Understanding the document bias in face matching.

Xinran Feng,A Mike Burton

doi:10.1177/17470218211017902

Xinran Feng, A Mike Burton

Open Access

https://doi.org/10.1177/17470218211017902

Copy DOI

Abstract

Matching unfamiliar faces is a well-studied task, apparently capturing important everyday decisions such as ID checks. In typical laboratory studies, participants make same/different judgements to pairs of faces, presented in isolation and without context. However, it has recently become clear that matching faces embedded in documents (e.g., passports and driving licences) induces a bias, resulting in elevated levels of “same person” responses. While practically important, it remains unclear whether this bias arises due to expectations induced by the ID cards or interference between textual information and faces. Here, we observe the same bias when faces are embedded in blank (i.e., non-authoritative) cards carrying basic personal information, but not when the same information is presented alongside a face without the card (Experiments 1 and 2). Cards bearing unreadable text (blurred or in an unfamiliar alphabet) do not induce the bias, but those bearing arbitrary (non-biographical) words do (Experiments 3 and 4). The results suggest a complex basis for the effect, relying on multiple factors which happen to converge in photo-ID.

Highlights

A large body of psychological evidence shows that viewers are highly error-prone when matching images of unfamiliar faces (e.g., Bruce et al, 1999; Megreya & Burton, 2006)
Viewers are generally unaware of their poor performance levels when matching unfamiliar faces (Zhou & Jenkins, 2020)
We explore the nature of this effect, focussing on the task-irrelevant written information available to viewers in a face matching task

Summary

Introduction

A large body of psychological evidence shows that viewers are highly error-prone when matching images of unfamiliar faces (e.g., Bruce et al, 1999; Megreya & Burton, 2006). McCaffery and Burton (2016) directly compared the performance of viewers matching pairs of face images presented as isolated images or with one embedded in a passport This simple manipulation did not affect overall accuracy but did significantly affect bias: participants were more likely to accept a pair of images as a “match” if one was embedded in a passport. Feng and Burton (2019) replicated the “passport bias” and showed that it occurred in other documents, such as driving licences and student ID cards—even though the latter are considered to carry much lower authority than passports They found that the bias was only present when the ID card carried personal information. Comparison of the performance on standard isolated face matching and the minimal card context will establish a baseline ID effect, independently of expectations induced by specific contexts, such as passports, driving licences, or workplace IDs

Participants

Design and procedure

Results and discussion