A computational and neural model of happiness

Paper out in Proceedings of the National Academy of Sciences (it's open access - anyone can download the paper by clicking on the PDF symbol) with my UCL colleagues Nikolina Skandali, Peter Dayan, and Ray Dolan.

We report the results of an fMRI experiment, two behavioral studies, and a large-scale smartphone-based replication study. We used computational modeling to show that happiness in our experiments is explained not by earnings, but by the combined influence of recent reward expectations and prediction errors resulting from those expectations. That means happiness depends not on how well things are going, but whether things are going better than expected. Here is our happiness equation:

In the fMRI study, we found that measurements of neural activity in a brain area called the striatum could be used to predict how much happiness would change as the result of the decisions that subjects made and the outcomes they received. In the smartphone experiment we showed that we could also predict happiness outside of the lab in 18,420 participants around the world. Thanks to everyone who played 'What makes me happy?' in The Great Brain Experiment on their phones (download the app if you would like to contribute to our ongoing research). Your data (221,040 happiness ratings) are in the paper!

You can read more about the study in the UCL Press Release and here: BBC News, Forbes, The Atlantic, Time, Washington PostTelegraph.

The paper can also be downloaded here. My research is supported by the Max Planck Society and the Wellcome Trust.