School of Human Sciences
Face and Voice Recognition Lab
Institute of Lifecourse Development
University of Greenwich
London
Twitter: @GRecognisers
Novel interactive face matching procedure
A research team from Nottingham Trent University, the University of Birmingham, the University of Greenwich, and the University of New South Wales has trialled a new interactive viewing procedure for face matching.
Smith, H. M. J., Andrews, S., Baguley, T., Colloff, M. F., Davis, J. P., White, D., Rockey, J. C., & Flowe, H. D. (2021). Performance of typical and superior face recognisers on a novel interactive face matching procedure. British Journal of Psychology, 112, 964-881. DOI: 10.1111/bjop.12499 (Read here - open access)
This new procedure supplies users with an interactive image of a face that they can rotate to different orientations. The results show that the interactive procedure reliably improved face matching accuracy for both typical and super face recognisers. Perhaps as importantly, the procedure improved accuracy at ruling out mismatching images.
The findings also show that the system even improves correct identification accuracy when image quality is low. As such, this interactive matching system may be particularly helpful in assisting with identification in CCTV images, which often relies on matching footage to mugshots.
Dr Harriet Smith, lead author and independent research fellow in psychology at NTU’s School of Social Sciences, said:
“These findings suggest that providing visual information about how a face looks from different viewpoints reduces human error, which has implications for any context where photo ID is used for identity verification, such as passport processing, border control, surveillance and CCTV.
Super recognisers are people with exceptional face recognition ability. While this skill is rare, some police forces now deploy such people to identification roles to try and mitigate the risk of incorrect face matching decisions. Our results suggest that this risk can be minimised further by using the interactive procedure.”
Follow the link below for a media article with further details:
Stop press
Some of the same researchers (this time not from the University of Greenwich) published a followup paper in Nature(Scientific Reports) in which the interactive face matching system was shown to be able improve line-up identification performance from police line-ups.
Flowe, H., Colloff, M., Seale-Carlisle, T., Karoğlu, N., Rockey, J., Smith, H. M., ... & Smith, L. L. (2021). Perpetrator pose reinstatement during a lineup test increases discrimination accuracy. Nature, 11, 13830. https://www.nature.com/articles/s41598-021-92509-0
A followup showing it increases accuracy if faces are from different ethnicities to the witness will be published soon in American Psychologist (pre-print here)
Colloff, M. F., Flowe, H. D., Smith, H. M. J., Seale-Carlisle, T. M., Meissner, C. A., Rockey, J. C., … Parganiha, A. (2020, July 7). Active exploration of faces in police lineups increases discrimination accuracy. https://doi.org/10.31234/osf.io/tvga4
Professor Heather Flowe describes the work here: -
The system featured in the Daily Mail – hence the cover photo.
Finally, we are applying for further research funds to further investigate the practical use of the system, and to better understand why it helps. The University of Greenwich will be a key partner in this project.
Research team
Dr Harriet Smith, Department of Psychology, Nottingham Trent University, UK
Dr Sally Andrews, Department of Psychology, Nottingham Trent University, UK
Dr Thom Baguley, Department of Psychology, Nottingham Trent University, UK
Dr Melissa Colloff, School of Psychology, University of Birmingham, UK
Dr Josh P Davis, School of Human Sciences, Institute of Lifecourse Development, University of Greenwich, London, UK
Dr David White, School of Psychology, UNSW Sydney, New South Wales, Australia
Dr James Rockey, Department of Economics, University of Birmingham, Birmingham, UK
Dr Heather Flowe, School of Psychology, University of Birmingham, UK
Comments