Iva Cheung fires a shot across the bow with a long blog post titled, “Why academic conference posters suck.”
Ahem. Obviously, I have thoughts on this.
Cheung begins by noting that there is not a lot of research on conference posters. This is true, but it is expanding. Melissa Vaught has been tracking this on Twitter with the #conferencetopub hashtag. Some fields, more on the health and medical side, are all over this.
She then makes the arguments that poster sessions are socially awkward. “No one’s quite sure what to do or how to react,” she writes. First, this is a problem with the entire concept of going to an academic conference, not just poster sessions. She even goes on to say, “posters save people with anxiety from having to speak in front of a crowd.” Second, this is a case of, “Your mileage may vary.” I have seen many people who know very clearly what to do and how to react. Some people feel awkward during any social interactions with new people. People can get better at this.
Institutions should have poster printing capabilities
Cheung argues that “posters are expensive.” That is not a problem with the poster format. That is a problem of institutional support. I have not paid for a poster in years, because my university has invested in a large format printer and paper. Department chairs and deans should realize that conference posters are a routine part of academic presentations, and invest accordingly.
Posters force you to think about what you’re doing
“Posters take an enormous amount of time to prepare,” Cheung writes, “whereas presentation slides can be (and frequently are) prepared on the flight over to the conference.” This is a feature of posters, not a bug. You have to think about your content in advance, and make hard decisions about what you are going to include. When you print a poster, your work is mostly done. Making a PowerPoint deck on the plane trip is rushed, half-assed preparation in comparison. If you make a PowerPoint deck on the plane, you still have to practice delivering the talk so that you don’t go over time. At least you should!
Cheung notes “travelling with a poster can be cumbersome. Because of their length, poster tubes technically exceed carry-on size restrictions(.)” I have no doubt this happens sometimes, but it seems to be vanishingly small. I know of nobody personally who’s had a problem taking a poster on a plane. Getting a laptop through security seems almost as bothersome.
Cheung’s next section has the header, “most academic posters are a visual nightmare.” Well, yeah, that’s what keeps this blog in business. But so are most PowerPoint talks.
We can do better on poster accessibility
Cheung’s most important argument is about the accessibility of posters. This is an important conversation, and one that I don’t think conference organizers and presenters think about enough.Posters are a visual medium, which poses a problem for someone with poor or no eyesight. Cheung argues that “oral presentations give people with visual disabilities immediate access to at least some of the content.” This is true if there is no presenter at a poster. If there is a presenter at a poster, however, the one-on-one nature of a poster presentation means that a presenter can more readily adjust the discussion to take into account the visual issues of the listener. A presenter giving a talk is unlikely to adjust the talk to accommodate anyone in the audience with a visual issue. (See this post about the experience of a blind colleague listening to conference presentations. Also see this post about a blind poster presenter.)
I have also seen posters incorporating 3D printed elements, which could make some aspects accessible to a visually impaired people in a way that a talk could not do.
Some conferences are videorecording talks, which can again make content available to visually impaired people. Some people are archiving posters online, and I thought standard text to voice tools would be able to help this problem.
I pulled up a PDF of my last poster, and asked Acrobat Reader to read out loud. Reader’s “read out loud” was not working for any document, but Acrobat Standard did read it. I learned that the kerning I did to make the poster look better disrupted the text recognition: it treated words where I had moved a letter as separate words. Hyphenated words were also read as separate words. I learned that if I was to archive posters, I should include a plain text version in the description. Like most issues around accessibility, this is not an unsolvable problem.
Cheung argues that posters are horrible for learning, citing ideas about glucose use that sounds rather similar to some contentious ideas about sugar and willpower. She argues that academic posters are too complex to learn from. I agree that most posters are too complex: this is, again, one of the reasons this blog exists. It is seems to me that any form of academic communication faces this problem.
Posters help start dialogue
Cheung suggests more short talks as alternatives to posters. This looks sensible on the face of it. Most people prefer talks, both as a presenter and an audience member. I love talks in the Ignite format. They can work well for small meetings. But I have extreme doubts that they can replace poster sessions or many meetings. The number of presenters is too large, and there is not enough space or time to accommodate them.
I also worry that a whole bunch of five minute talks will blur together in memory. It is hard to stand out when you have four or five talks an hour; imagine if you are sitting through 10 talks an hour. For eight hours. For several days.
Talks, by their nature, are synchronous “one to many” communications, typically with limited time for discussions. (And can you image the difficulty in people switching from room to room every five minutes?) Posters are more complex. Audience members can listen to the speaker at different times. The format permits conversations in ways that talks don’t.
The other main suggestion she has is for conference organizers to building in more networking time. Those hallway conversations are often the best thing about conferences. But just having “the explicit expectation that people with similar research interests can use that time to find each other and chat” is, perhaps, overly optimistic. Talking to strangers is hard. You can’t just put people in room and expect conversations to flow freely. Having a “social object” like a poster helps people identify others with similar interests, and gives them something to talk about.
Poster sessions fill a niche. Posters provide a straightforward way for a listener to identify who is working on topics they are interested in. (There is rarely enough time to read all the article titles in the abstract book, but you can easily scan rows of posters to find who is doing what.) Posters give people more opportunities to talk individually, and to take as much or as little time as possible.
I agree with a lot of Cheung’s points, but not the conclusion that we should kill all poster sessions. Let’s make posters better rather than abandoning them.
External links
The Zen of Presentations, Part 34: Lessons from the blind
The Zen of Presentations, Part 40: Lighting a fire under speakers
Hat tip to Mary Ellen Foster.
I'll go further on this...taking for example my attendance at the Division for Planetary Science meeting in October, which is a medium size general academic research conference (I think there were 850 registrants). Five days, three parallel sessions at any given time (save during plenaries), and talk *spacing* is 10 minutes, thus your talk is maybe 6-7 min with time for one or two questions and the speaker change over. So, as a speaker, my entire science interaction was crammed into that tiny mote of time, and on average 66% of the attendees weren't even in the room.
ReplyDeletePosters, however, were allowed to be up for the entire first four days of the meeting. Three of the afternoons had dedicated poster sessions clocking in over 2 hours each, and during each poster session there were 'highlighted' groups of posters where you were sort of guaranteed to have a presenter at the poster, and that presenter was guaranteed a flow of enthusiastic poster viewers. Plus, during other hours you could peruse the entire poster suite at your leisure and note ones to come back to, follow up on with the first author/presenter, etc.
This does present challenges, but the chance of a good poster making an impact is probably far higher in this case, since it is simply up and available so much longer than the few minutes of a talk. I actually posted about this here a few years ago, that posters hung at a conference like this actually need to take into account this dual nature of 'grab and hold while I present it' during the poster sessions, but being stand-alone enough to be viewed without a presenter to guide you over the course of the full meeting.
I think for myself I'm going to try to do a poster next year for these reasons. As always, for me a talk is attractive for exactly the (not great) reasons brought up in the column: I'm an inveterate procrastinator, so tossing the talk together during the meeting has great appeal. And that's exactly why I shouldn't do a talk.
Almost forgot, but this year's meeting was our first attempt to use digital posters, in this case the iPoster format. Only a select number of posters that specifically asked to be displayed this way were chosen (maybe 10ish? not clear). Such posters allow for direct interaction, and can include video, audio, etc. I think the reviews were very mixed for the iPoster format itself, but strong enthusiasm for the concept.
ReplyDelete
ReplyDeleteMy own experience with poster sessions over the years is one of limited interaction. Sometimes that have been in part due to poor set up (e.g. putting posters in side room, away from flow of traffic). There's sometimes the challenge of competing with the only slot that's open for attendees to get lunch. of course there are issues that could be addressed by organizers.
I've also seen short talks be very effective, but partly because equal to me was allotted for discussion and there was a single session. This means fewer talks, and really doesn't translate to large society conference scale. (FWIW the same conference had posters too, with 1.5-2 hours for poster sessions, with snacks provided and nothing concurrently scheduled.)
With regard to the research about conference posters, I think it's important to note that what I've found is largely about rates of conversion from conference presentation to publication. This tells us little to nothing about how effective conference presentations are for communication or comprehension.
Perhaps the biggest benefit of conference presentation is having an enforced deadline for thinking about the status of a project and how to communicate the information. That takes time, again regardless of format. IME acceptance of an abstract for a poster session seems to be a given for society meetings in the life sciences, and travel awards & support are often predicated on presenting at the meeting. So maybe the heart of the issue is treating posters as a checkbox for attendance.