Figure 1: Schematics of the experimental set-up and characterization of the single photon quantum-optical light source.

(a) Top: schematic of the set-up for generating single photons via SPDC. Light stimuli are triggered (T) by subjects and coupled into a single-mode fibre entering the dark chamber. The light is directed and focused onto the pupil (Maxwellian view) at an angle of 23° temporal to a highly attenuated red fixation light presented at the fovea. BBO, beta-barium borate crystal; F, bandpass and spatial filter; HWP, half-wave-plate; L, lens; ND, neutral density filter; PBS, polarizing beamsplitter. A flip mirror (FM) allowed for the switching of the detection mechanism from the EMCCD to a SPAD (semi-transparent, see Methods section for details). Bottom: schematic of the 2AFC protocol. A 1 ms light pulse is presented together with either the first or the second acoustic signal (intervals 1 and 2) that are separated by 800 ms. After the second acoustic signal, the subject gives their answer (1, 2) and a confidence rating (R1, R2 and R3; see Methods section for details.) (b) Numerical simulation of multiple-to-single photon ratio in SPDC as a function of mean photon pairs per pulse (that is, at different pump laser intensities). While the multiple-to-single photon ratio decreases as a function of mean photon pairs for both detection schemes, the EMCCD (solid line) exhibits an improved multiple-to-single ratio compared with the SPAD (dotted line), as the multi-pixel based detection in case of EMCCD allows for identification and of two and multi-photon events and their rejection. (c) Trade-off between multiple-to-single photon ratio and required number of experimental trials. Multiple-to-single photon ratio in SPDC using EMCCD-based detection as a function of mean photon pairs per pulse (black solid line – same as in b) exhibits an inverse relationship with the number of trials required to discriminate the performance of an ideal detector from random chance level (0.5) with a statistically significant probability (95% confidence interval; grey dotted line). The vertical grey bar indicates the mean number of photon pairs per pump pulse that was used in the experiments (see Methods section for details).