An in vivo Dissection, and Analysis of Socio-Affective Symptoms related to Cerebellum-Midbrain Reward Circuitry in Humans.

Emerging research in non-human animals implicates cerebellar projections to the ventral tegmental area (VTA) in appetitive behaviors, but these circuits have not been characterized in humans. Here, we mapped cerebello-VTA white-matter connectivity in humans using probabilistic tractography on diffusion imaging data from the Human Connectome Project. We uncovered the topographical organization of these connections by separately tracking from parcels of cerebellar lobule VI, crus I/II, vermis, paravermis, and cerebrocerebellum. Results revealed that connections from the cerebellum to the VTA predominantly originate in the right hemisphere, interposed nucleus, and paravermal cortex, and terminate mostly ipsilaterally. Paravermal crus I sends the most connections to the VTA compared to other lobules. We discovered a medial-to-lateral gradient of connectivity, such that the medial cerebellum has the highest connectivity with the VTA. Individual differences in microstructure were associated with measures of negative affect and social functioning. By splitting the tracts into quarters, we found that the socio-affective effects were driven by the third quarter of the tract, corresponding to the point at which the fibers leave the deep nuclei. Taken together, we produced detailed maps of cerebello-VTA structural connectivity for the first time in humans and established their relevance for trait differences in socio-affective regulation.

In vivo structural connectivity of the reward system along the hippocampal long axis.

Recent work has identified a critical role for the hippocampus in reward-sensitive behaviors, including motivated memory, reinforcement learning, and decision-making. Animal histology and human functional neuroimaging have shown that brain regions involved in reward processing and motivation are more interconnected with the ventral/anterior hippocampus. However, direct evidence examining gradients of structural connectivity between reward regions and the hippocampus in humans is lacking. The present study used diffusion MRI (dMRI) and probabilistic tractography to quantify the structural connectivity of the hippocampus with key reward processing regions in vivo. Using a large sample of subjects (N = 628) from the human connectome dMRI data release, we found that connectivity profiles with the hippocampus varied widely between different regions of the reward circuit. While the dopaminergic midbrain (ventral tegmental area) showed stronger connectivity with the anterior versus posterior hippocampus, the ventromedial prefrontal cortex showed stronger connectivity with the posterior hippocampus. The limbic (ventral) striatum demonstrated a more homogeneous connectivity profile along the hippocampal long axis. This is the first study to generate a probabilistic atlas of the hippocampal structural connectivity with reward-related networks, which is essential to investigating how these circuits contribute to normative adaptive behavior and maladaptive behaviors in psychiatric illness. These findings describe nuanced structural connectivity that sets the foundation to better understand how the hippocampus influences reward-guided behavior in humans.

Awake targeted memory reactivation doesn't work.

Memories are pliable and can be biased by post-encoding information. In targeted memory reactivation (TMR) studies, participants encode information then sleep, during which time sounds or scents that were previously associated with the encoded images are re-presented in an effort to trigger reactivation of the associated memory traces. Upon subsequent testing, memory for reactivated items is often enhanced. Is sleep essential for this process? The literature on awake TMR is small and findings are mixed. Here, we asked English-speaking adults to learn Japanese vocabulary words. During a subsequent active rest phase, participants played Tetris while sound cues associated with the vocabulary words were presented. Results showed that when memories were reactivated, they were either disrupted (Experiment 1) or unaffected (Experiments 2, 3). These findings indicate that awake TMR is not beneficial, and may actually impair subsequent memory. These findings have important implications for research on memory consolidation and reactivation.

Adolescents at risk for depression show increased white matter microstructure with age across diffuse areas of the brain.

Maternal history of depression is a strong predictor of depression in offspring and linked to structural and functional alterations in the developing brain. However, very little work has examined differences in white matter in adolescents at familial risk for depression. In a sample aged 9-14 (n = 117), we used tract-based spatial statistics (TBSS) to examine differences in white matter microstructure between adolescents with (n = 42) and without (n = 75) maternal history of depression. Microstructure was indexed using fractional anisotropy (FA). Threshold-free cluster enhancement was applied and cluster maps were thresholded at whole-brain family-wise error < .05. There was no significant main effect of risk status on FA. However, there was a significant interaction between risk status and age, such that large and diffuse portions of the white matter skeleton showed relatively increased FA with age for youth with a maternal history of depression compared to those without. Most tracts identified by the interaction were robust to controlling for sex, youth internalizing, in-scanner motion, neighborhood SES, and intra-cranial volume, evidence that maternal depression is a unique predictor of white matter alterations in youth. Widespread increases in FA with age may correspond to a global pattern of accelerated brain maturation in youth at risk for depression.

A transformer model for learning spatiotemporal contextual representation in fMRI data.

Representation learning is a core component in data-driven modeling of various complex phenomena. Learning a contextually informative representation can especially benefit the analysis of fMRI data because of the complexities and dynamic dependencies present in such datasets. In this work, we propose a framework based on transformer models to learn an embedding of the fMRI data by taking the spatiotemporal contextual information in the data into account. This approach takes the multivariate BOLD time series of the regions of the brain as well as their functional connectivity network simultaneously as the input to create a set of meaningful features that can in turn be used in various downstream tasks such as classification, feature extraction, and statistical analysis. The proposed spatiotemporal framework uses the attention mechanism as well as the graph convolution neural network to jointly inject the contextual information regarding the dynamics in time series data and their connectivity into the representation. We demonstrate the benefits of this framework by applying it to two resting-state fMRI datasets, and provide further discussion on various aspects and advantages of it over a number of other commonly adopted architectures.

Awake Hippocampal-Cortical Co-reactivation Is Associated with Forgetting.

Systems consolidation theories posit that consolidation occurs primarily through a coordinated communication between hippocampus and neocortex [Moscovitch, M., & Gilboa, A. Systems consolidation, transformation and reorganization: Multiple trace theory, trace transformation theory and their competitors. PsyArXiv, 2021; Kumaran, D., Hassabis, D., & McClelland, J. L. What learning systems do intelligent agents need? Complementary learning systems theory updated. Trends in Cognitive Sciences, 20, 512-534, 2016; McClelland, J. L., & O'Reilly, R. C. Why there are complementary learning systems in the hippocampus and neocortex: Insights from the successes and failures of connectionist models of learning and memory. Psychological Review, 102, 419-457, 1995]. Recent sleep studies in rodents have shown that hippocampus and visual cortex replay the same information at temporal proximity ("co-replay"; Lansink, C. S., Goltstein, P. M., Lankelma, J. V., McNaughton, B. L., & Pennartz, C. M. A. Hippocampus leads ventral striatum in replay of place-reward information. PLoS Biology, 7, e1000173, 2009; Peyrache, A., Khamassi, M., Benchenane, K., Wiener, S. I., & Battaglia, F. P. Replay of rule-learning related neural patterns in the prefrontal cortex during sleep. Nature Neuroscience, 12, 919-926, 2009; Wierzynski, C. M., Lubenov, E. V., Gu, M., & Siapas, A. G. State-dependent spike-timing relationships between hippocampal and prefrontal circuits during sleep. Neuron, 61, 587-596, 2009; Ji, D., & Wilson, M. A. Coordinated memory replay in the visual cortex and hippocampus during sleep. Nature Neuroscience, 10, 100-107, 2007). We developed a novel repetition time (TR)-based co-reactivation analysis method to study hippocampal-cortical co-replays in humans using fMRI. Thirty-six young adults completed an image (face or scene) and location paired associate encoding task in the scanner, which were preceded and followed by resting state scans. We identified post-encoding rest TRs (± 1) that showed neural reactivation of each image-location trials in both hippocampus (HPC) and category-selective cortex (fusiform face area [FFA]). This allowed us to characterize temporally proximal coordinated reactivations ("co-reactivations") between HPC and FFA. Moreover, we found that increased HPC-FFA co-reactivations were associated with incorrectly recognized trials after a 1-week delay (p = .004). Finally, we found that these HPC-FFA co-reactivations were also associated with trials that were initially correctly recognized immediately after encoding but were later forgotten in 1-day (p = .043) and 1-week delay period (p = .031). We discuss these results from a trace transformation perspective [Sekeres, M. J., Winocur, G., & Moscovitch, M. The hippocampus and related neocortical structures in memory transformation. Neuroscience Letters, 680, 39-53, 2018; Winocur, G., & Moscovitch, M. Memory transformation and systems consolidation. Journal of the International Neuropsychological Society, 17, 766-780, 2011] and speculate that HPC-FFA co-reactivations may be integrating related events, at the expense of disrupting event-specific details, hence leading to forgetting.

Setting boundaries: Development of neural and behavioral event cognition in early childhood.

The ongoing stream of sensory experience is so complex and ever-changing that we tend to parse this experience at "event boundaries," which structures and strengthens memory. Memory processes undergo profound change across early childhood. Whether young children also divide their ongoing processing along event boundaries, and if those boundaries relate to memory, could provide important insight into the development of memory systems. In Study 1, 4-7-year-old children and adults segmented a cartoon, and we tested their memory. Children's event boundaries were more variable than adults' and differed in location and consistency of agreement. Older children's event segmentation was more adult-like than younger children's, and children who segmented events more like adults had better memory for those events. In Study 2, we asked whether these developmental differences in event segmentation had their roots in distinct neural representations. A separate group of 4-8-year-old children watched the same cartoon while undergoing an fMRI scan. In the right hippocampus, greater pattern dissimilarity across event boundaries compared to within events was evident for both child and adult behavioral boundaries, suggesting children and adults share similar event cognition. However, the boundaries identified by a data-driven Hidden Markov Model found that a different brain region-the left and right angular gyrus-aligned only with event boundaries defined by children. Overall, these data suggest that children's event cognition is reasonably well-developed by age 4 but continues to become more adult-like across early childhood. RESEARCH HIGHLIGHTS: Adults naturally break their experience into events, which structures and strengthens memory, but less is known about children's event perception and memory. Study 1 had adults and children segment and remember events from an animated show, and Study 2 compared those segmentations to other children's fMRI data. Children show better recognition and temporal order memory and more adult-like event segmentation with age, and children who segment more like adults have better memory. Children's and adults' behavioral boundaries mapped onto pattern similarity differences in hippocampus, and children's behavioral boundaries matched a data-driven model's boundaries in angular gyrus.

Little brain, little minds: The big role of the cerebellum in social development.

Seminal work in the 1990's found alterations in the cerebellum of individuals with social disorders including autism spectrum disorder and schizophrenia. In neurotypical populations, distinct portions of the posterior cerebellum are consistently activated in fMRI studies of social cognition and it has been hypothesized that the cerebellum plays an essential role in social cognition, particularly in theory of mind. Here we review the lesion literature and find that the effect of cerebellar damage on social cognition is strongly linked to the age of insult, with dramatic impairments observed after prenatal insult, strong deficits observed after childhood damage, and mild and inconsistent deficits observed following damage to the adult cerebellum. To explain the developmental gradient, we propose that early in life, the forward model dominates cerebellar computations. The forward model learns and uses errors to help build schemas of our interpersonal worlds. Subsequently, we argue that once these schemas have been built up, the inverse model, which is the foundation of automatic processing, becomes dominant. We provide suggestions for how to test this, and also outline directions for future research.

Concussions in young adult athletes: No effect on cerebral white matter.

Introduction: The media's recent focus on possible negative health outcomes following sports- related concussion has increased awareness as well as anxiety among parents and athletes. However, the literature on concussion outcomes is equivocal and limited by a variety of diagnostic approaches. Methods: The current study used a rigorous, open- access concussion identification method-the Ohio State University Traumatic Brain Injury Identification method (OSU TBI-ID) to identify concussion and periods of repeated, subclinical head trauma in 108 young adult athletes who also underwent a comprehensive protocol of cognitive tests, mood/anxiety questionnaires, and high-angular-resolution diffusion-weighted brain imaging to evaluate potential changes in white matter microstructure. Results: Analyses showed that athletes with a history of repetitive, subclinical impacts to the head performed slightly worse on a measure of inhibitory impulse control and had more anxiety symptoms compared to those who never sustained any type of head injury but were otherwise the same as athletes with no history of concussion. Importantly, there were no group differences in cerebral white matter as measured by tract- based spatial statistics (TBSS), nor were there any associations between OSU TBI-ID measures and whole-brain principal scalars and free-water corrected scalars. Discussion: Our results provide support for the hypothesis that it is not concussion per se, but repetitive head impacts that beget worse outcomes.

SymCog: An open-source toolkit for assessing human symbolic cognition.

Symbol systems have a profound influence on human behavior, spanning countless modalities such as natural language, clothing styles, monetary systems, and gestural conventions (e.g., handshaking). Selective impairments in understanding and manipulating symbols are collectively known as asymbolia. Here we address open questions about the nature of asymbolia in the context of both historical and contemporary approaches to human symbolic cognition. We describe a tripartite perspective on symbolic cognition premised upon (1) mental representation of a concept, (2) a stored pool of symbols segregated from their respective referents, and (3) fast and accurate mapping between concepts and symbols. We present an open-source toolkit for assessing symbolic knowledge premised upon matching animated video depictions of abstract concepts to their corresponding verbal and nonverbal symbols. Animations include simple geometric shapes (e.g., filled circles, squares) moving in semantically meaningful ways. For example, a rectangle bending under the implied weight of a large square denotes "heaviness." We report normative data for matching words and images to these target animations. In a second norming study, participants rated target animations across a range of semantic dimensions (e.g., valence, dominance). In a third study, we normed a set of concepts familiar to American English speakers but lacking verbal labels (e.g., the feeling of a Sunday evening). We describe how these tools may be used to assess human symbolic processing and identify asymbolic deficits across the span of human development.

A missing link in affect regulation: the cerebellum.

The cerebellum is one-third the size of the cerebrum yet holds twice the number of neurons. Historically, its sole function was thought to be in the calibration of smooth movements through the creation and ongoing modification of motor programs. This traditional viewpoint has been challenged by findings showing that cerebellar damage can lead to striking changes in non-motor behavior, including emotional changes. In this manuscript, we review the literature on clinical and subclinical affective disturbances observed in individuals with lesions to the cerebellum. Disorders include pathological laughing and crying, bipolar disorder, depression and mixed mood changes. We propose a theoretical model based on cerebellar connectivity to explain how the cerebellum calibrates affect. We conclude with actionable steps for future researchers to test this model and improve upon the limitations of past literature.

Children show adult-like hippocampal pattern similarity for familiar but not novel events.

The ability to detect differences among similar events in our lives is a crucial aspect of successful episodic memory performance, which develops across early childhood. The neural substrate of this ability is supported by operations in the medial temporal lobe (MTL). Here, we used representational similarity analysis (RSA) to measure neural pattern similarity in hippocampus, perirhinal cortex, and parahippocampal cortex for 4- to 10-year-old children and adults during naturalistic viewing of clips from the same compared to different movies. Further, we assessed the role of prior exposure to individual movie clips on pattern similarity in the MTL. In both age groups, neural pattern similarity in hippocampus was lower for clips drawn from the same movies compared to those drawn from different movies, suggesting that related content activates processes focused on keeping representations with shared content distinct. However, children showed this only for movies with which they had prior exposures, whereas adults showed the effect regardless of any prior exposures to the movies. These findings suggest that children require repeated exposure to stimuli to show adult-like MTL functioning in distinguishing among similar events.

Distinct alterations in cerebellar connectivity with substantia nigra and ventral tegmental area in Parkinson's disease.

In Parkinson's disease (PD), neurodegeneration of dopaminergic neurons occurs in the midbrain, specifically targeting the substantia nigra (SN), while leaving the ventral tegmental area (VTA) relatively spared in early phases of the disease. Although the SN and VTA are known to be functionally dissociable in healthy adults, it remains unclear how this dissociation is altered in PD. To examine this issue, we performed a whole-brain analysis to compare functional connectivity in PD to healthy adults using resting-state functional magnetic resonance imaging (rs-fMRI) data compiled from three independent datasets. Our analysis showed that across the sample, the SN had greater connectivity with the precuneus, anterior cingulate gyrus, and areas of the occipital cortex, partially replicating our previous work in healthy young adults. Notably, we also found that, in PD, VTA-right cerebellum connectivity was higher than SN-right cerebellum connectivity, whereas the opposite trend occurred in healthy controls. This double dissociation may reflect a compensatory role of the cerebellum in PD and could provide a potential target for future study and treatment.

The fornix supports episodic memory during childhood.

Episodic memory relies on the coordination of widespread brain regions that reconstruct spatiotemporal details of an episode. These topologically dispersed brain regions can rapidly communicate through structural pathways. Research in animal and human lesion studies implicate the fornix-the major output pathway of the hippocampus-in supporting various aspects of episodic memory. Because episodic memory undergoes marked changes in early childhood, we tested the link between the fornix and episodic memory in an age window of robust memory development (ages 4-8 years). Children were tested on the stories subtest from the Children's Memory Scale, a temporal order memory task, and a source memory task. Fornix streamlines were reconstructed using probabilistic tractography to estimate fornix microstructure. In addition, we measured fornix macrostructure and computed free water. To assess selectivity of our findings, we also reconstructed the uncinate fasciculus. Findings show that children's memory increases from ages 4 to 8 and that fornix micro- and macrostructure increases between ages 4 and 8. Children's memory performance across nearly every memory task correlated with individual differences in fornix, but not uncinate fasciculus, white matter. These findings suggest that the fornix plays an important role in supporting the development of episodic memory, and potentially semantic memory, in early childhood.

The Social Cerebellum: A Large-Scale Investigation of Functional and Structural Specificity and Connectivity.

The cerebellum has been traditionally disregarded in relation to nonmotor functions, but recent findings indicate it may be involved in language, affective processing, and social functions. Mentalizing, or Theory of Mind (ToM), is the ability to infer mental states of others and this skill relies on a distributed network of brain regions. Here, we leveraged large-scale multimodal neuroimaging data to elucidate the structural and functional role of the cerebellum in mentalizing. We used functional activations to determine whether the cerebellum has a domain-general or domain-specific functional role, and effective connectivity and probabilistic tractography to map the cerebello-cerebral mentalizing network. We found that the cerebellum is organized in a domain-specific way and that there is a left cerebellar effective and structural lateralization, with more and stronger effective connections from the left cerebellar hemisphere to the right cerebral mentalizing areas, and greater cerebello-thalamo-cortical and cortico-ponto-cerebellar streamline counts from and to the left cerebellum. Our study provides novel insights to the network organization of the cerebellum, an overlooked brain structure, and mentalizing, one of humans' most essential abilities to navigate the social world.

A large-scale structural and functional connectome of social mentalizing.

Humans have a remarkable ability to infer the mind of others. This mentalizing skill relies on a distributed network of brain regions but how these regions connect and interact is not well understood. Here we leveraged large-scale multimodal neuroimaging data to elucidate the brain-wide organization and mechanisms of mentalizing processing. Key connectomic features of the mentalizing network (MTN) have been delineated in exquisite detail. We found the structural architecture of MTN is organized by two parallel subsystems and constructed redundantly by local and long-range white matter fibers. We uncovered an intrinsic functional architecture that is synchronized according to the degree of mentalizing, and its hierarchy reflects the inherent information integration order. We also examined the correspondence between the structural and functional connectivity in the network and revealed their differences in network topology, individual variance, spatial specificity, and functional specificity. Finally, we scrutinized the connectome resemblance between the default mode network and MTN and elaborated their inherent differences in dynamic patterns, laterality, and homogeneity. Overall, our study demonstrates that mentalizing processing unfolds across functionally heterogeneous regions with highly structured fiber tracts and unique hierarchical functional architecture, which make it distinguishable from the default mode network and other vicinity brain networks supporting autobiographical memory, semantic memory, self-referential, moral reasoning, and mental time travel.

Contingency of semantic generalization on episodic specificity varies across development.

Semantic memory-general knowledge of ideas and concepts-includes generalization processes that support inference. Episodic memory, on the other hand, preserves the specificity of individual events by binding together unique combinations of elements from an episode and relies on pattern separation to distinguish similar experiences. These two memory systems play complementary roles, supporting different mnemonic goals, but the nature and extent of their interdependence is unclear.1,2 Some models suggest that new information is encoded initially as hippocampus-dependent episodic memory and then, either through repetition or gist extraction, becomes semantic over time.3,4 These models also posit a neocortical route to semantic memory acquisition exists that can bypass the hippocampus.3 Both proposed routes are slow learning mechanisms, yet generalization can occur rapidly. Recent models suggest that fast generalization relies, in part, on the retrieval of individual but related episodes.5,6 Such episodic memory gating mechanisms render fast generalization contingent on the memory specificity of instances, a pattern that has been observed in adults.7,8 None of these models take into account the observation that generalization and episodic specificity have asynchronous developmental profiles, with generalization emerging years before episodic memory.9,10 We ask two questions about generalized and specific memory during early childhood: first, is rapid generalization contingent on remembering specific past memories? And second, does the strength or nature of this contingency differ across development? We found that the interdependence of generalization and episodic memory varies across development: generalization success in adults, but not in children, was contingent on context binding.

Understanding relational binding in early childhood: Interacting effects of overlap and delay.

Episodic memories typically share overlapping elements in distinctive combinations, and to be valuable for future behavior they need to withstand delays. There is relatively little work on whether children have special difficulty with overlap or withstanding delay. However, Yim, Dennis, and Sloutsky (Psychological Science, 2013, Vol. 24, pp. 2163-2172) suggested that extensive overlap is more problematic for younger children, and Darby and Sloutsky (Psychological Science, 2015, Vol. 26, pp. 1937-1946) reported that a 48-h delay period actually improves children's memory for overlapping pairs of items. In the current study, we asked how children's episodic memory is affected by stimulus overlap, delay, and age using visual stimuli containing either overlapping or unique item pairs. Children aged 4 and 6 years were tested both immediately and after a 24-h delay. As expected, older children performed better than younger children, and both age groups performed worse on overlapping pairs. Surprisingly, the 24-h delay had only a marginal effect on overall accuracy. Although there were no interactions, when errors were examined, there was evidence that delay buffered memory for overlapping pairs against cross-contextual confusion for younger children.

Pattern separation and pattern completion: Behaviorally separable processes?

Episodic memory capacity requires several processes, including mnemonic discrimination of similar experiences, termed pattern separation, and holistic retrieval of multidimensional experiences given a cue, termed pattern completion. Both computations seem to rely on the hippocampus proper, but they also seem to be instantiated by distinct hippocampal subfields. Thus, we investigated whether individual differences in behavioral expressions of pattern separation and pattern completion were correlated after accounting for general mnemonic ability. Young adult participants learned events comprised of a scene-animal-object triad. In the pattern separation task, we estimated mnemonic discrimination using lure classification for events that contained a similar lure element. In the pattern completion task, we estimated holistic recollection using dependency in retrieval success for different associations from the same event. Although overall accuracies for the two tasks correlated as expected, specific measures of individual variation in holistic retrieval and mnemonic discrimination did not correlate, suggesting that these two processes involve distinguishable properties of episodic memory.

The backbone network of dynamic functional connectivity.

Temporal networks have become increasingly pervasive in many real-world applications, including the functional connectivity analysis of spatially separated regions of the brain. A major challenge in analysis of such networks is the identification of noise confounds, which introduce temporal ties that are nonessential, or links that are formed by chance due to local properties of the nodes. Several approaches have been suggested in the past for static networks or temporal networks with binary weights for extracting significant ties whose likelihood cannot be reduced to the local properties of the nodes. In this work, we propose a data-driven procedure to reveal the irreducible ties in dynamic functional connectivity of resting-state fMRI data with continuous weights. This framework includes a null model that estimates the latent characteristics of the distributions of temporal links through optimization, followed by a statistical test to filter the links whose formation can be reduced to the activities and local properties of their interacting nodes. We demonstrate the benefits of this approach by applying it to a resting-state fMRI dataset, and provide further discussion on various aspects and advantages of it.