Thus, this neuron was excited when the monkey had to attend to the sample and store SKI-606 supplier it in working memory, but it showed little response to the same stimulus

when it was no longer behaviorally relevant. As a population, the sample response was significantly positive in both the large and the small reward trials in the DMS task (n = 66, p < 0.01, Wilcoxon signed-rank test) (Figure 4B, left), while it was not significantly different from zero in the control task (n = 50, p > 0.05, Wilcoxon signed-rank test) (Figure 4B, right). We reanalyzed the sample response in the DMS task using the same set of neurons across the two tasks (n = 50). The response was still significantly positive in the DMS task (p < 0.01, Wilcoxon signed-rank test). Even at the single-neuron level, 23 of the 66 neurons showed a significant excitation to the sample in the DMS task (21 neurons in the large reward trials, 12 neurons in the small reward trials, and 10 neurons in both of them) (p < 0.05, Wilcoxon signed-rank test). Their averaged activity showed an excitation to the sample for each reward size (Figure 4C, left), and the magnitude of the excitation was significantly larger in the large reward trials than in the small reward trials (large reward trials, mean ± SD = 2.4 ± 1.0 spikes/s; small reward trials, mean ± SD = 1.6 ± www.selleckchem.com/products/pf-06463922.html 1.3 spikes/s;

p = 0.014, Wilcoxon signed-rank test). Of the 23 neurons, 15 were also examined using the control task. Their averaged activity in the control showed little response to the sample (Figure 4C, right). Comparing the sample responses in the two tasks for each neuron (Figure 4D), the magnitude was significantly larger in the DMS task than

in the control task during the large reward trials (p < 0.01, Wilcoxon signed-rank test), with a similar trend occurring during the small reward trials (p = 0.19, Wilcoxon signed-rank test). The above data indicate that a group of dopamine neurons was excited by the sample aminophylline if the monkey had to retain the information about the sample in working memory. The activity of these neurons only reflected the need to use the information about the sample, not the specific information to be retained in working memory as follows. First, most of the neurons (18/23) did not represent the orientation of sample bar, which was the information that the monkey had to remember (p > 0.05, two-way ANOVA). Second, these neurons responded to the sample only phasically and did not show a persistent activation that would be necessary to retain the information during the delay period (Figure S2). These response patterns make a striking contrast with the object-selective and persistent firing of dorsolateral prefrontal neurons that have long been implicated in working memory (Rao et al., 1997 and Wilson et al., 1993). We found that only a subset of dopamine neurons signaled the sample information.