--- - |2 Abstract In item‐method directed forgetting, worse memory performance occurs for to‐be‐forgotten (TBF) than for to‐be‐remembered (TBR) items. However, recently TBF items have been found to be recognized more accurately than uninformative (UI) items not associated with any specific task. Here, we compare ERPs elicited by cues signaling the remember or the forget instruction with those elicited by uninformative cues. Participants were presented with a series of complex pictures, each followed by a symbolic TBR, TBF, or UI cue. On a yes–no recognition test, accuracy was higher for TBR than for both TBF and UI items. However, TBF items were recognized more accurately than UI items. ERPs elicited by the TBR cue were larger than ERPs elicited by both TBF and UI cues at the frontal P2, the late frontal negativity, and the late parietal positivity complex. This is consistent with more attention capture, selective rehearsal, and long‐term memory encoding of TBR. In contrast, both TBF and UI cues induced a larger frontal N2 than TBR cues. Critically, TBF elicited a larger late right‐frontal positivity than both UI and TBR cues and a larger late parietal positivity than UI cues. Moreover, the late right‐frontal positivity was correlated with better recognition performance. It may therefore reflect processing orientation rather than inhibition per se. Results suggest that, when compared to UI cues, both TBR and TBF cues are processed actively. Distinct processing is seen at right frontal and centroparietal positive ERPs that may mediate better subsequent recognition of TBF than of UI items. - Psychophysiology, Volume 55, Issue 10, October 2018.