Volume of the human septal forebrain region is a predictor of source memory accuracy

Abstract

Septal nuclei, components of basal forebrain, are strongly and reciprocally connected with hippocampus, and have been shown in animals to play a critical role in memory. In humans, the septal forebrain has received little attention. To examine the role of human septal forebrain in memory, we acquired high-resolution magnetic resonance imaging scans from 25 healthy subjects and calculated septal forebrain volume using recently developed probabilistic cytoarchitectonic maps. We indexed memory with the California Verbal Learning Test-II. Linear regression showed that bilateral septal forebrain volume was a significant positive predictor of recognition memory accuracy. More specifically, larger septal forebrain volume was associated with the ability to recall item source/context accuracy. Results indicate specific involvement of septal forebrain in human source memory, and recall the need for additional research into the role of septal nuclei in memory and other impairments associated with human diseases.

Fig. 1

Bilateral septal nuclei mask derived from Zaborszky et al. (2008) displayed on T1 magnetic resonance imaging template.

Fig. 2

Partial regression plots of standardized residuals and regression lines of memory test scores as a function of septal forebrain volume. Plots reflect multivariate relations between total (right and left) septal forebrain volume and (a) Total recognition discriminability (TRDI; ratio of correct hits to false positives), R² =0.187*; (b) Source recognition discriminability (SRDI; ratio of correct hits to List B false positives), R² =0.237*; and (c) Novel recognition discriminability (NRDI; ratio of correct hits to novel false positives), R² =0.115; with the variance accounted for by other independent variables in the models (age, intracranial volume) removed. *Significant at P <.05.