Bats surf storm fronts, and public perception of preprints
33:34 Share

This is a Science Podcast for January 10th, 2025. I'm Sarah Crespi. First up this week, as pre-print publications ramped up at the beginning of the COVID pandemic, so did media attention for these pre-peer-reviewed results.

But what do the readers of news reports based on preprints know about them? Jeff Brainard, an associate news editor here at Science, joins me to discuss the public perception of preprints in the news and how to inject skepticism into stories about them. Next, placing tiny tags on bats to follow them across Central Europe.

Former science intern and now researcher Edward Hermey revisits the podcast after 13 years away. He discusses the difficulty of tracking bats as they fly long distances at night and what new tagging technology is revealing about their migration patterns.

So today we're going to talk about the public perception of preprints, how they're covered by news outlets, and how they're treated the same or different to peer-reviewed research by the public. Associate News Editor Jeff Brainerd is here. He runs our News in Brief section and covers the ins and outs of scholarly publishing. Hi, Jeff. Welcome to the Science Podcast. Yeah. Hi, Sarah. Thanks for having me on today. Oh, sure.

To my mind, posting preprints, you know, finished manuscripts that have yet to be peer reviewed, they haven't undergone peer review. That's kind of been the norm in the physics world for a long time. I went and looked this up. The archive preprint server is for physics has was started in 1991. So this is a long standing practice over in that domain. But biology, medicine, they're a lot slower.

to post preprints. And this is for various reasons. But preprint server posting in medicine and biomedical science really accelerated during COVID, as did the news coverage of preprints. I even remember debates in our own newsroom about covering preprints during this time when everything was happening so fast and people were publishing so much. So Jeff, can you kind of talk about some of the tension there, like how reporters think about preprints and how they've kind of debated whether or not they should be covered or how they should be covered?

Yeah, I think that this debate really came to the fore during the COVID-19 pandemic. A lot of journalists and a lot of people reading their stories were uncertain about what preprints were and how reliable they were. And so because of the urgency of the pandemic, many scientists did

published scientific findings that they wanted to share with their colleagues as quickly as possible to help speed remedies and drugs and other public health measures. And

I think a lot of journalists felt compelled to report on those preprints before the manuscripts went through peer review because of the urgency, understandably. But in some cases, preprints that really reported shaky findings ended up getting perhaps more attention than they deserve. Early examples of that were drug treatments for COVID-19, such as hydroxychloroquine, and

And so in the wake of that, there has been some reassessment by journalists about when they should cover preprints and under what circumstances. And

Some light has been shed on this by a series of research studies by a social scientist named Alice Fleerackers, who is at the University of Amsterdam. And she interviewed journalists about their use of preprints and under what circumstances they would base a news story on preprints.

the findings in preprint. She interviewed about 30 journalists and got pretty mixed results in the sense that some journalists who were quite experienced with biomedical research and the subject matter felt that they were able to essentially vet and conduct some facsimile of some equivalent form of peer review by interviewing a number of scientists in the field.

Right. So this is an outside getting outsized comment, right? So asking another researcher. Yes. You've seen this manuscript. Do you think that it's going to hold water? That's right. And they were able to do that in a way that I think captured

caveats and enrich the reporting in some cases. In other cases, journalists, they were a little uncertain about whether they had the skills to conduct that kind of vetting or seek out enough expert opinion. So there was a practice adopted during the pandemic that journalists should

attach a kind of disclaimer, a notice. Typically, the note will say something like the preprint has not been peer reviewed or has not appeared in a scholarly journal or both. Yeah, we do that on the podcast too. We say this is not peer reviewed. This is, you know, from a preprint server or this is from a meeting. And sometimes that is also a similar situation. Yes, exactly. Do you want to talk a little bit about the further research that

scientists that you talked to did into the public perception of preprints? So reporters are handling it one way, but how does a reader handle it? What do they see? Yeah, it's an important question because scientists, when they post preprints, they know what a preprint is and they're assuming their colleagues know what a preprint is. But when a member of the public reads a news article that mentions a preprint, it's

do they always know what that means? And Alice Fleerockers, the social scientist I mentioned, she and her colleagues did a large survey study of U.S. adults to find out, do they know what a preprint means? And when they hear the term, what do they think it means? She published two studies in the last year describing her results. And in one of the studies, she reported to

that only about a third of her survey respondents could define what a preprint is in a way that's anything like what scientists would define it as. This idea of having gone through some vetting or appeared in a scholarly journal. So not many people knew what it was. And did they know that

they should question its credibility. So she tested this by giving the survey respondents different versions of the same news article that described a preprint about COVID-19 vaccines. And the original version said that the article was based on a preprint that had not been evaluated by outside experts. But she also gave a modified version that described the findings, but that didn't mention that the study was a preprint. And

Just mentioning that the study was pre-printed didn't make a difference. The respondents who were told this, they considered the findings to be as credible as the group that wasn't told. She also tested how credible the respondents found the findings by removing language that hedged

about the findings. How does a news writer say, this is something you should approach with skepticism, but I'm still telling you about it, you know, in a news story? She, Dr. Fleerachers, thinks that readers really should be told when there is uncertainty about scientific findings. And preprints may, on average, have a certain degree of uncertainty in

Because the authors are, in some cases, rushing to get out the findings, they may be preliminary, subject to further refinement. But it's not always the case that every preprint is necessarily inferior to an equivalent journal article on the same topic. There's been some research done comparing the preprint version of a manuscript to a

the final journal article version, in many cases, there are not large substantive differences. I mean, certainly peer review has its place in assuring the quality of the scholarly record, but it's not necessarily the case that there's a night and day difference in quality between preprints and journal articles.

Maybe this is an odd thing to say, but how do journalists make a story less credible? Or how do they point out that maybe a source like this, a preprint, would be less credible and that people should be skeptical? So in...

commenting about sort of the significance of her findings, Dr. Fleeracres encourages journalists and members of the public who are reading about their findings based on preprints to keep an open mind, but also be appropriately skeptical. Dr. Fleeracres encourages journalists to

include an appropriate amount of context so that readers can decide for themselves how much credibility to put in the findings reported in the

preprint. It's not an easy task, but she thinks it's doable if journalists consult an appropriate range of experts in the field who are not connected with the study. This is kind of a background question. Why was biomedical research so far behind? Why do they lag so far behind physics posting preprints on preprint servers? For a

a long time, many researchers, including ones in biomedicine, were perhaps somewhat skeptical about pre-printing as a legitimate form of scholarly communication when the model of peer-reviewed journals has been dominant for so long. So this hesitancy has persisted for some time, even though pre-printing was quite common, if not the dominant form of scientific communication among physicists. So I

I think it just shows that different scientific disciplines have norms about how they go about their work that are strongly held and maybe difficult to change. Do you think there was a concern that people would take what came out on a preprint server that, you know, had to do with human health? They would take it as the truth or correct without, you know, they were worried about what we're talking about today, but the

kind of emergency of the pandemic overcame those qualms? Yes. One of the reasons that preprints have been relatively slower to be

embraced and taken up by biomedical researchers is that when findings had a direct potential impact on public health and medical practice, where patients could potentially be harmed, for example, by a drug or other treatment that was not scientifically validated, scientists were mindful of that and perhaps reluctant to post their findings as preprints. But the COVID-19 pandemic kind of changed that calculus a bit.

And preprints in biomedical fields reached an all-time high during the pandemic. That number has come down slightly since then. I was going to ask, yeah. And there's also a role, I feel like, the way journal publishing works might have to play in this as well. So physics journals versus, say, a biomedical journal, like how much those editors or the publishers would care whether or not something had come out as a preprint before is also pretty different.

before the pandemic? Yes. Journals have an influence on whether authors pre-print their findings before submitting them to a journal for peer review and publication. And before the pandemic, some journals discouraged authors from pre-printing because they wanted to have the scoop in the

published form of the exclusivity of the article. But that's changed a little bit. I think the COVID-19 pandemic just showed that kind of moved the field a bit and the terrain in a way that journals have adapted now by relaxing some of those policies. Yeah, it's been interesting to see how much has changed.

So Jeff, since I have you here, I wanted to talk about another big story developing in the scholarly publishing world this year. The U.S. government is changing its open access rules. Can we talk about that? Yeah, actually, all of the U.S. agencies that fund research and spend more than a certain level of a year have required since the year 2013 that their grantees host webinars

the manuscripts resulting from this funded work in public repositories because the research was funded with taxpayers' money and the public had a right to read the results. There was a compromise reached in that year where the grantees and their publishers could request an embargo on the public release of these scientific papers of up to 12 months. And this was requested by the publishers for business reasons that...

They did not want to kind of lose their exclusivity that they have by putting these articles, at least initially, behind a paywall. Now there's a new policy that is being finalized as we speak and will be going to effect by the end of this calendar year. 2025, yeah. 2025. And it will require the immediate release in a federal public domain

repository of articles that result from federal funding. So that's a big change in U.S. policy and one that's causing some ripples. Researchers and their institutions and publishers are all looking at significant changes to make this happen. And not everybody's happy about it, but it's going to have potentially a big effect because something like 9% of

of all of the world's scientific papers are funded by the U.S. government. Right. Yeah. So that's a big swath of research that will suddenly be available as soon as what, it's submitted or accepted or when? The authors have to submit the manuscript to the repository as soon as it is accepted by a journal and it becomes public as soon as the journal article is published.

What about the data, the availability of the data as opposed to the published work? Yeah, making the data available is a key part of all of these policies. And that may be an even more far-reaching change and require bigger changes in practice by researchers and their institutions than the manuscript departments.

depositing requirement. And that's because the data sharing raises a variety of technical and logistical questions that a lot of researchers and the universities have not grappled with yet, as far as who formats the data in ways that it can be usable by others, and how much of the data and what kinds from a particular project

should be included. Is this a funded mandate? Like, is there a place to host this that's being paid for? That is a key question that universities have. The policies will allow for data to be deposited in an institutional repository or one that's acceptable to an agency. And there are standards for those data repositories, but

The extent to which any particular repository is going to be deemed sufficient for this purpose is still under discussion. One way that universities will be able to pay for these costs is through the indirect or overhead portions of federal research grants. But universities say that there's already a lot of demand on that revenue flow. All right, Jeff, this has been super fascinating. Thanks so much for talking with me. Well, it was a pleasure to talk with you, Sarah. Thanks. Jeff.

Jeff Brainerd is an associate news editor at Science, where he manages the news in brief section and reports on scholarly publishing. You can find a link to the stories we discussed at science.org slash podcast. Don't touch that dial. Up next, we talk about tracking tiny bats in the nighttime.

Morgan State University, a Baltimore, Maryland Carnegie R2 doctoral research institution, offers more than 100 academic programs and awards degrees at the baccalaureate, master's, and doctoral levels, is furthering their mission of growing the future, leading the world. Morgan continues to address the needs and challenges of the modern urban environment.

With a four-year quadrupling of research, more than a dozen new doctoral programs, and eight new National Centers of Excellence, Morgan is positioned to achieve Carnegie R1 designation in the next five years. To learn more about Morgan and their ascension to R1, visit morgan.edu slash research.

Just like some birds, some bats are known to migrate, seeking resources in different places with the change in seasons. But following these long nighttime treks has not been easy for researchers. Recently in Science, our former intern, Edward Hermey and colleagues, wrote about tracking common noctual bats, that's noctual, not nocturnal, as they flew more than a thousand kilometers across Central Europe.

Hi, Edward. Welcome back to the Science Podcast. Hi, it's great to be back. Yeah, so you were on the other side of the mic way back in the day. What do you remember about the Science Podcast in 2011? I loved it. I had such a fun time. Yeah, just getting a random paper every week to read and cold calling a scientist to ask them about it. But it's, yeah, some of my favorite memories from before my PhD was getting to the Science Podcast. It was so cool. Yeah.

Absolutely. And we still get to do that. We still get to talk to a random researcher every week. And I picked bats because bat migration is such an interesting field to research. It's hard to find them at night. What are some of the innovations that have let people begin to track these kind of tiny little furry light bodies?

that travel long distance at night. We have these new tags. They're called Icarus TinyFox Bat Tags. It's a fun name, but it kind of captures some of the aspects. It's a battery-powered tag that sends data over what's called the Sigfox network. And

For many previous tags, you would have to either follow the bats along the way or find them the next day and take the tag off to download the data. So you have to do the migration with the bat. That sounds like a lot of resources. And so there was a lot of effort from my group previously who have been studying these bats for over a decade now, where they actually followed them by plane to see these migrations. But you really can only follow one individual at a time.

And so with these new tags, we could actually have the data transmitted remotely using the Sigfox network, which can send roughly one message per day that kind of summarizes a lot of the behavior of the bat, but also gives us a location so we can track them as they move away from us. This also utilizes something called edge computing, which I think we've talked about a little bit on the podcast. Can you talk about how that played a role here? Yeah. So the tags are actually collecting a lot of different sensor data.

they can basically summarize that data on the tag,

and give us just the summary statistics. So we were getting the minimum and maximum temperatures, but also a summarized version of the acceleration that was a very useful proxy for energy expenditure. The idea is that it's less data, it's compressed into this tiny little package that it doesn't have to kind of transmit as much and it saves power. Exactly, exactly. Sigfox network has a limited bandwidth. And so you can only send about like, you know, an emoji's worth of data. So everything is compressed effectively.

And then we decompress it and know exactly which numbers correspond with what data.

Why is it important to get daily readings? I mean, can't you just say they started here and they ended here? You can learn a lot from just knowing the start and end points, but you really lack any understanding of what they're doing in between. And so a lot of previous work on migration actually came from data like that, where we have mark recaptures. You put a ring on an animal and you know where you released it, and then hopefully someone finds it in the future. But you kind of lose a lot of that

information of when exactly did they leave, how far did they go, what were they experiencing along the way, and maybe what kind of strategies they used to get around bad weather or a mountain range maybe. How did you get these tiny tags on the bats? We know where they sleep during the spring. We have roosts that are in these bat boxes that are hung on trees, and the bats use those to hibernate.

Once they come out of hibernation, they start to move from those boxes to what we call a stopover site, which is actually a wastewater treatment plant. And they move into the walls of that building. And so knowing where they're spending the day, we can catch them more easily. But it's still quite a challenge. And I'm asking all the people and friends that I have to come help me catch bats during this spring season every year. How long did you follow these bats and how many were you following? We tagged all.

about 100 individuals over three years. And the battery of the tag can last up to two months. But typically we only got data for around two to three weeks per individual. And so the tags were attached with glue predominantly, and the tags can fall off. The other thing that could happen is the tags just go outside of coverage. But we were able to follow, I think the longest individual was a

about five weeks of data, which is great. So, you know, having this relatively for bats, long-term data is fantastic to see sort of how they adjust to many different environmental changes and how they're moving over this large landscape.

What did you find were important factors in whether or not they migrated, stayed still, went faster, went higher? When we first looked at the data, we were surprised that there was just so much variability in when the bats were going. But if we plotted the migration days all together, we could see that there were certain peaks and that there were a lot of individuals doing certain things on the same day. And so we wanted to kind of see if there was any relationship to the environment. And when we did the model of the analysis, trying to predict

Whether or not a bat would migrate, we found that there was actually a strong relationship between wind speed and direction and temperature and pressure, suggesting that bats are actually migrating before these storm fronts. So in the spring, we have warm air coming and blowing typically northeast, which is the same direction the bats are trying to migrate. And it seems like they're surfing along the front of that storm.

And they can actually get this benefit from the wind support of the storm kind of pushing them along. That is very cool. I love this very poetic title about surfing on storm fronts for your paper. So, but you also talk a little bit in paper about flexibility and how that plays into their migration. What were they flexible about and what were some of the trade-offs that they faced? So we saw that bats really could migrate on almost any day. There was very few days when there was no migration happening at all, just showing that

the environmental conditions are only one component of the total migration strategy. There was also almost two months of windbaths from migrating over a relatively small scale. And so they can choose either to be these early migrants or late migrants. And when we kind of tried to explore that, so we found that late migrants actually seem to use more acceleration, which is often used as a proxy for energy expenditure than early migrants. So we suspected there's actually a trade-off. Perhaps late migrants are more

facing this extra cost by migrating later than the early migrants. Was this kind of what you were expecting to see in your bat migrations, this relationship with the weather and some of this stuff about early versus late migrators?

From previous work from our group, we had looked at when bats depart for migration, but not necessarily tracking them in relation to the weather. And in that work, we knew that wind and air pressure were important. But what was interesting is we actually had the wrong idea with air pressure. And so if you look at just the instantaneous air pressure that they're leaving under, you would actually say that it's higher pressure. But when you look at this pattern over the days leading up to migration, you realize that air pressure is dropping.

And so we weren't actually aware that there were these storms associated with that trigger for migration.

And I think the other really interesting thing is when previously my collaborators were following bats by plane to see their migrations, they had to make an assumption of what direction do they think the bats are going to migrate in. And so they flew just northeast. And if you're only looking in one place, you're only going to find them in that one place. But once we started having this unbiased way of tracking them, we realized that actually the migrations kind of fan out all over Europe.

which is fascinating because there doesn't seem to be any one specific corridor that they're taking. They really go a wide range across Central Europe. So the bats that you follow, these natural bats, they have this kind of complicated structure

migration interlocking with their breeding cycle, at least it seemed complicated to me. Like they have breeding grounds, they have where they raise their young, they have, they hold the sperm like through hibernation. There's a lot going on there. But if you were tracking other bat migrations, would you expect to see similar kind of links with weather or some of these trade-offs that you found with the noctules? We find in nature, a lot of animals try to use these energy saving strategies, but maybe if there's other

constraints on their time, they might have to ignore waiting for the best opportunity for the weather. If you really want to migrate as quickly as possible, then waiting for the best weather might not be the optimal strategy. And so depending on sort of the biology of the system, some species try to migrate as quickly as possible and some species seem to try to migrate with using as little energy as possible. And there's kind of different strategies that species use when they're doing these massive marathon migrations. Yeah, I know you're a bat guy and not a bird guy, but

Do you feel like there are some parallels between what you're finding with bat migrations and what has been found with bird migrations? Just are like, I think much better studied at this point. Yeah, so there is quite a lot already known about bird migration, but also there is surprisingly not that much known about small bird migration. So while there is a lot of

kind of passive monitoring through ringing efforts, where you can see the same type of stuff that we've found from bats as well. We're just beginning to use the same tags on other species to get bird migrations. So we're hoping to do this larger multi-species comparison as well to understand the similarities and differences between bats and a whole host of other species. We're really entering a new fun territory. Very cool. You got to be jealous of those bird people with all the citizen science observers, though, like just constantly pouring data forever.

for everybody. And they're not telling you about bats. I mean, I am very jealous. The more that we've been talking with bird migration researchers, like the more we also learn about bat migration because...

often they will kind of accidentally catch bats by putting up these nets as well. And so we get some of this additional information, phonology of when these migrations are happening. And I think there's just so many similarities between these behaviors that these Northeastern migration that we show in the bats is so similar to other species. And so it's really ripe for comparison. Now with bird migration, sometimes you'll get...

get, at least I will, get these alerts saying it's a big migration night. Turn off your lights. Like this is something that people should be aware of.

Are we going to see something like that for bats in the future? Are they affected by human activity the same way birds are? I would love for there to be a similar alert for bat migration. We definitely need more data like this and more collaborative efforts to try to build these systems. But I think it's definitely possible. And bats are threatened by human infrastructure quite often. So a big concern.

beyond whether or not we should turn off our lights is wind turbines. And so understanding when wind turbines are running and if bats are at risk from them during migration could be very important. So this definitely has some conservation implications. Why are wind turbines such a threat for bats? Is it just because it's just not something they have the kind of the sensory equipment to deal with? Basically, their echolocation is looking straight ahead and the wind turbines coming from the side and very fast.

They can either collide with the wind turbine, or I'm not exactly clear if this theory still holds, but also this barotrauma. So a strong pressure wave around the wind turbine moving can also damage their organs. And so just being anywhere near a wind turbine can be dangerous for a bat. But I think that in some cases they might even be attracted. Some species might be attracted to the wind turbines because they might think it's a tree and might be looking for a roost. And then that adds to this danger.

When we discussed covering your work on the podcast, I was promised some bat facts or some stories about the joys of working with bats. You know, since you were here at Science in 2011, I have followed your career a bit. And I saw at one point you were staying on a tiny island to watch bats. Is that right? Yeah. So my PhD work was kind of this crazy project. I was feeling a little ambitious and I decided to work on a bat that eats fish in the ocean. So this is the Mexican fish eating bat.

It literally flies out over the open ocean and catches fish. And so we knew very little about its foraging behavior. And so that was my PhD research, trying to uncover some of the social foraging aspects of how they find fish in the ocean. It was very incredible, but also quite challenging. Thank you so much, Edward. Yeah, no problem. Edward Hermey is a former science intern and is now a postdoctoral fellow in the Department of Migration at the Max Planck Institute of Anthropology.

of animal behavior, you can find a link to the paper we discussed at science.org slash podcast.

That concludes this edition of the Science Podcast. If you have any comments or suggestions, write to us at [email protected]. To find us on podcasting apps, search for Science Magazine, or you can listen on our website, science.org/podcast. This show was edited by me, Sarah Crespi, and Kevin MacLean. We had production help from Megan Tuck at Podigy.

Our show music is by Jeffrey Cook and Wenkui Wen. On behalf of Science and its publisher, AAAS, thanks for joining us.