YT YouTube2Text

YouTube2Text → Transcripts → Susan McConnell (Stanford): Designing effective scientific presentations

Susan McConnell (Stanford): Designing effective scientific presentations — Transcript

by Science Communication Lab · 7,576 words · 1,060 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00you're probably tuning into this talk
0:01because you're interested in improving
0:02your speaking skills and you know let's
0:05face it none of us really have had a lot
0:07of explicit training in presentation as
0:10scientists so okay we're not experts we
0:13may not be experts at public speaking
0:15but I would actually suggest to you that
0:18we are all experts at listening to talks
0:21think about the number of hours that
0:22each of us has sat in a chair in a
0:25lecture room in a a seminar in an
0:30Auditorium listening to journal clubs to
0:33all kinds of talks we've spent hours and
0:35hours and hours listening to talks so
0:38we're experts at knowing what a good
0:40talk is because we know what we want
0:42delivered as an audience member so if
0:45you think about it why do we go for to a
0:46talk well we're interested obviously in
0:49learning some uh about interesting and
0:51novel
0:52science we are going to listen to
0:55speakers who we think are credible and
0:58knowledgeable and doing novel work we
1:02are responsive to speakers who are
1:05enthusiastic and who just keep us awake
1:08and we want a talk that is well
1:10organized clear a talk that we can
1:12follow a talk that's not Laden down with
1:15jargon and a talk that gives us enough
1:18background to understand what's going on
1:20we want to be able to see the data
1:22understand it interpret it and then make
1:24our own decision about the science so
1:26we're all experts at listening to talks
1:29why can't we just translate what we know
1:31we want as an audience member into how
1:34we give a talk as individuals let's
1:37think about something how many times
1:39have we each seen a slide like this
1:42we've seen it a zillion times right
1:44we're sitting there in a chair listening
1:46to a seminar a slide like this comes on
1:49and we have no idea of what to look at
1:52are we supposed to read all of this text
1:55are we supposed to look at this figure
1:57or that figure or that figure where
2:00overwhelmed visually and we've all seen
2:01this slide a thousand times now I just
2:04want to make a quick disclaimer this was
2:06not my work and I'm not suggesting for a
2:09second that the authors of this really
2:11gorgeous paper would have shown a slide
2:13like that but if I were giving a journal
2:16Club on that topic I might have produced
2:18a slide like this for you in in trying
2:20to tell you about these data so when we
2:23think about what we know instinctively
2:26is just a disaster why do we keep doing
2:28it again and again and again again I
2:31think that we need to know some basic
2:33rules about PowerPoint and how to
2:35structure a talk in order to enable
2:38these visuals to work effectively as
2:40we're teaching as we're speaking about
2:42our own science as we're presenting
2:44Journal clubs so what I'm going to do is
2:47first of all go through some Basics
2:48about PowerPoint and then we're going to
2:50think about how you structure a talk in
2:53order to lead a member of the audience
2:55through the data in a way that they will
2:58understand each piece that you're
3:00presenting and be able to really
3:02understand what the take-home message is
3:04first of all what font should you use it
3:06turns out that some of the default fonts
3:08in PowerPoint and keynote are exactly
3:11wrong the font you should use is a
3:13sansera font what is a sansera font
3:16sansera fonts are fonts without little
3:18duber hickey lines and stuff at the
3:21bottom of letters and that's to be
3:23contrasted with sarap fonts which have
3:25all the little duber hickeys at the
3:27bottom of each letter so why do you use
3:29one and not the other well visual
3:32psychophysicists have um discovered that
3:36using a sansera font when you're
3:38projecting text onto a screen like this
3:41makes it easier for the audience to read
3:44the text quickly these fonts down here
3:47the Sarah fonts this is the font if you
3:49had a novel and you're going to read in
3:50bed it's very easy to read and that's
3:52why books are printed with Sarah fonts
3:55but you should choose and there are a
3:57range of choices that you can make one
3:59of the sandera fonts for your scientific
4:02presentations how big should your font
4:04be again be careful of some of the
4:06default font sizes one of the defaults
4:09in PowerPoint is astronomically huge
4:12bigger than this one right here it's
4:13like 42 it's way too big you don't need
4:15that so the range of font sizes that you
4:17should use are just shown here anywhere
4:19from 18 down to 36 with one exception we
4:23often in our scientific presentations
4:26we'll want to insert a reference to a
4:28published paper which isn't really the
4:30focus of the slide but we want to have
4:32it up there for scientific accuracy so
4:34under those circumstances you might use
4:37like a 14 point font to put your
4:40reference at the bottom of the slide
4:42other than that just work in this range
4:44and you'll be good avoid using capital
4:47letters this makes it clear why it's
4:49actually really hard to read all
4:51capitals and secondly in this age of
4:54email I think we all know that um
4:56capitalizing everything is the visual
4:58equivalent of shouting at someone so
5:00it's just not polite in addition when
5:03you're thinking about titles or how you
5:05would actually capitalize words use a
5:08sentence format in other words don't
5:10capitalize every word of your title or
5:13your um text down here having capitals
5:16of every word slows down the I and makes
5:19it difficult to read so remember the
5:22text and the slide is really there for
5:24the audience to be able to read quickly
5:26and
5:28easily color schemes are really
5:30important and thinking about the
5:32backgrounds of your slides now we've all
5:34seen a lot of slides that have very
5:37fancy backgrounds and in fact you want
5:40the simplest possible background for
5:42example a plain white background with
5:45dark letters either black or blue that
5:48works very well it's highend contrast
5:51and the only thing that appears on the
5:53slide is the information that your
5:55audience needs to understand this slide
5:58in the context of your talk
6:00avoid using the fancy patterned
6:04backgrounds you don't need your company
6:06or University logo on every slide you do
6:09not need a DNA double helix running down
6:12every slide all of that is extraneous
6:14maybe it might appear on your first
6:16slide but then get rid of it streamline
6:19it down to as simple as possible and
6:22think about the contrast of letters so
6:25dark against light works really well
6:27light against dark works very well a
6:29black background with white letters or
6:31bright yellow again high contrast easy
6:33for the audience to read when should you
6:36use which well if you were to give a
6:39talk for example at a very large
6:42Symposium with hundreds of people a very
6:44big room psychophysically in terms of
6:47and uh in terms of the projector the
6:49contrast is greatest if you use a dark
6:52background and light letters under that
6:54context so it's easier uh to project at
6:58high contrast
7:00if you're teaching in a small room or
7:03you're in a circumstance where you're
7:05worried about people staying awake um
7:07then I would use a white background with
7:10dark letters that is actually more
7:12effective however you do want to think
7:14about your science for example in my
7:16science I use a lot of uh fluorescent
7:20photo micrographs where I actually
7:22prefer a dark background because the
7:26contrast people I want people to be dark
7:28adapted so they can really see my data
7:30so in one of my scientific SE seminars
7:33regardless of whether I'm in a small
7:34room at a big room I use a black
7:36background it just works better for my
7:38science there are certain color
7:40combinations that you should avoid red
7:42green is bad this is incredibly
7:46difficult to look at in addition a large
7:49fraction of the population is red green
7:50color blind and finally red is a really
7:53angry color I actually saw an entire
7:55seminar once with a background color uh
7:58of the slides that was red and I was so
8:02agitated by 10 minutes into the talk so
8:04avoid that other combinations can be
8:07equally bad and this is bad not because
8:10the colors are unpleasant but because
8:12the contrast here is not big enough for
8:16an audience member to actually read the
8:19letters against this background this is
8:21important because in science we're often
8:25creating for example these cartoons of
8:28for example a signal trans pathway where
8:30you want each protein to be a different
8:32color so we're always looking for a new
8:34color for a new protein as we're going
8:36through a flow of a talk but we want to
8:39make sure that each protein that might
8:41in fact be this blue color would then
8:44have a label on top of it that's high
8:46contrast for example for this it might
8:48be black instead of green let's think
8:50now about how to lay out a slide
8:53visually in PowerPoint so that it's
8:54really easy for an audience member to
8:57follow the content of your talk first of
8:59all I strongly recommend that every
9:02slide have a heading at the top and the
9:05best heading is actually a statement a
9:07simple sentence that says in plain
9:09English what the bottom line of the
9:12slide is now you might say Sue excuse me
9:15this is not a statement and you're right
9:17okay busted this is we're really going
9:19through an outline for this particular
9:21talk but in my science talks every slide
9:24is headed by a
9:25statement if you're going to include
9:27text in a slide it's very very important
9:30to just limit the amount of text I
9:32really strongly urge you to use no more
9:36than two lines of text in a text block
9:39why is that the minute you show a big
9:42block of text like this I have lost you
9:44as an audience member you don't know at
9:47this point whether you should be reading
9:49all of this stuff or listening to me and
9:52if you as an audience member are now
9:54distracted and confused about what you
9:55should be doing I as a speaker have lost
9:58control of my talk so don't use things
10:01like this there's one exception there
10:03might be a wonderful quotation that you
10:05want to include in your talk for example
10:08a quote
10:09from uh the double helix or a quote from
10:12the beautiful writings of Roni kahal if
10:15you're a neurobiologist and under those
10:17circumstances it's great to include the
10:19full text what I recommend you do if you
10:21do include a quotation is rather than
10:23read it word by word instead sort of
10:26paraphrase it as you're going along you
10:27might say well the reason for limiting
10:30blocks of text to just two lines is that
10:32if it goes on forever people in the
10:34audience are going to have to make this
10:35huge effort to read it and that will
10:37preclude them from paying attention to
10:39what you're saying so you can paraphrase
10:41a quote that's very effective unless
10:43it's very poetic and you want to do a
10:45dramatic reading and that's fine
10:47too lists should be short try to limit
10:52your list to just three items avoid like
10:55mad long list and if you're going to
10:57have a list I strongly recommend that
11:00you use the animation feature in
11:02PowerPoint to unveil your list one at a
11:05time so when you're talking about item
11:07one in your talk that will appear and
11:10only when you're ready to talk about
11:12item two does that second item appear
11:14and likewise when you're ready to
11:16include item three that will come up as
11:18well be very generous in a slide with
11:22empty space it's more effective it's
11:24more powerful for the audience and this
11:27is why if you start to just load your
11:29slide with stuff it's visually
11:32overwhelming and often you can get into
11:34a situation we've all been in seminars
11:37before where something is very very very
11:39close to the edge of the slide and
11:41there's a slight misalignment between
11:44the projector and the screen that in
11:47that then results in stuff getting cut
11:49off and we've all been at talks where
11:50someone says oh I'm sorry if you could
11:52have seen the slide then you would be
11:55have been able to read it well just
11:56leave some space at the different edges
11:59the boundaries of the slide and don't
12:01forget about leaving some space at the
12:03bottom if you're in a room where
12:05everyone is sitting at the same level
12:08people in the back are trying to see
12:09over one another's head so if your stuff
12:12is too close to the bottom some people
12:14in the back can't see it so leave space
12:16on all three sides and a little space at
12:18the top again for the misalignment
12:20problem okay let's think about the style
12:22of these slides I urge
12:25you most wholeheartedly to include a
12:30simple image on every slide what most
12:34slides no every slide in every slide of
12:37one of my scientific presentations there
12:39is an image why is that here's the deal
12:43we audience members who are listening to
12:45a talk take in and process information
12:49in quite different ways from one another
12:52some of us are readers and that's the
12:54reason for having a sentence at the top
12:56of each slide some of us are going to
12:58take in information by reading others of
13:01us are very visual if we have an image
13:03on the slide then that appeals to the
13:06visual we're primates we're very
13:07visually oriented so we communicate
13:10through pictures and the third way we
13:12take in information is by listening the
13:14ideal the perfect PowerPoint slide
13:18represents a an absolutely synchronous
13:22match between a simple statement at the
13:25top a simple Visual and what you as a
13:28speaker are saying so that the same
13:31content is being delivered through all
13:34three channels at the same time without
13:37any
13:38distractions make your slide simple
13:41limit the number of stuff you include on
13:43each slide and just make one or two
13:45points from each slide now we've all
13:47heard this rule or maybe most of us have
13:50heard a rule how many slides should you
13:52show in a particular scientific
13:55presentation and the rule that I learned
13:57and the rule that perhaps you learned as
13:58well
13:59the rule is you show one slide per
14:02minute so if you're giving a 20 minute
14:03talk you show 20
14:05slides I would argue now that that might
14:08have been true when slides were actually
14:10on film and uh difficult to manipulate
14:13and there was no such thing as
14:15animations now that we have tools like
14:17PowerPoint and keynote that enable us to
14:20build content progressively in slides
14:23and to have things um that are simple
14:25simple points that we can easily change
14:28from talk to talk that rule doesn't
14:30really hold anymore and I don't think
14:32there is a firm rule I think it really
14:33depends on the amount of information
14:35that you're delivering in a talk so make
14:39your slide simple don't worry about this
14:41one slide per minute Rule and rather get
14:45feedback from uh as you practice a talk
14:48get feedback from your colleagues about
14:49whether you're overwhelming people with
14:51too much data and number of slides if
14:54your slides are simple you can show more
14:56than one per minute we have all seen
14:59talks where a slide like this comes up
15:01and what does the speaker do the speaker
15:03does this the speaker says I know the
15:05slide is really busy but the only thing
15:08I really want you to look at is this set
15:10of data right here we've been there
15:13right I've been there you've been there
15:15what has the speaker just told you the
15:16speaker has just said I'm too lazy to
15:19have created a slide that is actually
15:21tailored for what I want to show you and
15:23we as audience members are looking at
15:25all of this stuff we even if we're told
15:28to look at this I mean how many of us
15:30behave I personally I'm looking at this
15:32and I'm wondering about that so the
15:34speaker has not only um shown a little
15:36bit of disrespect and laziness to us in
15:38my opinion but in addition the speaker
15:41has lost us as audience members because
15:44we are now wandering around the slide
15:46and wondering we're having our own
15:48thoughts so if you're going to show a
15:51slide like this simply take the data
15:53that you want to show eliminate
15:55everything else and put that up and then
15:58explain
15:59the data completely explain to the
16:02audience what the axes are what each
16:05color is and remember that this is the
16:07very first time that anyone in the
16:09audience has seen this particular graph
16:12you've seen it a thousand times you get
16:14it intuitively like that but they won't
16:17so just show the data that you want to
16:19explain and explain the data that you
16:21show in other words here's the simple
16:23rule if you're not going to take the
16:25time to explain it get rid of it
16:27completely it doesn't belong along there
16:30I want to show you an example of a
16:32figure that I used to show in one of my
16:34scientific
16:35presentations and it's a brain and it's
16:38got some proteins and it shows the
16:40distribution of those proteins in the
16:42brain it's a pretty simple figure the
16:44content doesn't matter to us but I want
16:46to show you what this figure looked like
16:48before I put it into the talk in this
16:51format here's what the figure looked
16:52like in the actual paper that I took it
16:55from it had all of these extra labels
16:58that extraneous I was never going to
17:00explain that mge stands for the medial
17:03ganglionic Eminence because it's not
17:05relevant to the talk so what did I do I
17:07took this figure I put it into Photoshop
17:10and I got rid of all that leaving only a
17:13little Landmark the I to orient the
17:15audience to what we were saying so you
17:18can take complex figures export them and
17:22simplify them in order to fit perfectly
17:25with your
17:27presentation we have all all seen a lot
17:30of uh especially with the int
17:32introduction of keynote some very fancy
17:35transitions between slides what do I
17:37mean by transition well in my talk so
17:40far when I've Advanced the slides you
17:42simply see the new
17:44slide remember when keynote came out and
17:47the first time you were at a talk and
17:48the speaker Advanced to the next slide
17:51and the slide formed a box and the Box
17:54rotated in three dimensions and then a
17:56new slide appeared and you thought oh
17:58that's that's really cool that was wild
18:00and then the speaker got their advance
18:02and they Advanced to the second slide
18:04and the slide then did a big spiral and
18:08then the new slide appeared and by the
18:10third or fourth time that the speaker
18:11did that you felt
18:13nauseated we yeah so avoid these fancy
18:16transitions they are not in service of
18:18the talk there is an exception though
18:20there are some transitions that actually
18:22illustrate in a very subtle way actions
18:25here's an example this is a slide that
18:27I've shown in some of my uh
18:29presentations that have talked about um
18:31the migration of neurons in the
18:34developing brain and in this particular
18:36talk what I the point I wanted to make
18:38was that we would take out neurons from
18:40a very small region and culture them so
18:43I used a very simple whitee transition
18:45watch this it's subtle but there's the
18:48transition it actually wipes from one
18:50side to the other that sort of mimics
18:52the act of taking something out of the
18:53brain and putting it into culture so
18:56that's a sensible use another sensible
18:58use would be uh if you're going to zoom
19:00in on something if you're going to zoom
19:01in on this region you might use a zoom
19:03transition apart from that just advance
19:06from one slide to the next with a simple
19:10transition don't drown the audience with
19:14data what makes a talk memorable is when
19:18a speaker communicates a couple of
19:21take-home messages in a way that
19:24presents just enough data to be
19:27convincing but not so much data that the
19:30audience is just flooded with
19:33information and can't process it less is
19:37more and here's a way to think about it
19:39those of us who are biologists are used
19:41to culturing cells in minimum essential
19:44medium just enough of the critical
19:49nutrients and um uh elements and
19:53vitamins um that a cell needs to survive
19:56and nothing extra minim minum essential
19:59medium now if you take out any one of
20:02those elements in the medium the cells
20:05will
20:07die adding more doesn't add any value to
20:10your culture so if we think about
20:12minimum essential media we can also
20:16think about minimum essential data just
20:19the data that you need to convince the
20:22audience of The credibility of your
20:24science and no more that's really
20:27difficult and I think I think that's the
20:29most difficult decision that we have to
20:31make in terms of presenting a seminar
20:34what is just enough and where is the
20:36line between just enough and not too
20:39much for that experience helps a lot but
20:41also getting feedback from your
20:43colleagues when you're practicing your
20:44talk um to ask people can I get rid of
20:47this what can go all don't ask what can
20:50I add ask what can go to make this
20:52convincing and thorough and yet not too
20:55much so I think we all realize it's
20:58really easy to use PowerPoint badly
21:00right so we've seen a zillion examples
21:03like this of people presenting Journal
21:05clubs or seminars you see a slide like
21:08that we as audience members are visually
21:11overwhelmed we're lost we don't know
21:12what to look at we can't read this stuff
21:14it's too much to read we're
21:16distracted the minute a slide like this
21:18goes up the speaker has lost the
21:20audience completely so it takes a lot of
21:23work don't get me wrong it takes time to
21:25use PowerPoint or keynote effectively in
21:27a seminar because you have to think
21:29carefully about what you need to present
21:32what you need to present not what you
21:33want to present not what you love to
21:35present but what you need to present and
21:38to present it using clear simple
21:40graphics and a clear simple text so in
21:45fact let's go through an exercise let's
21:47take that previous slide from this
21:49lovely paper from JCB which I actually
21:51presented as uh for Journal Club many
21:54years ago let's take that previous slide
21:56which is figure two from that paper and
21:57let's see how we can break it down into
22:00its minimum essential components the
22:02first decision that we have to uh make
22:04in terms of looking at all the data in
22:06this figure is what stays and what goes
22:08what's the minimum essential data what
22:10are excuse my grammatical error there
22:13what are the minimum essential data that
22:15we need in order to present our Journal
22:16club we might decide at that point that
22:19this particular part of the figure panel
22:21B can go it's not essential but
22:22everything else is so how do we present
22:25that well let's deploy our PowerPoint
22:27rules in in this context first of all
22:30I'm going to show you how i' present
22:31panel a look at it for a second it's got
22:34some results in a gel and then it's got
22:36some images of cells how might I do that
22:38well here's the first figure or or
22:41here's the first slide in in uh
22:43presenting this figure notice that
22:45there's a sentence at the top and then
22:47I've taken the data from the figure and
22:50I've added some stuff over here I've
22:52added two labels that help the audience
22:55understand immediately that this is a
22:57PCR product and this is actually a
22:59western blot so I've added some stuff to
23:00be helpful to the audience but I've
23:02simplified their focus so there's a good
23:04match between showing this and having
23:07the text and what I would be saying if I
23:09were actually presenting a journal club
23:11and then using a simple animation I
23:13would then unveil the rest of this panel
23:17and I've added something for this
23:18audience down here showing you that this
23:20is actually a a photo micrograph of mdck
23:23cells so that's panel a presented in
23:26really two parts using a simple
23:28animation and some additions the next
23:31thing I might want to show actually
23:33comes from panel C of that uh figure and
23:37I've added a bunch of stuff here to be
23:39audience friendly first of all notice
23:40there's a sentence at the top a
23:42statement of what's going on I'm showing
23:45just one simple panel I've added a label
23:48and I've added this statement to help
23:50the audience that we're looking down at
23:51the surface of these cells from the
23:54Lumen I've also added some color coding
23:57so that they know
23:59what proteins are visualized in
24:01different colors that wasn't in the
24:02original figure I added that because who
24:04is the slide for it's for the audience
24:06the audience needs this and then using
24:08an animation we'll then show a side view
24:11of the same cells that's the control
24:12cells NOW Watch What Happens two things
24:15happen the sentence at the top changes
24:18because there's a new message and we've
24:19added the contrasting view of the cells
24:22that uh have this particular protein
24:24knocked down let's now think about the
24:28last part a new sentence that really
24:30states that this protein is essential
24:32for these cells to form little
24:33microvilli and in the absence of protein
24:36they don't so now we've presented that
24:39complicated figure in a few parts using
24:42some simple an animations in a very
24:44audience friendly format so what's the
24:47bottom line with PowerPoint make simple
24:49slides simple slides that make one or
24:51two points build your content
24:53progressively using animations rather
24:56than present everything all at once
24:58just show one and PowerPoint is and
25:01keynote they're all about control
25:03they're controlling what people see what
25:06you're saying and what they're reading
25:08and have all those match at the same
25:10time and remember if you're not going to
25:13talk about it just leave it out if
25:14you're not going to explain it it
25:16doesn't belong there at all it
25:19goes
25:20okay that's what I wanted to tell you
25:22about PowerPoint I hope that's helpful
25:25and now we want to transition into
25:27thinking more broadly about how to
25:29structure a talk so that its
25:32organization is clear to the audience
25:34and I think about it as almost taking
25:36someone by the hand and walking them
25:38through the science right and saying
25:40here's what I want you to see now and
25:42here's the path that we're following
25:43here's where we've just been and here's
25:45where we're going so a good talk is like
25:47a good paper it has a structure right a
25:50good talk starts out with a big question
25:53and then we build content over time we
25:55go through the meat of the talk and then
25:57we end with the conclusion that
26:00basically reaches back out to the big
26:01issues that we started same way as we
26:03write a paper why are we so bad at
26:05actually doing this it talks well if we
26:07discipline ourselves to do it I think we
26:09can do it better I'd like to show you an
26:11example from one of my own seminars of
26:13how I actually structure an introduction
26:16to define the really big question and
26:18then give enough background information
26:21to enable the audience to be able to
26:23follow the meat of the talk in in in the
26:26middle and as I'm doing that I also want
26:29to show you a trick that I've learned
26:32that I think is extraordinarily
26:34effective at giving good talks and
26:37that's an idea that I call something
26:39like a home slide or a Home image it's a
26:44picture maybe a cartoon an image of some
26:47sort that is going to signal to the
26:49audience that you're at a point in the
26:51talk where you're going to make a
26:53transition it's a little signal to the
26:54audience to perk up because you're going
26:56to tell them where where you've just
26:58been what it means and where you're
27:00going next so as I show you the
27:02introduction to one of my talks I'm
27:04going to build up into a home slide that
27:06you'll see a little bit later is going
27:08to come up again and again and again at
27:11transitions you might feel a little bit
27:13Jarred because now the background of the
27:14slide is going to change from white to
27:16black and this this Jarred feeling that
27:18you might experience is one of the
27:19reasons that I think it's actually quite
27:20important in a talk to keep this the
27:22same color background throughout the
27:24entire thing so here we go so here's the
27:27first slide from my standard seminar and
27:29notice first of all that my name and
27:31institution appear right here why is
27:33that well because people in the audience
27:34may be taking notes and they might want
27:36to write down your name and they'll want
27:37to know how to spell it correctly so
27:39it's a courtesy to them so as I start uh
27:42in my talk I start to discuss a kind of
27:45analogy between the brain which is what
27:47we study and a computer chip because
27:50each of them has to make you know really
27:53specific connections with one another uh
27:55the brain during development and the
27:56computer chip and a factory and I make
27:57kind of joke and see if people laugh and
27:59gauge the uh humor level of my audience
28:03at that point and then I show the
28:06audience the cells that actually work on
28:09the actual circuitry of the brain that
28:11my lab studies in the cerebral cortex
28:14and then I frame the major question that
28:16my lab is interested in which really has
28:18to do with the question of how it is
28:20that these individual neurons in the
28:21brain know what kind of connections to
28:24form what are the molecular mechanisms
28:26that guide those choices about sulfate
28:28and connectivity at this point then I
28:31introduced the two cell types that the
28:34talk will focus on to sets of neurons
28:36that you'll see have different colors
28:38and then I introduce two questions that
28:41the talk will address during the course
28:43of the next um however long it is maybe
28:46a 45 minute seminar one question having
28:49to do with the fate of the yellow cells
28:51and another question having to do with
28:53the fate of the blue cells and that
28:56basically then launches me into the
29:00actual data part of the talk this
29:03particular image right here is the home
29:07image or the home slide that I was
29:09referring to just a minute ago and as
29:12you'll see this image is going to
29:14reappear again and again and again in my
29:16talk during transitions and we'll see
29:18just in a couple of minutes how that can
29:20get played out and how it can be
29:22extraordinarily effective in enabling
29:25the audience to follow a talk okay so
29:28now we've built up an introduction and
29:29we're ready for the middle right so
29:31we're ready for the the middle of the
29:32talk which is the
29:34meat but there's a problem and that is
29:38that audiences have fairly limited
29:42attention spans this is an interesting
29:44graph so let's look at a little bit of
29:46data this is a plot of the percentage of
29:48a class that's paying full attention to
29:51a lecturer over the time during which
29:54the lecture is being delivered okay
29:56there's some good news right the the
29:58good news is right here in your
29:59introduction people are actually kind of
30:00tuned in and listening and here's the
30:04bad news which is that okay let's face
30:06it depending on how compelling the
30:09lecture is it might be here or it might
30:12be way down there but the bottom line is
30:15that at no in in under no circumstance
30:19are you ever going to get 100% audience
30:22attention for your entire talk why is
30:25that well think about being a member of
30:27the audience
30:28why does this happen it's because we're
30:31human we have thoughts we have thoughts
30:34that intrude on our attention span we
30:37have thoughts about oh my gosh did I
30:39really did I remember to load that gel
30:41or turn it off turn off the power supply
30:44um we have thoughts about whether we
30:46actually are going to be on time getting
30:49the kids from daycare we have thoughts
30:51about someone we just met you know we
30:53have all these random thoughts that
30:54intrude so we're human and we're going
30:56to space out we then as speakers have
30:59two choices one is we could say ah okay
31:02someone spaced out take no prisoners you
31:04spaced out forget it you're just you're
31:07never going to understand anything else
31:08I say or we can just acknowledge the
31:11fact that the audience is human that
31:13every single member of an audience is
31:17going to tune out at some point and we
31:20can build in a mechanism within our talk
31:22to enable them to catch up so obviously
31:26I kind of recommend the latter so so if
31:28the middle is the meat of the talk and
31:30it's the time when the audience is going
31:32to zone out at some point why because
31:34they're just people well what are we
31:36going to do about it what I suggest you
31:39do is to visualize the middle of the
31:42talk in the following
31:45way you're going to have a series of
31:48episodes or data Dives that you're going
31:51to present little stories and as you're
31:56presenting a particular Story You're
31:57going to start at a level that's fairly
31:59non-technical and you might actually get
32:01pretty deep into the data to a level
32:04that's really going to appeal just to
32:05specialist in your Fields now if we were
32:07to give a talk and start presenting our
32:09data and we've all been at seminars
32:11where people do this where we just go
32:13down into depth and then we just do data
32:15data data data data data data data data
32:17data data data data and then we conclude
32:19and it's over we've lost people along
32:22the way so this structure this visual
32:26structure here is a way way of planning
32:28your talk so that you can think about
32:32diving down into data and then coming up
32:34for air and it's at these transition
32:37points between data Dives that you use
32:40your home image to basically let the
32:43audience members catch up and know where
32:46you've just been what you're concluding
32:49from that part of the data and where
32:51you're going to go next so let's look at
32:54an example of one of those data dives
32:56from one of my seminars you've already
32:58seen the introduction to the seminar
32:59you've you've already seen how I
33:02introduced my home
33:04slide at this point in my talk I've gone
33:07through a review of the literature the
33:09First Data dive and I'm now beginning
33:11part two of the talk and here's the
33:12beginning of that section so once again
33:15you see this home image and you see the
33:16two questions we're halfway through the
33:18talk but I'm summarizing part one and
33:21getting ready to introduce the second
33:23set of questions at this point then I uh
33:27now
33:28uh start to explain that as we focus on
33:32the blue cells that we're going to be
33:34looking at the role of a separate
33:35protein that is actually expressed in a
33:38subset of neurons within the different
33:40layers and then I explain that we've
33:42generated knockout mice in which we've
33:44inserted a reporter into the cells that
33:47normally Express these this protein so
33:49that we can see their axonal connections
33:52and now we don't really need to focus on
33:54the data but look there's a sentence at
33:56the top that basically says what I just
33:58told you then we see a simple figure
34:00from the controls in which I would point
34:02out that what I want the audience to
34:04look at are the blue axons that label
34:07the connections between the two
34:08hemisphere that's in the control and
34:10then we look at the mutant the thing I'm
34:12going to emphasize is the absence of
34:14those connections here now we're going
34:16to look at it from the side same stuff
34:19sentence two simple images words will
34:22match the content where I really
34:24delivering the seminar to you and then I
34:27tell you what I just told you I just
34:29told you that the blue cells Express
34:31this protein that they form this type of
34:33connection and then I say you know in
34:36the first part of the talk I told you
34:38about another pathway so you might be
34:40wondering what is the relationship
34:42between these two Pathways so then I
34:44would go on to actually test that
34:46connection by uh presenting more data
34:50that would then build up into a small
34:52conclusion slide showing you that this
34:55protein represses the expression of this
34:58Gene so simple Graphics that really
35:02emphasize the bottom line the story that
35:05I'm telling you with the minimum
35:06essential data Okay so we've gone
35:09through the meet thinking about data
35:11Dives coming back making transitions now
35:13let's think about how to conclude a
35:15talk here's the good news is that as you
35:18say and now in conclusion look what
35:21happens to the attention level of your
35:23audience it's like oh it's it's over hey
35:26I better see what I just saw so that's
35:28good news now you've got one more really
35:30good chance to deliver your take-home to
35:33the audience
35:34effectively there's a danger to this
35:36though as well and that's the danger of
35:40hearing for example 40 minutes into a
35:42talk someone say oh to conclude and our
35:45we're like oh good you know actually
35:47gosh they're ending 20 minutes early
35:48that's great I can actually get a little
35:50bit more work done today so they St in
35:52conclusion blah blah blah blah blah blah
35:53blah blah blah and then they say and now
35:56in part two of of my talk and what's our
35:59response as an audience we're like oh
36:01dude no why is that because it was a
36:04false ending and it got us all excited
36:06so if you're going to conclude part one
36:08of your talk do the audience the favor
36:10of saying to to sum up this first part
36:13of the seminar and that way you won't
36:15create a false expectation but the good
36:17news is that you will perk up uh
36:19interest as you signal your conclusion
36:21and that means that you have one last
36:23chance to really reiterate your specific
36:27conclusion
36:28and most of us remember to do that but
36:30it's easy to forget to sort of back out
36:32to the big picture and return to the
36:35beginning to have your talk come full
36:37circle so that you're really ending uh
36:41by revisiting the big questions that you
36:43introduced at the very uh first part of
36:46the talk so again here's how I would do
36:48it in my talk just as a visual example
36:51of how one might think about this
36:53concluding up now sentence at the top
36:56the same diagram that you seen before
36:57about this repression and now adding
37:00other things you haven't seen the data
37:02but I would add that just to remind them
37:04that at least uh one of these mechanisms
37:06Works through the modulation of
37:08Chromatin they would have seen that
37:09slide already um I would then connect it
37:12back to the first part of the talk and
37:13talk about a little switch that had
37:15gotten set up um I would then refer back
37:18even earlier to a part of the talk that
37:20you didn't see now going back to the
37:21really big picture questions about how
37:24different types of neurons emerge over
37:26time many of these questions are still
37:28unanswered and therefore represent a
37:31frontier in my field I would then wrap
37:34up by acknowledging a course the
37:36colleagues that have contributed to the
37:37work over time and then here's a
37:40trick as you get to the very end of your
37:43talk rather than leaving up your
37:45acknowledgement slide have your
37:47conclusion slide appear at the very end
37:50for the question and answer period why
37:52is that well it helps the audience
37:54remember everything is in service of the
37:56audience and I actually have The
37:58Misfortune of working on three genes all
38:00of which you have names that end in two
38:02and none of which are memorable by
38:03anyone who doesn't work on them so by
38:06putting this picture up I'm enabling the
38:09audience to actually ask intelligent
38:11questions first you know if they're
38:13struggling to remember which gene they
38:14had a question about and secondly when
38:17I'm answering a question I can actually
38:19use the visuals to explain my answer and
38:21that turns out to be very helpful so I
38:24really recommend that at the end of your
38:26talk during Q&A put your summary slide
38:28back up again it's uh helpful for you
38:31and for the people attending your talk
38:33okay so what have I just told you about
38:35organizing a great talk first of all use
38:38PowerPoint wisely have very clean
38:42minimal slides secondly start out with a
38:46broad introduction so that everyone in
38:48the audience understands the big
38:50questions that are compelling in your
38:52field and then introduce the specifics
38:55gradually giving them just enough
38:57background information so that they can
38:59follow the data but not so much that
39:00they're
39:01overwhelmed third think of the talk as
39:04consisting of these episodes or little
39:07data Dives in which you're going to go
39:10at a point into your data remember the
39:12minimum essential data and then you're
39:14going to go into some depth but then you
39:16come up for erir and allow the audience
39:19to regroup see where you've just been
39:21frame the next question and then get
39:23there logically rather than just going
39:25through data data data data data which
39:27was just going to leave people in the
39:30dust a very effective way of making
39:34these transitions explicit is to design
39:37a home slide or a Home image that is a
39:41signal to the audience that you're at a
39:43point of transition and they will perk
39:45up and it will help them to understand
39:47the flow of your talk and finally your
39:50conclusion here is just the opposite of
39:52the introduction you start more specific
39:55with what you've learned but then end up
39:56broad again
39:58so is this all you need to know to give
39:59a great talk no way right I mean these
40:02are just some basic rules or suggestions
40:06or advice about how to create slides
40:09that are userfriendly audience friendly
40:12and how to structure your talk so that's
40:14well organized and very very clear and
40:17simple for an audience to follow but
40:19there are a lot of things that going
40:20into great giving a great talk um there
40:22is the whole performance aspect of
40:25speaking um there is the actual
40:27scientific content but at least in terms
40:31of the performance you can practice that
40:33you can and should rehearse your talks
40:35in fact the entire
40:39time from the first day of graduate
40:41school all the way well into being a
40:44tenur professor at Stanford I rehearsed
40:48every talk every lecture every Journal
40:51club that I gave every one of them and
40:53finally I got so comfortable with
40:55speaking I I don't actually have to do
40:57it anymore but that means rehearsing out
41:00loud and often in front of an audience
41:02dozens and dozens and dozens of talk get
41:05feedback and practice the way you're
41:07going to speak have the words rehearsed
41:10and get some input from your colleagues
41:13about whether you've identified
41:15effectively the minimum essential data
41:17for the talk secondly have yourself
41:20videotaped go and watch yourself and see
41:22what you do see if you're gesturing
41:24naturally and so on so those kinds of
41:26aspects of delivery are much more
41:28effectively done oneon-one rather than
41:30through a format like this another great
41:33resource that I highly recommend is a
41:35wonderful book by Michael Ally called
41:37The Craft of scientific presentations if
41:40you're going to have just a single book
41:41on scientific presentations in your
41:43bookshelf I personally would recommend
41:45this one it's a terrific model and it's
41:47actually the basis of a lot of the
41:49information that I've presented to you
41:50today well that's it thank you so much I
41:53hope this has been helpful and I wish
41:54you every success with your talks in the
41:56future

About this transcript

This page contains the full transcript of Susan McConnell (Stanford): Designing effective scientific presentations by Science Communication Lab, generated from the public captions YouTube serves with the video. The transcript has 7,576 words across 1,060 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.