Figure 3.

Parallel Aversive and Safety Memories Can Be Recorded as Depression of Odor-Specific Responses in Corresponding MBONs

(A and B) Imaging planes in MBON-α2sc (A) and MBON-α3 (B) dendritic fields and training and imaging protocol. A reduced CS+ odor-evoked response was observed in both MBONs.

(C–F) Imaging plane in the MBON-β′2mp dendritic field (C and D) or presynaptic terminals (E and F) and training and imaging protocol. Spaced training significantly reduced responses to CS− in MBON-β′2mp dendritic (C) and axonal fields (E), but not when CS+ followed CS− in reversed spaced training (D and F).

(G and H) Imaging plane in MBON-γ5β′2a dendritic field and training and imaging protocol. Neither spaced training (G) nor reversed spaced training (H) changed the odor-evoked responses of MBON-γ5β′2a. CS+ data correspond to average of experiments in which 50% of the trials used 4-methylcyclohexanol (MCH) as CS+ and 50% used 3-octanol (OCT) as CS+. The same applies for CS− data. Odor-evoked activity traces show means (solid line) with SEM (shadow). A black line underneath indicates a 5 s odor. Bar graphs display normalized area under the curve as means ± SEM. Individual data points are displayed as dots, and paired measurements are connected by stippled lines. Asterisks denote a significant difference between averaged responses to CS+ and CS−. See Figure S3 for non-normalized traces for the CS+, CS−, and third odor and Table S1 for statistics.