1 Aim: Much debate exists about how the brain transitions into an epileptic seizure. One 2 source of confusion is that there are likely to be critical differences between experimental 3 seizure models. To address this, we compared the evolving activity patterns in two, widely 4 used, in vitro models of epileptic discharges. 5Methods: We compared brain slices, prepared in the same way from young adult mice, 6 that were bathed either in 0 Mg 2+ , or 100µM 4AP, artificial cerebrospinal fluid. 7Results: We find that while local field potential recordings of epileptiform discharges in 8 the two models appear broadly similar, patch-clamp analysis reveals an important 9 difference in the relative degree of glutamatergic involvement. 4AP affects parvalbumin-10 expressing interneurons more than other cortical populations, destabilizing their resting 11 state and inducing spontaneous bursting behavior. Consequently, the most prominent 12 pattern of transient discharge ("interictal event") in this model is almost purely 13 GABAergic, although the transition to seizure-like events (SLEs) involves pyramidal 14 recruitment. In contrast, interictal discharges in 0 Mg 2+ are only maintained by a very 15 large glutamatergic component that also involves transient discharges of the interneurons. 16 Seizure-like events in 0 Mg 2+ have significantly higher power in the high gamma 17 frequency band (60-120Hz) than these events do in 4AP, and are greatly delayed in onset 18 by diazepam, unlike 4AP events. 19 Conclusions: The 0 Mg 2+ and 4AP models display fundamentally different levels of 20 glutamatergic drive, demonstrating how ostensibly similar pathological discharges can 21 arise from different sources. We contend that similar interpretative issues will also be 22 relevant to clinical practice. 23 24 Slice preparation: Male and female Emx1-Cre (B6.129S2-Emx1tm1(cre)Krj/J; Jackson 11 Laboratory stock number 5628), PV-Cre (B6;129P2-Pvalb/J; Jackson 12 Laboratory stock number 8069), and SOM-Cre (B6N.Cg.Ssttm2.1(cre)Zjh/J; Jackson 13 Laboratory stock number 18973) mice and C57/B6 mice (ages 3 -12 weeks) were used in this 14 study. Transgenic mice were back-crossed with the C57/B6 line maintained at Newcastle 15 University, and subsequently maintained on this C57/B6 background (Jackson Laboratory stock 16 number 000664). Mice were housed in individually ventilated cages in a 12 hours light, 12 17 hours dark lighting regime. Animals received food and water ad libitum. Mice were sacrificed 18 by cervical dislocation, brains removed and stored in cold cutting solution (in mM): 3 MgCl 2 ; 19126 NaCl; 26 NaHCO 3 ; 3.5 KCl; 1.26 NaH 2 PO 4 ; 10 glucose. For local field potential (LFP) 20 recordings, 400µm horizontal sections were made, using a Leica VT1200 vibratome (Nussloch, 21 Germany). Slices were then transferred to an interface holding chamber and incubated for 1-2 22 hours at room temperature in artificial CSF (ACSF) containing (in mM): 2 CaCl 2 ; 1 MgCl 2 ; 126 23 NaCl; 26 NaHCO 3 ; 3.5 KCl; 1.26 NaH 2 PO 4 ; 10 glu...