Halide perovskite solar cells demonstrated a dramatic increase in efficiency in the last decade mainly thanks to improvements in the bulk material quality, as well as the choice of suitable selective carrier transport layers (SCTLs). More recently, improvements have come from interface optimization strategies, such as defect passivation. Nevertheless, the interactions between the perovskite (PSK) and SCTLs are still unclear and will become more and more important to improve both efficiency and stability of PSK solar cells.In our study, we quantified the optoelectronic interactions at the SCTL-PSK interfaces employing a lateral heterojunction device (LHJ) (Figure 1A). The device is composed of a glass substrate, on which two different SCTLs (TiOx and NiOx) are deposited in a lateral electrode configuration with a well-defined 100 µm – long channel between them. The two SCTLs sit on two separate ITO electrodes that serve as contacts. We measured a ≈600 meV work function (WF) difference between the SCTLs, which constitutes the built-in voltage (Vbi) of our device. To complete the structure, a Pb-based perovskite layer is deposited on top, thus filling the channel and creating a p-i-n device which develops in the horizontal direction, i.e. in the sample plane. We tracked the surface electrostatic potential variation along the PSK channel employing high-spatial-resolution (≈11 µm) X-ray photoemission spectroscopy (XPS) both in dark short circuit, and under applied bias, by recording the I3d5/2 core level binding energy (BEI) of the PSK.We discovered that the substrate built-in voltage was still present on top of the PSK surface: a linear potential drop (Vdrop) of about 600 meV was recorded across the channel by XPS in dark, short circuit conditions, implying a nearly undoped PSK. Indeed, by fitting the data with a 2D drift-diffusion (DD) model, we found Ndop=1011 cm-3 to be the maximum PSK doping density able to sustain such a long-range linear drop (Figure 1B).Furthermore, we could control (and even invert) the potential drop by applying a voltage bias (Vapp) to the electrodes, which adds up to Vbi (Vdrop= Vbi- Vapp) (Figure 1C). Our results indicate that the tested PSK is nearly intrinsic and its Fermi level is controlled by the substrate over which it is deposited, and can be modulated by an external bias voltage.