In the human, visually evoked potentials (VEP) and cerebral oxygenation changes, as measured by near-infrared spectroscopy (NIRS), are assessed to elucidate the relation between electrophysiological and vascular responses to a checkerboard stimulus reversing at 3 Hz. Habituation of either response is analysed on two time scales. Within the 1-min stimulation period we find a decrease in P100N135-component amplitude, closely coupled to a decrease in the amplitude of the oxygenation parameters (concentration changes in oxygenated and deoxygenated haemoglobin, [oxy-Hb] and [deoxy-Hb]). The N75P100-component amplitude exhibits a different behaviour along the 1-min stimulation period. An initial increase is overridden by an overall decrease, the latter not reaching statistical significance. The analysis across the 13 successive stimulation blocks separated by resting periods of equal duration yields a trend for an decrease in the VEP-components' amplitude, not reflected in the vascular response. When calculating a ratio between the amplitude of the P100N135-component and the concentration changes in the haemoglobins a "coupling index" of a 0.2 microM decrease in [deoxy-Hb] and an increase of 0.6 microM in [oxy-Hb] is found per 1 microV increase in VEP-component amplitude. The ratio is the same irrespective of its assessment from the difference stimulation/ rest or from the habituation effect, i.e., the difference between the amplitudes at the beginning and towards the end of the stimulation period. Although supporting the notion that the coupling between neuronal activation and the vascular response exhibits linear aspects, our findings cannot be taken as a proof of such a linearity across all brain regions and activation types. On the contrary, tentatively calculating a coupling index for the data assessed in the visual system, we intend to stress the necessity to assess both neuronal and vascular response to allow for a comparison between different systems and conditions in whom neurovascular coupling is expected to be altered (Miller et al, 2001; Mechelli et al, 2001).