, 2000, 2004; Kim and Frank, 2009). Animals were

allowed to behave freely and were never forced to choose a particular trajectory. Errors were not rewarded, and after an incorrect choice of an outer arm, no reward was given until the animal returned to the center arm. Recordings began on the first day of exposure to T1. Animals ran on T1 for 3 days and then ran on both T1 and T2 from day 4 onward. Behavioral data were divided into four performance categories, based on the animals’ performance on each session. These categories roughly separate sessions into periods of (1) initial exposure to the task, (2) early learning, (3) early good performance, and (4) later good performance. The categories divided the sessions into (1) the first session animals performed at less than 65% correct, (2) the first session Vorinostat the animal performed between 65% and 85% correct, (3) the first session animals performed above 85% correct, and (4) subsequent sessions animals performed above 85%. Less than 65% was selected for the Lumacaftor mouse first category because all animals performed

at less than 65% on the first exposure to the task, the first session in T1. Above 85% was selected for the third and fourth category because all animals were able to perform the task in T1 at above 85% after many days of training. Because categories 1–3 are only for the first session in which animals reach the criterion, only one session per animal could be included in each category. Since more than one session per animal could be included in category 4, only the first such session per day was used to avoid counting cell pairs more than once per category. Data from all animals were included through exposure ten for T1 and exposure seven for T2. Exposures past these were excluded

because they represented data from three or fewer of the five animals. To detect SWRs, we recorded local field potentials (LFPs) from one channel of each tetrode, and SWRs were detected on all tetrodes in CA1. The LFP signal from these tetrodes was band-pass filtered between 150 and 250 Hz, and the envelope was determined by Hilbert transform. SWR events were detected if the envelope exceeded a threshold of mean plus three standard deviations for at least 15 ms on any tetrode in CA1. Events included times around the triggering event during which the envelope 4-Aminobutyrate aminotransferase exceeded the mean. We examined SWRs when animals were within 20 cm of the center well moving at a linear speed less than 1 cm/s. We also defined two measures to determine which cells to include in the analysis. Coactivation probability per SWR was the number of SWRs in which both cells in a pair were active, divided by the total number of SWRs. Activation probability per SWR was the number of SWRs in which a single cell was active divided by the total number of SWRs. Only cell pairs with coactivation of at least 0.01 or single cells with activation of at least 0.