The CFPT was created to assess face identity perception by minimizing memory demands (Duchaine et al., 2007). The design was inspired by the Farnsworth-Munsell color sorting task. Participants are presented with a target face and must sort six faces based on their similarity to the target faces. The six faces are morphs that have different percentages of the target face in them. Half of the trials involve upright faces; half involve inverted faces. The test generates large inversion effects and is strongly correlated with the CFMT (Bowles et al., 2009).
Duchaine, B., Germine, L., & Nakayama, K. (2007). Family resemblance: Ten family members. with prosopagnosia and within-class object agnosia. Cognitive Neuropsychology, 24: 419-430.
Bowles, D., McKone, E., Dawel, A., Duchaine, B., Schmalzl, L., Palermo, R., Wilson, C.E., & Rivolta, D. (2009). Diagnosing prosopagnosia: Effects of aging and participant-stimulus ethnic match on the Cambridge Face Memory Test and Cambridge Face Perception Test. Cognitive Neuropsychology, 26: 423-455
Lucia Garrido developed the Films Test to assess facial expression recognition in a more ecologically valid manner than the other expression tests we used. It is called the Films Test because Lucia selected stills from films that depicted natural expressions (Garrido et al., 2009).
Garrido, L., Furl, N., Draganski, B., Weiskopf, N., Stevens, J., Tan, G.C-Y., Driver, J., Dolan, R., & Duchaine, B. (2009). VBM reveals reduced gray matter volume in the temporal cortex of developmental prosopagnosics. Brain, 132: 3443-3455.
Participants are presented with three images simultaneously and must select the image with a face in it. The images are arrays of two-tone stimuli. Half the trials include an upright face and half an inverted face. The inversion effects are massive (around 50%).
Participants are presented with three face photos simultaneously and must select the image in which the face has been Thatcherized (eyes, mouth, or eyes & mouth flipped vertically relative to the rest of the face). Trials are half upright and half inverted. Like the previous task, the inversion effects are really strong (about 45%).