Pusher behaviour (PB) reflects some misrepresentation of verticality. However, its neural mechanisms are still unclear. The aim of this pilot study is to assess the perception of the subjective visual vertical in patients with PB using an orientation discrimination task relying on a psychophysical forced-choice procedure. A sample of eight patients with post-stroke hemiplegia, three of whom with (PB+ group) and five without (PB– group) a clear PB, and 10 matched healthy subjects, was selected. All participants were assessed with an orientation discrimination task based on the objective Two-Alternatives Forced Choice (2AFC) procedure, in which observers are forced to report whether an oriented stimulus that was tilted off-vertical by a varying amount (but never vertical) was tilted clockwise or counterclockwise from vertical. Participants' ability in detecting the visual vertical was estimated by calculation of bias and threshold, which represent deviations of the subjective vertical from the physical vertical and the sensitivity to orientation offsets away from vertical, respectively. While there were no significant differences between groups in terms of bias, statistical analysis showed a significantly higher threshold in the PB+ group compared to both PB– and control groups. Results suggest that vertical misrepresentation might be due to the presence, in patients with PB, of a lower signal-to-noise ratio in coding systems. Implications for clinical practice are discussed.