> When you mentioned Freud's theory being "there was a mental structure that not only made up plots etc for the other mental structure to 'watch'" I assumed that the mental structures were physically separate, not intertwined.

there isn't any scientific support for the following mental structures: the id, the ego, the superego, the dynamic unconscious. they are vivid metaphors but there are no such structures. they aren't neurological structures because there simply aren't any neurological structures that have those properties / play the relevant role. things get tricker if you consider them to be cognitive structures rather than neurological. the theory was that they were indeed supposed to interact. conflict between the id and the ego led to neuroticism (for example).

> With fMRI, straight females have a very particular area of the brain that "lights up" when she smells a man. Lesbian brains don't do that. Nor do men's.

one needs to be a bit careful with neurological findings. what they typically do is something along the lines of this: they take a group of people (ie 'straight female' or 'lesbian female' or whatever). the number of people in the group varies (sometimes it is fourteen, sometimes more, sometimes less). then they neuroimage their brains. then what they do is superimpose each brain on each other brain until they get the 'average straight female brain' or whatever. then they assign colours to numbers / frequencies and you get the pretty picture of the 'average straight female brain' or whatever. now, the first problem is that there is CONSIDERABLE variation between peoples brains. just generally speaking, i'm not talking about pathology or anything yet. i'm just saying that if you take a bunch of healthy people then you find CONSIDERABLE variation between their brains. so not a single person in the 'straight female brain' group actually does have a brain that looks like the picture of the 'average straight female brain'. as such... one can't necessarily posit 'that bit there lighting up' as a causal mechanism WITHIN THE INDIVIDUAL.

> As far as lie detecting goes, when people recall an event, many areas light up, some having to do with memory. When asked to make up a story, there were a couple other areas of the brain being used, in addition to some of the others used to recall a story. The researchers claimed lie detection was a piece of cake. Or, so says the reporter at livescience.com.

what happens when people believe they are remembering but when the researchers know they are confabulating or vice versa? i haven't heard of neuroimaging lie detecting before... once again i'd be very cautious about this because of the implications for neuroimaging being used in a court of law when people are giving testemony (with respect to allegations of sexual abuse, for example). one would want to be jolly sure that this was reliable before one relied on it. i personally would be really very surprised if this was a litmus test. researchers do often draw hyperbolic implications from their studies in order to secure research grants but i do think it pays to be cautious.

> Sociopaths have a problem where a certain portion of the brain does not work, but it does work in the rest of us.

well... we know they have a problem because they don't exhibit prosocial behaviour like the rest of us seem to do. of course there must differences in inner mechanisms that cause the behavioural differences.

> This part was used for empathy.

sociopaths lack empathy definitionally.

i wonder whether people with autism had similar neuroimaging results? people with autism are often thought to lack empathy though they don't engage in anti-social behaviour the way sociopaths do... what would it mean if they had similar brains? what would it mean if they had 'normal' brains in that respect?