Empathy

So, I see that Simon Baron-Cohen’s empathy test is making the rounds on Tumblr. That provides me with an opportunity to talk about a common misunderstanding about empathy.

The test, like most people’s thinking, is confusing at least two different phenomena that fall under the umbrella of “empathy”. Consider this: it has been said about both autists (high- and low-functioning) and psychopaths that their central characteristic is a lack of empathy. The stereotypical autist is a person doesn’t hurt a fly, if only because he lacks the social compass needed to do it; the stereotypical psychopath’s social compass is so well-developed that he effortlessly manipulates people for his own ends and purposes, regardless of the potential harm to the humans that become his instruments. Intuitively, it seems as if these two kinds of people shouldn’t be called variations on the same theme (lack of empathy).

And indeed, such is the case. There are at least two different concepts that are easily conflated under the empathy term. One is called cognitive empathy, which also commonly goes by the name “Theory of Mind (ToM)”. It is, roughly, the ability to predict and form a mental model of other people’s emotional state, their intentions and desires on the basis of nonverbal cues. Basically, social skills. Being able to tell when your conversation partner is bored by what you’re saying, for instance. Understanding the emotional state of someone by looking at their body language. Reading body language, tone of voice, context, and so on. This skill is present in psycopaths (unless they happen to have some other abnormality) and weak or not present at all in autistics and folks with nonverbal learning disabilities.

The other kind of empathy is affective empathy. If you understand that someone is being treated cruelly, if they are hurt and you feel sympathy for them, that is a sign of affective empathy. It has nothing to do with acquaring information about people’s emotional states, it is rather about how this knowledge (if/when you acquire it) affects you (if at all). This kind of empathy is weakened or nonexistent in psycopaths. If they are aware that they are hurting other people, that knowledge will often not move them to sympathy or regret. This kind of empathy may, in fact, be better than average in autistics. The abstract of one paper reads:

We administered the Interpersonal Reactivity Index (IRI), a multi-dimensional measure of empathy, and the Strange Stories test to 21 adults with AS [Asperger Syndrome] and 21 matched controls. Our data show that while the AS group scored lower on the measures of cognitive empathy and theory of mind, they were no different from controls on one affective empathy scale of the IRI (empathic concern), and scored higher than controls on the other (personal distress).

There is clearly a need to nuance here. We could go even further. ToM suggests that the defining characteristic of autists may the inability to form a working ToM — a mental map of other people’s emotional state, intents and so on. But then we can nuance this further by asking: is this due to an inability to form a Theory of (someone else’s) Mind, or is it due to the inability to acquire the information necessary to do this due to an inability to read nonverbal cues? Some suggest that autists in fact lack the conceptual machinery to form a working ToM, even if all the information they would need to do so was presented for them.

One common test to show this is to present the following scenario: let a child observe the following:

1. There is a room in which an object (e.g., a doll or a ball) is placed, while another person is watching and clearly (this is important) sees where the object is placed.
2. The person leaves the room.
3. The object is hidden somewhere else. The person who left the room clearly does not know where the object was hidden, or even that it was hidden.

Then ask the child where the person that saw the object be placed, but did not see it be hidden will look. Supposedly, all children over a given age (including children with Down’s, for instance) will say that the person will look where the object was originally placed, which is the correct answer, but autistic children will be a lot older before they grasp this, and will say that the person will look where the child saw it be hidden; and this is due to a failure to mentally differentiate between the child’s knowledge and another person’s knowledge — to form a working ToM. So perhaps there are people who lack the ability to form Theories of Mind given all information they might need, and others that can form working ToMs but are unable to extract the necessary information by reading body language.

Fascinating stuff, people. But the point with this post was to illustrate the difference between different kinds of empathy, and not do so in a wishy-washy New Age kind of way. This is especially important since empathy is often not used in a value neutral way: often, to say that such-and-such is empathic implies that they are a good person, while saying that such-and-such lacks empathy often implies that they are akin to a psychopath.