Abstract Phencyclidine (PCP), a non-competitive N-methyl-D-aspartate receptor antagonist, is known to produce schizophrenia-like psychosis in humans, including positive and negative symptoms as well as cognitive dysfunction. Moreover, acute administration of PCP can emulate corresponding behavioral symptoms in rodents. We investigated the effect of PCP and the possible rescuing potential of typical and atypical antipsychotic drugs (APDs) in vitro on spontaneously active neuronal networks. To this end, murine primary cortical cells were cultured on microelectrode arrays (MEAs). Concentration-response curves of PCP ranging from 0.01 to 200 µM were generated and network spike and burst rate as well as burst peak firing rate (PFR) and burst duration was measured in stable two-minute recordings. Measurements were done with and without pre-incubation with the APDs aripiprazole, clozapine, and haloperidol. We found a concentration-dependent network activity suppression reflected by a decrease in captured spike rate and network PFR upon PCP application relative to baseline. Preexposure with any of the three APDs mediated a right-shift of the PCP concentration-response curve (spike rate, PFR). However, as assessed by their IC50 values and Hill coefficients, the atypical APDs aripiprazole and clozapine exhibited a 20- to 30-fold protective potency—higher than the typical APD haloperidol (6-fold). In summary, a disruptive network effect of PCP as well as a protection by APDs could be demonstrated in the order of potency: clozapine > aripiprazole >> haloperidol. We propose this simple, noninvasive setup as a plausible electrophysiological model for testing current and future pharmaceuticals against schizophrenia-spectrum disorders.