Philadelphia Face Perception Battery (PFPB)

Assessment of face perception ability

A test of facial perception using naturalistic stimuli that minimizes reliance on memory and motor requirements to complete successfully. The battery consists of four tests of facial perception:

• Facial similarity: The participant is presented with three faces, one presented in the center top half of the screen while the other two are presented in the lower half of the screen side-by-side. The participant is instructed to choose which of the lower two faces is most similar (alike) to the face shown at the top of the screen. Responses are entered by using the computer mouse to click on the face that is determined to be more similar.
• Facial beauty: The participant is presented with two faces in the vertical center, side-by-side and is asked to choose which is the more attractive. Responses are made by using the computer mouse to click on the face that is determined to be the more attractive face.
• Face age: The participant is presented with two faces in the vertical center, side-by-side and is asked to choose which of the two faces is older. Responses are made using a computer mouse and clicking over the face determined to be older.
• Face gender: The participant is presented with one face in the top 2/3 of the screen and the words “male” and “female” in separate boxes in the bottom 1/3 of the screen. The participant is asked to determine if the face shown is male or female and use the computer mouse to click on the corresponding box.

Each test consists of 75 items which are presented in order of ascending difficulty. The tests are administered using an E-Prime script. The order of tests is randomized by the script for each administration.

E-prime, created by Psychology Software Tools Inc., is the shell program that is required to run the PFPB. A demo version of this shell program is available for piloting the program or running a very few patients. However, if a large number of patients/subjects will need to be tested, the full version will be required and this is a fairly large investment. Please see the link above for more information.

Instructions for each test are displayed by the program prior to administration. However, some participants may need further instructions or rewording of the instructions as shown below.

• Facial similarity: “You will see three faces on the computer screen. One will be located at the top (point to the top middle of the screen) and the other two will be located at the bottom of the screen (point to the bottom half of the screen). You must pick which of these two looks more like the one at the top. Use the mouse to click on that one.”
• Facial beauty: “You will see two faces on the computer screen. Pick which one you believe is more attractive or beautiful. Use the mouse to click on that face.”
• Face age: No additional instructions should be required.
• Face gender: No additional instructions should be required.

The program allows the administrator to discontinue the current test and continue on to the next one if the participant becomes unduly frustrated. Do this by pressing <shift> + <e> keys at the same time on the keyboard.

Approximate Time for Administration: 30-45 minutes

Use the provided Excel spreadsheet named “PFPB Analysis” on www.cfn.upenn.edu/aguirre/public/PFPB. Copy the last four columns of data located in the participant’s E-Prime data file and paste them in the colored columns of the spreadsheet in the order they are found in the data file. Update the participant’s initials/code, age, and education level. The accuracies and standard deviations are shown in the table next to the subject information. Standard deviations are calculated in respect to the control population described in Thomas et al (2007) and are corrected for the age and education level effects they found. Note: Make sure to delete all responses from a previous subject before copying and pasting a new subject's data. If a test must be stopped early (using the <shift> + <e> to exit) the data file will be shorter than other subjects who complete every trial and the past subject's data may remain after cutting and pasting.

The tests are freely available for non-commercial use. Do not distribute modified versions of the tests without the permission of the authors. Cite Thomas et al (2007) in any scientific application of the tests.

AL Thomas, K Lawler, IR Olson, GK Aguirre. (2008). Philadelphia Face Perception Battery (PFPB). Archives of Clinical Neuropsychology, 23(2), 175-187.