Effects of facial periphery on unfamiliar face recognition
Charlie D. Frowd
Antonio L. Manzanero
Springer Science+Business Media, LLC, part of Springer Nature 2018
Facial identification based on a comparison with a photo-ID is the most standard way to prove identity at security controls. Two
experiments are performed controlling for the presence or absence of within-contour facial periphery (masking) and its substi-
tution for an average periphery image (averaging), measuring matching accuracy, reaction time and signal-detection measures d’
an C. Experiment 1 compared face matching for the original periphery in the pair, its masking and its averaging using a sample of
frontal image pairs. Experiment 2 compared matches in the average condition with and without an apparent gap around internal
features. Results show that masking of facial periphery had a detrimental effect on unfamiliar face matching accuracy accom-
panied by an increased tendency to positive responses, while averaging composites yielded no difference as compared to the
original pair. No differences were found for matching in conditions with or without an apparent gap. The results suggest that face
periphery contributes to unfamiliar face matching accuracy through a holistic process which is disrupted when focusing exclu-
sively on the innermost features; the effect is dependent on the global structure of the face image and not related to low-level
details. These results should be considered in the context of improving unfamiliar face matching.
Keywords Face masking
It is widely accepted that face identification is part of every-
day life and is performed effortlessly. However this trust
placed on facial identification is not really justified. Kemp
et al. (1997) exposed the striking result that supermarket
staff would accept as valid half of the fraudulent cards pre-
sented when Bcheaters^ showed some similarity to a photo-
ID, and reject 10% of valid cards with minor variations in
appearance. Bruce et al. (1999) established that accuracy of
facial identification from video is far from ideal in good
viewing conditions (no time pressure or memory load re-
quirements). Reducing the task to a single yes/no decision
when both images are present is also problematic for unfa-
miliar faces (Megreya and Burton 2006), whose processing
is driven mainly by low level characteristics such as
illumination and view-point (Hancock et al. 2000).
Identification issues also occur among trained professionals,
such as passport officers, who incur in a 14% rate of false
acceptances to fraudulent photos (White et al. 2014b).
Meanwhile, large-scale automatic face recognition systems
do not provide a final solution, usually only providing a
gallery of matching candidates which a human operator
must select the most likely candidate; at best, these systems
reduce human matching decisions (White et al. 2015).
Therefore, there is a need to study the difficulty of unfamil-
iar face matching and to find methods for improving it.
The accuracy of face identification is related to a large
extent to differences between familiar and unfamiliar face
processing (Bruce et al. 2001). Familiar face processing
seems facilitated by an expertise in internal facial features.
This region includes eyes, nose and mouth is recognized
faster and in a more robust way than for unfamiliar faces
(Ellis, Shepherd, and Davies, 1979; Young et al. 1985).
Unfamiliar face recognition is affected more by changes in
external features - in particular hair (Bruce et al. 1999; Young
et al. 1985). The eyes and the nose seem to be the two most
important internal features for familiar face identification. In a
task involving changes to facial features, familiarization to
faces produces an increased sensitivity to the eye region,
* Rubén García-Zurdo
Universidad Complutense de Madrid, 28223 Madrid, Spain
University of Central Lancashire, Preston, UK