in my science, I use a lot of fluorescence, photomicrographs, where I actually prefer a dark background because the contrast, I want people to be dark adapted so they can really see my data. So in one of my scientific seminars, regardless of whether I’m in a small room or a big room I use a black background. It just works better for my science. There are certain color combinations that you should avoid. Red/green is bad: this is incredibly difficult to look at. In addition, a large fraction of the population is red-green color blind and finally, red is really angry color. I actually saw an entire seminar once with the background color of the slides that was red and I was so agitated by ten minutes into the talk, so avoid that. Other color combinations can be equally bad and this is bad not because the colors are unpleasant but because the contrast here is not big enough for an audience member to actually read the letters against this background. This is important because in science we’re often creating, for example, these cartoons, of, for example, of a signal transduction pathway, where you want each protein to be a different color. So we’re always looking for a new color for a new protein as we’re going through a flow of talk, but we want to make sure that each protein that might in fact be this blue color would then have a label on top of it that’s high contrast. For example, for this, it might be black instead of green. Let’s think now about how to lay out a slide visually in Power Point so that it’s really easy for an audience member to follow the content of your talk. First of all, I strongly recommend that every slide a heading at the top and the best heading is actually a statement, a simple sentence that says in plain english what the bottom line of the slide is. Now you might say, Sue, excuse me, this is not a statement, and you’re right ok, busted, this is really going through an outline for this particular talk, but in my science talks every slide is headed by a statement. If you’re going to include text in a slide, it’s very important to just limit the amount of text. I really, strongly urge you to use no more than two lines of text in a text block. Why is that? The minute you show a big block of text like this, I have lost you as an audience member. You don’t know at this point whether you should be reading all of this stuff or listening to me. And if you as an audience member are now distracted and confused about what you should be doing, I as a speaker have lost control of my talk. So, don’t use things like this, there’s one exception, there might be a wonderful quotation that you want to include in your talk. For example, a quote from The Double Helix or a quote from the beautiful writings of Ramon y Cajal if you’re a neurobiologist. And under those circumstances it’s great to include the full text. What I recommend you do if you do include a quotation, is rather than read it word by word, instead sort of paraphrase it as you’re going along. You might say, well, the reason for limiting blocks of text to just two lines is if it goes on forever, people in the audience are going to have to make this huge effort to read it and that will preclude them from paying attention to what you’re saying. So, you can paraphrase a quote, that’s very effective, unless it is very poetic and you want to do a dramatic reading, and that’s fine too. Lists should be short; try to limit your lists to just three items. Avoid like mad long lists and if you’re going to have a list, I strongly recommend that you use the animation feature in PowerPoint to unveil your list one at a time. So, when you’re talking about item one in your talk, that will appear. And only when you’re ready to talk about item two does that second item appear and likewise, when you’re ready to include item three, that will come up as well. Be very generous in a slide with empty space. It’s more effective, it’s more powerful for the audience. And this is why. If you start to just load your slide with stuff, it’s visually overwhelming and often, you can get into a situation, we’ve all been in seminars before, where something is very, very, very close to the edge of the slide and there’s a slight misalignment between the projector and the screen that then results in stuff getting cut off. And we’ve all been at talks, where someone says “oh, I’m sorry, If you could have seen the slide, then you would have been able to read it.” Well, just leaving some space at the different edges, the boundaries of the slide and don’t forget about leaving some space at the bottom. If you’re in a room where everyone is sitting at the same level, people in the back are trying to see over one another’s head, so if you’re stuff is too close to bottom, some people in the back can’t see it. So, leave space on all three sides and a little space at the top, again for the misalignment problem. Ok, let’s think about the style of these slides. I urge you most wholeheartedly to include a simple image on every slide. What, most slides? No, every slide. In every slide of one of my scientific presentations, there is an image. Why is that? Here’s the deal. We audience members who are listening to a talk take in and process information in quite different ways from one another. Some of us are readers and that’s the reason for having a sentence at the top of each slide. Some of us are going to take in information by reading. Others of us are very visual. If we have an image on the slide, then that appeals to the vision, we’re primates, we’re very visually oriented so we communicate through pictures. And the third way we talk in information is by listening. The ideal, the perfect power point slide represents an absolutely synchronous match between a simple statement at the top, a simple visual and what you as a speaker are saying. So that the same content is being delivered through all three channels at the same time, without any distractions. Make your slides simple, limit the number of stuff you included in each slide and just make one or two points from each slide. Now, we’ve all heard this rule, maybe most of us have heard a rule, how many slides should you show in a particular scientific presentation? And the rule that I learned and the rule that perhaps you learned as well, the rule is you show one slide per minute. So, if you’re giving a twenty minute talk, you show twenty slides. I would argue now that that might have been true when slides were actually on film and difficult to manipulate and there was no such thing as animations. Now that we have tools like PowerPoint and Keynote, that enable us to build content progressively in slides and to have things that are simple, simple points, that we can easily change from talk to talk, that rule doesn’t really hold anymore. And I don’t think there is a firm rule, I think it really depends on the amount of information that you’re delivering in a talk. So, make your slides simple, don’t worry about this one slide per minute rule and rather, get feedback from, as you practice a talk, get feedback from your colleagues about whether you’re overwhelming people with too much data and number of slides. If your slides are simple, you can show more than one per minute. We have all seen talks where a slide like this comes up and what does the speaker do? The speaker does this, the speaker says, “I know this slide is really busy, but, the only thing I really want you to look at is this set of data right here”. We’ve been there, right? I’ve been there, you’ve been there. What has the speaker just told you? The speaker’s just said, I’m too lazy to have created a slide that is actually tailored for what I want to show you. And we as audience members are looking at all of this stuff, we, even if we’re told to look at this, I mean, how many of us behave? I, personally, I’m looking this and I’m wondering about that, so the speaker has not only shown a little bit of disrespect, and laziness to us, in my opinion, but in addition, the speaker has lost us as audience members because we’re now wandering around the slide and wondering, we’re having our own thoughts. So, if you’re going to show a slide like this, simply take the data that you want to show, eliminate everything else and put that up. And then explain the data completely. Explain to the audience what the axes are, what each color is, and remember that this is the very first time that anyone in the audience has seen this particular graph. You’ve seen it a thousand times, you get it intuitively, like that, but they won’t. So, just show the data that you want to explain and explain the data that you show. In other words, here’s the simple rule: if you’re not going to take the time to explain it, get rid of it completely. It doesn’t belong there. I want to show you an example of a figure that I used to show in one of my scientific presentations. And it’s a brain and it’s got some protein and it shows the distribution of those proteins in the brain. It’s a pretty simple figure, the content doesn’t matter to us. But I want to show you what this figure looked like before I put it into the talk in this format. Here’s what the figure looked like in the actual paper that I took it from. It had all of these extra labels that were extraneous, I was never going to explain that MGE stands for the medial ganglionic eminence because it’s not relevant to the talk so, what did I do? I took this figure, I put it into Photoshop and I got rid of all of that, leaving only a little landmark, the eye, to orient the audience to what we were saying. So, you can take complex figures, export them, and simply them in order to fit perfectly with your presentation. We have all seen a lot of, especially with the introduction of Keynote, some very fancy transitions between slides. What do me mean by transition? Well, in my talk so far, when I’ve advanced the slides, you simply see the new slide. Remember when Keynote came out and the first time you were at a talk and the speaker advanced to the next slide and the slide formed a box, and the box rotated in 3 dimensions and then a new slide appeared and you thought “Oh, that’s really cool, that was wild!” and then the speaker got their advance and they advanced to the second slide and the slide then did a big spiral and then the new slide appeared and by the third or fourth time that the speaker did that, you felt nauseated. We, yeah, so avoid these fancy transitions, they are not in service of the talk. There is an exception though. There are some transitions that actually illustrate in a very subtle way, actions. Here’s an example. This is a slide that I’ve shown in some of my presentations that have talked about the migration of neurons in the developing brain. And in this particular talk, the point I wanted to make was that we would take out neurons from a very small region and culture them. So, I used a very simple wipe transition. Watch this, it’s subtle but, there’s the transition. It actually wipes from one side to the other. That sort of mimics the act of taking something out of the brain and putting it into culture. So, that’s a sensible use, another sensible use would be if you’re going to zoom in on something, if you’re going to zoom in on this region, you might use a zoom transition. Apart from that, just advance from one slide to the next with a simple transition. Don’t drown the audience with data. What makes a talk memorable is when a speaker communicates a couple of take-home messages in a way that presents just enough data to be convincing, but not so much data that the audience is just flooded with information and can’t process it. Less is more and here’s a way to think about it. Those of us who are biologists are used to culturing cells in minimum essential medium. Just enough of the critical nutrients and elements and vitamins, that a cell needs to survive and nothing extra. Minimum essential medium. Now, if you take out any one of those elements in the medium, the cells will die. Adding more doesn’t add any value to your culture. So, if we think about minimum essential media, we can also think about minimum essential data. Just the data that you need to convince the audience of the credibility of your science and no more. That’s really difficult and, I think, that’s the most difficult decision that we have to make in terms of presenting a seminar. What is just enough? And where’s the line between just enough and not too much. For that, experience helps a lot, but also getting feedback from your colleagues, when you’re practicing your talk, to ask people, can I get rid of this? What can go? Don’t ask what can I add, ask what can go? To make this convincing and thorough, and yet not too much. So, I think we all realize it’s really easy to use PowerPoint badly, right? So, we’ve seen a zillion examples like this, of people presenting journal clubs or seminars you see a slide like that, we as audience members, are visually overwhelmed, we’re lost, we don’t know what to look at we can’t read this stuff, it’s too much to read, we’re distracted. The minute a slide like this goes up, the speaker has lost the audience completely. So, it takes a lot of work, don’t get me wrong, it takes time to use PowerPoint or Keynote effectively in a seminar because you have to think carefully about what you need to present, what you NEED to present, not what you want to present, not what you’d love to present, but what you need to present and to present it using clear, simple graphics and clear, simple text. So, in fact, let’s go through an exercise. Let’s take that previous slide, from this lovely paper from JCB, which I actually presented for a journal club many years ago. Let’s take that previous slide, which is figure 2 from that paper, and let’s see how we can break it down into its minimum essential components. The first decision that we have to make in terms of looking at all the data in this figure is what stays and what goes. What’s the minimum essential data? What are, excuse my grammatical error there, what are the minimum essential data? that we need in order to present in our journal club. And we might decide at that point that this particular part of the figure, panel b, can go. It’s not essential, but everything else is. So, how do we present that? Well, let’s deploy our PowerPoint rules in this context. First of all, I’m going show you how I’d present panel a. Look at it for a second, it’s got some results in a gel and it’s got some images of cells. How might I do that? Well, here’s the first figure, here’s the first slide in presenting this figure. Notice that there’s a sentence at the top, and then I’ve taken the data from the figure and I’ve added some stuff over here. I’ve added two labels that help the audience understand immediately that this is a PCR product and this is actually a Western blot. So I’ve added some stuff to be helpful to the audience, but I’ve simplified their focus, so there’s a good match between showing this and having the text. And what I would be saying if I were actually presenting a journal club. And then, using a simple animation, I would then unveil the rest of this panel. And I’ve added something for this audience down here, showing you that this is actually a photomicrograph of MDCK cells. So that’s panel a, presented in really two parts, using a simple animation and some additions. The next thing I might want to show actually comes from panel c, of that figure. And I’ve added a bunch of stuff here to be audience friendly. First of all, notice that there’s a sentence at the top, a statement of what’s going on. I’m showing just one simple panel, I’ve added a label, and I’ve added this statement to help the audience that we’re looking down at the surface of these cells. from the lumen. I’ve also added some color coding so that they know what proteins are visualized in different colors. That wasn’t in the original figure, I added that because who is this slide for? It’s for the audience. The audience needs this. And then using an animation, we’ll then show a side view of the same cells, that’ s the control cells. Now watch what happens. Two things happen; the sentence at the top changes, cause there’s a new message, and we’ve added the contrasting view of the cells that have this particular protein knocked down. Let’s now think about the last part, a new sentence that really states that this protein is essential for these cells to form little microvilli and in the absence of protein, they don’t. So, now we’ve presented that complicated figure in a few parts using some simple animations in a very audience-friendly format. So, what’s the bottom line with PowerPoint? Make simple slides, simple slides, that make one or two points. Build your content progressively using animations, rather than present everything all at once. Just show one, PowerPoint and Keynote, they’re all about control. They’re controlling what people see, what you’re saying, and what they’re reading and have all those match at the same time. And remember, if you’re not going to talk about it, just leave it out. If you’re not going to explain it, it doesn’t belong there at all. It goes. Ok, that’s what I wanted to tell you about PowerPoint, I hope that’s helpful. And now we want to transition into thinking more broadly about how to structure a talk. So that its organization is clear to the audience. And I think about it as almost taking someone by the hand and walking them through the science, right, and saying, here’s what I want you to see now and here’s the path that we’re following, here’s where we’ve just been and here’s where we’re going. So, a good talk is like a good paper, it has a structure, right, a good talk starts out with a big question and then we build content over time, we go through the meat of the talk and then we end with a conclusion that basically reaches back out to the big issues that we started with. Same way as we write a paper, why are so bad at actually doing this at talks? Well, if we discipline ourselves to do it, I think we can do it better. I’d like to show you an example from one of my own seminars of how I actually structure an introduction to define the really big question and then give enough background information to enable the audience to be able to follow the meat of the talk in the middle. And as I’m doing that, I also want to show you a trick that I’ve learned that I think is extraordinarily effective at giving good talks. And that’s an idea that I call something like a home slide or a home image. It’s a picture, maybe a cartoon, an image of some sort, that is going to signal to the audience that you’re at a point in the talk where you’re going to make a transition. It’s a little signal to the audience to perk up because you’re going to tell them where you’ve just been, what it means and where you’re going next. So, as I show you the introduction to one of my talks, I’m going to build up into a home slide that you’ll see a little bit later, is going to come up again and again and again at transitions. You might feel a little bit jarred because now the background of the slide is going to change from white to black and this jarred feeling that you might experience is one of the reasons that I think it’s actually quite important in a talk to keep the same color background throughout the entire thing. So, here we go. So, here’s the first slide from my standard seminar and notice first of all that my name and institution appear right here. Why is that? Well, because people in the audience may be taking notes and they might want to write down your name and they’ll want to know how to spell it correctly, so it’s a courtesy to them. So, as I start in my talk, I start to discuss a kind of analogy between the brain, which is what we study and a computer chip, because each of them has to make really specific connections with one another, the brain during development and the computer chip in a factory. And I make kind of a joke and see if people laugh and gauge the humor level of my audience at that point. And then I show the audience the cells that I actually work on, the actual circuitry of the brain that my lab studies, in the cerebral cortex and then I frame the major question that my lab is interested in, which really has to do with the question of how it is that these individual neurons in the brain know what kind of connections to form, what are the molecular mechanisms that guide those choices about cell fate and connectivity. At this point then, I introduce the two cell types that the talk will focus on, two sets of neurons that you’ll see have different colors and then I introduce two questions that the talk will address during the course of the next however long it is, maybe a forty-five minute seminar. One question having to do with the fate of the yellow cells and another question having to do with the fate of the blue cells. And that, basically then, launches me into the actual data part of the talk. This particular image right here is the “home image” or the home slide that I was referring to just a minute ago. And as you’ll see, this image is going to reappear again and again and again in my talk during transitions. And we’ll see just in a couple of minutes how that can get played out and how it can be extraordinarily effective in enabling the audience to follow a talk. Ok, so, now we’ve built up an introduction and we’re ready for the middle, right? So, we’re ready for the middle of the talk, which is the meat. But, there’s a problem. And that is that audiences have fairly limited attention spans. This is an interesting graph, so let’s look at a little bit of data. This is a plot of the percentage of a class that’s paying full attention to a lecturer over the time during which the lecture is being delivered. Ok, there’s some good news, right? The good news is right here, in your introduction, people are actually kinda tuned in and listening. And here’s the bad news, which is that ok, let’s face it, depending on how compelling the lecturer is, it might be here, or it might be way down there, but the bottom line is that at no, under no circumstance are you ever going to get 100% audience attention for your entire talk. Why is that? Well, think about being a member of the audience. Why does this happen? It’s because we’re human! We have thoughts, we have thoughts that intrude on our attention span, we have thoughts about, oh my gosh, did I remember to load that gel or turn off the power supply, we have thoughts about whether we actually are going to be on time getting the kids from daycare, we have thoughts about someone we just met, you know, we have all these random thoughts that intrude, so we’re human and we’re going to space out. We then, as speakers have two choices, one is we could say, eh, ok, someone spaced out, take no prisoners, you spaced out, forget it. You’re just never going to understand anything else I say. Or, we can just acknowledge the fact that the audience is human, that every single member of an audience is going to tune out at some point. And we can build in a mechanism within our talk to enable them to catch up. So, obviously, I kind of recommend the latter. So, if the middle is the meat of the talk and it’s the time when the audience is going to zone out at some point, why? Because they’re just people! Well, what are we going to do about it? What I suggest you do is to visualize the middle of the talk in the following way. You’re going to have a series of episodes, or data dives, that you’re going to present little stories, and as you’re presenting a particular story, you’re going to start at a level that’s fairly untechnical and you might actually get pretty deep into the data. To a level that’s really going to appeal really just to specialists in your fields. Now, if we were to give a talk and start presenting our data, and we’ve all been at seminars where people do this, where we just go down into depth and then we just do data, data, data, data, data, data, data, data and then we conclude and it’s over, we’ve lost people along the way. So, this structure, this visual structure here, is a way of planning your talk so that you can think about diving down into data and then coming up for air and it’s at these transition points between data dives that you use your home image to basically let the audience members catch up and know where you’ve just been, what you’re concluding from that part of the data, and where you’re going to go next. So, let’s look at an example of one of those data dives from one of my seminars. You’ve already seen the introduction to the seminar, you’ve already seen how I introduced my home slide. At this point in my talk, I’ve gone through a review of the literature, the first data dive and I’m now beginning part two of the talk. And here’s the beginning of that section. So, once again, you see this home image and you see the two questions. We’re half way through the talk, but I’m summarizing part one and getting ready to introduce the second set of questions. At this point, then, I now start to explain that as we focus on the blue cells, that we’re going to be looking at the role of a separate protein that is actually expressed in a subset of neurons within the different layers. And then I explain that we’ve generated knock-out mice in which we’ve inserted a reporter into the cells that normally express this protein so that we can see their axonal connections. And now, we don’t really need to focus on the data, but look, there’s a sentence, at the top that basically says what I just told you, then we see a simple figure from the controls in which I would point out that what I want the audience to look at are the blue axons that label the connections between the two hemispheres. That’s in the control, then we look at the mutant, the thing I’m going to emphasize is the absence of those connections here. Now, we’re going to look at it from the side, same stuff, sentence, two simple images, words will match the content, were I really delivering the seminar to you. And then, I tell you what I told just told you. I just told you that the blue cells express this protein, that they form this type of connection and then I say, you know, in the first part of the talk, I told you about another pathway. So, you might be wondering, what is the relationship between these two pathways? So, then I will go on to actually test that connection, by presenting more data that would then build up into a small conclusion slide showing you that this protein represses the expression of this gene. So, simple graphics that really emphasize the bottom line, the story that I’m telling you, with the minimum essential data. Ok, so we’ve gone through the meat, thinking about data dives, coming back, making transitions. Now let’s think about how to conclude a talk. Here’s the good news, is that as you say and now, in conclusion, look what happens to the attention level of your audience. It’s like, oh, it’s over, hey, I better see what I just saw. So, that’s good news, now you’ve got one more really good chance to deliver your take-home to the audience effectively. There’s a danger to this though, as well, and that’s the danger of hearing for example, forty minutes into a talk someone say, “Oh, to conclude”, and our, we’re like, “oh, good, you know, actually, they’re ending twenty minutes early, that’s great, I can actually get a little bit more work done today.” So, they say in conclusion, blah blah blah, blah blah blah, and then, they say “and now in part two of my talk” and what’s our response as an audience? We’re like, “oooh, dude, no”. Why is that? It’s because it was a false ending. And it got us all excited. So, if you’re going to conclude part one of your talk, do the audience of favor of saying, to some sum up this first part of the seminar, and that way you won’t create a false expectation. But, the good news is, you will perk up interest as you signal your conclusion, and that means that you have one last chance to really reiterate your specific conclusions and most of us remember to do that. But it’s easy to forget to sort of back out to the big picture. And return to the beginning, to have your talk come full circle, so that you’re really ending by revisiting the big questions that you introduced at the very first part of the talk. So, again, here’s how I would do it in my talk, just as a visual example of how one might think about this. Concluding up, now, sentence at the top, the same diagram that you’ve seen before, about this repression and now adding other things you haven’t seen in the data, but I would add that, just to remind them, that at least one of these mechanisms works through the modulation of chromatin, they would have seen that slide already. I would then connect it back to the first part of the talk and talk about a little switch that had gotten set up. I would then refer back even earlier to a part of the talk that you didn’t see, now going back to the really big picture, questions about how different types of neurons emerge over time. Many of these questions are still unanswered and therefore represent a frontier in my field. I would then wrap up by acknowledging, of course, the colleagues that have contributed to the work over time and then, here’s a trick, as you get to the very end of your talk, rather than leaving up your acknowledgements slide, have your conclusion slide appear at the very end. For the question and answer period. Why is that? Well, it helps the audience. Remember everything is in service of the audience and I actually have the misfortune of working on three genes, all of which have names that end in two and none of which are memorable by anyone who doesn’t work on them. So, by putting this picture up, I’m enabling the audience to actually ask intelligent questions. First, you know, if they’re struggling to remember which gene they had a question about and secondly, when I’m answering a question, I can actually use the visuals to explain my answer. And that turns out to be very helpful. So, I really recommend that at the end of your talk, during Q and A, put your summary slide back up again. It’s helpful for you and for the people attending your talk. Ok, so what have I just told you about organizing a great talk? First of all, use PowerPoint wisely. Have very clean, minimal slides. Secondly, start out with a broad introduction, so that everyone in the audience understands the big questions that are compelling in your field and then introduce the specifics gradually, giving them just enough background information so that they can follow the data, but not so much that they’re overwhelmed. Third, think of the talk as consisting of these episodes, or little data dives, in which you’re going to go at a point into your data, remember the minimum essential data, and then you’re going to go into some depth, but then you come up for air and allow the audience to regroup, see where you’ve just been, frame the next question, and then get there logically, rather than just going through data, data, data, data, which was just going to leave people in the dust. A very effective way of making these transitions explicit is to design a home slide or a home image that is a signal to the audience that you’re at a point of transition and they will perk up and it will help them to understand the flow of your talk. And finally, your conclusion here is just the opposite of the introduction. You start more specific with what you’ve learned, but then end up broad again. So, is this all you need to know to give a great talk? No way! Right? I mean, these are just some basic rules or suggestions or advice about how to create slides that are user-friendly, audience-friendly, and how to structure your talk so that it’s well organized and very, very clear, and simple for an audience to follow. But there are a lot of things that go into giving a great talk. There’s the whole performance aspect of speaking. There’s the actual scientific content, but at least in terms of performance, you can practice that, you can and should rehearse your talks. In fact, the entire time, from the first day of graduate school, all the way, well into being a tenured professor, at Stanford, I rehearsed every talk, every lecture, every journal club, that I gave. Every one of them. And finally, I got so comfortable with speaking I don’t actually have to do it anymore. But that means rehearsing, out loud and often in front of an audience, dozens and dozens and dozens of talks, get feedback and practice the way you’re going to speak. Have the words rehearsed and get some input from your colleagues about whether you’ve identified effectively the minimum essential data for the talk. Secondly, have yourself videotaped. Go and watch yourself and see what you do. See if you’re gesturing naturally and so on. So, those kinds of aspects of delivery are much more effectively done one on one, rather than through a format like this. Another great resource that I highly recommend is a wonderful book, by Michael Alley, called The Craft of Scientific Presentations. If you’re going to have just a single book on scientific presentations in your bookshelf, I personally would recommend this one. It’s a terrific model and it’s actually the basis of a lot the information that I’ve presented to you today. Well, that’s it! Thank you so much, I hope this has been helpful and I wish you every success with your talks in the future.

You May Also Like

About the Author: Oren Garnes

78 Comments

  1. This was a good talk. I like the 'data dives' suggestion and coming back to a home slide. It's almost like breaking a lecture up into parts, making it easier for someone to pay attention to the next 'data dive' even if they spaced out on a previous one. It sucks when you lose track of something, and then everything else following that point in the talk doesn't make any sense.

  2. Thanks very much, very helpful. However, I'd just like to remeber that besides PowerPoint and KeyNote, there's also Beamer for Ubuntu people.

  3. one comment about "less is more". I think a common strategy used by a lot people is that put the redundant slides after the end of the talk. In this way, if someone in the audience bring up a question that could be perfectly answered by those slide, voilà.

  4. Great and very helpful presentation that I have ever seen. I learn many things and at the same time I was gotten positive feeling .Thanks Dr.McConnell

  5. This is what I'm looking for. I will have scientific presentation at International seminar next Tuesday. Thank you so much for the tutorial 🙂

  6. Thank-you for such an effective presentation on presentations!  You have covered all the points I would have myself, and now I will only need to have the students (non English speaking PhDs)  watch this video, do some followup tasks and then we can move forward more efficiently.  Thanks!

  7. thank you so much, this has been a wonderful help! I'm presenting my first ever seminar for my honours so more than a little nervous, but this has been amazing. Thank you!!

  8. This presentation was quite good, from a presentation standpoint. From a design standpoint some things could be added:

    Comic sans is a badly designed ineffective font that doesn't work well when anti-aliased, please don't use it, people in the audience will loath your presentation if you use it.

    Limit yourself to one, at max two fonts; this applies to figures to. Many fonts will distract the viewer and look bad.

    No boxes, don't put boxes around figures, images or text, it serves no purpose and just makes the presentation look bad

    You absolutely don't have to limit your text colour to blue and black, I would avoid primary colours and go for darker tones.
    For light backgrounds and lighter tones for dark backgrounds, use adobe kuler to check contrasts.

    Limit your colour use, two or three colours is enough, try to apply this to figures, this might be hard in some figures, avoid primary colours and use adobe kuler to find nice complementary colour selections.

    A nicely designed presentation will make it more memorable and will make people pay more attention, a effective presentation need not be ugly.

  9. I'm looking for videos to help me on a presentation for October and this video had been really helpful. Thanks!

  10. This is the best presentation ever i seen and learnt about power point presentation and trully helpful to improve slides presentation. She touches all the sides of power point presentation. Thank you so much for the video

  11. Thank you so much! that was really helpful..
    I'm having my first paper presentation this week and the tips in your video were really helpful.. This was one of the best presentations I've ever seen..
    Really good job.,!

  12. This is incredibly useful! I have sent your video to all of my trainees and have changed my own talks with your advice in mind. Thanks so much!

  13. Thank you and much appreciated. I hope move inspiring videos from you in the future. By having this might give impact for my final year project. wish me luck.:)

  14. Amazing presentation. I'd like to know which softwares were used to unify powerpoint with the video image of the presenter in one single screen. I noticed the presenter was apparently looking at a screen on her right side, so she could be sure exactly where to indicate on the screen in order to, at the end, look like she really had a powerpoint screen behind her.

  15. 0:00 Intro
    3:05 Font style/size
    5:29 Color contrast
    8:50 Layout (heading, text, lists)
    11:19 Use of empty space
    12:20 Simple image on each slide
    13:51 How many slides to show
    14:57 Avoid busy slides
    19:10 Data: Don't overdo it
    21:21 Minimum essential components
    25:19 Structure of a good talk
    26:30 Use of home slide
    29:27 Meat & Taters; Keeping the audience's attention
    31:45 The specificity dive
    35:07 Conclusions
    37:40 Conclusion/Q&A slide
    38:33 Conclusion of this actual presentation

  16. "You're probably tuning into this talk because you're interested in improving your speaking skill."

    Nah, I just gotta watch this for class.

  17. I wish this was an obligatory presentation to watch by everyone in STEM. How pretty the life would become? 🙂 Hehe. No more talks where you just sit agitated because you cannot follow the speaker most of the time.

  18. What's so hilarious is that at about 30 minutes into her presentation she switched to a dark screen and I found myself getting very sleepy and I almost stopped the video. Then she goes back to the white screen, shows a bell chart, and starts talking about how the attention span of the audience goes down over a period of time. That's EXACTLY how my attention span was during her presentation. But it was a good presentation though.

  19. This is so helpful, my advisor made us watch the whole video 🙁 , but we still liked it. Also, I think it's a good video to watch when tripping.

  20. What do you think about the strategy to motivate the audience in an, ehm, old school fashion way even more?
    In times where some students do not know the number of sides of a triangle, I might suggest this approach:
    https://youtu.be/AjmbgZ2wZvk
    Have a smile and a nice day!

Leave a Reply

Your email address will not be published. Required fields are marked *