Fig. 3

Empirical quantile–quantile (Q–Q) plots of the normalized spike count distributions at retrieval for remembered and forgotten items. As in Figs. 1 and 2, the distributions consisted of normalized spike counts from each neuron in response to each trial across all patients and sessions. Targets were divided based on their memory judgment: remembered (a, b) and forgotten (c, d). The foil-by-neuron response distribution (x-axis) was compared to corresponding target-by-neuron response distributions in the hippocampus (a: remembered, 60,991; c: forgotten, 47,787 recordings) and amygdala (b: remembered, 84,033; d: forgotten, 65,763 recordings). In panel (a), the sharp deflection toward the y-axis was evident for remembered targets in the hippocampus and a small deflection in the opposite direction was observed in the amygdala (b). This deflection was not present for forgotten items in the hippocampus (c) or the amygdala (d). The skewness values between the two distributions (remembered targets vs. foils) differed significantly between brain regions (hippocampus vs. amygdala), with a significant difference observed in the hippocampus only (B = 10,000, p < 0.05, two-tailed). No visual (Q–Q plots) or statistical evidence of a difference in skewness between the forgotten target vs. foil distributions was observed in either the hippocampus or amygdala. Refer to Table 3 for detailed statistical reporting. Supplementary Fig. S3 reports Q-Q plots for remembered and forgotten items with the top 0.25% of both the target-by-neuron and foil-by-neuron distributions removed. The deflection observed in panel (a) disappeared after removing the top 0.25% of data, indicating that relatively few neurons fired strongly in response to remembered targets compared to foil items, and only within the hippocampus. *p < 0.05.