I have consistently, over the course of my life, heard people describe sociopathy and related mental illnesses as being caused by a lack of empathy. This, intuitively, seems wrong, since that seems like a massively important brain function, that really ought to have a major and extremely visible effect on your thinking. Now, obviously it does have a serious impact (amoral behavior, etc), but it seems rather unlikely to me that someone like this really shouldn't be able to mask themselves as normal. (I'm also not sure why lack of empathy would make you want to dissect squirrels, but that seems like a side issue). 

The upshot is that I'm seriously confused about what these mental disorders are, and how they work. Do these individuals have the ability to empathize but not sympathize? I'm not sure how that would work, but I'm not at all an expert on cognitive science. Is the standard explanation for these disorders just wrong? Are these people genuinely figuring out what humans care about by looking? 

(As a side note, if it's the last one, has anyone considered getting a sociopath to work on FAI? Bringing someone who can't be trusted into an enterprise is a risky move, but if there genuinely are people in the world who have spent their entire lives practicing working out human emotions without feeling them...)

New Comment
36 comments, sorted by Click to highlight new comments since:

(I'm also not sure why lack of empathy would make you want to dissect squirrels, but that seems like a side issue)

Many people (like me) find it naturally interesting to dissect watches; empathy may refrain that instinct when it comes to living things (though I suspect a lot of neurotypical kids enjoy torturing bugs).


Psychopaths generally feel most emotions. What they may not experience, at least the way non-psychopaths do, is the emotion of fear and as a result have trouble recognizing facial expressions of fear in others, are bad at predicting behaviors that will scare people and have reduced amygdala activity during moral judgments. See the work of Abigail Marsh for one.

I doubt a sociopath would provide advantages over and above high functioning autistics.

Edit: Also, the whole ASPD thing is just a DSM clusterfuck. It in no way carves reality and the joints and definitely includes more than just psychopaths. But it does look like there are a group of people who do bad things and also have an empathy deficit. As you might imagine, it's a really difficult thing to study.

Also, the whole ASPD thing is just a DSM clusterfuck. It in no way carves reality and the joints

I suspect this may be true of most of the DSM. Here is a relevant article, excerpted by Bryan Caplan here.

Scott Aaronson had a post time ago about a common mistake: Empathy does not equal Sympathy.

A much-belated side-note for anyone who happens to come by:

Interestingly, the words 'empathy' and 'sympathy' have not been used consistently across different academic disciplines. (This has little direct significance to Aaronson's post; I am merely remarking on a use of language that may indirectly cause confusion.) I was doing a quick review of the psychopathy literature, and found that the word 'sympathy' is not used at all – it seems to be described instead by 'affective empathy'.

Batson (2009) wrote in detail about this in The Social Neuroscience of Empathy (bold emphasis mine):

Among philosophers, coming to feel as the other feels has often been called “sympathy,” not empathy (Hume, 1740/1896; Smith, 1759/1853). Among psychologists, it has been called “emotional contagion” (Hatfield, Cacioppo, & Rapson, 1994), “affective empathy” (Zahn-Waxler, Robinson, & Emde, 1992), and “automatic emotional empathy” (Hodges & Wegner, 1997).


Other-oriented emotion felt when another is perceived to be in need has not always been called empathy. It has also been called “pity” or “compassion” (Hume, 1740/1896; Smith, 1759/1853), “sympathetic distress” (Hoffman, 1981, 2000), and simply “sympathy” (Darwall, 1998; Eisenberg & Strayer, 1987; Preston & de Waal, 2002; Sober & Wilson, 1998; Wispé, 1986).


It would simplify matters if empathy referred to a single object and if everyone agreed on what that object was. Unfortunately, as with many psychological terms, this is not the case. Both empathy and sympathy (the term with which empathy is most often contrasted) have been used in a variety of ways. Indeed, with remarkable consistency exactly the same state that some scholars have labeled empathy others have labeled sympathy. I have discerned no clear basis—either historical or logical—for favoring one labeling scheme over another. The best one can do is recognize the different phenomena, make clear the labeling scheme one is adopting, and use that scheme consistently.

These illnesses are definitely not well understood, but you may have characterized sociopathy reasonably when you described it as lack of sympathy without lack of empathy. How they can come apart is indeed not intuitively obvious, but lots of the ways the brain works are unintuitive. Sociopaths seem to be the mirror image of autistics, who seem to lack empathy but not sympathy (autistics have great difficulty understanding the feelings of others, but seem as motivated as anyone to avoid hurting others when they can figure out how).

What is the case for labelling them "Illnesses"?

Something to understand about Sociopathy/Psychopathy/ASPD is that your mental model has been effected by selection bias (also availability bias). The only ASPD people you hear about are those with problems. The ones you hear about do not just have ASPD. ASPD magnifies other problems, be they biological, social, or circumstantial.

ASPD + Conduct Disorder causes the type of psychopathic serial killers popular culture is familiar with. ASPD + a brilliant mind and educational opportunity means you'd make a great lawyer or CEO. A guy with ASPD who gets addicted to drugs is more likely to rob or steal to get a fix. A guy with ASPD who sells used cars is more likely to be a great salesman because he pushes for the sale even when it's against the interest of his customers. A gang member with ASPD is more willing to use violence or kill for the gang. etc.

People can go their entire lives without finding out that they've got ASPD, let alone others finding out. The problem with Psych statistics is that they can't get a cross section of the population; they can only infer from the subset of people who need psychological help. It's textbook selection bias.

To go into analogy mode, think of humans as web browsers. A normal, functioning human is Internet Explorer 9; it receives the information and acts on it the way you'd expect. An autistic human would be an out of spec browser from 1992; it tries it's best but it can't render everything right and doesn't quite know what's going on, even though it would like to. A human with ASPD is like Firefox with the Adblock addon; it knows exactly what the website wants it to do and exactly what's expected of it, but it refuses to render or be effected by advertisements. If you've got other addons (mental disorders) though, the interaction between the two will make the browser more likely to catastrophically fail.

A normal, functioning human is Internet Explorer 9; it receives the information and acts on it the way you'd expect.

I was going to say this is a bad example, since IE9 still doesn't handle a lot of things the way it ought to, but humans are crazy the world is insane so whatevs.


Beat me to it. Here's a précis:

James Fallon is a neuroscientist who researches the brain physiology of criminality. It turns out his brain scans are indistinguishable from those of psychopathic killers, his ancestry includes violent killers, he has all the high-risk alleles for violence, and his family, friends, and in particular, his psychiatric collaborators with expertise in the behavior of psychopaths all characterize him as kind of a sociopath -- possessing a certain glib charm but lacking in the ability to genuinely connect.

Repeating a comment of mine from last Fall:

1) "Empathy" as the emotional interest in others, could also be called "moral sense", the kind of empathy that sociopaths are said to lack.

2) "Empathy" as the ability to identify emotionally with others, the set of instincts or "firmware" that make interpersonal communications and interactions go smoothly, the kind of empathy that autistics are said to lack.

3) "Empathy" or "imaginative identification with others", a more intellectual version, part of the definition of an intelligent being which is associated with metalaw. The ability to intellectually and purposely imagine yourself in the place of another, even a very different other, such as an extra-terrestrial alien, hence its association with metalaw.

My understanding is, as with most characteristics, that psychopathy is not an either / or, but rather a "to what degree" . Give a psychopathy test to anyone and they will score somewhere between highly psychopathic and a low level psychopath ( even if at certain times ) . I think that the character traits involved, including a lack of empathy for the suffering of others, are the result of an emotional shut down or wall of defence brought about do to various levels of trauma and emotional pain experienced. At the root are psychological survival tactics, however misguided, as well as pain avoidance. Psychopaths learn that goal achievement for them is increased by taking feelings out of the equation. The harder the door is slammed on their feelings and the longer it remains so, the less likely it will be reopened. So psychopathy is not caused by a lack of empathy but rather, a lack of empathy, to varying degrees, is a resultant characteristic of emotional disconnect.

Actually psychopathy is notable as a category in that the distribution is very much bimodal. The majority of people get very low scores on the tests. A few get very high ones.

Any chance of getting even a minor explanation as to whether or not there is any validity in the information that I have presented?

After reading a few dozen papers and books on the topic recently, I don't recognize your above portrait of psychopathy as being very accurate, FWIW.

has anyone considered getting a sociopath to work on FAI?

Since mild cases of ASPD are hard to distinguish from mild cases of Asperger's, and a lot of people with Asperger's are in programming, that doesn't seem like an off the wall idea.


Frankly, I'm more worried about the opposite problem. Might we need more neurotypical people working on FAI?

Extrapolated volition of a sociopath would be bad bad news.

This doesn't seem to me to address MinibearRex's proposal.

We don't want to extrapolate the sociopath's volition; we want the sociopath to extrapolate our volition. The idea is that sociopaths have experience with thinking objectively about humans' volition.

I don't disagree (similarly to how the proverbial fox might know many ways for a predator to get into the henhouse, and may seem like a good guard candidate for that reason, if the obvious problem with this were to be solved).

Maybe not.
Given that we don't have a clear way of predicting the output of a (of the?) volition-extrapolating process given its inputs, I'm not too confident about that.

The idea with FAI is the burden of proof is not on me :). The burden of proof is on the psychopath.

I'm not at all sure what "burden of proof" means in this context. Can you unpack that a little?

Assuming you believe the FAI project is coherent, what you want to do is prove safety (of AI code, of people involved, etc.) not prove lack of safety. So if I say a sociopath is bad news, and you are not sure -- well it is not my job to convince you! It is the sociopath's job to convince you (s)he's safe.

My personal opinion is sociopaths are badly badly broken, possibly not even entirely human.


How exactly are you defining "entirely human" such that 1 to 3 percent of the population of H. sapiens fails to qualify?

Well, I am not sure, since I do not know what it is like to be a bat. But it is my understanding from descriptions of psychopathy that psychopaths do not have the same kind of inner life that an ordinary human being has. (This is not a behavioral test, but then psychopaths will not do so well on behavioral tests either, since so many of them behave like monsters).


Others in this thread have already addressed some of it, so I'll just point you here:


and here:


and here:


and call it good.

(From personal experience, knowing some people who'd probably score as "mildly psychopathic", there's probably an inner-experience distinction in the same sense that there is with other strong neurological variations, and it can be more or less marked, but the idea that they're "not fully human" seems more like a comforting rationalization, a way of trying to cope with the often profoundly antisocial behavior that high-profile psychopaths appear casually capable of.)

Ah, I see. OK, thanks for clarifying.

For my part, if deciding whether a given FAI project is safe to turn on ever comes down to dueling intuitions and personal opinions, that alone seems like sufficient evidence to conclude that the project is not safe to turn on. (I would say the same thing about a bridge.)