Deliberate Disguise in Face Identification

Eilidh Noyes, Rob Jenkins

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Facial image comparison is difficult for unfamiliar faces and easy for familiar faces. Those conclusions are robust, but they arise from situations in which the people being identified cooperate with the effort to identify them. In forensic and security settings, people are often motivated to subvert identification by manipulating their appearance, yet little is known about deliberate disguise and its effectiveness. We distinguish two forms of disguise— Evasion (trying not to look like oneself) and Impersonation (trying to look like another person). We present a new set of disguised face images (the FAÇADE image set), in which models altered their appearance to induce specific identification errors. In Experiment 1, unfamiliar observers were less accurate matching disguise items, especially evasion items, than matching undisguised items. A similar pattern held in Experiment 2, in which participants were informed about the disguise manipulations. In Experiment 3, familiar observers saw through impersonation disguise, but accuracy was lower for evasion disguise. Quantifying the performance cost of disguise reveals distinct performance profiles for impersonation and evasion. Evasion disguise was especially effective and reduced identification performance for familiar observers as well as for unfamiliar observers. We subsume these findings under a statistical framework of face learning.
Original languageEnglish
Pages (from-to)280-290
Number of pages11
JournalJournal of Experimental Psychology: Applied
Volume25
Issue number2
Early online date7 Jan 2019
DOIs
Publication statusPublished - 1 Jun 2019

Fingerprint

Learning
Costs and Cost Analysis

Cite this

@article{c533d8debcfa47909f503661ad7f53e3,
title = "Deliberate Disguise in Face Identification",
abstract = "Facial image comparison is difficult for unfamiliar faces and easy for familiar faces. Those conclusions are robust, but they arise from situations in which the people being identified cooperate with the effort to identify them. In forensic and security settings, people are often motivated to subvert identification by manipulating their appearance, yet little is known about deliberate disguise and its effectiveness. We distinguish two forms of disguise— Evasion (trying not to look like oneself) and Impersonation (trying to look like another person). We present a new set of disguised face images (the FA{\cC}ADE image set), in which models altered their appearance to induce specific identification errors. In Experiment 1, unfamiliar observers were less accurate matching disguise items, especially evasion items, than matching undisguised items. A similar pattern held in Experiment 2, in which participants were informed about the disguise manipulations. In Experiment 3, familiar observers saw through impersonation disguise, but accuracy was lower for evasion disguise. Quantifying the performance cost of disguise reveals distinct performance profiles for impersonation and evasion. Evasion disguise was especially effective and reduced identification performance for familiar observers as well as for unfamiliar observers. We subsume these findings under a statistical framework of face learning.",
keywords = "Disguise, Face matching, Face recognition, Identification, Security",
author = "Eilidh Noyes and Rob Jenkins",
year = "2019",
month = "6",
day = "1",
doi = "10.1037/xap0000213",
language = "English",
volume = "25",
pages = "280--290",
journal = "Journal of Experimental Psychology: Applied",
issn = "1076-898X",
publisher = "American Psychological Association Inc.",
number = "2",

}

Deliberate Disguise in Face Identification. / Noyes, Eilidh; Jenkins, Rob.

In: Journal of Experimental Psychology: Applied, Vol. 25, No. 2, 01.06.2019, p. 280-290.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Deliberate Disguise in Face Identification

AU - Noyes, Eilidh

AU - Jenkins, Rob

PY - 2019/6/1

Y1 - 2019/6/1

N2 - Facial image comparison is difficult for unfamiliar faces and easy for familiar faces. Those conclusions are robust, but they arise from situations in which the people being identified cooperate with the effort to identify them. In forensic and security settings, people are often motivated to subvert identification by manipulating their appearance, yet little is known about deliberate disguise and its effectiveness. We distinguish two forms of disguise— Evasion (trying not to look like oneself) and Impersonation (trying to look like another person). We present a new set of disguised face images (the FAÇADE image set), in which models altered their appearance to induce specific identification errors. In Experiment 1, unfamiliar observers were less accurate matching disguise items, especially evasion items, than matching undisguised items. A similar pattern held in Experiment 2, in which participants were informed about the disguise manipulations. In Experiment 3, familiar observers saw through impersonation disguise, but accuracy was lower for evasion disguise. Quantifying the performance cost of disguise reveals distinct performance profiles for impersonation and evasion. Evasion disguise was especially effective and reduced identification performance for familiar observers as well as for unfamiliar observers. We subsume these findings under a statistical framework of face learning.

AB - Facial image comparison is difficult for unfamiliar faces and easy for familiar faces. Those conclusions are robust, but they arise from situations in which the people being identified cooperate with the effort to identify them. In forensic and security settings, people are often motivated to subvert identification by manipulating their appearance, yet little is known about deliberate disguise and its effectiveness. We distinguish two forms of disguise— Evasion (trying not to look like oneself) and Impersonation (trying to look like another person). We present a new set of disguised face images (the FAÇADE image set), in which models altered their appearance to induce specific identification errors. In Experiment 1, unfamiliar observers were less accurate matching disguise items, especially evasion items, than matching undisguised items. A similar pattern held in Experiment 2, in which participants were informed about the disguise manipulations. In Experiment 3, familiar observers saw through impersonation disguise, but accuracy was lower for evasion disguise. Quantifying the performance cost of disguise reveals distinct performance profiles for impersonation and evasion. Evasion disguise was especially effective and reduced identification performance for familiar observers as well as for unfamiliar observers. We subsume these findings under a statistical framework of face learning.

KW - Disguise

KW - Face matching

KW - Face recognition

KW - Identification

KW - Security

UR - http://www.scopus.com/inward/record.url?scp=85061092703&partnerID=8YFLogxK

U2 - 10.1037/xap0000213

DO - 10.1037/xap0000213

M3 - Article

VL - 25

SP - 280

EP - 290

JO - Journal of Experimental Psychology: Applied

JF - Journal of Experimental Psychology: Applied

SN - 1076-898X

IS - 2

ER -