Human beings are ready to make moral judgments about how they would behave or feel others should behave in various circumstances, and there is a growing body of knowledge from moral psychology and neuroscience that suggests that there may be innate universal principles to which human beings adhere. However, when faced with concrete circumstances in real life how do people actually act? – do they act in conformity with their abstract principles or do the specific circumstances override these? Everyone would agree that in principle it would be a moral requirement to throw a switch, if able to do so, that would save the lives of five people in danger. The majority of people would also agree that they would throw the switch even if this would accidentally kill one other person who just happened to be in the wrong place at the wrong time – but thereby saving the five. However, given a group of five actual people, and an uninvolved bystander, whom you might have seen and talked with beforehand, what action would actually be taken?

This problem, the possible disjunction between abstract moral judgements and actual behaviour, cannot be studied experimentally; people can only be asked what they would do in the circumstances. However, many years of research using the technology of immersive virtual reality (IVR) demonstrates that people tend to respond to virtual situations and events as if these were real. Therefore IVR can be used to create situations in which people are exposed to specific circumstances where they are not merely asked about how they might behave, but actually have to take action. The primary aim of the proposed research is to explore the relationship between moral judgment and behaviour by exploiting IVR. Based on prior research, we will explore whether there is dissociation between how people decide what is morally right or wrong and what they do in actual moral situations. Evidence of dissociation would enable us to explore the precise cause of this difference between judgments and actions.

A person in an immersive virtual reality encounters an incident of bullying – the virtual character on the right is attacking the one on the left. Even though the person knows that this situation is not actually happening he would tend to have similar thoughts and feelings as in reality, and may actually intervene to try to stop the fight. The photograph itself is illustrative of the idea rather than of an actual experiment (courtesy Aitor Rovira and David Swapp, UCL).

The research will provide laboratory based experimental data in a field of research where otherwise the data available is either opinion based, or based on historical records. The problem the research addresses is fundamental – the roots of our actions in the face of moral challenges. This is not only important scientifically, but also in a practical sense with very wide repercussions for the understanding of behaviour in society and international relations. This includes topical issues such as supposed justifications of torture – here doing harm to one in order to apparently save many is a variant of one particular type of moral dilemma. Similarly, so-called ‘collateral damage’ when people are killed, for example, in a bombing raid because their house was near a ‘legitimate target’ is a variant of another.


The starting date of this project is 09/11/2009.

This project is funded by leverhulme trust

This page last modified 15 November, 2009 by [Sylvia Pan]

Text size

Computer Science Department - University College London - Gower Street - London - WC1E 6BT - Telephone: +44 (0)20 7679 7214 - Copyright © 1999-2005 UCL

Search by Google
Link to UCL home page