The fitted sigmoid curves were used to compare distributions with a Kolmogorov-Smirnov test. The learning rates were estimated from the slopes of the sigmoid curves. Epigenetics Compound Library molecular weight The duration of the learning impairment after SCH23390 was measured as the sum of the duration of postinjection blocks showing slower learning curves and

smaller learning rates than baseline blocks. Electrode penetration sites were determined using MRI scans obtained before surgery. The recording chamber was positioned stereotaxically over the left lateral PFC of each animal overlying the principal sulcus (i.e., the dorsolateral and ventrolateral portions of the PFC were equally accessible). The location of the principal sulcus could also be mapped out neurophysiologically (absence of cells and low-amplitude LFP signals). Electrophysiological signals were recorded simultaneously from 7–15 dura-puncturing tungsten microelectrodes (FHC Instruments), located 1 or 2 mm away from the

infusion cannula. Electrodes were lowered each day either independently or in pairs and were advanced using custom-made screw-driven minimicrodrives mounted on a plastic grid (Crist Instruments) with spacing of 1 mm between adjacent locations. Neuronal activity was amplified, filtered, and stored using an integrated multichannel recording click here system (Plexon Neurotechnology Research Systems). To minimize any sampling bias of neuronal activity, we did not prescreen activity for any visual responsiveness. Electrodes and cannula were advanced until the activity of neurons was isolated well from several electrodes, and then data collection through began.

From each electrode, we simultaneously recorded spiking activity and the LFP. Both signals were referenced to ground. The spike signal (passband 154–8.8 kHz) was threshold triggered to separate neuronal spikes from background noise, and individual spike waveforms were stored at 40 kHz. LFPs (passband 0.7–300 Hz) were recorded continuously with a sampling rate of 1 kHz. Postinjection blocks were classified as washout when learning was unimpaired (not different from baseline). We note that this was not dependent on a literal washout of the drug or a recovery of neural activity to the baseline state. The dopamine D1-like receptor antagonist SCH23390 was purchased from Sigma/RBI and dissolved in commercially available sterile saline (0.9% NaCl) at 10 μg/μl under strict sterile conditions and stored at −20°C. The pH was corrected to be around 6.0. For control experiments, we used commercially available sterile saline (pH 5.5). The day of the recording, an aliquot of SCH23390 was thawed. A plastic tube (Tygon microbore) was chemically sterilized and connected to a sterile cannula that had been previously attached to a microdrive on the recording grid. Cannulas were Hamilton needles (30 GA, inner diameter 0.16 mm and outer diameter 0.31 mm) with bevels of 45°.