This study investigated diurnal non-image forming (NIF) effects of illuminance level on physiological arousal in parallel to NIF effects on vigilance and working memory performance. We employed a counterbalanced within-subjects design in which thirty-nine participants (mean age = 21.2; SD = 2.1; 11 male) completed three 90-min sessions (165 vs. 600 lx vs. 1700 lx at eye level) either in the morning (N = 18) or afternoon (N = 21). During each session, participants completed four measurement blocks (incl. one baseline block) each consisting of a 10-min Psychomotor Vigilance Task (PVT) and a Backwards Digit-Span Task (BDST) including easy trials (4–6 digits) and difficult trials (7–8 digits). Heart rate (HR), skin conductance level (SCL) and systolic blood pressure (SBP) were measured continuously.The results revealed significant improvements in performance on the BDST difficult trials under 1700 lx vs. 165 lx (p = 0.01), while illuminance level did not affect performance on the PVT and BDST easy trials. Illuminance level impacted HR and SCL, but not SBP. In the afternoon sessions, HR was significantly higher under 1700 lx vs. 165 lx during PVT performance (p = 0.05), while during BDST performance, HR was only slightly higher under 600 vs. 165 lx (p = 0.06). SCL was significantly higher under 1700 lx vs. 165 lx during performance on BDST easy trials (p = 0.02) and showed similar, but nonsignificant trends during the PVT and BDST difficult trials. Although both physiology and performance were affected by illuminance level, no consistent pattern emerged with respect to parallel changes in physiology and performance. Rather, physiology and performance seemed to be affected independently, via unique pathways.