1. Surgically resected tissue from the tip of the human temporal lobe of seven patients undergoing temporal lobectomy was employed to study functional properties of GABAergic inhibition mediated through activation of GABAA receptors, using patch-clamp recording techniques in acutely isolated neurons and in slices of human temporal cortex. 2. Human temporal cortical pyramidal neurons from surgically resected tissue could be acutely isolated with the use of conventional methods. These neurons appeared normal in morphology, in their intrinsic membrane properties, and in their response to application of exogenous gamma-aminobutyric acid (GABA). 3. Application of GABA to acutely isolated human temporal cortical neurons elicited a large current with an average reversal potential of -65 mV, presumably mediated through a GABAA-activated chloride conductance. Application of varying concentrations of GABA generated a concentration/response relationship that could be well-fitted by a conventional sigmoidal curve, with an EC50 of 25.5 microM and a Hill coefficient of 1.0 4. Coapplication of the benzodiazepine clonazepam and 10 microM GABA augmented the amplitude of the GABA response. The concentration dependence of this benzodiazepine augmentation could be best-fitted by an equation assuming that the benzodiazepine interacted with two distinct binding sites, with differing potencies. The high-potency site had an EC50 of 0.06 nM and maximally contributed 38.5% augmentation to the total effect of clonazepam. The lower potency site had an EC50 of 16.4 nM, and contributed 66.1% maximal augmentation to the overall effect of clonazepam. These data derived from adult human temporal cortical neurons were very similar to our findings in adult rat sensory cortical neurons. 5. The effects of equimolar concentrations (100 nM) of clonazepam, a BZ1 and BZ2 agonist, and zolpidem, a selective BZ1 agonist, on acutely isolated human temporal cortical neurons were also investigated. Zolpidem and clonazepam were equally effective (71.5 vs. 65.0%, respectively) in potentiating GABA responses elicited by application of 10 microM GABA. This suggests that many of the functional benzodiazepine receptors in these neurons were of the BZ1 variety. 6. GABAergic synaptic inhibition was also studied with the use of patch-clamp recordings in slices of human temporal cortex. Extracellular stimulation at the white matter/gray matter border elicited compound synaptic events in layer II-V cortical neurons. These events usually consisted of an early excitatory postsynaptic potential (EPSP) and a late multiphasic inhibitory postsynaptic potential (IPSP). Application of either clonazepam or zolpidem (both at 100 nM) to the slice during extracellular stimulation reversibly augmented the late compound IPSP. 7. Spontaneous IPSPs were also recorded in approximately 50% of human temporal cortical neurons. These events did not have a preceding EPSP and were usually monopolar, with a single exponential rise and decay. This supported the idea that these events were triggered by spontaneous activity of GABAergic interneurons. Bath application of either clonazepam or zolpidem (both at 100nM) to the slice during ongoing spontaneous IPSP activity increased the amplitude and lengthened the time constant of decay of these events. 8. To our knowledge, this is one of the first detailed characterizations of the functional properties of GABAA-mediated inhibition in human cortical neurons using patch-clamp recordings in both isolated cells and slices of resected temporal cortex. Isolated pyramidal neurons exhibited GABAA-mediated currents that were comparable in many aspects with GABA currents recorded from adult rat cortical neurons, including similar GABA concentration/response curves, and similar two differing potency site effects for clonazepam augmentation of GABA currents. In addition, evoked and spontaneous IPSPs recorded in human cortical neurons appeared similar to IPSPs in rat cortical