Moral Decision-Making and the Ventromedial Prefrontal Cortex

According to one popular theory on the role of affect in executive function, the Ventromedial Prefrontal Cortex (VMPFC) is the interface between affect and decision-making. Affective responses to possible avenues of actions arise from locations such as the amygdala, and are weighed into the decision-making process. In this view, then, the abnormal behaviour that is observed after lesions to the VMPFC is caused by a lack of affective inputs into decision-making, with intact rational cost-benefit analysis.

Koenigs et al (2007) decided to investigate this further, by comparing the performance of a group of patients with lesions to the VMPFC to healthy controls, and an additional group with lesions elsewhere in the brain. The participants were presented with dilemmas which were either non-moral (e.g., would you take the slow, scenic route, or the fast, boring route), impersonal and moral (e.g., choosing whether to lead dangerous smoke into a room where a single person is, or letting the smoke get into a room where three persons are), or personal and moral (e.g., choosing to save 5 patients by taking another patient’s organs against his will).

As the figure above shows, patients with lesions to the Ventromedial Prefrontal Cortex respond “too rationally” to personal moral dilemmas. Healthy controls generally opted not to act in these dilemmas (perhaps an example of what Kahneman called omission bias), while VMPFC patients generally endorsed immoral but ultimately rational behaviours.

I think this story is particularly interesting because this is a case where VMPFC patients are in a sense more rational than the rest of us. Objectively, if you want to save as many lives as possible, you should act as the VMPFC patients do. This is unusual, because the behaviour of frontal lobe patients is generally not best described as rational. For example, in studies using the Iowa gambling task, patients with the same type of brain damage show an inability to weigh cost versus benefits – the VMPFC patients show a tendency to use a high-risk, high-reward strategy that is ultimately less successful (and thus, less rational) than an alternative low-risk, low-reward strategy.

So it would seem that the behaviour of the VMPFC patients in the study by Koenigs et al (2007) can’t be explained by saying that the patients are acting solely on a cold, rational cost-benefit analysis, without input from affective and moral components. If it were this simple, VMPFC patients would not be impaired at the Iowa gambling task.

One way of explaining why VMPFC patients are impaired at both these tasks is to assume a general insensitivity to fear. The case of choosing to save lives through involuntary surgery and the case of choosing a high-risk strategy when gambling both reflect situations where most of us are probably deterred to some extent by fear of the consequences. To quote Koenigs et al (2007):

In the absence of an emotional reaction to harm of others in personal moral dilemmas, VMPC patients may rely on explicit norms endorsing the maximization of aggregate welfare and prohibiting the harming of others. This strategy would lead VMPC patients to a normal pattern of judgements on low-conflict personal dilemmas but an abnormal pattern of judgements on high-conflict personal dilemmas, precisely as was observed.

If you want to know more, the Neurophilosopher beat me to blogging this article. There is also a write-up in the New York Times.