We recently investigated the binding properties of the antagonists [3H]‐raclopride and [3H]‐spiperone to intact Chinese hamster ovary cells expressing recombinant human D2long‐dopamine receptors (CHO‐D2L cells). Compared with saturation binding with [3H]‐raclopride, raclopride reduced [3H]‐spiperone binding with to low potency in competition binding experiments. The present findings illustrate the ability of spiperone to inhibit [3H]‐raclopride binding non‐competitively. While raclopride only decreases the apparent KD of [3H]‐raclopride in saturation binding experiments, spiperone only decreases the number of sites to which [3H]‐raclopride binds with high affinity. Also, while the IC50 of raclopride depends on the concentration of [3H]‐raclopride in competition experiments, this is not the case for spiperone. Kinetic studies reveal that the binding of raclopride at its high affinity sites does not affect the association of subsequently added [3H]‐spiperone nor the rebinding of freshly dissociated [3H]‐spiperone to the same or surrounding receptors. Yet, spiperone does not affect the dissociation rate of [3H]‐raclopride and raclopride does not affect the (genuine) dissociation rate of [3H]‐spiperone. The easiest way to interpret the present findings in molecular terms is to assume that D2L‐receptors or their dimeric complexes possess two distinct binding sites: one with high affinity/accessibility for [3H]‐raclopride and the other one with high affinity/accessibility for [3H]‐spiperone. The ability of bound spiperone to inhibit high affinity raclopride binding while the reverse is not the case suggests for the occurrence of non‐reciprocal allosteric interactions. These new findings could point at the occurrence of allosteric interactions between different classes of D2‐receptor antagonists.