Hippocampal CA1 place cells encode intended destination on a maze with multiple choice points

Abstract

The hippocampus encodes both spatial and nonspatial aspects of a rat's ongoing behavior at the single-cell level. In this study, we examined the encoding of intended destination by hippocampal (CA1) place cells during performance of a serial reversal task on a double Y-maze. On the maze, rats had to make two choices to access one of four possible goal locations, two of which contained reward. Reward locations were kept constant within blocks of 10 trials but changed between blocks, and the session of each day comprised three or more trial blocks. A disproportionate number of place fields were observed in the start box and beginning stem of the maze, relative to other locations on the maze. Forty-six percent of these place fields had different firing rates on journeys to different goal boxes. Another group of cells had place fields before the second choice point, and, of these, 44% differentiated between journeys to specific goal boxes. In a second experiment, we observed that rats with hippocampal damage made significantly more errors than control rats on the Y-maze when reward locations were reversed. Together, these results suggest that, at the start of the maze, the hippocampus encodes both current location and the intended destination of the rat, and this encoding is necessary for the flexible response to changes in reinforcement contingencies.

Figure 1.

A, Schematic representation of the concatenated Y-maze. In pretraining, two of the four goal boxes contained reward in each block of 10 trials. In this example, boxes 2 and 4 are rewarded in the first 10 trials and boxes 1 and 3 in the second 10 trials. B, Representation of the binned areas used to examine the influence of goal destination on place cell firing. C, Example of the paths taken by a rat on 20 consecutive trials. The paths are ballistic and reflect little hesitation at choice points. The rat returns to the same goal on every trial until it is not rewarded and then immediately chooses another box. If that box is rewarded, it returns to this location in subsequent trials. This illustrates the ability of the rat to learn in one trial that the reward location has changed. D, Frequency distribution of place fields on the maze. Higher peaks and warmer colors indicate higher numbers of place fields.

Figure 2.

Electrode placement. The arrow indicates the mark of the electrode tip in the dorsal hippocampus.

Figure 3.

CA1 place cells in the start box encode intended destination. A, Four examples (1 on each row) of cells with place fields in the start box that fired predominantly on journeys to one of the four goal boxes. The left column shows all of the paths for a single recording session, with red dots indicating the spikes from one neuron. The shaded gray box is the place field assessed for intended trajectory. The middle left column shows the data separated into journeys to each goal box. The middle right column shows the average firing rate of the cell in the start box on journeys to each of the goal boxes. The right column shows the cluster, waveforms, and autocorrelogram of the cell. Calibration: horizontal lines, 300 μs; vertical lines, 100 μV. B, An example of a cell that had similar firing rates in the start box on journeys to all goal boxes.

Figure 4.

CA1 place cells before the final choice point encode intended destination. A, Two examples (1 on each row) of CA1 place cells with place fields before the final choice point that fired predominantly on trials to one of the two possible goals. The left column shows all of the trials from a single session, with the spikes from an individual neuron represented as red dots. The shaded gray box indicates the area of the maze with the place field of interest. The middle left column shows journeys to each goal box separately. The middle right column shows the average firing rate in the gray shaded area for journeys to the two possible goal boxes. The right column shows the cluster, waveforms, and autocorrelogram of the cell. Calibration: horizontal lines, 300 μs; vertical lines, 100 μV. B, An example of a cell that had similar firing rates on journeys to both goal boxes.

Figure 5.

A, Proportion of place fields in the start of the maze (areas 1 and 2) and after the start areas, but before the second choice point (top pie chart). These are broken down into the proportions of goal-sensitive and goal-nonsensitive cells in each of these two regions (bottom pie chart). B, Distribution of F ratios for goal-sensitive and goal-nonsensitive fields in the start box and first common stem of the maze (areas 1 and 2).

Figure 6.

Place field changes within a session. A, Five examples (1 on each row) of cells that developed robust place fields after several trials. The left column shows the whole session, with red dots indicating spikes fired by an individual neuron. The right column shows the individual trials to a specific goal box (the first one is on the left). B, Two examples (1 on each row) of cells that initially had robust place fields but ceased to fire at some point during the session. The left column shows the whole session, with red dots indicating spikes fired by an individual neuron. The right column shows the individual trials to a specific goal box (the first one is on the left). C, Two examples of cells (1 on each row) whose fields changed locations over maze trials.

Figure 7.

Performance on the double Y-maze task. A, The mean ± SEM number of correct choices made by the hippocampal lesion and control group in the first 10 trials of each session (reward locations constant) in the 15 postsurgery sessions (for clarity, the data are presented in 3-session blocks). The dashed line indicates the chance performance level. B, The mean ± SEM number of correct choices made in the second 10 trials of the same sessions after the reversal in reward location in the same session blocks. C, The mean ± SEM number of correct choices made in the first 20 trials of each session (reward locations constant) by the hippocampal lesion and control group in the three 40-trial testing sessions. D, The mean ± SEM number of correct choices made in the second 20 trials of the same sessions after the reversal in reward location. E, Average performance of the hippocampal lesion and control groups on each trial. Each data point is the mean ± SEM percentage of correct choices for a given trial across the 15 postsurgery testing sessions.

Figure 8.

Performance on the alternation task. The mean ± SEM number of correct choices (of 10) in the 21 training sessions for both the hippocampal lesion and control groups are plotted. For clarity, the data are presented in blocks of three sessions. The dashed line is the number of correct responses expected by chance.