**References** Lefebvre, C. D., Marchand, Y., Smith, S. M., & Connolly, J. F. (2007). Determining eyewitness identification accuracy using event‐related brain potentials (ERPs). Psychophysiology, 44(6), 894-904. Lefebvre, C. D., Marchand, Y., Smith, S. M., & Connolly, J. F. (2009). Use of event-related brain potentials (ERPs) to assess eyewitness accuracy and deception. International Journal of Psychophysiology, 73(3), 218-225. Meixner, J. B., & Rosenfeld, J. P. (2014). Detecting knowledge of incidentally acquired, real-world memories using a P300-based concealed-information test. Psychological science, 25(11), 994-2005. Tanaka, J. W., Curran, T., Porterfield, A. L., & Collins, D. (2006). Activation of pre-existing and acquired face representations: the N250 event-related potential as an index of face familiarity. Journal of Cognitive Neuroscience, 18(9), 1488-1497. **Rationale for the study and design** Our goal is to devise a CIT-style test that would be of immediate use to police, especially in situations involving uncooperative eyewitnesses. In criminal investigations, it is very difficult to prosecute the individual responsible when there are no eyewitnesses, and even when eyewitnesses step up and make an identification, they are notoriously unreliable. Although some eyewitnesses genuinely make mistakes in suspect identification, other eyewitnesses seek to intentionally deceive police. This could be because they are either trying to protect a family member or friend from prosecution, or their hatred for the criminal justice system and police drive them to obstruct investigations. Uncooperative eyewitnesses pose a threat to police officers because even with specialized training, there is no way to accurately detect deception. Future applications of this study would compare p300 waveforms to overt responses to face stimuli and lineups; deception would be detected when waveforms suggesting recognition do not elicit the same outright response from the participants. The Concealed Information Test (CIT) works in a variety of situations, and was applied in the form of eyewitness identification by Lefebvre et al twice, but not perfectly. A weakness in the researcher’s 2 studies stemmed from their use of group-wise comparisons. Since criminal investigations are often conducted on a one-on-one basis, using group-wise comparisons in research and trying to apply them to individual-wise and real-life situations is difficult. Therefore, this study will use individual-wise comparisons to distinguish a single participant’s targets from foils. This deception detection research would involve the ability to distinguish a single participant’s targets from foils rather than using group-wise comparisons. The transition from traditional group-wise comparisons to individual-wise comparisons is key to increasing the external validity of applied memory and cognition research. *We hypothesize that:* 1. Participants will demonstrate recognition for the culprit and for the Joe face, compared to the foils. 2. This recognition will be evident when observing p300 ERP waveforms, and recognition will still be shown when participants are asked to lie. *Data collection done to date:* Pilot study run November 2016 [Pilot Breakdown][1] [Pilot Data (raw data also uploaded)][2] *Collaborators:* Mario Baldassari, Stephen Lindsay **Timeline** Brainstorming started in January 2016, an undergraduate's independent study created first prototype of the stimulus presentation program in EPrime in spring 2016. The program was tweaked through to the fall and an initial pilot sample without EEG recording was collected in November 2016. A pilot with EEG was conducted in spring 2017, and the main experiment will be run from May to September. **Participants** Undergraduates at the University of Victoria *Want enough participants for statistical power, and to account for future technical difficulties and exclusion parameters. 25 participants in each condition, based on the 20 recruited for Lefebvre, who manipulated lying within s's and had 40 exposures to the culprit. *Exclusion Parameters:* 1. Individuals who are instructed to lie via button-presses but do not (these may be recoded into the truth condition where possible) 2. Individuals who answer one of the final, simultaneous lineups incorrectly 3. Individuals who profess to or seem to the experimenter to be not paying attention, falling asleep, etc. (will be decided before viewing data) 4. Individuals in the truth-telling condition who fail to ID the culprit more than 80% of trials (possibly lower % based on discretion after seeing # of p's eliminated but before viewing summary data) 5. Trials with missing data (RT, ERP due to artifacts) 6. Trials in which participants make an unintended response **Procedure** Participants will watch a short crime video, then watch it a second time while narrating it aloud. Both times, they will be instructed to pay attention to the culprit "Because we will ask you to identify him later." The purpose of the study will be explained, and if we are in the later stages of the experiment (e.g. Experiment 2 or later), an EEG cap will be affixed. Following a Joe/No Joe paradigm adapted from Tanaka et al. (2006), participants will be randomly assigned a face from the group and asked to study it for as long as they like. Then, in a series of 6 yes/no recognition memory practice trials (4 cycles) they will be instructed to identify the face (called “Chris”). If they fail to identify Chris more than once or produce more than one false alarm, this phase will be repeated. After minimum performance is successfully reached, participants will be given one more opportunity to study Chris and advance at their own discretion. In the next phase, a set of 12 new faces matching the description of the culprit (none of which were used during Chris training) will be cycled 40 times with appearances by Chris and the culprit in each cycle. Participants will be told that there will be two faces they will recognize, Chris and the culprit, and to press a button every time they see them. Each trial will begin with a fixation cross, with a varying inter-stimulus interval between the cross onset and the face onset of 650-850ms. Faces will appear on the screen for 1.5 seconds; a gray screen will appear after the response is given and will be replaced by a new fixation cross after 1s. Reaction times will be measured from the time a face stimulus appears on the screen to the time the button is pressed. Face order will be different for each cycle based on a fixed random order that will show differently for subjects in a 12-cycle set. All lineups will contain the culprit. Participants in the first few pilot tests will be asked to ID the culprit, after which the main sample of participants will all be asked to “Imagine that the culprit is a friend who you want to save from going to jail. Thus, when you see him, you will not press the button to identify him.” To avoid risk of muting EEG amplitude for the second presentation, faces will not be repeated without at least 2 intervening faces. (Schweinberger, Pickering, Jentzsch, Burton, & Kaufmann (2002); Trenner, Schweinberger, Jentzsch, & Sommer, 2004). Upon completion of the 40 cycles, participants will be presented with a classic culprit-present simultaneous lineup for the learned face "Chris" and a separate culprit-present lineup for the criminal. Participants who were asked to lie will not be asked to do so any longer, thus we will treat these identifications as proof that the witness did, in fact, recognize both faces during the previous task. *Pilot Procedure* Participants in the pilot study will watch the video with the experimenter in the room and will be instructed to narrate it out loud. After a second viewing in which they are instructed to watch the criminals face “because you will be asked to identify him later,” they will complete the rest of the procedure as outlined above except that we will not collect their EEG data. A second pilot study will be conducted later, in which participants’ reaction times will be analyzed for indications of truth or lying. **Planned Analyses** Pilot study: Overall and individual average RT's for culprit, Chris, and foils. Additional analyses of accuracy to determine that exposure conditions are sufficient to learn the faces. Main study: Comparison of bootstrapped P300 (amplitude or area) a la analyses in Meixner & Rosenfeld (2014). **To be included in this registration** - Videos - Faces for test phase and EEG observation phase - Lineups - EPrime run files Contact Information: For more information please contact Mario Baldassari (mjbldssr@uvic.ca) or Steve Lindsay (slindsay@uvic.ca). [1]: https://osf.io/s786d/ [2]: https://osf.io/85bc5/