Fig. 5: Comparison between generation performance and spatial relative position or structural connections.

a shows the definition of a score calculated based on the relative angle from a certain region of interest (in this case, the Left Frontal Dorsal region). The score at the region of interest was set to zero, and the score was added by one for every shift of one angle from the region of interest. However, the score for the region completely opposite to the region of interest was reduced to +2 because that region is located at the same slice surface as the region of interest and adjacent to the region of interest. b illustrates the definition of the values required to calculate the strength of the structural connections between the square recording regions where electrodes were placed. We downloaded the original connection strengths and atlas data from open data shared by Allen institute (https://connectivity.brain-map.org), and reconstructed the connection matrices. In this illustration, the two squares represent the areas (e.g., Slice A, Slice B) where electrical measurements were made. In addition, two examples of connections between them are expressed as pipes. The connection strength between the atlas areas included in the square region of the electrical recording was calculated for each connection as a normalized value in terms of the percentage of the area, expressed as a dark red area in Fig. 5b, of the intersection of the region of the electrical recording with the area of the atlas at both ends of the connection. The normalized value as a percentage of the area, expressed as a dark red area in Fig. 5b, was calculated for individual connections (Refer to the method section for details). In the square region of the electrical recording, the normalized value, expressed as a dark red area, was calculated for each connection as a percentage of the area of the intersection of the region of the electrical recording with the area of the atlas at both ends of the connection). The normalized value as a percentage of the area was calculated for individual connections (Refer to the method section for details). Then, the connection strength between the regions of electrical recording was calculated as an averaged quantity of connection strength in all connected pairs of atlas areas at both ends. c~l plot the prediction results for the relative angles between the measurement regions. Among them, d~g plot scatter plots with the predicted firing rates for the relative angles between the regions, and c summarizes the correlation values and p values for the points between Angle = 0 and Angle = 1 in d~g as a four-bar graph corresponding to the order from d–g. Among d–g, d represents results to excitatory neurons in the left hemisphere, e to inhibitory neurons in the left hemisphere, f to excitatory neurons in the right hemisphere, and g to inhibitory neurons in the right hemisphere. i through l are plotted as scatter plots with sharpness, prediction performance of Synchronization score, as the vertical axis relative to the relative angles between the measurement regions. The difference in the meaning of the x axis in the i~l means results to the first quadrant of the left hemisphere, j to the third quadrant of the left hemisphere, k to the first quadrant of the right hemisphere, and l to the third quadrant of the right hemisphere. h The correlation values and p values for the points at Angle = 0 and Angle = 1 in i~l are summarized as bar graphs. n~q and s~v are the same as the d~g and i~l, except that the horizontal axis is the connection strength. Then, the correlation values and p values at n~q are summarized in m, and the correlation values and p values at s~v are summarized in r. The significance level is set at about p = 0.01 and the dotted lines are overlaid, and it can be read that the p values corresponding to s and v are much lower than that level.