Fire Science Show
Fire Science Show
176 - The Myth of Panic with Daniel Nilsson
You are not supposed to use the word 'panic' in the context of human behaviour in fires, yet this episode contains 196 instances of it. Why? because we try to get to the bottom of the thing! Can panic be both a myth and a reality?
Join us as we challenge the age-old narratives of panic in emergencies with our distinguished guest, Professor Daniel Nilsson, from the University of Canterbury in Christchurch, New Zealand. We unravel the misconceptions surrounding human behaviour during fires, spotlighting the harmful effects of outdated myths perpetuated by media and literature. By debunking these myths, we aim to reshape fire safety engineering, focusing on realistic human reactions and informed decision-making.
Our conversation takes you across cultural landscapes, from the Western world to the Soviet Union, questioning national claims of panic susceptibility and the portrayal of panic as a contagious force. Challenging the historical context, we try to figure out what panic is once you apply the scientific method to understand it.
We also try to assess historical events like the Victoria Tower evacuation and the Love Parade tragedy, we underscore how engineering failures, not irrational behaviour, often lead to crowd disasters. Professor Nilsson and I delve into the psychology of decision-making during evacuations, emphasising the power of clear communication to prevent chaos and save lives.
Finally, we reflect on the profound research of Quarantelli, who redefined panic through thousands of disaster case studies, revealing the complexity of human responses in crises. We explore how cultural contexts influence perceived panic behaviour by differentiating between non-rational and irrational actions. We aim to enlighten fire safety professionals and the broader audience on the importance of replacing the myth of panic with empathy and evidence-based strategies, fostering a safer and more understanding approach to emergency responses.
Further reading?
Sure.
Daniel's paper in SFPE Europe magazine - is panic a myth or reality?
Proloux and Sime paper where they challanged the limiting of information (1991)
And if you really got into this, Jonathan Sime's PhD
Also, a ton of material on Human Behaviour and evacuation is waiting for you in the Uncovered Witness project!
Thank you to the SFPE for recognizing me with the 2025 SFPE Fire Safety Engineering Award! Huge thanks to YOU for being a part of this, and big thanks to the OFR for supporting me over the years.
----
The Fire Science Show is produced by the Fire Science Media in collaboration with OFR Consultants. Thank you to the podcast sponsor for their continuous support towards our mission.
Hello and welcome to the Fire Science Show. Perhaps I'm talking a little bit too much about my travels recently, but indeed the trip to New Zealand was something else and besides participating in the Brilliant Fire NZ conference, I also had the chance to stay for a few days at the University of Canterbury. For me, it was a very important trip, as research from Canterbury inspired me to become a scientist and I've always looked up to these guys. So it was fantastic to meet up with all the colleagues from New Zealand Charlie ,A ndy B Aatif, Ton , pete Thompson and also some guests they had Bronwynn Forrest and Patrick Van Hess. Like so many people I've met and I've enjoyed all those meetings thoroughly, and among those, one who hosted me was Professor Daniel Nilsen. I've spent three days with Daniel at Canterbury and, of course, I took the chance to interview him.
Wojciech Wegrzynski:Daniel is one of the lead human behavior and fire experts and actually also someone from whom I've learned about the craft of engineering systems for human evacuation. First I was in 2015, where he led a course on evacuation with late Rita Fahy in Copenhagen. Anyway, for a long time I was trying to set up an interview with Daniel and we were thinking about appropriate topic and you know, one thing that is very interesting in the space of human evacuation is this phenomenon or myth of panic. That is something that always starts a very emotional discussion in social media when someone uses the term panic. Panic is widely used to describe fires and evacuations in the media, but the question is is it really there? Is it a real thing? Can we blame stuff on panic? Actually, many scientists already know that you cannot, but still there are a lot of myths around the phenomenon of panic. In this episode we go deep on panic and we try to some extent debunk the myths. We try to illustrate where they came from. So we're going through ancient literature from early 20th century to see how it was first described in publications and then see how our understanding of panic evolved. And also what's really really important and that's the fundamental part of the episode we discuss what it means to engineering, what decisions are being taken wrongfully because of the concept of panic, which is incorrect, and what decisions can be taken better if we start to acknowledge the human behavior on a higher level on the way that it should be acknowledged in fire safety engineering. So in this episode we're trying to replace a myth of panic with fire safety engineering Another episode that I had the chance to record live in person.
Wojciech Wegrzynski:The dynamic is a little bit different than when we record online. I kind of enjoy it. Perhaps in future I'll be able to do more like this. But anyway, enough of talking, let's spin the intro and jump into the episode. Welcome to the Firesize Show. My name is Wojciech Wegrzyński and I will be your host.
Wojciech Wegrzynski:This podcast is brought to you in collaboration with OFR Consultants. Ofr is the UK's leading fire risk consultancy. Its globally established team has developed a reputation for preeminent fire engineering expertise, with colleagues working across the world to help protect people, property and environment. Established in the UK in 2016 as a startup business of two highly experienced fire engineering consultants, the business has grown phenomenally in just seven years, with offices across the country in seven locations, from Edinburgh to Bath, and now employing more than a hundred professionals. Colleagues are on a mission to continually explore the challenges that fire creates for clients and society, applying the best research experience and diligence for effective, tailored fire safety solutions. In 2024, ofr will grow its team once more and is always keen to hear from industry professionals who would like to collaborate on fire safety futures. This year. Get in touch at ofrconsultantscom.
Wojciech Wegrzynski:Okay, hello everybody. I'm here today with Professor Daniel Nielsen. Hey, daniel, hi, what a nice change to be in person. For me that's perhaps the third time it happens. Really good, and thanks for the invitation to the University of Canterbury. In this episode we're going to discuss panic, one of the favorite things of all behavioral fire scientists. So often it's referred to as the P word. Why does it annoy a behavioral scientist that much?
Daniel Nilsson:Well, I think there are a lot of myths concerning panic and it's been described in a lot of different ways in the literature. So, looking back, panic is mentioned in like historic literature quite a bit and there are a lot of claims made about panic, claims that some might be true and many of them are actually myths. And these myths, they seem to live for a very long time in our awareness and sometimes we make suboptimal decisions based on myths about panic and we're afraid of something which might not actually happen.
Wojciech Wegrzynski:So you think there's real consequences of this non-factual idea of?
Daniel Nilsson:what panic is, Absolutely. I've seen too many cases where people are afraid of panic vaguely defined as panic something they're afraid of and the consequences that that would result in, and then they try to limit information to people. Okay, and that's not really very good, since if you limit the information to people, then they can't make informed decisions. So hence, evacuation should, in theory, take a long time.
Wojciech Wegrzynski:And is there any medical definition of what panic would be if we go outside of fire?
Daniel Nilsson:There are definitions of panic. So you have definitions of panic. In the stock market you have anxiety and panic attacks. That's quite different from fire and there are definitions of panic.
Wojciech Wegrzynski:Is in the fire situation we're gonna get there yeah, yep, you've prepared some material from historical papers, like where we can go through the how it started, like where the myths came from.
Daniel Nilsson:That's, that's absolutely, and I thought I'd start off by reading some sections out of books just to give you a bit of an idea of how panic has been described, and one of the publications that I often use in my my courses when I teach on the topic of panic is the one by croker, and croker wrote a publication in 1917 1917 okay, that's a long time ago yeah, and croker was a fire chief in new york, I believe, and and wrote a book based on his experience of being a fire chief, and this is just one example of a series of events that happened, according to croker describing what he describes them, or he mentions, as panic.
Daniel Nilsson:so he says that a girl sitting at the nearest machine or table sees it so, so the fire. She leaps up and yells fire, shrieking as hard as she can. In the space of a breath the room is in pandemonium. The girls 60 or 70 of them are all on their feet, jamming each other against the rows of machines, yelling, trampling on each other. In their frantic efforts to get to the exit they ordinarily use an elevator or a staircase. In the flimsy material the fire is spreading fast, but it is still the merest infant fire, making a good deal of smoke and a little more heat, but still capable of quick extinction, even with a chemical extinguisher. On the floors above and below that in which the fire is, panic also holds sway. The garment workers, the feather workers, the shirt waste workers, remembering other horrors, stampede without any cause, like sheep determined to be slaughtered.
Daniel Nilsson:Croker then continues and says perhaps you think this is an exaggerate or a sensational account of a hypothetical case. It is not Just. This very thing happened in March 1911 at the Triangle Waste Fire in Washington Place. This is one of the more famous fires in the history of fire engineering. Now, if you look at this description, there are several things that are mentioned to describe what panic is. There seems to be an indication of an overreaction. Yeah, yelling running. Exactly. So there is a fire. It's a small fire, but people are overreacting Exactly. He also claims that it's not particularly logical behavior. They don't really take in all the information and sort of go for the best option. They don't consider their options.
Wojciech Wegrzynski:I'm quite surprised that he mentions people on other flows are also reacting, also reacting in the same way, and sort of.
Daniel Nilsson:There is an element With a stampede, stampede, and that's the other thing is that in many of these old publications, and Stampede Stampede, and that's the other thing is that in many of these old publications and you can see it in new publications as well the authors use analogies and sort of compare it to animal behavior. So a stampede, we lose rationality, we behave like animals, which is very graphical. But the question is if that's actually a good explanation.
Wojciech Wegrzynski:If you put it into the context of the times when it happened. You don't have modern fire alarms, you don't have big trust in firefighters' response, you certainly do not have modern escape pathways, and people are living in times where they get news from newspapers and they read about those terrific fires in which hundreds of people perished every now and then. We're still in the age of great fire of XYZ city, right, yeah? So I think a responsive person in 1917, hearing about a fire being in the circumstances of overcrowded room, witnessing it firsthand, with no real good escape routes, that's pretty different from a responsive 21st century user of an airport.
Daniel Nilsson:Well, possibly there is a difference in the type of behavior, but still, that people would respond or overreact in this way, even when the fire is very small, it's still to me very unlikely. The other thing in Croker's description is that there's almost this link between panic and the casualties. So panic causes casualties, Stampede, Stampede. People get injured. And the final question I ask myself when I read this was was Croker actually there? Did Croker witness this? If he's a fire chief in New York, was he in the factory when it happened? Or is this looking at the consequences of the fire and drawing your own conclusions?
Daniel Nilsson:And maybe providing a poetic description or a book right, exactly, but this is not the only account. There are other accounts as well, so Phillips in 1951, so a bit later said that panic has been the cause of more loss of life than burning by fire or suffocation by smoke. Okay, quite a powerful statement. To say that more people die from panic than actually from smoke inhalation, which we know today is, is not true people die from smoke, yeah, but but then again, okay, again the context 1951.
Wojciech Wegrzynski:Yeah, and I've interviewed david purser. He told me that this brought uh interest to him because firefighters told him in 1940s people would not die from smoke, they would go out of the room, take five breaths and they're good. Yeah, and suddenly you start having people dying. So this is before the plastics era.
Daniel Nilsson:It is, but it's still with things burning. So so, just carbon monoxide. So it is sort of in a lot of these publications, panic is considered dangerous and hence something that needs to be prevented.
Wojciech Wegrzynski:Well, here it says that it has been the source of more loss of life than the fires themselves.
Daniel Nilsson:So then, you need to consider it in design and, as we mentioned before, some of these publications claim that panic can happen even if there's no real cause, like a real small fire or something that people might just believe there is a fire and can actually cause panic. So Phillips continues and says panic may be aroused when there is not the least danger from fire and in an undisciplined rush to escape many may suffer serious injury or death. So you have another claim here that people will overreact. It doesn't have to be a danger at all and people will run towards exit and are crushed in exits since there are too many people running at the same time.
Wojciech Wegrzynski:So in this record it's more like the panic is the real danger, not the fire itself, exactly, so perhaps better to not notify, yeah, and sort of looking at what triggers panic.
Daniel Nilsson:Philips does give some explanations, yeah, like incidents that may lead to panic include the sight of smoke or flames, the smell of burning, the sound of escaping gas or any unexpected noise or happening, whether the danger is real or imaginary.
Daniel Nilsson:So we have a bunch of publications that says anything can cause panic, and if panic happens we get really bad outcomes. It is interesting, though, that in some of these publications you look at the characteristics of people and who are more prone to panic than others. Phillips again says that the possibility of panic arising will depend to a large extent upon the composition of the assembly of people. Panic is most infectious this is something you find in a lot of the publications that it's like a disease it infects from one person to another, and the loss of control by a few weaker-minded persons may rapidly affect the whole. A crowd consisting of children and young persons may be more prone to panic than an assembly of older persons of a reasonable degree of intelligence. Okay, quite powerful wording. It's from the 1950s, but basically saying that some people are more prone to panic than others, what kind of scientist was?
Daniel Nilsson:Phillips, was he a behavioral scientist, fire scientist, I guess fire person Okay, and this is actually true for many of these publications that it is often someone who might be multidisciplinary, have an opinion about panic.
Wojciech Wegrzynski:So from this record of Philips. It's not just that. It's caused easily by a slight view of smoke, a slight view of fire or any other trigger. It can happen spontaneously in a group of people without you having any control of that. It's infectious, it spreads.
Daniel Nilsson:Like a disease. Yeah, so if one person panics, then everyone will panic. Yeah, yeah, they'll have a mass disaster.
Wojciech Wegrzynski:It causes a loss of control by a few weaker-minded people and rapidly affects the whole. Okay, so indeed, like this description from those early, early materials that they do reassemble the image of panic that media would, would employ. So it's like kind of uh, I would say cultural description of what panic could be right, which we know that it's not very factual when you compare that with science.
Daniel Nilsson:Yes, there are definitely some elements and some myths that you need to consider. There have been sort of as you saw from Phillips. They claim that some individuals are more prone to panic than others. There have been claims that that is actually a national difference, so some nationalities might panic more than others. So bird and docking, for example, claim that, yes, you would have different tendencies to panic in different countries. Bird and docking also have a funny quote about British crowds. They actually claim that British crowds panic less than other crowds in the world and they're British. So they say that other cases could be quoted for the good behavior of British crowds. It might be added that foreigners have been known to say that the British rarely panic because they are fundamentally unimaginative.
Wojciech Wegrzynski:Since they don't have any imagination, then the British don't panic and they behave quite well in emergencies. I'm also not so sure about good behavior of British crowds.
Daniel Nilsson:Yeah, so there are a lot of those claims.
Daniel Nilsson:But, an interesting thing though, we talked about publications from different parts of the world and we basically looked at the Western world during the Cold War and pre-Cold War, but they actually had the same types of claims in the Soviet Union. They actually had the same types of claims in the Soviet Union. So the USSR, where Reutemann in 1975 wrote that panic spreads rapidly over the whole or a large part of the public and induces people to flee from the danger zone in any way they can. This leads to the use of physical force by everyone, or at least many of the individuals trapped in the premises. So he even adds physical force that you're fighting to get out, and says that this could happen when one person gets panic and then it spreads, similar to disease. So even in a country where they didn't really have a lot of communication, I would suspect with the UK or the US, where Croker was from they still have this concept of panic in the Soviet Union. And, as I said, cold War I don't expect too much, but perhaps you know.
Wojciech Wegrzynski:I wonder to what extent those observed because, as you mentioned, there's multiple people, multiple records stating very similar things. I would guess it must have been to some extent observed behavior. Perhaps people to some extent behaved like that and it's just qualification of the action that has been misleading. Perhaps people were in a very emotional response to fire and not necessarily the state of panic.
Daniel Nilsson:One potential explanation of it, which we will discuss as well, is that you have difficulty estimating how other people behave. You know how you behave, but you have difficulty understanding how other people behave, and if you can't understand someone's behavior, they must be panicked.
Wojciech Wegrzynski:And since you don't understand why they behaved in a certain way, Do we have access to any records like voice records or video records of events like that? Or interviews with people, but not like personal statements, more like objective video material.
Daniel Nilsson:Absolutely. We've done experiments where we tried to give people different type of information to see how they behave. Yeah, to try and test out some of these myths, like can you tell them there's a fire, for example, can you not? And then we've observed their behavior. Reutemann continues to give some examples of cases where you've had panic. Arguably Probably Reutemann wasn't actually there, since he gave examples in the US and other places. Argues that it can cause crowd crush, same as we heard from the other.
Daniel Nilsson:Yeah, stampede. He also says that panic may also arise when there is no real danger to life.
Wojciech Wegrzynski:So you see the same type of trend Again.
Daniel Nilsson:It's re-emphasis of the exactly Can be caused by the smallest, faintest smoke exactly someone just saying fire and everyone panics, but also claims which many of the other publications claim that it can be prevented, so we can reduce the risk of panic. Now, a funny addition this is publication from the soviet union. They had political editors. So it does say in this publication that modern theaters and movie houses, clubs and cultural centers and other public buildings constructed in the USSR provide comfortable conditions and guarantee the safety of the public in such premises and halls by providing suitable passages, doors and staircases. Besides, the administrative and operational staff of such buildings are trained to forestall panic and effectively extinguish any fire in the building. Okay, so it wasn't a problem in the Soviet Union, it was just outside of the Soviet Union.
Wojciech Wegrzynski:It was a problem, but I expect that to be the political editor who went in and added but it's interesting that they've already recognized there would be operational things like the staff being able to forestall the panic, even though it's so easily triggered. Exactly, if it's so easily triggered and so overwhelming, how can a single person of a staff stop it right?
Daniel Nilsson:yeah, no, I agree, it's sort of. It's something that can be caused by anything. So the the logical question here is how can you prevent something which can happen without any cause? Yeah, what can you do?
Wojciech Wegrzynski:if it's just prevented exactly.
Daniel Nilsson:So it's sort of a counter-argument to the argument of panic happening, since how can you prevent something that can happen, irrespective of so? It could be a small thing or nothing at all, so you couldn't really prevent that, but they still claim that you can prevent it.
Wojciech Wegrzynski:So is this something that leads to those dangerous things like limiting the information, like?
Daniel Nilsson:this attempt to prevent Absolutely so in many of these publications. Going back to Phillips again, phillips writes that in some buildings a warning system of lights or other means which would not attract the attention of the public is used to give early warnings of an emergency to staff who are then in a position to be of material assistance in maintaining an orderly evacuation of premises. So basically what Phillips is saying that since there is a risk of panic, then we limit the information to the people. Okay, tell the staff we're trained, and then the staff can deal with the people. But then again, if people panic for no reason at all, the staff trying to evacuate people could trigger panic right.
Wojciech Wegrzynski:Yeah.
Daniel Nilsson:So how does that help really? But there is that tendency or that recommendation in many publications to actually limit the information that is given to people, and this is, in my view, this is actually quite bad. Now, the publication that really was relied on a lot was this that you see here he's holding a book and it looks old.
Wojciech Wegrzynski:There's Manual of Safety Requirements in Theaters and Other Places of Public Entertainment that you see. Here he's holding a book, yeah, and it looks old as manual of safety requirements in theaters and other places of public entertainment, issued by home office london's his majesty stationary office 1935. Yeah, and he stole it from swedish library exactly.
Daniel Nilsson:So this is.
Daniel Nilsson:There's a funny story which um around us and so I, when I did my phd.
Daniel Nilsson:I really needed to quote this and I really needed to see it, and it wasn't really available anywhere in sweden, so I had to get it on loan from the uk and it was the most quirky and strange loan I've ever had. I got it on a one day loan from the national library, or I think it was was either Oxford or Cambridge where they sent it, and they even called our librarian to tell them to sit beside me when I read it, since it was a national treasure and I could never lose it. So I did that and I read it. And then the next week I went down into our library and there was actually a book that had been donated to us from the Swedish Fire Protection Agency. They changed their library and they had donated a lot of books, and it was donated. It was in our basement at Lund University and I was clicking through the book and there it was. So we had a national treasure in our that no one knew about.
Daniel Nilsson:But it was really good, since now I have the source material. So this is what was relied on and it is. I mean, it is official, it's His Majesty's stationery office, it has an official stamp on it and what it says in this book is that it is recommended that any telephone for the use for the purpose of communicating with the fire brigade shall be so situated that members of the public cannot overhear a call and be alarmed. This might easily cause panic. So an official document saying you can't tell people since then they will panic, means that engineers would look at this and say, well, I need to limit the information to people.
Wojciech Wegrzynski:So, as an important comment to the listener, you're hearing recalls from a very, very old past of what has not even been 5-7 engineering then and which to some extent have transponed themselves to the modern times. However, today we know that early communication with the building occupants is critical. The response is fundamental to limit the time it takes for the pre-evacuation of people and essentially this is the fundamental thing that allows to do 5-step engineering, Absolutely. So it's kind of interesting to see that this idea of panic or these descriptions you've shown, they go like 180 degrees, like opposite direction of what we would consider good 5-step engineering today.
Daniel Nilsson:Yeah, they do. But it's not just old publications. Like in publications that are written today, you still have claims about panic and panic happens, and we're doing this to prevent panic. So, even though many of the publications I've read are from early 1950s or 1934, you still have this myth of panic, this thing that people are afraid of, that we're trying to prevent. Now, the first time this was actually systematically criticized was by Jonathan Syme in the 1980s. So Jonathan Syme wrote a publication called the Concept of Panic, where he went through the description of panic and how it was used in the field and started criticizing it, since we use panic without actually knowing what panic meant.
Daniel Nilsson:That was the main criticism in this publication. So someone might say well, this was a fire here, A lot of people died, there was panic. But in those accident investigations the investigator didn't really understand that people's perspective. They just saw the outcome and they said this must have been panic. They didn't try and interview people to figure out how did you make decisions in that situation, and did you make a panic decision or not?
Wojciech Wegrzynski:So that was one of the criticism. It's quite interesting because let's brace them for a second you were tasked with a fire where you had, let's say, 50 people died in a narrow passage Yep. So how can you tell a panic from just the crowd crash incident? How can you tell a panic from people who just lost their way and were unable to evacuate, or they were maybe taking care of others in that room?
Daniel Nilsson:And I would even ask the question of was the outcome of the fire even related to the panic? Since many times when you have crowd crash situations, then it might actually be something else.
Daniel Nilsson:It might be a too narrow passage yeah, and people trying to funnel situation exactly a funnel and you get an increase of the crowd pressure, and this was one of sime's points. Like, panic is often used as a scapegoat. You have a fire, a lot of people die. Well, you need an explanation for why a lot of people die. They're're dead, so they panicked. It's their fault, they died. So you're using a skateboard goat. You're basically blaming the people for the outcome of the fire by saying that you panic, and one of the problems of this is it takes away the focus from the real reason. Okay, since you look at it and you say 50 people died in this opening, they must have panicked. Case closed.
Daniel Nilsson:We don't have to look at changing our design to do wider exits not a funnel, for example, since we've found the reason, and that was a reason which, if we go back to the old publication, which is impossible to stop anyway.
Wojciech Wegrzynski:It was just a freak accident, though, if you have trained personnel in the USSR, then yes, maybe so, maybe so. So why did Syme start? Did you have any chance to ever talk with him about that?
Daniel Nilsson:I haven't had a chance to talk to Jonathan Syme, so Jonathan Syme passed away before I got into this area. But I think the reason was one of the reasons was the fact that panic was used as a reason for limiting information about the fire to people, which is bad. And another thing that Syme highlighted in his publication was also the role of the media. So panic in the headline sells paper Okay, without actually a definition. And you sort of have panic and the reader reads that and you go oh panic, oh yeah, interesting. Today we would say clickbait. Yeah, pretty much Everyone who reads that paper probably has a different definition of what panic is, but everyone is interested to read the paper. So that was another reason. And media he saw often used panic without actually describing what Sime would say is a panic situation. It was just people who are stressed, trying to get out quickly, but saying panic sells papers. So media had an important role in this. So let's go to science definition maybe. Well, it's.
Daniel Nilsson:The thing with jonathan syme is that jonathan syme didn't actually offer a definition, just said that the way we use it today is not suitable okay, so just observation just observation and he criticized some of the things like irrational behavior is often used like they don't act in a rational way, whereas Simon would argue that, no, given their perspective, they probably did what they thought was right. But irrational means that people act. They have all the information and they do the opposite. And he didn't really see that. You don't really see that in fire situations you can see people having a lot of stress put on them and making simplified decision-making, but not necessarily having the perfect information and saying, well, that's the exit I should take, but I'm going to take that one which is worse. Okay, so people don't really do that in a fire situation. So hence irrational might not be the right word to use in a definition of panic. So there are a couple of myths related to panic that are important to know about.
Daniel Nilsson:As an engineer, one of the things we've talked about is limit the information for panic. Now this has been dismissed. You should tell people about the fire, you should tell them what is happening, and I guess the first publication to look at that was by Canterbury and Sime. They did interviews with survivors from fires and homes and hospitals and they asked people about recognition, their location, when they realized there was a fire, tried to get them to tell the interviewer about their behavior and their perception, and they developed something which is called the behavior sequences. And the behavior sequence is just a series of behaviors that are typical for an evacuation situation, and the main thing of the behavior sequences is that initial part of the behavior sequences, where people receive information.
Daniel Nilsson:They either ignore the cue or they investigate and search for more information until they're convinced they should evacuate, since you either ignore, which means it takes longer, or you get that first cue and then you go. I need to confirm and I need to process and I need to make sure that this is actually a fire. That always takes time and that's why we have pre-movement time, or at least part of the pre-movement time is that recognition time. So people never hear one cue and they go. I'm going to evacuate and take zero seconds Always takes a significant time for them to process it, to make a decision.
Wojciech Wegrzynski:Always takes a significant time for them to process it, to make a decision. So in the previous thought process it was thought that you will see the fire and you may panic, or if you're sufficiently British or intelligent, you will not Instantly. Instantly right and now we understand there's a series of chain of events.
Daniel Nilsson:And with this behavior sequence sort of framework, the more information you can give people to a limit, I mean you can't overload them with information but if you give them clear information, they should, in theory, get out of this loop quicker Okay, since they make an informed decision quicker. So, for example, if you smell smoke, there is no alarm. You need to explore what's happening. If you have a voice alarm that tells you there is a fire, it tells you which floor it's on and it tells you to evacuate, you might not have to look as long for additional information.
Wojciech Wegrzynski:So it's quicker. That is an exercise that I like to give to people like a thought experiment. Imagine you're walking through a corridor of your office and you hear a fire alarm. You would most likely be confused about it in a way that you would not know. Is it a real fire alarm? Is it a drill? Was something planned today? But if you end up in the same situation, walking through that corridor, and you smell burned plastic from a cable and you hear the alarm, that must be a fire somewhere, right?
Daniel Nilsson:Absolutely, and I've seen this from my own experience when we evacuated an office building and we just used their alarm, which was a ringing bell, and I was a bit confused, since everyone said that they checked the clouds or the cloud, and I didn't really understand what they meant. But they actually checked the intranet. That was the first thing. So they heard an alarm and then they were wanting to check is this a drill or is this a real fire? So they checked the internet and then, when they didn't say anything about a drill, then okay, it's probably real. So that's part of that behavior sequences and that's a delay. So the more information you can give them, the better, the quicker. And Jonathan Syme so Prue and Syme actually did an experiment, a really interesting experiment, where they evacuated a metro station and they had the opportunity to try different types of alarms. So they had one alarm which was a ringing bell, which is not great if you don't train people what it means. They had staff running around telling people to evacuate. They had a public announcement system saying evacuate, but no real reason. They had staff plus this PA system which told people to evacuate, and then they had a PA system which added information about what has happened plus what people should do, okay, and the outcome was Bell. Well, they had to terminate the experiment before everyone was out, since no one, there were still people on the platform.
Daniel Nilsson:When you had staff running around, it was a matter of how quick the staff was and how quickly they were running, but it took about eight minutes. When you had a public announcement system where you didn't say what has happened, just that they should evacuate, it actually took 11 minutes. It's a long time. And when you had staff plus the same PA system, then it was more efficient than the staff. So staff only took eight minutes. You add a PA system, it takes seven minutes.
Daniel Nilsson:But the thing was that when they had a PA system which said what had happened and what you should do, then it was by far the shortest time six minutes to evacuate the station without any staff interaction, since you told them what had happened and they don't have to look for additional information. And then you told them what had happened and they don't have to look for additional information. And then you told them what to do and they did that. Okay. So telling people what is happening and so telling people what they should do is good, since you reduce that time it takes for people to respond initially.
Daniel Nilsson:And I've done similar experiments and others have done as well where we've done equivalent experiments with voice alarms and, and in one voice alarm we say that there is a fire and in another one we say that we don't mention fire, and in none of these cases we ever get stressed behavior or panic. People are quite calm. They behave in a very similar way, and this is just one example. I did it at Lund University, but others have done it in other places and don't typically see any panic or stress reaction in those situations where the stimuli are quite mild.
Wojciech Wegrzynski:Have you ever seen intense emotional response in an experiment?
Daniel Nilsson:The only time I've seen an intense emotional response was when we've done experiments in a smoke-filled tunnel with acetic acid.
Daniel Nilsson:So that was irritating smoke. It was slightly irritating smoke. The thing, though, in that particular case where the person was stressed, it was probably more linked to claustrophobia, sort of being in an enclosed environment, rather than it being fire smoke. I've seen stress as well, when we did an evacuation in Victoria Tower, where you see a heightened level of stress but you don't have people running or rushing to exit eating. No, no, stampeding. Those experiments was with one person at a time. They just walk maybe a bit faster, but I haven't seen any any running really. So that's a rational response. Right, it is. You should move quicker when there is a fire.
Daniel Nilsson:Now, another myth is that crowd crush is the same as panic.
Daniel Nilsson:Okay, and the problem with that is you see a crowd crush and then you say, well, that's equivalent to panic, but there have been so many cases of crowd crush that are unrelated to panic.
Daniel Nilsson:People might be extremely stressed as they're being crushed, but that's different from the stress and their behavior leading to the crowd crush. In many cases, what causes a crowd crush is narrow openings, really high densities, counterflow. If you have a funnel where a lot of people are moving and everyone's pushing a bit, but then towards the edge or the end of the funnel, you get an increase of pressure. So an example would be the love parade, where you had opposing flows. You had a, basically a tunnel that they had to go through, and the flow of people going in was simply bigger than the tunnel capacity, and hence you got the crown crusher. That does not mean that people were panicking to get out through that tunnel, but it just meant the flow led to a situation where you got a crash. So in that case you wouldn't be able to say well, we had a crash, hence panic.
Wojciech Wegrzynski:The important distinction is that panic is something non-preventable in a way, while crowd crash is preventable by engineering.
Daniel Nilsson:Crowd crash is preventable. I guess we have to figure out first what panic actually is. That's an interesting question. So does panic exist Now? The panic the way it's been described in historical literature probably doesn't exist, but that doesn't mean that panic doesn't exist. It's just that we need a clear definition of panic. But it's not the same as Crowdcrush. It's not the same as Crowdcrush Exactly.
Daniel Nilsson:It doesn't end with Stampede. No, exactly, and I would argue it rarely ends with Stampede. But we'll come back to the definition of panic, which is also the reason why I think it very rarely leads to a Stampede. Okay, let's go further.
Daniel Nilsson:Another myth was that panic spreads as a disease.
Daniel Nilsson:Yeah, people from top floor reacted the same way, exactly, and that has been dismissed in many fires. So one example that I remember is the World Trade Center bombings. One of the things that was discovered there was that as people were evacuating and they were evacuating down staircases and they were filled with smoke and lighting had gone out a lot of people expressed frustration, and you can say that that's sort of expressing their anxiety the seed of panic, according to the old definition, rather than that spreading through the crowd as wildfire or a disease. You actually had people calming each other, so saying to others like you'll be fine, we'll do this together, we'll walk down together, and it was a very orderly evacuation and we've seen this quite a bit that people might express this frustration and stress, but then you have people around it, so, rather than being reinforced, many times it's actually dampened and people feel security by being there with others who try and calm them down. Now, this does not mean that we don't consider the behavior of others. So we have the theory of social influence.
Daniel Nilsson:So you look at others and you understand their behavior, or you look at their behavior and you understand the situation. So we have the normative social influence. That's about the fear of being different. It's like an inhibitor. And then we have informational social influence. So if you see people who are stressed, you probably take the situation more seriously. That's a big step from taking a situation more seriously to stampeding example losing control. So yes, we do have some social influence influencing how we there's a good behavior, but it's not Absolutely.
Daniel Nilsson:And if there is a fire alarm and someone is running past your office, you probably think it's a more serious fire than if someone is walking past your office. So, yes, there is that element, but in my view, probably not to the point of causing you to lose your thinking altogether and just run out, which would be the traditional definition of panic. So an interesting question. You've raised this a couple of times what is actually panic and does it?
Wojciech Wegrzynski:exist.
Daniel Nilsson:What's the definition of panic? Now, the one that I often go to when I teach my students about panic is one by Quarantelli from 1954. Now, quarantelli defined panic in a slightly different way. What was it? Quarantelli was a researcher who looked at disasters, so how people behave in disasters.
Daniel Nilsson:So not just fire, Not just fire any type of disaster, and so the definition is not just fire related, it's just sort of in disasters, not just fire, any type of disaster, and so the definition is not just fire-related, it's just sort of in disasters Now, and Quarantelli derived the definition rather than set. So usually as a researcher, you might define it and then you apply it. Quarantelli did it differently. So he looked at a thousand or several thousand cases where people's behavior in disasters were described, and whenever he discovered something which he, based on his gut feeling, would say I would say this is panic, he saved it. And then he read all of these publications and it was only a handful, it wasn't that many publications and tried, based on what he saw, to define panic.
Daniel Nilsson:And the definition that he eventually wrote down was that panic is an acute fear reaction, marked by the loss of self-control, which is followed by non-social and non-rational flight behavior. And there are several things in this definition that are worth mentioning. Acute Okay, very, very. So acute would be the first one Very, very short-lived. So what Quarantelli saw was seconds of heightened anxiety and stress.
Daniel Nilsson:People might temporarily lose their self-control, so not a whole moment running around we're talking five seconds, maybe two to five seconds, maybe a bit more in extreme situations. He also said that there was always an emotion, a fear reaction. Okay, so there is an emotional component. That seems to make sense. He also said that there's usually some type of non-social behavior. Now, quarantelli was very careful in the word choice here. Non-social is not antisocial. Antisocial would be fighting to get out, so not aggressive, but just not adhering to the social norms.
Daniel Nilsson:Exactly, not considering the fact that you're there with others rather than hurting others to get ahead, and that's quite different. And you might remember some of the myths like people were fighting to get out.
Daniel Nilsson:That's the definition of panic. According to Quarantelli, that doesn't really happen, but you might not consider the others. You just walk, but you're not fighting to get out first, since that's competitive behavior. But this is non-social. He also said non-rational. Rather than irrational. Irrational would be you have perfect information and then you make the wrong decision. Okay, and people don't do that. But you can do non-rational, which means almost like a reaction. You don't take in all the information. You see something in front of you and go for it.
Wojciech Wegrzynski:You don't make a Like pick a spurter exit because you think this one is.
Daniel Nilsson:Or since that's the only one you perceive, okay, since you don't perceive the one that is closer, okay, and that's a non-rational. It's like tunnel vision, so that could definitely happen, according to Corintelli. Quarantelli also highlighted that it's flight behavior. So running away rather than freezing or apathy, he said it is some type of flight behavior. Okay, now, what Quarantelli used to illustrate this is two examples that I remember. One is a woman who was doing the dishes in the kitchen and there is an explosion. I think it was a tire blowout, okay, but before this happened there had been some explosions in the area and the woman was in the kitchen with her child, but when this tire blew out, she ran out into the lawns. Is that what you're supposed to do if there's a gas explosion? And forgot the fact that she had a child? Now, that is non-social, since she forgot she had a child in the crib, but she didn't hit the kid going out, since that would be anti-social. Yeah, so it's just non-social.
Daniel Nilsson:You forgot the fact that you were there with someone else, and the example that corintelli uses for the non-rational was a man from memory that was sort of doing some gardening I think it was with a wheelbarrow and then there was a plane that came down and he was afraid that that plane would crash on him and basically what the man what did was, without considering anything, he just ran. Now, when a plane is coming towards you, you have no idea if it's going to hit you or not, or it's going to miss you, so there's no point running really, you can just stand still. But in that particular situation running seems sensible since you just want to get away. So that would be non-rational, not irrational. It's not running towards the crash site of the plane, so that would be the non-rational. So the bit of tunnel vision or narrowing of the cognitive math happens. So that's the definition by Crantelli. Now the question is given this definition, does panic happen?
Wojciech Wegrzynski:It's interesting because if you consider it at a level of emotional response of a single human being within the crowd, I think, like this it could happen, like people would have different emotional responses, right, you can imagine someone traumatized by a fire in the past who has to relieve a fire even again, you know, and they might have, let's say, a panic attack, whatever, that probably bad warning, but their response in here would be justified, you know, because of their experiences. You might have someone who's just really like scared, or who's in some sort of environment they don't know. Like, imagine if a fire alarm happened to you while I don't know, being in Japan, you don't understand what's happening. You know the social norms are different, everything's different. I mean, I would be stressed as hell, right, so I can imagine a response. But then would it spread to the Japanese people surrounding me? Highly unlikely.
Daniel Nilsson:Probably not, and probably what would happen in that particular situation is you have an earthquake in Japan or something like that that you're unused to, or something like that that you're unused to, and then for a period of five seconds, you're really stressed and you don't know what to do. But then, after five seconds, you calm down, you take your jacket, you go out and you stand outside the building and you seek help. So for that small time period, yes, maybe you do have panic, but from a fire engineering or an earthquake engineering perspective in this case, does that matter? That's a pre-movement time engineering perspective.
Wojciech Wegrzynski:In this case does that matter? That's a pre-movement time?
Daniel Nilsson:Yeah, exactly.
Wojciech Wegrzynski:And I would definitely. Okay, it's hard to say the indefinitives, but I would rather not start a fight with the Japanese surrounding me or start a stampede, right. So I can resonate with the emotional response of humans. That sounds very human, right, but this record of stampede, people running around, people not knowing what to do, shouting and stuff that sounds very Hollywood-ish, you know.
Daniel Nilsson:Yes.
Wojciech Wegrzynski:It's kind of like you know a car falling from the cliff and always exploding, right? Yeah, yes, a car can fall from a cliff and be damaged. The explosion is a Hollywood thing, yeah, and here as well. Like the emotions could be there.
Daniel Nilsson:It could, but there are cases. So, for example, the station nightclub has been used as an example of where you could potentially have had something which is similar to Corintel's definition.
Daniel Nilsson:This is actually research done by Aguirre, who I think was one of Corintel's students, and his and Aguirreirre students, where they looked at the behavior of people evacuation the station, evacuating the station nightclub and the station nightclub, as you know, rapidly developing fire, really bad conditions and people were highly stressed and we have video feedback from that right exactly, and what, from memory, they do mention, they do see in that, is that some people were there with someone else and you might be holding hands as you evacuate, but at some point, with smoke coming over your head, you actually let go and then you evacuate on your own and for five seconds you've let go and then you're looking for your partner. That could actually be the definition of panic, since in that situation you have non-social I'm just going to focus on getting out. Yeah, and also in that particular situation, letting go for a second or a fraction of a second could mean you lose someone in the crowd, right?
Daniel Nilsson:yeah, so it can have an impact on you personally, but in the station nightclub, looking at the design, it was the narrow openings, the lack of windows and the quick fire, the quick fire growth. Those were the real factors that influenced a lot of the losses, rather than people panicking If we said that, they would panic and that's why they died.
Wojciech Wegrzynski:again, going back to the scapegoat, we're blaming the people for dying, which we know that that wasn't the reason. It was an environment that was unescapable?
Daniel Nilsson:Yeah, absolutely. But an interesting question is going back to the literature. We have these claims about panic from the UK, from the US, from Russia, and it seems to be something we're afraid of. So the question is why do we have this myth of panic and this preconceived idea of what panic is? And one of the reasons I I think is that we can't really understand why other people make decisions. We understand why we make decisions, but we don't understand why other people make decisions. Now, in order to test that, I did experiments during my PhD, which I actually never published, but I did experiments where I went to IKEA and I had people listen to a different type of voice messages and as part of that, I also asked them like, if you would have heard this voice message at IKEA, just imagine yourself and you heard this message, would you panic?
Daniel Nilsson:And I didn't give them a definition of panic, I just asked them would you panic? And what do you think people would reply if they no, no, I wouldn't panic, exactly right. But then I asked them okay, that person, random person there what's the likelihood of them panicking? They don't seem that stable. I don't know how they're thinking, so they probably panic, and that was the trend. I never panic, they panic. I know how I behave. I don't know how they behave, hence they panic, since they're acting totally irrational from my perspective, since I don't understand how they make decisions. So this is, I believe, one of the reasons why the myth of panic is there, since you don't trust other people, you don't know how they make decisions.
Daniel Nilsson:A really good example of this is one that Syme gives. So Syme, and this is based on a real fire that happened where elderly people were trapped by a fire. So these were maybe 60 plus or thereabouts. So elderly people who met in a venue and then there was a fire and the corridor filled up with smoke and they couldn't evacuate and their behavior, as described by someone else, was that the elderly residents. They heard the alarm and they go into a state of panic. They are very loud, they were very frightened, they were yelling and at one point they were throwing something out the window. So the person who watched them said that that's total panic.
Wojciech Wegrzynski:That's the observer's description.
Daniel Nilsson:Okay, that's the observer's description of them. It's total panic. They threw something out the window. And then Syme then explains that, well, when you talk to the people who were trapped, they were trapped. They were elderly, some with respiratory issues, so they weren't going to go out into the smoke and they were afraid of losing their lives. They were high up in a building and they tried to sort of argue among each other how should we evacuate, and one person came up with the idea well, we can jump up, but it's quite far and my hip isn't great, so let's throw a chair out the window to see what happens to the chair.
Daniel Nilsson:And they did, and the chair broke and they decided we're not going to jump out, since that's going to happen to my hip If the chair breaks. My hip is great. So from their perspective, that was a rational decision to Extremely rational. It's a scientific experiment to see if you can jump out the window. But from an external perspective, that's panic, since it's totally irrational. I can't understand what they're doing. They're screaming, they're throwing out chairs. I can't understand it. Hence, it must be this concept of panic that they're displaying, but in this case, it was highly rational. It might have been in a stress situation, but they made an informed decision based on an experiment they had done.
Wojciech Wegrzynski:So this is why I think that we have this myth of panic, since we struggle to understand how other people make decisions. But let's now focus on real consequences for real fire engineers. Okay, on the one hand, you can just consider this an annoying nuisance of the press media, like click-baiting titles. On the other hand, it causes real damage. So if it's victim blaming, it makes it difficult to figure out the real reason for damage. What would be the consequences for engineers and how engineers should approach the concept of panic?
Daniel Nilsson:Well, I think, first of all, if panic is discussed at a building meeting, you probably have to make sure everyone is on the same page. And then it's about thinking this concept of all if panic is discussed at a building meeting, you probably have to make sure everyone is on the same page. And then it's about thinking this concept of panic. It's not the panic as such, it's actually what you want to do to prevent something which might not be a danger. So sometimes you might have a fire engineer who says, no, we can't tell people what's happening, since you will have mass panic and you will have a crowd crush. Then you have to pick that apart and say, actually that's really bad, since that's going to make the situation worse. But sometimes someone says, well, we might get panic, so let's make the exits wider. That's actually good. But I'll still have a discussion at the meeting of what is panic. What does the research say? Is it something we should be afraid of? And rather than focusing on panic, let's just make the exits wider and tell people what's happening.
Daniel Nilsson:But it is so ingrained, like what do we call the bars to open the doors? Panic bars, panic bars. Since people panic, they can't be trusted. So if they don't know how to operate a door, they just run into it and it will open. That's a panic bar, but it's based on the myth of people not making rational decisions and running into a door without using the handbook and then it opens automatically.
Daniel Nilsson:Now, on the grand scheme of things, panic bars are maybe neutral. I don't think they make the situation worse. They make it slightly better, maybe, but they're also quite tricky to understand how to use if you're not used to it. So, probably neutral, but some things we do are actually bad, and some things we do to prevent this myth of panic or that panic from happening is bad. So, rather than making the situation unclear by using panic, a concept that we don't have the same definition of, maybe we should just look at what does the research say in terms of what should we tell people? Should we make exits wider? Should we avoid funnels? Should we make exits wider? Should we avoid funnels? Should we make s exits easily open?
Wojciech Wegrzynski:yeah, so so you just recommend good fire engineering. That that's. That's it? Yep. And how how do you feel about us stepping into the role of debunking the myth of panic when we see that in media? Because every now and then you see that happening in the media. Yep, yeah, on linkedin it's like every other week. Someone describes some even as a panic and it's being jumped by a bunch of fire scientists. Do you think it's our role? Is it necessarily important? Not, necessarily.
Daniel Nilsson:I think in the research community we need to have a common understanding of panic and particularly we can't, in my view, have researchers saying that they're doing this to prevent panic when they don't have a clear definition of what they're trying to prevent. So we definitely need a clear definition in the research community to be able to speak the same language and not just throw panic around as a word that just sells scientific papers. What media does and what people understand might be difficult to influence, and media will still be using panic, and I've seen examples of panic being used in the heading and then it's not used in the article at all. But that's probably since the heading might be set by someone else than the one who actually wrote the paper or the article. But I think the main thing is in the research community and also in the fire engineering practice, that we're not using panic without actually knowing what we mean and we're not using panic to introduce things that are bad for the design.
Wojciech Wegrzynski:I think that the only time where I would be really inclined to step in would be when the panic is brought up as the reason for some consequences, absolutely where it is kind of obvious that the true reason may be somewhere else.
Daniel Nilsson:Yes, like if you would have a situation like a crowd crash situation and I would say, oh, they've panicked and they died, I think I would be inclined to step in absolutely if there is an accident investigation, as we mentioned before, using panic as a scapegoat and basically blaming the people without actually exploring what was wrong in the design, that's also a bad use of panic, since we are basically blaming the people for their behavior. And if we go back to the old publications, we're blaming people for behavior that was not preventable, since they've been caused by anything. Case solved, we can move on. But no, we have to learn from disasters. So panic is not particularly useful in accident investigation unless we interview people and we try and understand. You break it apart Exactly and then we're probably fine. Same as Quarantelli, I was initially very stressed and didn't know what to do, but then I calmed down and I made an informed decision based on the information I had.
Wojciech Wegrzynski:I think looking at it from that perspective is useful and helpful. I hope after this interview, a lot of people understand much better why people online would be so pissed about someone using the word panic Absolutely. Also for us fire engineers, it's important that we understand this so we take better engineering decisions in our everyday engineering. Very good Thanks, daniel, good job. Thank you Good job on ripping the myth of panic apart. It was a pleasure to do an interview in person for once. Yes, it was great to meet in person, and that's it.
Wojciech Wegrzynski:So far in the episode the word panic has been used 182 times and that's probably the most I've ever used the word In my life. I feel liberated to be able to say panic, panic, panic, panic Nah, just kidding. It's important that we understand, or we have a common understanding what panic is and it's just a word. You know it's us who give meanings to words. Words is just a sound and the word panic on its own doesn't mean anything. It's us who decide what meaning we put behind the word. We need to understand the difference between short, emotional, irrational response to a situation from stampeding and spreading like a wildfire type of mind virus or whatever that makes people do stuff that are general. Panic is not crowd crush. Panic is not a main driver of evacuation. Panic could be there. It could lead to some troubles for individuals. It could lead to losing your loved ones when you try to evacuate together. It could lead to taking perhaps a wrong decision or not the most optimal decision, in a split second, but it is not a driver of evacuation at large, and this is important and if you view it from this lens, you will be able to really engineer your buildings in a better way. And one thing that's really, really important panic can be used to blame victims, and this is, in my opinion, the worst thing. If someone perished in the fire and then we try to figure out how to prevent deaths of further individuals in similar situations and someone blames it on panic, it's kind of the victim's fault and it's not the fault of those who engineered the systems that failed the person. And from that we learn nothing. From that we lose all the learnings from a tragedy and we should not do that. We owe it to the society. We need to learn, we need to strive to do better, we need to learn on our mistakes and we need to design engineering systems that provide safety to everyone around. And, yeah, if we blame it on phenomenon that does not even exist, probably, or at least not in a way that media describe it. Don't fall for Hollywood panic. The cars when they fall from the cliffs, they don't immediately explode and crowds do not start running and stampeding at the first sight of fire. Let's split the Hollywood reality or the movie reality from the reality.
Wojciech Wegrzynski:Anyway, I hope you've enjoyed my interview with Daniel Nielsen. I also had the privilege to spend multiple hours interviewing Daniel for my other project, uncover Witness Fire Science Revelations. It's also a podcast. It has eight episodes on means of escape and human behavior in fires. We'll also talk about the myth of panic a bit in that podcast. It's more like educational series, more like fire fundamentals. I show in Fire Science Show and I'm sure if you enjoy Fire Science Show you would enjoy that one. There's so much more material on human behavior in fires that you will not find in the 5th Sense Show. So if you are interested, if you enjoyed me talking with Daniel, if you enjoy how Daniel presents knowledge, then that podcast would be a good thing for you and for today's. That would be it and I'm back to my reality. Fixed, my jet lag. I'm catching up with the work and getting ready for next episode that's gonna air next Wednesday, of course, thank you.