This project provides the data and code notebooks that to accompany the manuscript **Beyond average: Using face regression to study social perception**. There, an algorithm is described that addresses some common theoretical issues in the use of composites in face research, and illustrates its flexibility.
The algorithm is implemented using Python, and a dedicated package is available for free download and use. Installation instructions can be found at [Github][1].
The Jupyter notebooks provided here act as a user guide to the functionality of the package. In addition, there are notebooks that describe the more general analysis of data that was carried out for the behavioural experiments in the study.
A pre-print of the manuscript is available.
29/08/2018 - Data reanalysed and reuploaded to reflect changes in the functioning of the package, ensuring compatibility.
[1]: https://github.com/alexjonesphd/python-face-regressor