In the article, “Menopause status and within-group differences in chronological age affect the functional neural correlates of spatial context memory in middle-aged females,” by Arielle Crestol, Sricharana Rajagopal, Rikki Lissaman, Annalise A. LaPlume, Stamatoula Pasvanis, Rosanna K. Olsen, Gillian Einstein, Emily G. Jacobs, and M. Natasha Rajah, which appeared on pages 8756–8768 of the December 13, 2023 issue, the datamat creation for the fMRI data subset was erroneously sampled. The authors note the following:“The PLS datamats (i.e., onset timing files) for 16 Pre-Meno and 6 Post-Meno females contained some shifted event onsets. To correct this unintentional error, we corrected the datamats for these participants and rerun the B-PLS analysis. In the rerun analysis, the sign of the PLS solution flipped. Such sign flips are an arbitrary result of the resampling with permutation and bootstrapping and do not affect the interpretation of the results given that the direction of the relationship between brain–behavior correlations and singular image remain the same (McIntosh and Lobaugh, 2004, pg. S252).
“The updated analysis reveals the same contrast effects at comparable percentage cross-block covariance in LV1 (15.74% cross-block covariance accounted for) and LV2 (10.54% cross-block covariance accounted for), and both LVs are significant at p < 0.01. For LV 1 (p = 0.001), the brain–behavior correlation profile was generally the same in Pre-Meno females. In Post-Meno females, the patterns of correlations related to age, spatial context retrieval accuracy, and recognition accuracy effects previously observed in EncEasy are now observed in EncHard.⇓ As such on page 26, line 601 should refer to EncHard and state, “…at higher levels of task demands.” In addition, previously the activation peaks reported in Table 2 and in Figure 3 for lateral PFC extended both dorsally and ventrally, and the peak in the inferior parietal cortex (IPC) was primarily in the angular gyrus, BA 39. Now, the lateral PFC extends ventrally, and the IPC activation includes angular gyrus, BA 39, and extends to BA 19 and IPC BA 40. In addition, both positive and negative saliences are now observable. Importantly, the main effects reported originally in positive saliences and the general conclusions for LV1 remain unchanged.
“LV 2 (p = 0.006) had the same brain–behavior correlation profile as Post-Meno females. However, LV2 did not have a strong representation from Pre-Meno females⇓ as before. The activation peaks reported in Table 3 and in Figure 4 are slightly changed, with both positive and negative saliences now being above the threshold. As a result, the main effect no longer includes IPC but remains for the parahippocampal gyrus and occipitotemporal cortex. Despite these changes, the conclusions for LV2 did not change. Finally, in this reanalysis, LV3 and LV10 have p = 0.047, 8.38% cross-block covariance (previously p = 0.072, 8.88% cross-block covariance), and p = 0.038, 4.29% cross-block covariance (previously p = 0.069, 3.95% cross-block covariance), respectively. However, given the low p- and cross-block covariance values for these two LVs, they do not contribute significantly to the results. Below we present the corrected Tables 2 and 3 and updated Figures 3 and 4.
“In conclusion, reanalysis of the data does not change the main conclusions of the published study.”
The corrected Figures 3 and 4, as well as Tables 2 and 3, appear below, and the article has been corrected online. These errors do not affect the conclusions of the paper, and the online version has been corrected to account for the shifted event onsets and subsequent reanalyses.
Local maxima of negative and positive saliences identified in LV 1
Local maxima of negative and positive saliences identified in LV 2
