AskTog: Interaction Design Solutions for the Real World
Interaction Design Section   Living Section   About Bruce Tognazzini

AskTog, October 2004

Panic! How it Works and What To Do About It

• The twin-engine private pilot, upon suffering a partial engine malfunction, instantly shuts down one engine and feathers its propeller (turning the blades to minimize wind resistance), just as he was trained. He then struggles valiantly to reach the nearby runway, even as the plane makes a slow spiral turn into the ground. After the crash, the National Transportation Safety Board (NTSB) determines he shut down and feathered the propeller of the engine that was still functioning perfectly.

• An inexperienced scuba diver attempts to draw breath and realizes he is not getting any oxygen. He bolts for the surface, his mouth closed tightly to hold the little precious air he still has. As he rises, the now-expanding air destroys his lungs and, by the time he reaches the surface, he is, for all practical purposes, dead. His tank is found half-full and his equipment shown to have been working perfectly.

• A single-engine pilot, flying solo, is distracted by a terrifying banging sound. She investigates the noise, while forgetting the most important rule: "First, fly the plane." Her aircraft makes a smooth downward glide directly into a backyard barbecue party. The NTSB later discovers that the passenger door of the aircraft had been closed with the metal end of the unused passenger seat belt on the outside of the plane. The wind was whipping the buckle benignly against the sturdy metal side of the plane.

• A driver, upon realizing his car has begun to roll toward a group of people, slams on his accelerator, instead of brake, mowing down the people. He realizes something is dreadfully wrong with the car but is certain in his mind his foot is on the brake. He presses ever-harder.

• A computer-software test subject spends the first seven minutes of the test, rather than carrying out the task she’s been asked to perform, slowly massaging the top of her mouse with her index finger. Later, she reports that her mouse at home has a ball in the top.

• A US President, informed that his nation is under attack, spends an equal number of minutes staring at his hands, listening to a Kindergarden teacher tell a story. Later, he reports he was afraid the children would be frightened if he left the room suddenly.

• A woman rises quietly from the table during a formal dinner, disappearing into the bathroom, where she silently dies. The autopsy discovers an easily-dislodged piece of food in her windpipe. She has died of embarrassment.

All of the people in these real events were perfectly normal. Their actions were neither “crazy,” nor “stupid,” nor “cowardly,” nor any of the other labels people who have never experienced this phenomenon love to heap upon those who do. Their actions were the simple and predictable result of panic, a perfectly normal human reaction.

Panic is not only misunderstood, we too often fail completely to take it into consideration. Reuters (5 Aug 2004: “Deer in Headlights Pose Dilemma for Motorists”) reported this sage advice from Ann Dellinger, a CDC epidemiologist, on what to do when you discover a deer in the path of your speeding car:

“Drivers who came across animals on the roads should consider a number of factors, including road and weather conditions, before deciding whether to take evasive action.”

Excellent advice if you’re going three miles per hour at the time. But it’s not only the speed that gets in the way of making such a carefully-weighed decision, it’s that the driver is as likely to be “frozen in the headlights” as the panicked deer. Even if the driver did have more than a couple of seconds to go through all the steps necessary to carry out Ms. Dellinger’s detailed analysis, his brain may very well have “locked up.”


Join my intensive (and fun!) lecture/ workshop course. Sign up now!

Interaction Design course: Go from zero to interaction designer in just three days.

You may be coming in cold from engineering or graphic design. You may already be an interaction designer wanting to "fills in the blanks," establishing a more solid theoretical and practical base. I've aimed this course at all of you, covering the needs of both individual contributors and managers.

Join me as I teach the Apple method and show you how to not only organize for and come up with successful designs, but sell them to engineering and upper management.

It's intensive, yes: A one-semester-equivalent with a strong, real-world bias. However, we have a lot of fun along the way, and you'll leave having worked with a team to design and build a complete project, so you will have not only learned, but experienced everything taught.

User Experience Conference Website There's more than my course at an NN/g conference. You'll find a breadth of other specialized courses and networking opportunities that will put you and your company at the leading edge of the design curve.

The Nature of Panic

The Ancient Greeks blamed the woodland sprite, Pan, for panic. He would follow people through the forest, causing frightening rustling noises in the bushes until the travelers would be running blindly in fear, resulting in cuts, scrapes, and contusions. This continued until the invention of another mischievous sprite, called “Attorney,” who would jump out of the bushes and sue Pan for all those cuts, scrapes, and contusions, putting the poor sprite out of business.

Today, we know panic as “ a discrete period of intense fear or discomfort that is accompanied by at least 4 of 13 somatic or cognitive symptoms... often accompanied by a sense of imminent danger or impending doom and an urge to escape...or desire to flee from wherever the attack is occurring.”

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th Edition [DSM-IV]. Washington, DC, American Psychiatric Association, 1994, p. 394-403

The symptoms the DSM-IV list are:

  1. palpitations, pounding heart, or accelerated heart rate
  2. sweating
  3. trembling or shaking
  4. sensations of shortness of breath or smothering
  5. feeling of choking
  6. chest pain or discomfort
  7. nausea or abdominal distress
  8. feeling dizzy, unsteady, lightheaded, or faint
  9. derealization (feelings of unreality) or depersonalization (being detached from oneself)
  10. fear of losing control or going crazy
  11. fear of dying
  12. paresthesias (numbness or tingling sensations.
  13. chills or hot flushes

The natural role of the anxiety leading to panic appears to be to guide us toward immediate evasive action, be it flight or fight. Panic ensues when we are unable to formulate an effective evasive action, we choose the wrong evasive action, the evasive action is ineffective, or the evasive action goes terribly wrong in ways we do not understand.

Panic is not necessarily noisy. The woman who died in the bathroom alone, rather than "raising a fuss," died rather than facing the embarrassment of letting people know of her predicament.

Technology vs. Nature

A common element runs through all the cases mentioned at the head of this article—technology. We were “designed” by evolution to work effectively within our natural environment.

When we create technologies that do not conform to the rules of the natural world, we increase the odds of panic. Thus, the trackball user, faced with the embarrassment of attempting to use a trackball with no ball (e. g., a mouse) while someone witnesses her incompetence, panics.

When we create technologies that expand the natural world, such as scuba gear, people experience new phenomena, such as the illusion of lack of air brought on at depth by the increasing density of the air they are breathing.

When we create technologies that are extremely complex and do not provide comprehensive feedback for each and every possible error, such as a seat belt left unbuckled, people have a tendency to drive their aircraft into garden parties.

When we create technologies where similar actions produce dissimilar results, such as placing a brake and accelerator pedal side-by-side, to be actuated in the identical manner by the identical limb, people will periodically die.

(In defence of said placement, it does ensure that people don't press the accellerator and brake at the same time, which would also produce undesirable results, particularly if the driver, realizing the situation, quickly removed his foot from the brake.)

Counterforces to Panic

We have at least three ways to reduce or eliminate the effects of panic.

1. Competence Training

Trainers attempt to build competence in the individual, so that panic never ensues. The scuba student who is told of the illusion brought about by high air density, then taken to depth so he can experience it, will be less likely to panic upon re-experiencing the same feeling.

Most of scuba training centers on potential panic situations, most notably loss of air. The theory is that by practicing the procedures over and over again under safe, controlled circumstances, that, should a true emergency arise, students will be prepared to address them without the heightened physiological complications of panic.

Flying students go through similarly endless rounds of “engine out,” “engine on fire,” and other sorts of emergency procedures, ensuring that, later in life, should a true emergency arise, they will know exactly what to do.

This training becomes almost fanatical in the case of helicopter pilots, who have approximately three seconds to carry out a single, critical action in case of any hint of engine failure (lowering the “collective” control to cause the blades to autorotate). Failing to accomplish this task results in the helicopter taking on the aerodynamics of a set of car keys.

2. Product Design

It is possible to design products that reduce the incidence of and resulting effects from panic. Had a seat belt-unfastened warning light been present on the Beachcraft Bonanza that crashed the garden party, the disaster wouldn’t have taken place. The pilot, when scanning her instruments, would have seen the tell-tale light and not been leaning over, feeling around the floor of the aircraft to pinpoint the source of noise when the plane hit the house.

Instead, the FAA relies on pilot training to ensure that on each and every flight, pilots never, ever fail to fasten every seat belt on the plane, in both occupied and unoccupied seats. This very training heightens the chance of panic, since we pilots “know,” when trouble arises, the seat belts are already accounted for. After all, other pilots might forget something like that, but not us.

(I've never spoken to a twin-engine pilot who didn't explain to me that the damning statistics for twin-engine fatalities did not apply to him.)

In the case of the mouse, I would not start labelling all mice as “not a trackball device” (warning labels must always have at least one meaningless extra word such as "device"), but we have plenty of other opportunities to reduce user panic.

Most such panic arises in the computer world during initial learning, a particularly critical time for us between purchase and the time the check clears. Other panic arises when things go wrong.

Initial Learning Anxiety

Many users who hit one “bump” during their exploration of a website or application will panic and back off. As anyone who has worked in tech support will attest, they immediately lose their ability to read, resulting in RTFM (read-the-manual) syndrome, reducing them to a quivering mass barely able to dial the 800 number.

Most websites don’t suffer from this tech support problem. Why? Because users just simply go somewhere else. The penalty for these bumps in the road is not only instantaneous and permanent, but invisible. They typically don’t bother to let you know why they are leaving, and even if they do write a complaint in that cute little postage-stamp-sized box on your customer-support form, your people are usually offering the now-long-gone user a work-around, instead of reporting the problem to you

Steps to Reduce Initial Learning Anxiety

Initial learning anxiety may be costing your company a lot of money. Here’s how to attack the problem

A. Identify the sources

How do you find elements of a computer software design that are causing anxiety and panic? The usual way—user testing.

User testing is actually ideal for discovering fearful elements of a website or traditional application because your test-subject users are already frightened. They typically find themselves in unfamiliar surroundings and, regardless of all your gentle prefacing, believe they are being watched by these perfect strangers to see exactly how stupid they are. What could be more panic-provoking?

When explaining to test subjects how to think aloud, ensure that you also cover “feeling aloud,” so that you are getting a stream of shifting emotions, not just a critical analysis.

Even this is not enough. Most people are loath to report negative emotions other than anger. Many people, particularly women, are adept at sensing other's emotions. Have people with this demonstrated skill watch users during the study, making particular note of the user's play of emotions.

Also bear in mind that fears are naturally heightened during tests. Reducing test subject’s fears to zero may be over-design. Your actual users may start out in a more comfortable state.

B. Offer a gentle initial learning curve

After you have discovered the bumps, try to eliminate them. Where you can’t eliminate them, smooth them out. Offer better prompts that users can understand. Break big steps into a series of little ones.

C. Ensure early success experiences

Make sure that the first page or pages are dead-easy to use. If people are successful early-on, it will give people the confidence to continue.

Product Design for When Things Go Wrong

Most of our panics on computers are in slow motion, and don’t usually result in death. They can, however, result in destruction. When I started out in computers, disk drives were really, really big—not in capacity, but in physical size. You would plop a disk platter on the drive to use it, replacing it with another platter when done.

Some folks said a “head crash” was when the read/write head would land on the platter, causing a furrow that would render the platter unusable. Not true. A real head crash next required that you, quick, got ahold of the backup platter and placed it on the same drive mechanism and had the now-damaged head trace a furrow in the backup drive. The crash was complete when you then placed the “grandfather” disk on the same defective drive, ensuring a complete loss of data.

Today, things are not much different. People get a message stating, “There are a lot of interesting things on this disk, but I seem to have misplaced my glasses, so I can't read it right now. Would you like me to utterly destroy all your work for the last year?”

Oh, I’m sorry. That’s not actually what it says. It says, “Disk damaged. Initialize?”

Too bad people don’t have a clue what “initialize” means. In a single click, they greatly reduce or eliminate the possibility of recovering their information through a wide variety of available software and hardware options. Anyone writing messages like the above needs to be initialized.

3. Panic Desensitization

A third approach to reducing or eliminating panic is through desensitization. You don’t hear a lot about it, because we don’t do much panic desensitization. Instead, we do nice, safe simulations that we think work as a substitute. They don’t.

Subjective vs. Objective Competence

One reason we’ve been lulled into believing that simple simulations can substitute for real panic training is that the trainees involved truly believe they have been trained.

A Study of Panic in Recreational Scuba Divers by David F. Colvard, MD and Lynn Y. Colvard, PhD (The Undersea Journal, First Quarter 2003) showed that fewer than 15% of new divers bolted for the surface when they first experienced panic. 81% were able to fall back on their training.

This sounds like good news: 81% were able to fall back on their training! All those hours in the swimming pool worked!

The problem is that this first occasion for panic was typically not life-threatening in itself. It was generated by fear of the unfamiliar. If we look at the most life-threatening situation of all, the outcome is quite different.

(The researchers discovered an interesting gender split: two thirds of the women were able to recognize an offer of help, whereas only a little over a third of the men were able to do so.)

Buddy Breathing

We all spend a lot of time in scuba class learning to “buddy-breath.” When you run out of air, you swim to your partner, slash your finger under your throat, then use puppy-dog eyes until your buddy takes the breather out of his mouth and hands it to you. You take a single breath and hand it back, and so forth, all the while you both use the other hand to keep hold of each other, while swimming languidly toward the surface.

That’s the way it works in the pool, and that’s the way it works in 15 feet of clear water. The way it works 100 feet down is that one or both of you usually end up dead. Why? Panic. If you’ve got a decent instructor, he’ll also teach you how to get the heck away from your airless buddy after he freaks out, before he has a chance to kill you, too.

The solution to all this is realistic simulation. Some modern flight simulators come very close. An immersive simulation can cause a pilot who ultimately crashes in such a simulator to walk away exhibiting all the psychological and physiological symptoms of a pilot in a real craft, less the trauma of the explosion and fire, of course.

Such simulators are not available to private pilots, of course, resulting in the grim, but incontrovertable statistic that twin-engine aircraft suffer four times the fatality rate of single-engine aircraft. It’s not that the single-engine pilot doesn’t panic, it’s just that he doesn’t have a lot of decision-making to do. He has a clear realization that we’ve ended the flying portion of our trip, and it is time to put up the tray tables and adjust our seats to the full forward position for landing.

The twin-engine pilot is not so lucky. He has all kinds of things he has to do immediately and flawlessly, like identifying a defective engine and feathering the prop. He has decisions to make, like should I try to limp back to the airport or land in that field near that guy in the single-engine plane? He has done these maneuvers and made these decisions over-and-over in his training, but never in a state of panic, and panic is what he is feeling right now. It would behoove the government to give twin-engine pilots access to realistic simulators. It would keep these pilots from dropping on citizens’ heads.

Automobile drivers also should be required to complete emergency-maneuver training, such as skid pad training. They should also be trained in “trick cars” that allow the instructor to switch the brake pedal into an accelerator pedal, so they can experience the panic of believing they are braking when they are not, along with appropriate steps to take should unexpected acceleration happen “in real life”. (This training could be done on a neighborhood street with the kids at play, but, personally, I would recommend it instead be done on a nice, wide runway, empty because all the pilots are doing simulator training.)

Training Soldiers

The military doesn’t mess around when it comes to panic-training the infantry.

They start out by making the troops terrified, not of the enemy, but of their own commanders. The devil they do know becomes 10 times more threatening than the devil they don’t know. They then demonize fear, by calling it cowardice. They show film after film of dead guys, lots and lots of dead guys, to desensitize the troops. After that, they start with the bullets.

While in basic training, the soldiers are required to crawl across a field on their bellies like snakes, while real, honest-to-god bullets whiz over their heads. If they lift their heads too high, they are immediately rewarded with their death benefits.

Once they reach the battle theater, they go through a formal process called, “blooding,” wherein they are sent into their first battle. Armed with the knowledge there is no way back, since their commanders will ruin them if they succumb to “flight,” they fight. Only after they are “blooded,” are they considered useful battle troops.

Training the President

The ideal president has been “blooded” in battle during his youth. Such a man would be unlikely to spend half the time it takes a nuclear missile to reach and destroy a US city staring at his fingers.

It’s become increasingly rare for such a man to be elected President, however, particularly since we unavoidably faced peace for a couple of decades.

Final Thoughts

Panic is a far more universal state than people realize. It may be noisy and obvious; it may be quiet as a church mouse; it may only be detected by the sudden absence of its victims.

We need to expect it, design for it, allow for it, and test for it. We need to understand how it affects test subjects and users alike.

As for the case of the woman who slipped off to the bathroom to die: This was not a rare instance. While most choking men and boys tend to break into an instant game of Charades, consisting of a lot of pointing to their throats, women and girls in our society have a distressing tendency to wander off alone and die, out of politeness. If you see someone begin to slip away from the table without a word, get them to say something. If they don't comply, give them a hug, a great, big, Heimlich hug. You could save a life.

For a deeper understanding of panic, read the story of my own slow-motion panic that continued to unravel for more than 18 hours in my companion article, Anatomy of a Panic: A Case Study

The case-study panic was a serious one, and illustrates how, if you once set a user's panic off, they can continue to wander down wrong paths for hours afterwards, turning an initially-unpleasant situation into one that, in my case, bordered on fatal.

After you've read my case study, please go on to read Marcin Komorek's own case study involving cell phones and sunrises. It illustrates the subtle level of panic we tend to induce in the users of our electronic devices. If people such as myself and Marcin, who are intimately involved in technology, can be induced to panic, imagine every-day users.

No, wait! Don't imagine them! Test! Set up usability studies where the test subjects face significant time constraints to complete tasks and set them to work trying to use your products/websites. You'll see by their body language, even before they start hitting all the wrong buttons, that they've entered panic. Find out why, and fix it!

A Reader Response


Panic is of great interest to me. I believe there are a huge number of injuries or deaths that would not happen if not for panic. I know a man, a amateur scuba diver who used to be very active in the recreational scuba community, who co-wrote the first really thorough study of fatalities in recreational diving. He's an attorney, and his coauthor is a physician and another expert diver. So they know their stuff.

My acquaintance, Ralph Singer, said they studied one hundred fatal accidents, and are positive that ninety-five of them could have been prevented if the victim had not panicked. Me, I'm a martial arts student and an amateur student of assaults, riots, disasters and the like. And it's amazing how much of staying in one piece in such situations, from a sucker puncher or a violent burglar, to a panicked crowd, a riot, or an extended disaster like a huge earthquake or war, depends on keeping your head.

Tog, to me, all this falls under the broader category of preparedness, and THAT falls under the even broader category of what might be called mind set—facing reality and taking safety seriously, rather than the more usual human head-in-the-sand approach. Being a "chicken" pilot—which attitude I heartily endorse, by the way—means you take the potential danger seriously, such that, for instance, if the whole point of a weekend trip is to take a ride in someone's plane, and you can't approve of the safety arrangements, you would happily stay on the ground even while some friends made fun of you.

Having the right mind-set, to me, includes it all. So you would never ride in the plane that killed John Denver. That's part of it. But another part of it is something you touched on in your John Denver article: Being trained and/or experienced enough when you do a particular flight that you are well within your safety envelope. My point is that it's not just the cautiousness to not go past a well-understood safety envelope that shows this wise mind-set. Taking the time to educate and train yourself is another sign; and in the final analysis is a result of the same mind-set.

Hang-glider pilots have a very high instance of serious, often fatal, accidents. One acquaintance who works with the wheelchair basketball league told me that around ninety-five percent of the athletes in the league are former hang-glider pilots! But this is not at all due to "inherent" danger of hang gliding. Like the Long E-Z, there are inherent elements of danger in hang gliding. But with very few exceptions, those injuries happen because the pilots push the limits of their skill and/or their aircraft past the point where they are really in control.

I was at a glider meet with my father up in Mojave, it must have been thirty years ago. A bunch of glider pilots—my father was too old to compete—were in a race. Interestingly there was not a single accident among the gliders. Some miles away, there was a big meet with ultralight aircraft, gyrocopters and the like, and we heard sirens of ambulances several times a day. Too many guys who bought a toy and didn't want to spend the extra year gaining the necessary skill and experience before they tried something wild.

(There was a near-miss when an Air Force jet got too close to a glider. The Air Force pilot vehemently denied that he got anywhere near any of the gliders, which all the Air Force hot dogs had been warned about. However when they inspected his plane they found the wing tip wheel from the glider imbedded in the leading edge of his jet fighter. Yipes!)

The accident rate among sailplane pilots is very very low, and nearly all of the fatalities happen when someone knows he's pushing the limits.

Back to scuba diving. Ralph Singer, my instructor/expert friend, says that panic can be almost completely prevented by enough training, which too few people bother with. He himself recommends an entire year of snorkeling, with plenty of "free" dives, so you develop HABITS of functioning while holding your breath, and then lots and lots and lots of training on scuba gear in safe conditions. Then you will automatically be able to act intelligently when suddenly you are short of breath and feel like you are drowning.

Regarding self defence, an outstanding book called, The Gift of Fear, by Gavin De Becker, says that most victims of violent assaults, including murders, have chances to act very effectively to escape, especially before the situation goes too far, but they don't listen to their fear in time. Or they listen too much—they live in terror, or they panic.

He says that if people put the same energy into dealing with danger that they do into denial, most victims would in fact not be victims. Of course some preparation is important; taking time to inform yourself of elementary facts and precautions, and perhaps learning some self defence techniques. But the most important weapon of self defence is the brain.

Of course martial arts training can save you by enabling you to fight successfully. But, Brian, most martial artists are not saved by fighting. Ninety-nine percent of the time when a martial artist is saved by their training it's by recognizing a dangerous situation and fleeing from it shamelessly. Being willing to be "chicken." We don't hear the statistics because it's mostly the actual assaults we hear about. But the way trained martial artists AVOID TROUBLE is pretty remarkable. I'm sorry the public is not more aware of it. Sigh.

Panic. Yeah. In things like the aftermath of an earthquake, or getting lost in the woods, etc.; in more extended emergencies, of course panic is one of the big killers. I believe the best way we have to forestall panic is to prepare ahead of time. Perhaps one way to look at panic is that it's what the organism does when it feels it has no options left and blind lashing out is better than nothing. I believe a vital element in panic is previous denial, when the person either has told themselves it wouldn't happen at all or it wouldn't happen to them, or they admitted intellectually while denying it emotionally.

It's infamous that there are many smokers who say that they know they have a risk of cancer but they don't care because everyone is going to die today someday—and when they are diagnosed with lung cancer they, yes, they panic. Their superficial acceptance was in fact part of a denial system.

-Jay Maupin

Tog's Response

Jay's father was the designer of the world's first carbon fiber composite glider, the Carbon Dragon, a particularly elegant design that is still held in high regard today.

Read also my companion Article: Anatomy of a Panic: A Case Study, which follows a real-life panic victim through 18 hours of slowly-evolving disaster.

Reader Antonomasia submitted a link to another article of interest, The Art Of Failure: Why Some People Choke And Others Panic, which differentiates between "choking"—thinking too much—and panic—thinking too little. It's worth reading.

Another Reader Response

Good day.

I have just finished reading "Panic! How it Works and What To Do About It", and I am left with the impression that you are a SCUBA diver (or if you're not, you seem know about it).

You talk about good training includes the experience of panic-generating situations.

I cannot agree more with you. I'll give you a personal example, about "raptures of the deep" since I am a SCUBA diver.

I managed to experience a "chamber dive", where a bunch of guys were put in basically a big pressure cooker, then pressurized down to 100 feet for some 15 minutes. Hilarity ensues, because we were all "high" with the raptures...

I was glad I experienced it, because some weeks later, at one of the hardest spots to dive in the world (vertical wall, random tidal currents, four degree water), I was narced right before I was starting to lose all my air some 120 feet down (frozen regulator).

If I had not experienced the chamber dive, I would not have known I was narced... But I managed to calculate that I could safely reach the surface in the time all my air would run out, which I did, slowly and safely.

I am not saying that the chamber dive experience saved me there, but it definitely helped, by making me recognize the symptoms of being narced and thus adjusting my thinking "speed" accordingly...

-Marc Dufour, Montréal

Tog's Response

Actually, I'm both a SCUBA diver and a pilot. I'm licensed to travel from 200 feet below sea level to 20,000 feet above (clearly demonstrating laxity in our licensing systems).

The kind of exercise Marc discussed is available to not only SCUBA divers, but pilots, who can suffer similar cognitive impairment at high altitude in the absence of auxiliary oxygen. The problem is that few people know it, and even fewer take advantage of it.

Have a comment about this article? Send a message to Tog.

Previous AskTog Columns >

Don't miss the next action-packed column!
Receive a brief notice when new columns are posted by sending a blank email to

return to top

Contact Us:  Bruce Tognazzini
Copyright Bruce Tognazzini.  All Rights Reserved