I'd now like to remind you of a syndrome we discussed in my first lecture, the Capgras delusion. So, the patient has been in a head injury, say a car accident. He seems quite normal in most respects, neurologically intact, but suddenly starts saying his mother is an impostor. She's some other woman pretending to be my mother. Now why would this happen, especially after a head injury? Now remember, he's quite normal in all other respects.

Well, it turns out in this patient the wire that goes from the visual areas to the emotional core of the brain, the limbic system and the amygdala, that's been cut by the accident. So he looks at the mother and since the visual areas in the brain concerned with recognising faces is not damaged, he says, Hey it looks just like my mother. But then there is no emotion because that wire taking that information to the emotional centres is cut. So he says, If this is my mother how come I don't experience any emotions? This must be some other strange woman. She's an impostor. Well, how do you test this?

It turns out you can measure the gut-level emotional reaction that someone has to a visual stimulus - or any stimulus - by measuring the extent to which they sweat. Believe it or not, all of you here - if I show you something exciting, emotionally important, you start sweating to dissipate the heat that you're going to generate from exercise, from action. And I can measure the sweating by putting two electrodes in your skin, changes in skin resistance - and if skin resistance falls, this is called the Galvanic Skin Response. So every time anyone of you here looks at tables and chairs, there's no Galvanic Skin Response because you don't get emotionally aroused if you look at a table or a chair. If you look at strangers there's no Galvanic Skin Response. But if you look at lions and tigers and - as it turns out - if you look at your mother, you get a huge, big Galvanic Skin Response. And you don't have to be Jewish, either. Anybody here, looking at your mother, you get a huge, big Galvanic Skin Response when you look at your mother.

Well, what happens to the patient? We've tried this on patients. The patient looks at chairs and tables, nothing happens. But then we show him a picture of his mother on the screen, no Galvanic Skin Response. It's flat - supporting our idea that there's been a disconnection between vision and emotion.

Now the Capgras delusion is bizarre enough, but I'll tell you about an even more bizarre disorder. This is called the Cotard's syndrome, in which the patient starts claiming he is dead. I suggested that this is a bit like Capgras except that instead of vision alone being disconnected from the emotional centres in the brain, all the senses, everything, gets disconnected from the emotional centres. So that nothing he looks at in the world makes any sense, has any emotional significance to this person, whether he sees it or touches it or looks at it. Nothing has any emotional impact. And the only way this patient can interpret this complete emotional desolation is to say, Oh, I'm dead, doctor. However bizarre it seems to you, it's the only interpretation that makes sense to him.

Now Capgras and Cotard are both rare syndromes. But there's another disorder, a sort of mini-Cotard's that's much more commonly seen in clinical practice (those of you here who are psychiatrists know this, or psychologists). It's called Derealisation and Depersonalisation. It's seen in acute anxiety, panic attacks, depression and other dissociative states. Suddenly the world seems completely unreal - like a dream. Or you may feel that you are not real - Doctor, I feel like a zombie. Why does this happen? As I said, it's quite common.

I think it involves the same circuits as Capgras and Cotard's. You've all heard of the phrase, playing possum. An opossum when chased by a predator suddenly loses all muscle tone and plays dead. Why? This is because any movement by the possum will encourage the predatory behaviour of the carnivore - and carnivores also avoid dead infected food. So playing dead is very adaptive for the possum.

Following the lead of Martin Roth and Sierra and Berrios, I suggested Derealisation and Depersonalisation and other dissociative states are an example of playing possum in the emotional realm. And I'll explain. It's an evolutionary adaptive mechanism. Remember the story of Livingstone being mauled by a lion.

Dr. Livingston, (picture courtesy of John Murray, Publishers)

He saw his arm being ripped off but felt no pain or even fear. He felt like he was detached from it all, watching it all happen. The same thing happens, by the way, to soldiers in battle or sometimes even to women being raped. During such dire emergencies, the anterior cingular in the brain, part of the frontal lobes, becomes extremely active. This inhibits or temporarily shuts down your amygdala and other limbic emotional centres, so you suppress potentially disabling emotions like anxiety and fear - temporarily. But at the same time, the anterior cingular makes you extremely alert and vigilant so you can take the appropriate action.

Now of course in an emergency this combination of shutting down emotions and being hyper-vigilant at the same time is useful, keeping you out of harm's way. It's best to do nothing than engage in some sort of erratic behaviour. But what if the same mechanism is accidentally triggered by chemical imbalances or brain disease, when there is no emergency. You look at the world, you're intensely alert, hyper-vigilant, but it's completely devoid of emotional meaning because you've shut down your limbic system. And there are only two ways for you to interpret this dilemma. Either you say the world isn't real - and that's called Derealisation. Or you say, I'm not real, I feel empty - and that's called Depersonalisation.

Epileptic seizures originating in this part of the brain can also produce these dreamy states of Deralisation and Depersonalisation. And, intriguingly, we know that during the actual seizure when the patient is experiencing Derealisation, you can obtain a Galvanic Skin Response and there's no response to anything. But once he comes out of the seizure, fine, he's normal. And all of this supports the hypothesis that I'm proposing.

OK, finally let's talk about another disorder, the one that jumps into people's minds when they think of madness - namely schizophrenia. These are patients who have bizarre symptoms. They hallucinate, often hearing voices. They become delusional, thinking they're Napoleon - or George Bush. Or they're convinced the CIA has planted devices in their brain to control their thoughts and actions. Or that aliens are controlling them.

Psycho-pharmacology has revolutionised our ability to treat schizophrenia, but the question remains: why do they behave the way they do? I'd like to speculate on this based on some work we've done on anosognosia (denial of illness) - which you see in right-hemisphere lesions - and some very clever speculations by Chris Frith and Sarah Blakemore. Their idea is that unlike normal people, the schizophrenic can't tell the difference between his own internally-generated images and thoughts versus perceptions that are evoked by real things outside.

If anyone of you here conjures up a mental picture of a clown in front of you, you don't confuse it with reality partly because your brain has access to the internal command you gave. You're expecting to visualise a clown, that's why you see it and you don't hallucinate. But if the mechanism in your brain that does this becomes faulty, then all of a sudden you can't tell the difference between a clown you're imagining and a clown you're actually seeing there. In other words, you hallucinate. You can't tell the difference between fantasy and reality.