GFMT2: A psychometric measure of face matching ability

Abstract

We present an expanded version of a widely used measure of unfamiliar face matching ability, the Glasgow Face Matching Test (GFMT). The GFMT2 is created using the same source database as the original test but makes five key improvements. First, the test items include variation in head angle, pose, expression and subject-to-camera distance, making the new test more difficult and more representative of challenges in everyday face identification tasks. Second, short and long versions of the test each contain two forms that are calibrated to be of equal difficulty, allowing repeat tests to be performed to examine effects of training interventions. Third, the short form tests contain no repeating face identities, thereby removing any confounding effects of familiarity that may have been present in the original test. Fourth, separate short versions are created to target exceptionally high performing or exceptionally low performing individuals using established psychometric principles. Fifth, all tests are implemented in an executable program, allowing them to be administered automatically. All tests are available free for scientific use via www.gfmt2.org.

Match me if you can: Evidence for a domain-general visual comparison ability

“…Participants completed 80 face comparison trials (40 match and 40 nonmatch) from the Glasgow Face Matching Task 2-Short Form (GFMT2-S; White et al, 2021). The GFMT2-S is an updated version of the GFMT that was created to be more difficult and representative of real-world face identification tasks (i.e., variation in head angle, pose, expression, and image quality) than the original task.…”

A new approach to diagnosing and researching developmental prosopagnosia: Excluded cases are impaired too

“…Also, why are they reporting problems in daily life but not on our objective tests? Could another measure beyond the CFMT, FFT and CFPT be more accurate in capturing their daily difficulties with faces, e.g., Prosopagnosia Index (Shah et al, 2015 ), the Oxford Face Matching Task (Stantic et al, 2022 ), Glasgow Face Matching Test (Burton et al, 2010 ; White et al, 2022 ), Benton Test (Murray et al, 2021 ; Rossion & Michel, 2018 ; Wang et al, 2020 ) or UNSW Faces Test (Dunn et al, 2020 )? All these questions are unanswered because excluded cases are rarely reported upon by researchers.…”

Statistical feature training improves fingerprint-matching accuracy in novices and professional fingerprint examiners

“…Face comparison Participants completed a standardized test of face comparison as a baseline control task: the Glasgow Face-Matching Task-2 (GFMT2-SA and SB; White et al, 2021 , p. 2; see upper panel of Fig. 1 ) where participants view two faces side by side and were asked ‘are these images of the same person or two different people?’ on each trial.…”

“…We then selected the 80 trials (40 match and 40 non-match) that had the highest item-to-test correlations and then divided the trials into two equally difficult versions of the test (comprising 20 match and 20 non-match trials each). We used this method as it identifies trials that are most predictive of overall test performance and provides an overall estimate of a trial’s contribution to test reliability (Guilford, 1954 ; see also White et al, 2021 and Wilmer et al, 2012 ).…”

“…The fingerprint comparison trials from previous experiments were used, and we also added extra trials to create three fingerprint tasks for the pre-training, post-domain-general training, and post-domain-specific training phases with equal numbers of trials in each. To do so, we divided the trials from Experiment 1 ( n = 80) and 10 additional trials (the 10 next highest item-to-test correlations from the Growns and Kukucka ( 2021 ) pilot data; Guilford ( 1954 ); see also White et al ( 2021 ) and Wilmer et al ( 2012 ) into three equally difficult tasks ( N = 90; 30 trials per task). It is important to note that the rare minutiae may not have occurred in this task at the exact same frequencies as they do in the general population (e.g.…”