Predicting Face Recognition Performance in Unconstrained Environments

Phillips, P. Jonathon; Yates, Amy N.; Beveridge, J. Ross; Givens, Geof H.

doi:10.1109/cvprw.2017.83

Cited by 2 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We emphasize that in order to screen for potentially superior face processing skills multiple ecologically meaningful and challenging tasks must be employed to assess distinct aspects of face processing (Bate et al, 2018;Fysh, 2018;Herzmann et al, 2008;Ramon, 2018a;Ramon et al, 2019aRamon et al, , 2019b. To this end, accuracy scores recorded for face matching under "best-case scenarios" (Fysh, 2018) are insufficient, including, for example, moderate time periods between image creation, or the use of well-lit frontal faces as required for reliable computer-based identity matching (Phillips, Yates, Beveridge, & Givens, 2017;Phillips et al, 2018). Especially when tasks require simple same/different decisions under virtually no time constraints (e.g., 30 s here, or even 3 months in Phillips et al (2018)), high performance accuracy cannot provide a reliable basis to identify or validate allegedly superior abilities.…”

Section: Discussionmentioning

confidence: 99%

Normative data for two challenging tests of face matching under ecological conditions

et al. 2020

View full text Add to dashboard Cite

Background: Unfamiliar face processing is an ability that varies considerably between individuals. Numerous studies have aimed to identify its underlying determinants using controlled experimental procedures. While such tests can isolate variables that influence face processing, they usually involve somewhat unrealistic situations and optimized face images as stimulus material. As a consequence, the extent to which the performance observed under laboratory settings is informative for predicting real-life proficiency remains unclear. Results: We present normative data for two ecologically valid but underused tests of face matching: the Yearbook Test (YBT) and the Facial Identity Card Sorting Test (FICST). The YBT (n = 252) measures identity matching across substantial age-related changes in facial appearance, while the FICST (n = 218) assesses the ability to process unfamiliar facial identity despite superficial image variations. To determine the predictive value of both tests, a subsample of our cohort (n = 181) also completed a commonly used test of face recognition and two tests of face perception (the long form of the Cambridge Face Memory Test (CFMT+), the Expertise in Facial Comparison Test (EFCT) and the Person Identification Challenge Test (PICT)). Conclusions: Focusing on the top performers identified independently per test, we made two important observations: 1) YBT and FICST performance can predict CFMT+ scores and vice versa; and 2) EFCT and PICT scores neither reliably predict superior performance in ecologically meaningful and challenging tests of face matching, nor in the most commonly used test of face recognition. These findings emphasize the necessity for using challenging and ecologically relevant, and thus highly sensitive, tasks of unfamiliar face processing to identify high-performing individuals in the normal population.

show abstract

Section: Discussionmentioning

confidence: 99%

Normative data for two challenging tests of face matching under ecological conditions

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The EFCT and PICT have been reported to provide normally distributed performance accuracy scores and according to some are among the most commonly used tools to assess face processing ability, both in and outside laboratory settings (cf. Phillips, Yates, Beveridge & Givens, 2017).…”

Section: Introductionmentioning

confidence: 99%

Normative data for two tests of face matching under ecological conditions

Stacchi¹,

Huguenin-Elie²,

Caldara³

et al. 2019

Preprint

View full text Add to dashboard Cite

How do humans process unfamiliar faces, and how can we reliably identify individuals that are most proficient at it? Motivated by its relevance in applied contexts, much empirical work has sought to answer these questions. Controlled laboratory tests have been developed to understand the contribution of different variables and inter-individual differences. However, such face processing tests involving stimuli derived from ideal conditions, or manipulations and tasks that are not representative of real life, may lack ecological validity. This crucial consideration is often overlooked when laboratory tests are used to predict real-life proficiency. The present study followed the rationale that traditionally used controlled tests should be paired with more realistic, and ecologically meaningful ones – in terms of the stimuli and tasks performed. Testing large and heterogenous samples, we standardized two underused tests of facial identity matching: the Yearbook Test (YBT; Bruck et al., 1991) and the Face Identity Card Sorting Test (FICST; Jenkins et al., 2011). These procedurally simple tests mimic real-life challenges in face perception, as they assess unfamiliar facial identity matching across superficial image changes, or substantial age-related change in appearance. Beyond providing normative data, we describe how performance measured by these tests relates to that observed on more commonly used tests of face recognition (CFMT+; Russell et al., 2009) and perception (EFCT; PICT; White et al., 2015). Our findings suggest that (i) the YBT and FICST are easy-to-use, preferable alternatives to pairwise face matching tasks, which are prone to speed-accuracy trade-offs; (ii) challenging and ecologically valid tests should complement highly controlled measures when aiming to identify individuals with superior face processing abilities for real-life purposes.

show abstract

Predicting Face Recognition Performance in Unconstrained Environments

Cited by 2 publications

References 14 publications

Normative data for two challenging tests of face matching under ecological conditions

Normative data for two challenging tests of face matching under ecological conditions

Normative data for two tests of face matching under ecological conditions

Contact Info

Product

Resources

About