In the eyewitness identification literature, stress and arousal at the time of encoding are considered to adversely influence identification performance. This assumption is in contrast with findings from the neurobiology field of learning and memory, showing that stress and stress hormones are critically involved in forming enduring memories. This discrepancy may be related to methodological differences between the two fields of research, such as the tendency for immediate testing or the use of very short (1–2 hours) retention intervals in eyewitness research, while neurobiology studies insert at least 24 hours. Other differences refer to the extent to which stress‐responsive systems (i.e., the hypothalamic–pituitary–adrenal axis) are stimulated effectively under laboratory conditions. The aim of the current study was to conduct an experiment that accounts for the contemporary state of knowledge in both fields. In all, 123 participants witnessed a live staged theft while being exposed to a laboratory stressor that reliably elicits autonomic and glucocorticoid stress responses or while performing a control task. Salivary cortisol levels were measured to control for the effectiveness of the stress induction. One week later, participants attempted to identify the thief from target‐present and target‐absent line‐ups. According to regression and receiver operating characteristic analyses, stress did not have robust detrimental effects on identification performance.