This study investigated eyewitness identification using ERPs. Twenty participants completed two eyewitness lineup tasks (standard and deception conditions). For the standard condition, participants tried to accurately identify the culprit, whereas in the deception condition, they were asked to deceptively conceal their recognition of the culprit. Identification rates based on P300 patterns were calculated using two different individual analysis procedures (A and B) that varied in stringency. Correct identification rates for the standard condition were 100% for both procedures A and B. For the deception condition, correct identification rates of the concealed culprit were 90%, and 70% respectively for procedures A and B. Data from a prior study [the culprit-absent condition from Lefebvre, C.D., Marchand, Y., Smith, S.M. & Connolly, J.F., 2007. Determining eyewitness identification accuracy using event-related brain potentials (ERPs). Psychophysiology, 44, 894–904.] was reanalysed to investigate differences in false identification rates based on procedures A and B. False identifications were substantially higher when using procedure A (29%) versus procedure B (0%). Overall, superiority was found for procedure B compared to procedure A based on Grier's A'.