Would you kill one person to save the lives of others?

Are morals based on facts and knowledge, or are they grounded in emotions? In other words, if you programmed a computer with the right information, would it make the same moral judgements as a person? ‘No’, is the answer suggested by new research suggesting some moral decisions are based on emotions rather than explicit moral rules.

Six patients with damage to their ventral medial prefrontal cortex (VMPC), an area involved in emotions, were recruited. The patients, who have normal IQ and knowledge of social and moral norms, were presented with: impersonal moral choices (Do you divert a runaway trolley off its track, saving the lives of five workmen but imperilling a single workman on the other track?); personal moral choices (Do you kill a fellow hostage in order that you and eight children can escape from terrorists?); and non-moral decisions (Do you buy a branded product or buy a supermarket’s own version?).

The choices made by the patients with prefrontal cortex damage were compared with those made by healthy controls and by patients with damage to parts of the brain not associated with emotion.

The groups didn’t differ in how they made impersonal moral choices and non-moral decisions. Crucially, however, the patients with prefrontal cortex damage tended to answer more ruthlessly than the other participants when it came to 13 out of 21 personal moral decisions. These decisions tended to pit the welfare of the majority against the participant having to commit deliberate harm to others. Repulsion at committing such deliberate harm caused the control participants to sacrifice the well-being of the majority, whereas the patients with prefrontal cortex damage tended to make more ‘utilitarian’, logical choices, harming one person to save the many.

“What is absolutely astonishing about our results is how selective the deficit is” said co-author Marc Hauser. “Damage to the frontal lobe leaves intact a suite of moral problem solving abilities, but damages judgements in which an aversive action is put into direct conflict with a strong utilitarian outcome”.

Koenigs, M., Young, L., Adolphs, R., Tranel, D., Cushman, F., Hauser, M. & Damasio, A. (2007). Damage to the prefrontal cortex increases utilitarian moral judgements. Nature.

Link to full list of the moral decisions (pdf).

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

5 thoughts on “Would you kill one person to save the lives of others?”

  1. Does anyone else find it weird that the author used the words “deficit” and “damages judgements”, and if so do you have ADHD? I thought the people with prefrontal cortex damage were making the obviously morally correct choices and the ‘normals’ were disgustingly selfish in indulging their squeamishness. I have ADHD and my prefrontal cortex functions are quite abnormal, for better and worse.(The link to the decision list is dead)

  2. Thanks anon – I’ve mended the link.I agree the author’s choice of wording does sound strange, given that the brain-damaged patients made decisions that were in the interest of the greater good. I guess he means they have a deficit in the sense that their choices were different from people whose brains are intact. Their judgments had become less human, even if in fact they were objectively better judgments.

  3. I would say the 'brain-damaged' reponse was more human than the control group. I have no problem with the study, it is the wording at the end I also find troubling. The point at which using a greater logical facility is deemed less 'human' is the most intriguing aspect of this article. It is an unneccessary simplification that should not, in my opinion, be made. The facts should just be presented without the 'moralising'.

  4. To follow up, I believe the most 'human' response to a moral situation is – firstly, to process the inevitable emotional response – secondly, to use logic to analyse this response. To not do so is, in my eyes, 'animalistic'. Using the limited language available, the brain damaged response is far more 'human' than the 'normal' one. A 'normal' response may in fact be defective by it's very assumption of superiority.

Comments are closed.