The association of phenylboronic acid (no unpaired electron, compound 1) with the free radical phenyl nitronyl nitroxide (PNN, S = 1/2, compound 2) constitutes an interheteromolecular hydrogen bonding system displaying ferromagnetic intermolecular interactions. We have investigated its spin density distribution to visualize the pathway of these magnetic interactions. This complex crystallizes at room temperature in the monoclinic space group P2(1/n). The unit cell includes one pair (1 + 2). The molecule (1) bridges two radicals (2) by hydrogen bonds OH ON: the two different hydrogen bond lengths are quite similar (1.95 and 1.92 Angstrom). Infinite chains of this run along the b-axis. In this structure the methyl groups of the PNN are randomly distributed in two different configurations. Below T = 220 K the compound undergoes a crystallographic phase transition due to the ordering of these methyl groups. We have determined the low-temperature structure using both X-ray and neutron diffraction. The new space group is pi. The global structure is preserved and infinite chains still run along the b-axis, but the unit cell now comprises two different pairs (1 + 2) instead of one, with four different hydrogen bond OH ON distances: 1.96 and 1.84 Angstrom for the first pair, 1.96 and 1.91 A for the second pair. The spin density of this complex was measured at T = 1.8 K (H = 4.6 T) by polarized neutron diffraction. The data were treated using both maximum entropy approach and wave function modeling. As in the isolated PNN, the main part of the spin density is located on the O-N-C-N-O fragment of each radical in the unit cell. However, compared to the isolated case, a significant difference exists: a large unbalance is observed between the two oxygen atoms of each radical. Moreover, a positive contribution is found on the two hydrogen atoms involved on the OH ON hydrogen bonds of each phenylboronic acid molecule. The stronger contribution corresponds to the longer hydrogen bonds. On the radical the stronger reduction is observed on the oxygen atoms involved in the shorter hydrogen bonds. The experimental results are compared to those obtained by density functional theory (DFT) calculations: on the whole, the experimental effects have been reproduced. However, if there is a good qualitative agreement, from the quantitative point of view, the DFT results are still: very far from the experimental ones.