Fig. 5: WM effects of optogenetic activation of the mPFC→dmStr pathway.

a Virus injections to induce ChR2 expression in the mPFC→dmStr pathway and fiber implantation site in the left hemisphere. b Validation of pathway-specific activation induced by ChR2 activation. Left: example electrophysiological trace from a single site on the electrophysiological probe implanted in PrL; Right: Box-whisker plot of mean spike rates for Light Off/On conditions shows minimum-maximum range and 25th and 75th percentiles (black line: median, red line: mean, red crosses: outliers; ***p = 7.95 × 10⁻⁷, 32 units from n = 3 mice in mPFC; two-sided Wilcoxon signed rank test). c Optogenetic activation of the mPFC→dmStr pathway during the encoding (E), maintenance (M), and retrieval (R) periods in Cre-dependent ChR2-expressing mice (n = 9 mice) did not change performance during any period of the task (Bonferroni corrected p-values for Left Sample Run trials: E: 0.9 M: 1.0, R: 1.0; Right Sample Run trials performance also did not change in all periods of the trial: p(E) = 1.0, p(M) = 0.16, p(R) = 0.38 (two-sided Wilcoxon signed rank test with Bonferroni correction). d Optogenetic activation of the mPFC→dmStr pathway partially rescued WM impairment induced by MK-801 injection when applied during the WM maintenance period (performance increased from 48.1 ± 7.8% to 65.9 ± 7.2%; Left Sample Run trials, p(E) = 1.0, *p(M) = 0.011, p(R) = 1.0 (two-sided Wilcoxon signed rank test with Bonferroni correction); Right Sample Run trials performance also did not change in all trial periods; p(E) = 1.0, p(M) = 1.0, p(R) = 1.0 (two-sided Wilcoxon signed rank test with Bonferroni correction). For a given task period, ChR2 activation was performed in 50% of trials in a random fashion. Black lines represent individual sessions. Error bars show s.e.m. Gray shaded error areas represent the s.e.m. spread during Light Off encoding condition. Source data are provided as a Source Data file. ns, non-significant.