Correction to: Nature Neuroscience https://doi.org/10.1038/s41593-023-01350-3, published online 8 June 2023.

In the version of the article initially published, there was an error in Fig. 2h and several editing errors in the figure legends. In Fig. 2h, the topmost data point in the SNI / Oxy graph was cut off and has now been restored. In the Fig. 1e legend, the sentence “Significance (P < 0.05) was calculated by means of two-way ANOVA with Tukey’s post hoc test for b-d and Sidak’s post hoc test for direct comparison of Sham-Oxy versus Sham-Sal” has been removed. In the Fig. 2c legend, “t = 2.912, d.f. = 45, P = 0.0111” has been deleted; in the Fig. 2h,i legend, “two-way ANOVA interaction, F1,25 = 0.3945, P = 0.3945; Sidak’s multiple comparisons, SNI-Oxy versus SNI-Sal t = 2.486, d.f. = 25, P = 0.0395)” and “Data represent the mean +/–s.e.m.” have been deleted. In the Fig. 2f legend now reading in part “Sham two-way ANOVA interaction F1,54 = 0.002843, P = 0.9577; Sidak’s multiple comparisons Sham-Sal t = 2.484, d.f. = 54, P = 0.032; Sham-Oxy t = 2.311, d.f. = 54, P = 0.0488; SNI two-way ANOVA interaction F1,58 = 9.51, P = 0.0031; Sidak’s multiple comparisons SNI-Sal t = 4.489, d.f. = 58, P < 0.0001)” previously read “…F1,56 = 0.01304, P = 0.9095; Sidak’s multiple comparisons Sham-Sal t = 2.839, d.f. = 56, P = 0.0126; Sham-Oxy t = 2.331, d.f. = 56, P = 0.0462; SNI two-way ANOVA interaction F1,58 = 9.51, P = 0.0031; Sidak’s multiple comparisons SNI-Sal t = 4.489, d.f. = 58)”. In the Fig. 6d,e legend, “F1,36 = 10.6” previously read “F1,36 = 5.269”; in Fig. 2f,g legend, two sentences have now been added: “oxycodone withdrawal led to hypersensitivity relative to Sham-Sal (RBC1HI-treated and -untreated) mice” and “repeated measures two-way ANOVA interaction F12,136 = 3.438, P = 0.0002; Tukey’s multiple comparisons WD7 Sham-Sal-Veh vs Sham-Oxy Veh q = 4.008, d.f. = 170, P = 0.0262; Sham-Sal-RBC1HI vs Sham-Oxy-Veh, q = 5.403, d.f. = 170, P = 0.0011; OD10”; and one sentence has been removed from the Fig. 6k legend: “Significance was calculated by means of two-way ANOVA with Bonferroni’s post-hoc test”. In the Fig. 7f legend, “F3,52 = 1.554, P = 0.2116” replaces “F1,52 = 4.787, P = 0.0332”, while “Significance was calculated by means of two-way ANOVA with Bonferroni’s post hoc test” has been removed. In the Extended Data 4 legend, “paired t-test, t = 7.326, d.f. = 4, P = 0.0018” and “**P < 0.01” have been added. These changes have been made to the HTML and PDF versions of the article.