Purpose: To date no systematic method has been used for characterising the residual visual capacity of blindsight subjects that would allow comparison and generalisation across all subjects. We aimed to provide such comparison and apply it to a large number of cortically blind patients to address both the question of prevalence and characteristics of blindsight. Methods: Detection of suprathreshold gratings in a two alternative forced-choice paradigm provides a criterion free measure of spatial processing within the field defect. The presence of a band-pass spatial channel has previously been reported in one blindsight subject. We have applied the same methodology to study eight cortically blind subjects with V1 lesions. Magnetic Resonance Imaging was used to determine the extent of brain damage in all cases. Results: We have demonstrated the presence of blindsight in seven out of eight subjects tested. The eighth subject had a relative large lesion extending outside of V1. The spatial channels mediating blindsight appear to have a band-pass frequency response characteristics in two subjects and low-pass in five subjects. The higher cut-off frequencies in all cases were below 4 cycle/degree. The lower cut-off frequency of the band-pass channel was approximately 0.5 cycle/degree. Conclusion: Blindsight in cortically blind subjects may not be a rare phenomenon. Our study also indicates that the spatial channels giving rise to blindsight performance are tuned to low spatial frequencies, consistent with mediation by subcortical processes.