Fig. 3-S3

Note that fish were imaged ~20 min after embedding, thus initial food deprivation time is already 20 min. Hence, the initial reduction in LH active cell count, which occurs within 30 min ( Figure 2) may not be observable using this imaging method. ( a) Fish 1 and 2 were imaged using volumetric imaging for 115 min, whereas fish 3 and 4 were imaged only at a single plane, and for a slightly shorter time period of 90 min (see images in ( b)) (i): Mean Δf/f across the entire (both lobes) of the cH, mLH and lLH (i.e. raw) show increases in baseline fluorescence over time. (ii) Mean Δf/f with baseline subtracted (i.e. detrended). Since a rising baseline over long imaging periods is difficult to interpret (see text for discussion), we also display detrended traces. ( b) (i): Average intensity projection images showing imaged regions with ROIs outlined. (ii) Spike-triggered averages based on extracted lLH calcium spikes (from detrended traces) usually reveal an accompanying reduction in cH calcium fluorescence (Δf/f). ( c) Calcium spike frequency (spikes/min, left) and calcium spike amplitude (Δf/f %, right) for each ROI averaged over 5 min bins throughout the imaging session for the above four fish. Colored lines are the means, shaded areas reflects SEM. ( d) Over the entire imaging period, calcium spike frequency (left) was significantly higher in the cH as compared to the mLH (p = 0.014) and lLH (p = 0.014). Calcium spike amplitude (right) was also significantly higher in the cH as compared to the mLH (p = 0.014), but not the lLH (p = 0.057), asterisks denote p<0.05, one-tailed Wilcoxon rank-sum test.