Extended Data Fig. 1: OFL-related behavioural experiments.
a, Apparatus for OFL, with DEM present during conditioning. b, Trial-wise freezing in OBS and DEM during OFL conditioning. N = 17 OBS-DEM pairs. c, Higher CS-related freezing in DEM than OBS (*P < 0.0001) during OFL conditioning (left); more CS-related rearing (*P < 0.0001) and grooming (*P = 0.0024) in OBS than DEM during OFL conditioning (right). P values from unpaired t-tests. N = 10 OBS-DEM pairs. d, No increase in CS-related OBS freezing versus pre-CS when partition between OBS and DEM is opaque. N = 9 mice. e, Minimal increase in CS-related OBS freezing versus pre-CS (#P = 0.0312) during conditioning, not retrieval, when DEM is not shocked. P value from paired t-test. N = 9 mice. f, Anxiety-related behaviour and social interaction testing 1-2 days after OFL. g, Normal OBS performance, versus unconditioned controls, in the elevated plus-maze test. n = 6 mice/group. h, Normal OBS performance, versus unconditioned controls, in the novel open field test. n = 6 mice/group. i, Normal OBS social interaction with a novel conspecific mouse, versus unconditioned controls. n = 6 mice/group. j, Similarly higher CS-related freezing versus pre-CS in animals directly conditioned alone (conditioning #P = 0.0001, retrieval #P = 0.0087) or in the presence of a conspecific (conditioning #P = 0.001, retrieval #P = 0.0010). P values from Bonferroni post hoc tests following mixed-model ANOVA. n = 7 mice/group. Two-tailed statistical tests were used. Data shown as means ± SEM.
