In the learner group analyzed 24 hr after training, the peak of the signals in the left telencephalon appeared later than that in the left optic tectum (Figure 2A, fourth row of the right panel, blue and pink lines). For the statistical comparison, the peak times for each Metformin mouse individual were identified from a curve fitted by the least-squares method and compared between the left tectum, and the left and right telencephalon (Figure 2B). We found a significant difference between the peak time of the tectum and that of the left or right telencephalon (Figure 2B, left telencephalon: p = 0.0350, unpaired t test,

right telencephalon: p = 0.0044, unpaired t test). There was no significant difference between the left and right telencephalon (p = 0.6516, unpaired

t test). The peak time of the left tectum activity in learners measured 24 hr after the last training (LTM) is not significantly different from that of four other conditions: cue-alone, shock-alone, cue-shock unpaired, and learner 30 min (STM) (Figure 2C). Figure 2D graphically illustrates the C59 datasheet fluorescence changes in the learner and control fish in each telencephalic hemisphere. The learner fish showed an increased fluorescence change and therefore response to cue presentation in both telencephalic hemispheres (Figure 2D, pink line). There were no obvious peaks in fluorescence changes in control fish (Figure 2D, green line). Altogether, we could identify the calcium signals specifically related to the long-term memory for learned avoidance

behavior in zebrafish telencephalon. To investigate whether the activated areas are necessary for the learning and retrieval of the behavioral program, we performed bilateral ablation of these areas before and after the training (Figures 3A1 and 3A2). The ablation site was determined based on the average coordinates of the activity centers in the imaging experiments (Figure 3E, ablated, black circles; see Experimental Procedures). In the control group, we ablated the anterior-most region of the telencephalon (Figure 3E, sham, open circles). The ablation sites were confirmed by Nissl staining after testing was complete (Figure 3B). We first performed the operation medroxyprogesterone on the first day, trained fish on the second day, and examined the memory retrieval in a short-term period (STM, 30 min after the last training) and a long-term period (LTM, 24 hr after the last training) (Figure 3A1). As shown in Figure 3C, the activated area-ablated fish could reach learning criterion in the active avoidance paradigm during training as efficiently as sham-operated animals (Figure 3C, sham S1 versus abl S1, p = 0.2642 unpaired t test; sham S3 versus abl S3, p = 0.3659, unpaired t test). Moreover, when the memory retrieval was tested 30 min after the last training, the ablated fish could remember the task as well as the sham-operated fish (Figure 3C, sham STM versus abl STM, p = 0.2604, unpaired t test).