Background: Public spaces are usually designed with respect to various patient populations, but not Parkinson's disease. The objective of this study was to explore what type of easily applicable visual cueing might be used in public spaces and some interiors to improve gait in people with Parkinson's disease. Methods: Thirty-two patients with freezing of gait walked an 8-meter track on 6 different floor patterns in single- and dual-task conditions in random sequence. The reference pattern was a virtual large transverse chessboard, and the other patterns differed either in size (small floor stones), orientation (diagonal), nature (real paving), regularity (irregular), or no pattern. Time, number of steps, velocity, step length, cadence, and dual-task effect were calculated. The number and total duration of freezing episodes were analyzed. Results: Virtual, large, transverse floor stones improve time (P = 0.0101), velocity (P = 0.0029), number of steps (P = 0.0291), and step length (P = 0.0254) in Parkinson's disease patients compared with walking on no pattern. Virtual floor stones were superior in time and velocity to the real ones. Transverse floor stones were better than diagonal, whereas regular pattern stones were superior to irregular in some gait parameters. Subjectively, the reference pattern was preferred to the irregular one and to no pattern. No direct effect on freezing of gait was observed. Conclusions: Parkinson's disease patients may benefit from floor patterns incorporating transverse oriented large rectangular visual cues. Because public space can be regulated with respect to people with medical conditions, the relevant legislative documents should be extended to allow for parkinsonian gait disorder. © 2019 International Parkinson and Movement Disorder Society. © 2019 International Parkinson and Movement Disorder Society