Correction to: Nature Communications https://doi.org/10.1038/s41467-025-60080-1, published online 27 May 2025

Following publication of this article, some patch clamp recording raw data were found to be in error, resulting in mistakes in Fig. 3j, Supplementary Fig. 5c–f and associated Source data. In the Results section “eCB system controls plasticity at VTg-PMd synapses and behavioral adaptation”, the original sentence reading “We first measured the effect of single or five-time 2MT exposure conditioning on VTgPV neurons’ excitability (Supplementary Fig. 5a) and found no difference among groups (Supplementary Fig. 5b–f)” has now been amended and extended as follows: “We first measured the effect of single or five-time 2MT exposure conditioning on VTgPV neurons’ excitability (Supplementary Fig. 5a). Since we recorded the membrane potentials while holding the cell at about –45 to –60 mV with no spontaneous firing (already in the method). Although baseline membrane potentials differed among the various groups (Supplementary Fig. 5d), this only reflects variations in fundamental intrinsic and extrinsic electrophysiological properties. Crucially, when we applied simulated current injection to test excitability, the number of action potentials elicited showed no difference among groups (Supplementary Fig. 5c), indicating that the level of cellular excitability remained unchanged. These data suggest that learning-induced suppression of VTgPV neurons does not affect their excitability at the somatic level, but could be due to changes at the VTgPV → PMdCCK synapses.” The text, figures and source data are now amended in the HTML and PDF versions of the article.