Stress can impair cognitive performance, as commonly observed in cognitive performance anxiety (CPA; e.g., test anxiety). Cognitive theories indicate that stress impairs performance by increasing attention to negative thoughts, a phenomenon also known as threat-interference. These theories are mainly supported by findings related to self-report measures of threat-interference or trait anxiety. Our main aim was to test, for the first time in a single study, the hypotheses that acute CPA-related stress negatively affects both working memory (WM) performance and objectively assessed threat-interference during performance. In addition, we aimed to assess the validity of a new stress-induction procedure that was developed to induce acute CPA. Eighty-six females were randomly assigned to a CPA-related stress group (n = 45) or a control group. WM performance and threat-interference were assessed with an n-back task (2-back and 3-back memory loads), using CPA-related words as distracters. The stress group showed higher state anxiety and slower WM performance. Both effects were moderated by trait CPA: the effects were stronger for individuals with higher trait CPA. Finally, trait CPA moderated the effect of stress on threat-interference during higher cognitive load: individuals with higher trait CPA in the stress group showed higher threat-interference. We conclude that acute CPA increases threat-interference and impairs WM performance, especially in vulnerable individuals. The role of threat-interference, cognitive load, and trait anxiety should be taken into account in future research. Finally, our method (combining our stressor and modified n-back task) is effective for studying stress-cognition interactions in CPA.