Fig. 4

Social memory is conserved but social approach in familiar environment is deficient in Scn2a KO mice. a In the “sociability” phase of the three-chambers task, Scn2a^KO/+ and WT mice spent significantly more time investigating the stranger mouse than the empty cage, and the time spent on the stranger’s cage was slightly yet significantly longer in the Scn2a^KO/+ group (genotype-side interaction: F_1,50 = 9.524, p < 0.01). b In the “social memory” phase of the task, mice from both groups showed a significant preference for the novel mouse over the familiar one and no significant genotype effect was observed (genotype-side interaction: F_1,50 = 0.059, NS; side effect: F_1,50 = 52.740, p < 0.001; genotype effect: F_1,50 = 0.117, NS). c In the direct interaction social memory paradigm (assessed in an unfamiliar environment), animals from both groups gradually habituated to the first intruder. A significant gain of interest was observed upon inserting a new female. No significant differences were however observed between Scn2a^KO/+ and WT mice (genotype-trial interaction: F_4,39 = 0.383, NS; trial effect: F_4,44 = 7.840, p < 0.001; genotype effect: F_1,11 = 3.217, NS). d In the tube test for social dominance, Scn2a^KO/+ mice that underwent the testing battery won significantly more encounters than their WT littermates (Chi-squared test; p < 0.05). The winning rate was however at a random level in naïve Scn2a^KO/+ mice (i.e., that never experienced behavioral tasks). e When exposed to a young intruder male in their home cage, the latency of Scn2a^KO/+ mice to the first sniffing contact was significantly longer than in the WT group. f The total number of sniffing event was also decreased in the Scn2a^KO/+ group whereas the number of direct contact, chasing and aggression events was comparable between Scn2a^KO/+ and WT animals. g Monitoring social behavior for 24 h showed a comparable interaction time in WT-WT and Scn2a^KO/+-Scn2a^KO/+ pairs without significant genotype effect in either the active (nighttime) or inactive (daytime) periods (genotype-time interaction: F_23,207 = 1.026, NS; time effect: F_23,207 = 20.760, p < 0.001; genotype effect: F_1,9 = 2.888, NS). Str stranger mouse, Emp empty cage. Three-chambers: WT: N = 23, KO: N = 29; direct interaction social memory: WT: N = 11, KO: N = 12; tube test: WT (test battery): N = 13, KO (test battery): N = 13, WT (naïve): N = 12, KO (naïve): N = 12; resident-intruder: WT: N = 13, KO: N = 13; 24-h monitor home-cage social interaction monitoring WT: N = 10 pairs, KO: NO = 10 pairs. Values are expressed as mean ± standard error of the mean. Statistical significance was assessed using two-way ANOVA (using “side” as a within-subject factor, a, b) or two-way repeated measures ANOVA (c, g) followed by Sidak’s multiple comparison post-hoc test, chi-square (d) or Student t test (e, f) with significance set at *p < 0.05, **p < 0.01, and ***p < 0.001