Recent reports indicate that the subjective ability of humans to discriminate between polarization E-vector orientations approaches that of many invertebrates. Here, we show that polarization-modulated patterned stimuli generate an objectively recordable electrophysiological response in humans with normal vision. We investigated visual evoked potential (VEP) and electroretinographic (ERG) responses to checkerboard patterns defined solely by their polarization E-vector orientation alternating between ± 45°. Correcting for multiple comparisons, paired-samples t-tests were conducted to assess the significance of post-stimulus deflections from baseline measures of noise. Using standard check pattern sizes for clinical electrophysiology, and a pattern-reversal protocol, participants showed a VEP response to polarization-modulated patterns (PolVEP) with a prominent and consistent positive component near 150 ms (p < 0.01), followed by more variable negative components near 200 ms and 300 ms. The effect was unrecordable with visible wavelengths >550 nm. Further, pseudo-depolarization negated the responses, while control studies provided confirmatory evidence that the PolVEP response was not the product of luminance artefacts. Polarization-modulated patterns did not elicit a recordable ERG response. The possible origins of the PolVEP signals, and the absence of recordable ERG signals, are discussed. We conclude that evoked cortical responses to polarization-modulated patterns provide an objective measure of foveal function, suitable for both humans and non-human primates with equivalent macular anatomy.