Matching and Mismatching Manipulated and Unmanipulated Passport Photographs in a Six-Second Time Frame
Nicholas Panayotakos
Ryerson University
Dr. Julia Spaniol
Based on the study of Benedikt Emanuel Wirth and Claus-Christian Carbon (2017) we conducted a similar, but shorter, experiment in a classroom of 71 participants. As the participants, our job was to observe two passport photographs of unfamiliar faces, side by side, in a six-second time frame period and we were to determine if the individuals portrayed in the pictures were matching or not. The photographs were either manipulated using three different styles of manipulation (paraphernalia, distinct features, hairstyles), or there were entirely unmanipulated. This process was repeated 48 times, for a maximum of 24 match trials and 24 mismatch trials.
Keywords: unfamiliar faces, match and mismatch, manipulated and unmanipulated passport photographs, six-second time frame

Matching and Mismatching Manipulated and Unmanipulated Passport Photographs in a Six-Second Time Frame
In our daily lives, we are ought to, rather frequently, provide proof of who we are. This occurs in several situations, usually, situations where we focus on security more than customary, such as cross-continental flights, crossing borders, or even entering government buildings (e.g. embassies). Our methods of identification, thus far, function through photographs that also fulfill certain requirements (e.g. lighting, face orientation) and are accompanied by our biographical data (e.g. passport). Therefore, providing proof of who we are is playing a vital role in our lives and, if not done correctly, countless negative results may come forth (e.g. illegal immigration, fraud).
Our condensed experiment, which was adjusted for our classroom, used the Wirth and Carbon (2017) study, titled "The effects of professional experience and time pressure on passport-matching performance", as a foundation in an attempt to replicate its results. In this report, I will elaborate on the findings of three studies as well as the experiment that was conducted in our classroom and, also, I'll provide a synopsis of our goal and the methodology used in our experiment. My main focus will be the Wirth and Carbon (2017) study, followed by the study conducted by Jennifer M. McCaffery and Mike A. Burton (2016), and lastly, a study by David White, Richard I. Kemp, Rob Jenkins, Michael Matheson, and Mike A. Burton (2014). All three studies have a common emphasis, the difficulty of matching unfamiliar faces, specifically, in scenarios such as airport security checks.
Although in the Wirth and Carbon (2017) study it was indicated that police officers were far better in determining matches and mismatches, — yet, there was still a substantial error margin — the McCaffery and Burton (2016), and White et. al (2014) studies showed no significant increase in accuracy in comparison to various individuals who also took part in the studies that had no professional experience in unfamiliar face matching. Our classroom experiment showed similar results, with no significant difference between our participants and police officers, particularly in the mismatch results.
The Wirth and Carbon (2017) study was split into two experiments. One focused entirely on matches and mismatches of unfamiliar faces, and the other followed the exact same structure but with a difference in the methodology. The difference was that there was the addition of a time limit (25min to check all passengers). As noted in the study, the most significant result from the first experiment was that all of the participants were relatively tolerant with their decisions (Wirth & Carbon, 2017) even though the consequences could end up being highly severe.
The experiment conducted in our classroom, although it had striking similarities (e.g. we used the exact same photographs) with the study of Wirth and Carbon (2017), it was adjusted accordingly to fit the properties of a classroom (e.g. limited time and resources). Another similarity, however, was that we used the same methodology of one the experiments that used a local time-limit, again, for classroom efficiency reasons. In our case, just a mentioned in the abstract, it was six seconds for every pair of photographs.
The McCaffery and Burton (2016) study conducted three experiments. Firstly, participants had to determine if a face matched with a passport containing biographical data accompanied by a second face next to the frame. The data could be valid or invalid. Secondly, the participants are presented with three types of data,