Last week, I wrote about a simple idea: far from being an automatic reflex, empathy often requires a choice to engage with others’ emotions. Moreover, people often refuse this choice, because empathy can be challenging, painful, costly, or all three. Instead of meeting these challenges, we often keep our distance from others’ suffering, tune out the opinions of people with whom we disagree, and generally empathize only when it is convenient.

Viewing empathy as a choice, I think, reframes the science of this phenomenon. Psychologists tend to focus on people’s ability to empathize—for instance, measuring how people measure on traits such as emotional intelligence. On this view, some people are simply good at reading others, and some are not. I call this the “Roddenberry Hypothesis” because it is embedded in the character structure of Star Trek, The Next Generation (indisputably the best television show of all time). TNG’s cast includes Deanna Troi—a Betazoid known throughout the galaxy for sharing and understanding others’ feelings—and Data—an android who can no more empathize than you or I can jump tall buildings. Scientists long shared Roddenberry’s assumption that empathy is (1) automatic, and (2) stable. Increasingly, I believe that a more important trait is not how good a person is at empathizing, but how motivated they are to engage with others in the first place.

This changes, among other things, the way we view psychiatric conditions featuring deficits in empathy. The most famous of these are of course autism spectrum disorders (ASD). In a landmark work almost 20 years old, Simon Baron-Cohen characterized ASD as a form of “Mindblindness,” or an inability to understand others’ experiences. Baron-Cohen’s description of this deficit was dramatic and bleak. Here is his take (borrowed from Alison Gopnik) on how a dinner party must feel to a person with ASD:

“At the top of my field of vision is the blurry edge of a nose, in front are waving hands… Around me bags of skin are draped over chairs, and stuffed into pieces of cloth, they shift and protrude in unexpected ways… Imagine that these noisy skin bags suddenly moved towards you, and their noises grew loud, and you had no idea why, no way of explaining them or predicting what they would do next.”

In essence, this argument implies that individuals with ASD, like Data, are constitutionally incapable of empathizing. Although Baron-Cohen and others have backed off of this position, Mindblindness continues to dominate much research and popular conceptions surrounding ASD. Recently, however, a different model has begun to garner attention in autism science. This viewpoint—best described by Coralie Chevalier, Robert Schultz and their colleagues—holds that individuals with ASD might be able to engage with social cues, but less motivated to do so. Early in development, children with ASD might not feel as rewarded by the act of engaging with others as do typically developing children. This, in turn, might reduce ASD individuals’ motivation to “practice” reading others’ thoughts and emotions. Like anything else that one doesn’t practice, a lack of experience with other minds eventually could produce symptoms of Mindblindness.

This social motivation theory further suggests that when ASD individuals are motivated to empathize, they may do so unexpectedly well. And indeed, a spate of neuroimaging and behavioral studies have shown that in high motivation situations—such as when viewing family members as opposed to strangers, or being instructed specifically to attend to others’ internal states—individuals with ASD show improved (though not necessarily “normal”) ability to infer others’ states and more typical patterns of brain activity in response to social cues.

A more recent study paints a similar motivational picture of psychopathy. James Blair and others have sometimes framed this disorder as a functional cousin of ASD: whereas individuals with ASD may find it difficult to understand others’ internal states, individuals with psychopathy appear to understand, but not share others’ emotions. This callousness, in turn, underlies the unfortunate streak of violent and criminal behaviors that characterize psychopathy. But—as with ASD individuals’ difficulties understanding others—we can ask the question: are individuals with psychopathy lacking the ability to share others’ emotions, or simply the motivation to do so?

Christian Keysers and his colleagues took a clever neuroscience approach to this question in a paper published last month (Keysers speaks about the work in an interview here). They built on loads of data demonstrating that people activate similar neural systems when they experience pain and when they observe others in pain (a phenomenon I’ve coined the Bill Clinton effect). Keysers and colleagues found that individuals with psychopathy failed to engage these regions as much as control subjects when watching videos of others experience painful events. However, this study featured a fascinating twist: in a second condition, both criminals and control subjects were instructed to explicitly try to empathize with the people in the videos they watched. Under this condition, individuals with psychopathy engaged these regions more than they had without instructions, and almost as much as control subjects. These data, again, suggest that oftentimes deficits in empathy can be reduced through the right motivational “triggers.”

One important message to not take away from this work is that the deficits associated with ASD and psychopathy are in some way these individuals’ fault. Both conditions clearly involve early-developing differences in people’s responses to social cues. However, broadly abandoning the idea of empathy as an automatic ability that people either have or don’t have—and instead viewing it as a choice—can change the way we can see these disorders, and the ways we think about treating them.