Deepfake Detection in Super-Recognizers and Police Officers
Using videos from the Deepfake Detection Challenge (cf. Groh et al., 2021), we investigated human deepfake detection performance (DDP) in two unique observer groups: Super-Recognizers (SRs) and "normal" officers from within the 18K members of the Berlin Police. SRs were identified either via previously proposed lab-based procedures (Ramon, 2021) or the only existing tool for SR identification involving increasingly challenging, authentic forensic material: beSure® (Berlin Test For Super-Recognizer Identification; Ramon & Rjosk, 2022). Across two experiments we examined deepfake detection performance (DDP) in participants who judged single videos and pairs of videos in a 2AFC decision setting. We explored speed-accuracy trade-offs in DDP, compared DDP between lab-identified SRs and non-SRs, and police officers whose face identity processing skills had been extensively tested using challenging. In this talk I will discuss our surprising findings and argue that further work is needed too determine whether face identity processing is related to DDP or not.
Characterising the brain representations behind variations in real-world visual behaviour
Not all individuals are equally competent at recognizing the faces they interact with. Revealing how the brains of different individuals support variations in this ability is a crucial step to develop an understanding of real-world human visual behaviour. In this talk, I will present findings from a large high-density EEG dataset (>100k trials of participants processing various stimulus categories) and computational approaches which aimed to characterise the brain representations behind real-world proficiency of “super-recognizers”—individuals at the top of face recognition ability spectrum. Using decoding analysis of time-resolved EEG patterns, we predicted with high precision the trial-by-trial activity of super-recognizers participants, and showed that evidence for face recognition ability variations is disseminated along early, intermediate and late brain processing steps. Computational modeling of the underlying brain activity uncovered two representational signatures supporting higher face recognition ability—i) mid-level visual & ii) semantic computations. Both components were dissociable in brain processing-time (the first around the N170, the last around the P600) and levels of computations (the first emerging from mid-level layers of visual Convolutional Neural Networks, the last from a semantic model characterising sentence descriptions of images). I will conclude by presenting ongoing analyses from a well-known case of acquired prosopagnosia (PS) using similar computational modeling of high-density EEG activity.
Super-Recognizers: facts, fallacies, and the future
Over the past decade, the domain of face identity processing has seen a surging interest in inter-individual differences, with a focus on individuals with superior skills, so-called Super-Recognizers (SRs; Ramon et al., 2019; Russell et al., 2009). Their study can provide valuable insights into brain-behavior relationships and advance our understanding of neural functioning. Despite a decade of research, and similarly to the field of developmental prosopagnosia, a consensus on diagnostic criteria for SR identification is lacking. Consequently, SRs are currently identified either inconsistently, via suboptimal individual tests, or via undocumented collections of tests. This state of the field has two major implications. Firstly, our scientific understanding of SRs will remain at best limited. Secondly, the needs of government agencies interested in deploying SRs for real-life identity verification (e.g., policing) are unlikely to be met. To counteract these issues, I suggest the following action points. Firstly, based on our and others’ work suggesting novel and challenging tests of face cognition (Bobak et al., 2019; Fysh et al., in press; Stacchi et al., 2019), and my collaborations with international security agencies, I recommend novel diagnostic criteria for SR identification. These are currently being used to screen the Berlin State Police’s >25K employees before identifying SRs via bespoke testing procedures we have collaboratively developed over the past years. Secondly, I introduce a cohort of SRs identified using these criteria, which is being studied in-depth using behavioral methods, psychophysics, eye-tracking, and neuroimaging. Finally, I suggest data acquired for these individuals should be curated to develop and share best practices with researchers and practitioners, and to gain an accurate and transparent description of SR cases to exploit their informative value.