The brain’s functional architecture and its links to emotion processing and episodic memory during the resting-state and an encoding task

Episodic memory, i.e., the conscious memory for personally experienced events within a particular spatio-temporal context, and emotion processing represent salient features of the clinical picture of many mental disorders, and are tightly intertwined with cognitive functioning.
Over the past three decades, cognitive neuroscience research using functional magnetic resonance imaging (fMRI) has seen two major shifts: firstly, a shift away from region- based to network-based modeling of brain function, and, secondly, more credit is given to the state of the brain during rest. Alongside this progress, substantial advances in fMRI technology and analysis methods have been achieved. There has also been an increase in awareness of the importance of statistical power and inter-individual variability in brain function. The acknowledgment that the brain works as a network consisting of interacting subnetworks and brain regions, a concept falling under the term functional connectivity, has been a milestone in the field of fMRI-based cognitive neuroscience.
This PhD project sought to combine these developments to investigate the brain’s functional architecture in relation to episodic memory and emotion processing in healthy young adults.
To investigate whether exposure to emotionally aversive pictures affects subsequent resting-state networks differently from exposure to neutral pictures, Study 1 incorporated a resting-state EEG-fMRI study (n = 34), in which we focused on investigating (i) patterns of amygdala whole-brain and hippocampus connectivity, (ii) whole-brain resting-state networks, and (iii) the amygdala's regional low-frequency fluctuations, all while EEG-recording potential fluctuations in vigilance. Despite the successful emotion induction, none of the resting-state measures was differentially affected by picture valence.
In Study 2, the major aim was to address the extent to which differential brain responsivity during encoding might explain inter-individual differences in episodic memory performance. For this purpose, we analyzed a large sample of adults (n = 1,434) who underwent a picture encoding task in a single MR scanner. Complementing a voxel-based with a network-based approach, Study 2 found that responsivity in some of the regions implicated in successful encoding, as well as responsivity of six functional connectivity networks (FCN), were associated with inter-individual differences in memory performance.
Alongside these two core projects, two other studies, based on the sample from Study 2, looked into emotional memory. Study 3 aimed at investigating the involvement of functional activation clusters and their interactions in the effect of emotional memory enhancement (n = 1,418). The results underline the involvement of cortico-cerebellar interactions and localized cerebellar activity. Finally, Study 4 used prediction modeling to investigate FCN responsivity to negative pictures and its links to behavior (n = 1,147). The primary objectives were to (i) assess which FCNs are important for negative picture encoding, (ii) detect subjects with peculiarities in FCN responsivity, and (iii) find links between peculiarities between this measure and emotion-related behavioral phenotypes. The main finding was that individuals with peculiarities in network responsivity to negative pictures tend to also have peculiarities in an emotion-related behavioral phenotype.
Overall, this work has strengthened the central assumption that the brain functions as a network during states of resting and episodic memory encoding. The temporal unfolding of emotion processing seems to be not reflected in a subsequent state of rest (Study 1). Brain activity during a task of encoding can be summarized to robust FCNs. The responsivity of six of these FCNs is linked to later free recall performance of the encoded stimuli (Study 2). The cerebellum and its interactions with the cerebrum are involved in emotional memory enhancement (Study 3). Among the healthy, there are individuals with peculiar emotion responsivity on a neurofunctional level (Study 4), a finding which sets the path for future studies to investigate whether it could be a neurofunctional marker for being in an antecedent stage to develop a mental disorder. Together, the findings point to a multifactorial nature of complex behavior. Identifying neurofunctional markers based on inter-individual differences in memory and emotion processing may open the possibility for studying associations with other phenotypes, such as psychological traits, genetic or epigenetic makeup, or individual metabolomic profiles. Our work contributes to the growing understanding of the neurofunctional underpinnings of emotion processing and episodic memory.