Extended Data Fig. 6: Differences in SLIs cannot be explained by a biased sample of bimodal units.
From: Allothetic and idiothetic spatial cues control the multiplexed theta phase coding of place cells
a, Histograms of theta phase preference for example units classified as unimodal, bimodal, and unclassified ( > 2 peaks) units. Pyramidal cells in the deep layer of CA1 tend to exhibit strong firing in both the early and late phases of theta (bimodal cells), while cells in the superficial cell layer tend to fire only in the late phases of theta37,100 (unimodal cells). X-axis shows the theta phase repeated for two cycles. b, Number of units classified in each category. No significant difference in the distribution was observed among gain groups. This finding suggests that gain manipulation affects the relative size of the second lobe while sparing the modality of theta phase preference; for instance, bimodal units maintain their bimodality despite a smaller second lobe during gain manipulation. χ2(6) = 4.45, P = 0.616; chi-squared test. c, Smoothed phase coding plots for units classified as unimodal and bimodal cells. The plot for the bimodal cells showed a larger second lobe. d, Shuffling analysis to test if the difference in the sampling of unit modality can explain the observed SLIs. The decrease in SLIs in the low and high gain conditions was greater than the null distribution. The finding complements that of b, together demonstrating that the effect of gain on the size of the second lobe is not due to the change in the theta modality of the units or a biased sampling of unimodal/bimodal units across gain groups. Low: P <= 1.00 ×10−3, Mid: P = 0.229, High: P <= 1.00 ×10−3. Multiple comparisons correction was not applied. ***: P < 0.001.
