Learning context and the other-race effect

Strategies for improving face recognition

Jacqueline G. Cavazos, Eilidh Noyes, Alice J. O'Toole

Research output: Contribution to specialist publicationArticle

2 Citations (Scopus)

Abstract

People recognize faces of their own race more accurately than faces of other races—a phenomenon known as the “Other-Race Effect” (ORE). Previous studies show that training with multiple variable images improves face recognition. Building on multi-image training, we take a novel approach to improving own- and other-race face recognition by testing the role of learning context on accuracy. Learning context was either contiguous, with multiple images of each identity seen in sequence, or distributed, with multiple images of an identity randomly interspersed among different identities. In two experiments, East Asian and Caucasian participants learned own- and other-races faces either in a contiguous or distributed order. In Experiment 1, people learned each identity from four highly variable face images. In Experiment 2, identities were learned from one image, repeated four times. In both experiments we found a robust other-race effect. The effect of learning context, however, differed depending on the variability of the learned images. The distributed presentation yielded better recognition when people learned from single repeated images (Exp. 1), but not when they learned from multiple variable images (Exp. 2). Overall, performance was better with multiple-image training than repeated single image training. We conclude that multiple-image training and distributed learning can both improve recognition accuracy, but via distinct processes. The former broadens perceptual tolerance for image variation from a face, when there are diverse images available to learn. The latter effectively strengthens the representation of differences among similar faces, when there is only a single learning image.
Original languageEnglish
Pages169-183
Number of pages15
Volume157
Specialist publicationVision Research
PublisherElsevier Ltd
DOIs
Publication statusPublished - 1 Apr 2019
Externally publishedYes

Fingerprint

Face recognition
Experiments
Testing

Cite this

Cavazos, Jacqueline G. ; Noyes, Eilidh ; O'Toole, Alice J. / Learning context and the other-race effect : Strategies for improving face recognition. In: Vision Research. 2019 ; Vol. 157. pp. 169-183.
@misc{92a0e40276fb45f88668671d6ca4a91a,
title = "Learning context and the other-race effect: Strategies for improving face recognition",
abstract = "People recognize faces of their own race more accurately than faces of other races—a phenomenon known as the “Other-Race Effect” (ORE). Previous studies show that training with multiple variable images improves face recognition. Building on multi-image training, we take a novel approach to improving own- and other-race face recognition by testing the role of learning context on accuracy. Learning context was either contiguous, with multiple images of each identity seen in sequence, or distributed, with multiple images of an identity randomly interspersed among different identities. In two experiments, East Asian and Caucasian participants learned own- and other-races faces either in a contiguous or distributed order. In Experiment 1, people learned each identity from four highly variable face images. In Experiment 2, identities were learned from one image, repeated four times. In both experiments we found a robust other-race effect. The effect of learning context, however, differed depending on the variability of the learned images. The distributed presentation yielded better recognition when people learned from single repeated images (Exp. 1), but not when they learned from multiple variable images (Exp. 2). Overall, performance was better with multiple-image training than repeated single image training. We conclude that multiple-image training and distributed learning can both improve recognition accuracy, but via distinct processes. The former broadens perceptual tolerance for image variation from a face, when there are diverse images available to learn. The latter effectively strengthens the representation of differences among similar faces, when there is only a single learning image.",
keywords = "Face recognition, Learning context, Other-race effect, Training",
author = "Cavazos, {Jacqueline G.} and Eilidh Noyes and O'Toole, {Alice J.}",
year = "2019",
month = "4",
day = "1",
doi = "10.1016/j.visres.2018.03.003",
language = "English",
volume = "157",
pages = "169--183",
journal = "Vision Research",
issn = "0042-6989",
publisher = "Elsevier Ltd",

}

Learning context and the other-race effect : Strategies for improving face recognition. / Cavazos, Jacqueline G.; Noyes, Eilidh; O'Toole, Alice J.

In: Vision Research, Vol. 157, 01.04.2019, p. 169-183.

Research output: Contribution to specialist publicationArticle

TY - GEN

T1 - Learning context and the other-race effect

T2 - Strategies for improving face recognition

AU - Cavazos, Jacqueline G.

AU - Noyes, Eilidh

AU - O'Toole, Alice J.

PY - 2019/4/1

Y1 - 2019/4/1

N2 - People recognize faces of their own race more accurately than faces of other races—a phenomenon known as the “Other-Race Effect” (ORE). Previous studies show that training with multiple variable images improves face recognition. Building on multi-image training, we take a novel approach to improving own- and other-race face recognition by testing the role of learning context on accuracy. Learning context was either contiguous, with multiple images of each identity seen in sequence, or distributed, with multiple images of an identity randomly interspersed among different identities. In two experiments, East Asian and Caucasian participants learned own- and other-races faces either in a contiguous or distributed order. In Experiment 1, people learned each identity from four highly variable face images. In Experiment 2, identities were learned from one image, repeated four times. In both experiments we found a robust other-race effect. The effect of learning context, however, differed depending on the variability of the learned images. The distributed presentation yielded better recognition when people learned from single repeated images (Exp. 1), but not when they learned from multiple variable images (Exp. 2). Overall, performance was better with multiple-image training than repeated single image training. We conclude that multiple-image training and distributed learning can both improve recognition accuracy, but via distinct processes. The former broadens perceptual tolerance for image variation from a face, when there are diverse images available to learn. The latter effectively strengthens the representation of differences among similar faces, when there is only a single learning image.

AB - People recognize faces of their own race more accurately than faces of other races—a phenomenon known as the “Other-Race Effect” (ORE). Previous studies show that training with multiple variable images improves face recognition. Building on multi-image training, we take a novel approach to improving own- and other-race face recognition by testing the role of learning context on accuracy. Learning context was either contiguous, with multiple images of each identity seen in sequence, or distributed, with multiple images of an identity randomly interspersed among different identities. In two experiments, East Asian and Caucasian participants learned own- and other-races faces either in a contiguous or distributed order. In Experiment 1, people learned each identity from four highly variable face images. In Experiment 2, identities were learned from one image, repeated four times. In both experiments we found a robust other-race effect. The effect of learning context, however, differed depending on the variability of the learned images. The distributed presentation yielded better recognition when people learned from single repeated images (Exp. 1), but not when they learned from multiple variable images (Exp. 2). Overall, performance was better with multiple-image training than repeated single image training. We conclude that multiple-image training and distributed learning can both improve recognition accuracy, but via distinct processes. The former broadens perceptual tolerance for image variation from a face, when there are diverse images available to learn. The latter effectively strengthens the representation of differences among similar faces, when there is only a single learning image.

KW - Face recognition

KW - Learning context

KW - Other-race effect

KW - Training

UR - http://www.scopus.com/inward/record.url?scp=85044997424&partnerID=8YFLogxK

U2 - 10.1016/j.visres.2018.03.003

DO - 10.1016/j.visres.2018.03.003

M3 - Article

VL - 157

SP - 169

EP - 183

JO - Vision Research

JF - Vision Research

SN - 0042-6989

PB - Elsevier Ltd

ER -