Reunion behavior following stressful separations from caregivers is often considered the single most sensitive clue to infant attachment patterns. Extending these ideas to middle childhood/early adolescence, we examined participants' neural responses to reunion with peers who had previously excluded them. We recorded event‐related potentials among nineteen 11‐ to 15‐year‐old youth previously classified on attachment interviews (11 secure and 8 insecure‐dismissing) while they played a virtual ball‐toss game (Cyberball) with peers that involved fair play, exclusion and reunion phases. Compared to secure participants, dismissing participants displayed a greater increment in the N2 during reunion relative to fair play, a neural marker commonly linked to expectancy violation. These data suggest a greater tendency toward continued expectations of rejection among dismissing children, even after cessation of social exclusion. In turn, the link between self‐reported ostracism distress and neural signs of negative expectancy at reunion was moderated by attachment, such that self‐reports were discordant with the neural index of expectancy violation for dismissing, but not for secure children. A video abstract of this article can be viewed at http://www.youtube.com/watch?v=qzoRel5c-4s. Reunion behavior following stressful separations from caregivers is often considered the single most sensitive clue to infant attachment patterns. Extending these ideas to middle childhood/early adolescence, we examined participants' neural responses to reunion with peers who had previously excluded them. We recorded event‐related potentials among 19 11‐to15‐year‐old youth previously classified on attachment interviews (11 secure and 8 insecure‐dismissing) while they played a virtual ball‐toss game (Cyberball) with peers that involved fair play, exclusion and reunion phases.