People with atrial fibrillation are at an increased risk of stroke, cognitive decline and dementia. We recently identified that individuals with atrial fibrillation exhibit a reduced cerebrovascular reactivity to carbon dioxide, indicative of a diminished cerebral vasodilatory reserve. In this study we sought to determine whether neurovascular coupling (NVC) is blunted in atrial fibrillation in comparison with age-matched, hypertensive and healthy control participants. Posterior (PCA) and middle cerebral artery (MCA) flow velocity (Vm), along with beat-to-beat mean arterial pressure (MAP), were measured during a NVC assessment consisting of five cycles of visual stimulation (reading) for 30 s followed by 30 s with both eyes-closed in 12 patients with atrial fibrillation (69  yr; mean [SD], 3 women), 13 patients with hypertension (67  yr, 4 women) and 12 healthy control participants (63  yr, 4 women). Cerebrovascular conductance indices (CVCi) were calculated as PCA Vm/MAP and MCA Vm/MAP. NVC was denoted by the peak response observed during visual stimulation and is expressed as a percentage change from the eyes-closed period. Resting PCA Vm and MCA Vm were not different between the atrial fibrillation (31  and 49  cm/s), hypertension (33  and 55  cm/s), and healthy control (36  and 55  cm/s) groups (P>0.05). Visual stimulation evoked an increase in PCA CVCi in all groups, but the magnitude of the hyperemic response was blunted in patients with atrial fibrillation (18  %) and hypertension (17  %), in comparison with healthy controls (26  %) (P<0.05). Visual stimulation evoked a greater increase in MCA CVCi in the atrial fibrillation group (17  %), than patients with hypertension (10  %) and healthy controls (13  %) (P<0.05). Thus, when peak PCA CVCi responses were considered following subtraction of the peak MCA CVCi responses (i.e., the “non-selective” effect of the NVC test), a significantly diminished response was observed in the atrial fibrillation group (1  %) compared the hypertension (7  %) and healthy control groups (13  %) (P<0.01). In summary, individuals with atrial fibrillation exhibit a blunted NVC response to visual stimulation when compared to age-matched, healthy control and hypertensive participants. Additional studies are required to identify the mechanisms for the attenuated cerebral perfusion response of the occipital lobe to visual stimulation in atrial fibrillation (e.g., attenuated neuronal activation, altered cerebrovascular vasodilatory signaling).