Active place avoidance task

An individual animal was placed on a metal 82-cm-diameter disk-shaped arena (Figure 2A). The arena was elevated 76 cm above the floor and centered in a 3 × 4 m² room surrounded by opaque curtains and various items. The animal was constrained to remain on the disk by a 40-cm-high transparent wall that allowed the animal to see the surrounding room through the wall. The position of the animal was tracked every 17 ms from an overhead camera using digital video spot tracking software (Tracker; Bio-Signal Group).

PCP impairs cognitive control. Shown are the schematics of the active place avoidance tasks on a continuously rotating arena that creates two distinct frames of spatial information, one defined by stationary room cues and another defined by rotating arena cues. A, In the two-frame, Room+Arena− task variant, the rat must avoid a stationary shock zone (red) by using (✓) the stream of relevant “Room” information from the stationary frame while ignoring (X) the stream of irrelevant “Arena” information from the rotating frame. Using one class of information while ignoring another requires cognitive control. B, In the one-frame, Room+ task variant, the irrelevant Arena information can be minimized by shallow water to attenuate the arena cues for place localization and reduce the demand for cognitive control. C, Inset, Experimental design. Rats were trained for several days on one task variant. After optimal performance, the rats received systemic administration of either a control or PCP dose 30 min before testing. The moderate dose of PCP causes hyperlocomotion, as measured by the distance that rat walked during both task variants. D, Inset, Representative paths from a single rat; red indicates the shock zone and shocks. The moderate dose of PCP impairs familiar two-frame (n = 10), but not one-frame (n = 4) place avoidance, as measured by the number of errors and entrances to the shock zone. E, PCP did not change responses to shock, measured as the change from the initial experience of shock before PCP treatment. The number of shocks/entrance estimates the rat's willingness to escape shock. F, Infusing PCP into both hippocampi. Inset, Experimental design, n = 11. Intrahippocampal PCP does not cause hyperlocomotion. G, The injection but is sufficient to impair familiar two-frame Room+Arena− place avoidance. Inset, Representative paths from a single rat. H, Room+Arena− avoidance deficit after intrahippocampal infusion was related to the magnitude of the change in medium-frequency gamma oscillation amplitude in the hippocampal LFP, but not the changes in theta or slow gamma oscillation amplitudes (n = 8). Correlations and regression lines are for fits only to the PCP data. Error bars indicate ± SEM. Control, low, moderate, 0, 3, 5 mg/kg PCP doses, respectively. *p < 0.05 relative to all groups. Bold type indicates significant correlations.

Both a two-frame Room+Arena− task variant (Fig. 2A) and a one-frame Room+ task variant (Fig. 2B) were used. In both tasks, the arena rotated at 1 rpm and an unmarked 60° shock zone was fixed in stationary room coordinates. To avoid shock, the animal had to use the relevant stationary distal room cues to localize itself and the positions of shock. Because the arena was rotating, the animal also had to ignore the irrelevant information from the rotating spatial frame. In the Room+Arena− task variant, the irrelevant cues were scents and fecal boli that were stable on the arena surface, but in the Room+ variant, their salience was reduced by shallow water (∼2 cm saline) on the arena surface (Wesierska et al., 2005). Therefore, in the Room+Arena− task variant, the demand for cognitive control to ignore the salient irrelevant arena cues was high but the demand was low in the Room+ task. All other cognitive and motivational aspects of the two tasks were similar. Avoidance was reinforced by a constant current foot-shock (0.3 mA, 50 or 60 Hz, 500 ms). The shock was delivered across the rat on the grounded arena surface via a low impedance stainless steel shock electrode that was implanted subcutaneously between the rat's shoulders so the major voltage drop was across the rat's paws. Under control of software, shock was delivered when the animal was in the shock zone for 500 ms. The shock was repeated every 1.5 s until the animal left the shock zone. The spatial resolution of the position measurements was ∼3 mm.

The goal was to test the effect of PCP on active place avoidance after the task had been learned and the memories established. Rats were first habituated to the environment during a 10 min trial on the stationary arena with no shock. The shock and arena rotation were then turned on and rats were given 5 10 min training trials with 10 min intertrial interval. The 5 trials/d training protocol continued on subsequent days until the rat's performance reached a criterion number of shock zone entries, which was ≤2 in 2 consecutive trials. The day after reaching criterion, each rat received a dose of PCP by either systemic injection or intrahippocampal infusion 30 min before testing in conditions that were identical to the training. The rats had a wash-out period of at least 48 h between different doses of PCP.

The time series of the animal's positions was analyzed. The distance that the animals walked was used to quantify locomotion. This measure increased after systemic PCP administration because the drug causes hyperlocomotion. The number of entrances into the shock zone was counted to estimate place avoidance. This measure decreases as the animals learn to avoid shock and can be interpreted as an index of the ability to perform the conditioned avoidance and judicious retrieval of the place avoidance memory in both task variants (Wesierska et al., 2005; Lee et al., 2012). To estimate the animal's willingness to escape shock under PCP, we measured the number of shocks that were received each time the shock zone was entered. Because this measure depends on the subject's prior experience with shock, we computed the difference in the shocks/entrance during the PCP trial compared with the initial trial before any experience with shock and PCP. In addition to comparing the measure of place avoidance in the high and low cognitive control task variants, this analysis of shocks/entrance served as additional evidence for evaluating whether PCP impaired place avoidance per se or merely the willingness to avoid shock itself.