A. House lights signaled the rat to nose poke into the center odor port and wait 500 ms before odor delivery. The odor indicating the size (large or small) of the reward to be delivered at the end of the trial. Odor presentation lasted 500ms and was followed by a 250–500 ms post-odor variable delay, which ended with the onset of directional cue lights. Directional lights flashed for 100 ms on either the left or right, instructing the rat to respond to the left or right fluid well, respectively. After the entering the correct fluid well rats were required to wait 500–1000 ms before reward delivery. B. There were four possible reward size and response direction combinations; large-left, large-right, small-left and small-right.

A. Latency to exit the odor port after directional cue lights had been extinguished. B. Percent correct scores as a function of all trials in which a choice was made to one of the fluid wells. C–D. Correlation between reaction time and percent correct scores for large and small reward trials, respectively. Asterisks: planned comparisons revealing statistically significant differences (t-test, p < 0.05). Error bars indicate standard errors of the mean (SEM). E. Location of recording sites. Gray dots represent final electrode position. Gray box marks extent of recording sites. NAc, Nucleus accumbens core. NAs, Nucleus accumbens, shell. N = 6 rats.

Lines represent the average firing rate over the course of the trial for all four conditions. Activity is aligned to odor onset. Each tick mark indications one action potential. Gray bars represent the analysis epoch, which is referred to as the ’odor epoch’. The odor epoch encompasses activity (750 ms) after odor onset prior to any knowledge about response direction (i.e. cueing of directional light).

A. Population firing based on preferred/non-preferred reward outcomes and direction (n = 229). In this plot, for each neuron, direction and outcome were referenced to the max response before averaging. Black lines correspond to the preferred outcome; gray corresponds to the non-preferred outcome. Thick lines represent the preferred direction; thin-dashed lines represent the non-preferred direction. Zero is the time of odor onset. Directional light cues were presented 750–1000 ms after odor onset. Gray bar indicates odor analysis epoch. B. Averaged normalized activity for each odor-responsive neuron. Hotter colors indicate higher firing. Each row represents activity of one neuron, which were sorted by firing in the preferred outcome/direction. C. Distribution of size indices determined by subtracting small reward from large reward trials and dividing by the sum of the two for activity during the odor epoch (gray bar in A). Black bars represent the number of neurons that showed a significant difference between large and small reward trial types during the odor epoch (ttest; p < 0.05).

Population firing under four different sequences of trials averaged across direction (n = 229). Table illustrates four trial types. Thick = pervious outcome preferred; Thin = previous outcome nonpreferred; Black = current trial preferred; Gray = current trial nonpreferred; Solid = No prediction; Dashed = predictable trial (Single-dashed = predicted preferred; Multi-dashed = predicted non-preferred). Insets represent average activity during the pre-odor (500 ms before odor) and the odor epoch (odor onset plus 750 ms) across these 4 conditions. Error bars, SEM. B–D: Left. Blown up versions of lines in ‘A’. Analysis comparing each pair are presented to right of the line histograms. B. Distribution reflecting the difference between pre-odor cue firing (500 ms; gray bar) when the predicted reward was to be large compared to when the reward was not predictable (predicted big minus no prediction; thin black dashed minus thin solid gray). Thus, positive values indicate higher firing when the predicted reward was large. Reward that preceded trials used in this analysis were small reward trials. C. Distribution reflecting the difference between pre-cue firing when the preceding reward was large compared to when the preceding reward was small (big minus small divided by big + small; Thick solid black minus thin solid gray). Thus, positive values indicate higher firing when the previous reward was large. The value of the reward on the current trial for this analysis could not be predicted (50/50). D. Distribution reflecting the difference between pre-cue firing when the preceding reward was large and the predicted reward was small compared to when the preceding reward was small and there was no prediction (Thick gray dashed minus thin solid gray). Black bars represent the number of neurons that showed a significant difference between respective comparisons (ttest; p < 0.05). E. Correlation analysis between distributions in B and C. F. Correlation analysis between distributions in B and D.

A–B. Correlation between firing rate (odor epoch) and reaction time (light offset to odor port exit) for large and small reward trials during all recording sessions. C–D. Correlation between firing rate and percent correct scores for large and small reward trials during all recording sessions.