Importance of examining stimulus type in fMRI studies of sex differences in memory recall.

In their article, Spets and Slotnick (2020) conduct a metaanalysis on fMRI studies examining sex differences during long-term memory recall. They conclude that sex differences in brain activity during long-term memory recall do indeed exist and that collapsing across sex is inappropriate in the field of cognitive neuroscience. Their activation likelihood estimation meta-analysis revealed several regions where males were reported to have greater activation than females during long-term memory tasks. There were no regions where females showed greater activation than males. The authors interpret these results as supporting the neural efficiency hypothesis – that males require greater neural effort to achieve equivalent behavioral performance as females. The lack of female > male activity suggests that females engage in more efficient processing. This interpretation suggests that measures of motivation and task engagement are critical to include. It could be that sex and motivation are confounded and that males are simply less engaged in the task and therefore require more effort to perform it. Indeed, increased default mode network activity is associated with a lack of task engagement (Greicius & Menon, 2004). Increased hippocampal activity has been found in depressed relative to healthy individuals as they recall autobiographical memories and this has been interpreted as the increased effort required by depressed participants to recall a specific memory (Young et al., 2014). These results lend support to the interpretation that men require more effort to achieve the same behavioral outcome. The effort hypothesis only makes sense, however, in the context of no female > male activity. Indeed, the lack of female > male differences in the meta-analysis is surprising, considering that all but one study reported significant female> male activations (St Jacques et al., 2011). It is possible that the lack of significant female > male activity is due to collapsing across stimulus type. Indeed, in most of the studies of autobiographical memory, memory for stimuli that have a self-referential component, female > male activity was found in prefrontal regions including the DLPFC and cingulate cortex (Compere et al., 2016; Piefke et al., 2005; Young et al., 2013). When examining memory for faces and shapes, female > male activity was found in posterior regions such as the parietal lobe (Ino et al., 2010; Spets et al., 2019; Spets & Slotnick, 2019). Thus, stimulus type may indeed play a critical role in the observed sex differences – with women engaging more prefrontal regions during self-referential memory recall and more posterior regions for item recall. The interpretation that men are exerting more effort becomes less convincing when considering these female > male activations. Indeed, these results suggest that women are exerting more cognitive control/emotion regulation during autobiographical memory recall and that they may be using more linguistic than visual strategies when engaging in item/spatial memory recall. This interpretation is congruent with the finding that women consistently outperform men in verbal tasks while men outperform women in spatial tasks (Weiss et al., 2003). In this context, the increased hippocampal activity found in men across stimulus types could be interpreted as an increased use of visual-spatial processing, given hippocampal activity has also consistently been found when visual-spatial strategies are employed (Bird & Burgess, 2008). In conclusion, Spets and Slotnick (2020) provide strong evidence that sex differences in neural processing exist. However, whether these differences reflect more effort required on the part of males to achieve similar performance to females or whether the sexes are employing different strategies to achieve the same performance remains to be determined. Including measures of motivation and effort, as well as collecting self-report information on what strategies participants are employing during fMRI is critical for interpreting reported sex differences.