Schneier Mulls Psychology of Security

bednarz writes "Cryptography expert Bruce Schneier says security decisions often are much less rational than one would prefer. He spoke at the RSA conference about the battle that goes on in the brain when responding to security issues. Schneier explains 'The primitive portion of the brain, called the amygdala, feels fear and incites a fear-or-flight response, he pointed out. "It's very fast, faster than consciousness. But it can be overridden by higher parts of the brain." The neocortex, which in a mammalian brain is associated with consciousness, is slower but "adaptive and flexible,"'"

Just look to government....

Too bad the Department of Homeland Security doesn't have a neocortex.

Re:Just look to government....

Another way of looking at the amygdala is to consider it a Priority Interrupt Controller. Other parts of the brain evaluate success or impending failure of certain goals, such as survival, and the amygdala chooses the strongest and most important issues and flags them for highest attention. This can be overriden by conscious rationality, but that is slower. I believe the amygdala evolved to handle fast decisions needing urgent attention or the cave bear would eat you by the time you reasoned out how to rationally respond.

Re:Just look to government....

Okay. I'll look to government. I'll even be bipartisan... or antibipartisan :)

We have two parties that have issues with threats to the world, after all. The Republicans have Terrorism, and the Democrats have Global Warming. Both are real and significant threats, but neither of them really gets addressed in the healthiest way possible. There's a lot of focus on OMG-deadly high-profile terrorist attacks, and on OMG-deadly consequences of global warming. Both parties have their people propose some ridiculously broad, sweeping changes to deal with the problem which would negatively impact everyday lives; fortunately, the more ridiculous ones are more likely to fail. And, of course, both parties are willing to throw money at people who claim to have some sort of solution to their problem, whether or not it's actually anything real, meaningful, or worthwhile (like the latest stupid XYZ antiterrorist technology rollout, or the latest bio-fuel legislation/subsidy).

No, they're not the same thing, but one can draw worthwhile parallels, and both parties would benefit by comparing themselves to the other, shaping their actions to avoid these excesses.

Re:Just look to government....

There's a lot of focus on OMG-deadly high-profile terrorist attacks, and on OMG-deadly consequences of global warming.

Terrorism could cause a tiny handful of people to die. Warming could cause a mass extinction. Do you understand what I mean by mass extinction? I mean http://en.wikipedia.org/wiki/Mass_extinction [wikipedia.org].

One of these is a minor annoyance to the human species. The other is the end of life as we know it. Some have even suggested that run-away global warming caused Venus to become the hell-hole it is today. These are very different problems.

You are right that politicians, in general, care more about the appearance of solving problems than actually solving problems. But don't equate global warming with the relatively trivial issue of terrorism.

Oblig. joke

Too bad the Department of Homeland Security doesn't have a neocortex.

That's alright, they have a neoconcortex instead!

Sorry, couldn't help myself. You may now mod this post into oblivion...

Most people cannot define "security".

Bruce has more at his website.
http://www.schneier.com/essay-155.html [schneier.com]

As he says, we really should have two different words for the "feeling of security" and "security".

Amydala feels fear

There is always Anakin to the rescue.

Brain region for thinking about security

Most thinking about security seems to be centered in the nullcortex.

42

Which is why "Don't Panic!" is such good advice.

It makes sense

From the article: There is a "feeling versus reality," Schneier said. "You can feel secure but not be secure. You can be secure but not feel secure. The primitive portion of the brain, called the amygdala, feels fear and incites a fear-or-flight response, he pointed out. "

That is why the real amygdala hides in the background pretending to be a mere attendant like the pitutary gland and communicates with a prominantly placed fake-amygdala using elaborate signals and esp communication. All these scientists have been fooled into studying the fake-amygdala. So they underestimate the real security of the brain. Let someone assassinate the fake-amygdala in a spaceport in Coruscant and suddenly you will see the real amygdala emerge from the shadows and assume the role as the rightfully elected Queen of Naboo.

Re:It makes sense

Security through obscurity never works. ;)

repeat?

Didn't we have an article about this already? Oh wait, that was about the fact that he was going to speak at the con, now we have an article about the talk he just gave?

Good lord, I want that guys press agent!

Tom

Security - 100%

Is there something like "100% Security" for anything? I doubt it.

Schneier says no, but that's not his aim

In his essay [schneier.com] he tells a little joke about aiming for 100% security:

I remember in the weeks after 9/11, a reporter asked me: "How can we prevent this from ever happening again?" "That's easy," I said, "simply ground all the aircraft."

100% security has never been his aim. His aim, AFAICT, is to distinguish real security from BS, so we can evaluate the costs and tradeoffs and then make smart choices.

More on this philosophy:

The truth is that we're not hopelessly bad at making security trade-offs.[...]There are several specific aspects of the security trade-off that can go wrong. For example:

      1. The severity of the risk.
      2. The probability of the risk.
      3. The magnitude of the costs.
      4. How effective the countermeasure is at mitigating the risk.
      5. How well disparate risks and costs can be compared.

The more your perception diverges with reality in any of these five aspects, the more your perceived trade-off won't match the actual trade-off.

Irritating.

It never fails to annoy me when people take snippets of theoretical psychology and redistribute them as truth. Scientists' views of which parts of the brain are responsible for which characteristics of human life change on almost a daily basis, yet phrases such as "language centre" or "mammalian brain" are constantly being used in a way that presents them as definite fact.

It seems unnecessary to incorporate impressive-sounding terms into a speech that, quite honestly, seems to be stating the obvious. Increasing or decreasing security is a response to fear; fear is an emotion and, therefore, decisions that use it as a base will not be purely rational, but will have emotional bias, like every other human decision. You don't need vague descriptions of brain "impulses", and such, to prove that.

A point easily proven

People care more about problems that they can't control than ones they can prevent.

For example: Airplanes. How many people feel more secure behind the wheel of a car than on a long flight with turbulence?

Put your hands down, now the sheer probability of getting into a car accident in one's lifetime (if one drives) is a miniscule number below one. Death statistics are somewhere around 1 in 237 of a car type accident. The odds of an airplane death are like 1 in 5051 source [nsc.org]

However, people are freakishly nervous about planes... So, by induction (the bane of an engineer's existance) we can extrapolate (another fancy bane) that security people will ignore the dangerous mundane and fixate on the extraordinary rarity.

Difficulty Factor

I think that the true dichotomy of the situation that plays out in people's minds is 1) Spend the time and effort to secure this system the right way and stress out all the time 2) Be lazy and just do what will get us by and at least buy us some security even if it's security through obscurity.

instant vs. considered responses

Malcolm Gladwell's book "Blink" [amazon.com] talks a lot about the differences between first impression and actual, thoughtful reaction to a situation, including some interesting studies on what happens when the two conflict and how measurement of the effects of those conflicts on reaction time can tell us a lot about how the brain is processing material. There's controversy around some of his conclusions but I strongly recommend the book and everything else Gladwell has written.

Consciousness?

This may be a little off the point of TFA but I am not sure what the author means when he refers to the part of the brain that is related to "Consciousness". Neither am I clear on what the author means by the term "consciousness" here. Is consciousness, per the author, limited to the brain? What about the emotional, vital parts of a human being and what about that "inner voice" or daemon or feeling many people talk about? Are those not parts of consciousness? Coz then I would like to make them a part of the equation on which people base their fears, also.

No doubt I will be flamed for this

But, I wonder how much attention we would pay to a psychologist speaking on computer science. Is the only qualification to speak on anything non-technical, the ability to pick up a first year text book and leaf through it?

As much as I respect Schenier, I would no sooner trust his assertions on psychology, than I would trust those of Dr. Phil. If he had co-written a couple of articles with someone relevant and had them published in a proper journal things would be different. But after reading the (otherwise impressive) list of his publications it is clear that this is not the case.

So flame me, but Schenier has little authority when he speaks of psychology.

Too complicated

His view is far too complicated. The essence of security is: people think they are secure. They happily type their data into web sites without considering where it goes because in most cases, they have no clue what systems are in operation. Past the words "computer", "database", and "Internet (or Web)" the average person has no concept of how any of it works. Someone, their bank say, sends them a link to a website -- the first problem is, they really have no way to verify it is from their bank, other than going to their local branch and asking, which seems to be beyond anyone's capability. Now, once they've accepted that the link is "legitimate", whether it is or not, they plow ahead and begin banging on the keyboard and typing in their info. Screens come and go, they are admonished occasionally when they don't enter something right, and finally some message pops up thanking them and that's that. Whether the whole transaction was legitimate or not never enters into it.

"Security" is a misnomer -- you are no more secure against possible data theft or manipulation on the Internet than you are physically safe crossing the street in a crosswalk. The only security you can have is in being vigilant in what you do and following up everything you do to make sure it is legitimate. Past that, you're on you own.

Security and panic disorder

As someone who suffers with panic disorder and who is also a network security person by trade (CISSP consultant, unfortunately) I can attest to the irrationality of the "lower" brain. Persons with my condition frequently find it unbearable to do something as simple as stand in line or get in an elevator as even the smallest perceived loss of freedom is enough to send the heart rate soaring. On a particularly bad day I had to excuse myself from a post office line for 15 minutes to pretend to be filling out an address on an envelope as the impending "Next!" and being faced with some "official" was making my hands shake. For the rest of you there is no more mundane experience than mailing a parcel!

All that said, I think Schneider's comments about the amygdala are a bit misplaced. The horrendous waste of security resources in this country -- the 3oz limitation on liquids for example -- do not originate from a panicky, palm sweating reaction but rather a much more calculated, if reactive, decision to make the average person feel like something is being done. If you want to talk about the amygdala and security, talk about one's reaction to a stranger approaching you in the park at night with a "hey buddy, come here a second." Corporate and government security policies are hashed out in nauseatingly arduous sessions with many "expert" consultants who throw out their usual spiel to justify their oversized fee.

Bruce would do better to argue that we need to account for our tendency to implement security schemes which favor the perception of effectiveness rather than true scenario effectiveness. Then again, he is a cryptographer, we can't expect him to be an expert on all things security. Injecting bits of psychology is tempting but runs the risk of being disingenuous. He loses a little credence in my view.

Obvious action item here

The primitive portion of the brain, called the amygdala, feels fear and incites a fear-or-flight response

Hire River Tam [wikipedia.org] as your chief security officer.

Got that right

"The primitive portion of the brain, called the amygdala, feels fear and incites a fear-or-flight response, he pointed out. "It's very fast, faster than consciousness. But it can be overridden by higher parts of the brain."

But rarely is, in ninety-eight percent of the known cases, i.e., humans.

"The neocortex, which in a mammalian brain is associated with consciousness, is slower but "adaptive and flexible,"

Again, rarely - about two percent of the known cases at best.

Chimpanzees simply don't do well with the fear of death. You can blame evolution, but facts are facts.

Overridden by higher parts of the brain

As a programmer I find (in regards to security) that fear is often overridden by laziness