Mnemonic discrimination-the ability to distinguish between existing memories and similar new inputs-supports accurate memory in the face of interference (e.g., in a card game, recognizing that the card you are currently seeing, the King of Hearts, is distinct from the King of Diamonds you encountered earlier in the same hand). Although hippocampal pattern separation has been emphasized as a key mechanism for this process, recent theoretical frameworks and empirical findings suggest that surrounding medial temporal lobe and cortical regions also contribute by resolving interference. An important source of interference in everyday life is temporal interference: similar events may compete not only because they overlap perceptually, but also because they are separated by intervening experiences (e.g., distinguishing the King of Hearts you are currently holding in your hand from the King of Diamonds you encountered earlier in the game becomes increasingly difficult as additional cards are played and additional hands intervene between the two episodes). Here, we examined how temporal interference impacts mnemonic discrimination and its neural correlates using high-resolution fMRI and a continuous recognition task in which the lag between an item's first presentation and its subsequent target or similar lure was systematically manipulated (10, 60, or 140 intervening trials). Behaviorally, increasing lag impaired both mnemonic discrimination and target recognition. At the neural level, lag effects were not observed in hippocampal subfields. Instead, temporal interference modulated activity in extrahippocampal MTL regions including parahippocampal and perirhinal cortex, as well as distributed cortical regions including frontoparietal control areas. Importantly, lag-related modulation was process-specific, with partially dissociable patterns for successful mnemonic discrimination and target recognition. Together, these findings demonstrate that temporal interference engages a distributed MTL-cortical network that supports interference resolution during mnemonic discrimination.

Rodent models of human adolescent binge drinking persistently reduce adult neurogenesis and induce innate immune signaling cascades, suggesting long-lasting dysfunction in the hippocampal neurogenic niche. Nevertheless, many neuroprogenitors survive, and the impact of adolescent ethanol on the mature phenotype of these surviving neurons is unknown, resulting in an unstudied cellular population that could impact hippocampal function. To label a cohort of immature neurons, we used doublecortin-CreERT2 transgenic mice crossed with Rosa26-CAG-tdTomato reporter mice, resulting in tamoxifen-induced labeling of newborn neurons. Female offspring underwent adolescent intermittent ethanol (AIE) (5 g/kg ethanol or water by gavage, 2-day-on/2-day-off from postnatal day 29-58 ± 1), followed by a multi-month ethanol-free period. AIE increased dendritic branch width and cellular migratory distance, increased hilar axonal varicosities, reduced dendritic spine density, and shifted action potential generation with escalating current injections in tdTomato+ cells. Although tdTomato+ expression was also decreased in AIE relative to CON mice, there was no significant change in the number of tdTomato+ somata. AIE also reduced microglial territory in the polymorphic and molecular layers despite no change in cell number. Furthermore, microglia entwined in the dendritic trees of surviving adolescent neuroprogenitors exhibited increased sphericity, potentially indicative of a shift towards a more reactive morphological phenotype. Surviving neuroprogenitors exhibited no changes in labeling of the vesicular acetylcholine transporter, despite robust cholinergic innervation proximal to the mature dendritic tree that reduced rapidly with increasing distance from dendritic arbors. These results suggest that AIE causes lasting changes in the architectural and physiological phenotype of surviving adolescent neuroprogenitors, which may be related to disrupted neuronal-microglial dynamics and have critical consequences for both hippocampal circuitry and function.

A neurocomputational theory of scene and memory construction during memory recall, prospective thinking, and autobiographical memory is described. The theory describes the nature of the representation of scenes using spatial view cells in a continuous attractor network in the medial parahippocampal cortex, and how the location in the scene being represented can be shifted by self-motion eye position update even when the scene is not visible. The theory builds on the evidence for humans for a ventromedial visual cortical scene pathway via the prostriate cortex and ventromedial visual cortical regions to the medial parahippocampal scene (or place) area where scenes are represented. The approach builds on a theory about how visual features in scenes are used along this pathway to build a continuous attractor network representation of scenes in the medial parahippocampal cortex that represents the relative spatial locations of features or landmarks in the scene. The theory shows how episodic memory can be built in an attractor network in hippocampal CA3 that learns associations between the location in a scene ("where"), the object or person at that location in the scene ("what"), and the reward/affective value conveyed to the hippocampus from the orbitofrontal cortex in part via the ventromedial prefrontal cortex and posterior cingulate cortex. The theory of scene construction then shows how backprojections from hippocampal CA1 via the entorhinal cortex to the medial parahippocampal cortex can recall or construct the scene representation in the medial parahippocampal cortex, the parahippocampal scene (or place) area; and how whole memories including visual imagery of objects, rewards, and their locations in scenes can be constructed by iterative neocortical-hippocampal-neocortical processing.

Viral infections during critical periods of pregnancy trigger inflammatory responses that alter offspring brain development and behavior, resulting in psychiatric disorder-associated phenotypes. This condition is commonly referred to as prenatal stress induced by maternal immune activation (MIA). Here, we investigate the effects of MIA on the neurophysiological properties of the dentate gyrus-CA3 circuit, a hippocampal circuit critical for cortical input processing and memory formation, and increasingly implicated in psychiatric disorders. Pregnant Sprague-Dawley dams were systemically administered polyinosinic:polycytidylic acid (poly I:C), a synthetic double-stranded RNA that elicits a viral-like immune response by activating the Toll-like receptor 3 (TLR3)-NF-kB signaling pathway. Extracellular recordings were performed on acute slices of male offspring at postnatal days 61-80. MIA increased the antidromic population spike amplification at the mossy fiber (MF)-granule cell layer response without modifying the antidromic responses at the CA3-CA3 commissural-associational fibers. Paired pulses with short interstimulus intervals revealed decreased levels of phasic and tonic GABAergic inhibition. In addition, MIA reduced coupling between presynaptic fiber volleys and MF field excitatory postsynaptic potentials (MF fEPSP). At the level of synaptic plasticity, MIA impaired frequency-dependent facilitation and decreased MF fEPSP paired-pulse facilitation, two hallmark forms of short-term plasticity at the MF-CA3 synapse. Lastly, MF-mediated long-term potentiation (MF LTP) and long-term depression (MF LTD) showed altered magnitudes in MIA slices. Collectively, our results indicate that MIA alters neuronal processing and output in the DG-CA3 hippocampal network, supporting the contribution of maternal inflammation to long-term circuit dysfunction associated with viral infections during pregnancy.

Intimate partner violence-related brain injury (IPV-BI) predominantly affects women and often reflects both structural trauma from mild traumatic brain injury (mTBI), as well as ischemia/hypoxia from non-fatal strangulation (NFS). This polytrauma can impair cognitive function, particularly learning and memory processes, which may contribute to the continuation of the cycle of violence. This study investigates the effects of mTBI and hypoxic/ischemic injury on hippocampal long-term potentiation (LTP), a biological model of learning and memory, to determine if the combination of mTBI+NFS would exacerbate deficits in LTP. Female rats were randomly assigned to either Sham, mTBI, NFS, or mTBI+NFS groups to assess acute injury metrics and LTP capacity 7 days after injury. At the time of injury, NFS and mTBI+NFS groups exhibited significant reductions in blood oxygen saturation, heart rate, and poor neurological assessment metrics. Isolated mTBI and NFS impaired LTP in the CA1 hippocampal region while mTBI+NFS increased synaptic excitability in the CA1 yet impaired CA1 LTP. Contrary to our hypothesis, the combined mTBI+NFS did not exacerbate deficits, but these findings highlight that even a single episode of mTBI or NFS is sufficient to impair hippocampal LTP in the female CA1 region.

The ventral regions of the hippocampus play a crucial role in emotional responses through projections to the hypothalamus. In this context, the ventral hippocampus sends extensive projections to components of the medial hypothalamic defensive circuit, where projections to the anterior hypothalamic nucleus (AHN) influence both anxiogenic and innate fear responses, as well as contextual fear reactions. This study aims to examine the branching projection pattern of hippocampal cells that directly project to the AHN and identify the complete range of brain regions affected by key hippocampal outputs that drive fear and anxiety. To accomplish this, we first explored the full range of hippocampal cells that project directly to the anterior hypothalamic area using an AAV retroviral vector carrying the enhanced yellow fluorescent protein (EYFP) reporter gene. We found that the hippocampal cells projecting to the anterior hypothalamus were located primarily along the entire rostro-caudal axis of the ventral subiculum. Next, by combining an AAV retro vector injection into the AHN to insert Cre recombinase specifically in the subicular neurons projecting to the anterior hypothalamic region and subicular injections of an AAV vector expressing membrane-bound GFP and the presynaptic protein synaptophysin fused to the red fluorescent protein mRuby in a Cre-dependent manner, we investigated the collateral branching pattern of subicular neurons that project to the AHN. Our results demonstrated that ventral subicular neurons projecting to the anterior hypothalamus exhibit an extensive branching projection pattern that connects to several brain regions involved in various functions, including memory, motivation, arousal, circadian rhythmicity, neuroendocrine regulation, and goal-oriented behavior. This reveals that ventral subicular anterior hypothalamic-projecting cells display a widespread branching projection pattern, uniquely positioning them to influence complex, integrated responses while driving anxiety and fear behaviors.

The hippocampus is central to memory and spatial planning. A prominent candidate mechanism supporting navigational decision making is hippocampal theta sequences-brief (~120 ms) place-cell sequences representing future trajectories-which have been interpreted as planning signals based on their alternation between maze arms in T-maze tasks. However, recent work suggests that intrinsic left-right alternation in medial entorhinal cortex (MEC) theta sequences may underlie this phenomenon. Here, we test this hypothesis using a model of MEC theta dynamics in a T-maze and show that standard Bayesian decoding yields hippocampal theta sequences that alternate between arms. We then analyzed existing simultaneous MEC-hippocampal recordings from Vollan et al. (2025). We found tight coupling between MEC and hippocampus, along with coordinated left-right alternation, while the animal was performing an alternation task. Notably, this conjoint alternation occurs not only at choice points but also during inbound and forced-turn trials. These findings suggest that left-right alternation in hippocampal theta sequences is a ubiquitous phenomenon driven by MEC dynamics, challenging the interpretation that these specifically reflect intentional decision-making or planning processes only.

Physical activity is found to promote healthy brain aging that is often indexed by hippocampal structure and function; however, evidence from cross-sectional studies remains limited and inconsistent, particularly for hippocampal morphology. Discrepancies in the literature may be explained by a historical focus on volumetric analyses, which obscure nuanced changes in cortical geometry, particularly in heavily folded regions of the hippocampus. In this study, we conducted a cross-sectional analysis to examine whether moderate-to-vigorous physical activity (MVPA) was associated with hippocampal macrostructure (thickness and surface area) and microstructure (T1w/T2w ratio) among 555 cognitively healthy adults (ages 36-90 years) from the Human Connectome Project-Aging cohort. Weekly MVPA was measured using the International Physical Activity Questionnaire, and participants were categorized as low-active (≥ 0-< 150 min/week), moderately active (≥ 150-< 300 min/week), or highly active (≥ 300 min/week) according to the World Health Organization 2020 guidelines on physical activity. Hippocampal thickness, surface area, and T1w/T2w ratio were derived from unfolded hippocampal surfaces using HippUnfold. Multivariate general linear models tested linear and quadratic age effects and MVPA × age interactions separately for each hemisphere, controlling for demographic and health covariates. Among the low-active group, age was associated with widespread multivariate differences in hippocampal macrostructure and microstructure, characterized by lower regionally specific thickness and surface area and by a higher T1w/T2w ratio. MVPA moderated age-related changes in the left hippocampus, with significant effects observed in the moderate- and high-MVPA groups. These effects were characterized by greater surface area and T1w/T2w ratio and were associated with lower thickness values. These results suggest that individuals who meet or exceed physical activity recommendations exhibit distinct age-related differences in hippocampus macrostructure and microstructure compared with those who do not.

Traditionally viewed as a mere relay station of the hippocampal formation, the subiculum is emerging as a computational hub of the medial temporal lobe, essential for multiplexed neural representations guiding behavior. While foundational animal research has revealed that its sophisticated cytoarchitecture allows for conjunctive coding of spatiotemporal information, our understanding of how such neural properties translate to human cognitive function remains incomplete. The evolutionary expansion of its anterior region and the enhancement of its anatomical complexity and widespread neocortical connectivity imply that the function of the human subiculum may extend beyond the spatial domain. In this review, we examine how the subiculum may serve as a scaffold for goal-directed spatiotemporal computations in humans and speculate on its role in supporting abstract cognitive functions such as mental imagery and episodic simulation.

Parkinson's disease (PD) encompasses motor and neuropsychiatric symptoms (NPS) including depression, anxiety, and apathy. Cognitive subtypes-PD with mild cognitive impairment (PD-MCI) and PD with dementia (PDD)-and the related Dementia with Lewy Bodies (DLB) exist within the spectrum of Lewy body disease (LBD). While NPS are a major comorbidity, established imaging biomarkers are not available, nor is the neuronal basis well known. Additionally, prevalence of NPS may be linked to cognitive status. The hippocampus is implicated in the development of NPS and may undergo microstructural changes that are not reflected in volume changes. High-resolution diffusion tensor imaging (DTI) was used to identify hippocampal changes in LBD compared to healthy controls (HC) and examine the relationship with NPS within the LBD group, both through association with DTI measures as well as comparison of cognitive subgroups. The hippocampus was manually segmented in 38 LB spectrum participants (aged 58-87 years, 32% female) and 35 HC (aged 50-83 years, 54% female) on high-resolution (1 × 1 × 1 mm[3]) DTI. The LBD group (LBD-All) was divided into LBD-Cognitively Impaired (LBD-CI) and LBD-Cognitively Unimpaired (LBD-CU). Hippocampal volume, mean diffusivity (MD), and fractional anisotropy (FA) were extracted and compared across groups. Relationships between significantly different hippocampal measures and depression, anxiety, and apathy were explored within the LBD-All group. Hippocampal FA was lower in LBD-All relative to HC; however, no differences in MD or volume were observed. Hippocampal FA was positively associated with depression scores within LBD-All. There were no differences in volume, MD, or FA between LBD-CI and LBD-CU. Lower hippocampal fractional anisotropy in the LBD spectrum relative to HC potentially reflects microstructure alterations; however, higher hippocampal FA in LBD with more severe depression suggests more complex microstructural changes such as altered synaptic density, Lewy neurite organization, or neuroplastic changes.

The study of cross-species differences in the organization of the cerebral cortex received significant attention during the past century. However, the evolutionary principles governing changes in brain connectivity and their impact on cognition remain poorly understood. In the current study, we analyzed differences in anatomical input from the cerebral cortex to the hippocampal region across six species spanning more than 100 million years of mammalian evolution-the tenrec, rat, cat, marmoset, macaque and human. Our results demonstrate that, unlike direct transmodal cortical input, direct unimodal cortical input to the hippocampal region was selectively eliminated with the increase in brain size. Additionally, input from primary sensory regions was eliminated prior to input from non-primary sensory regions. Our findings suggest that hippocampus-related processes in different species operate on fundamentally different types of information, potentially underpinning cross-species differences in hippocampus-dependent cognition. Furthermore, our findings outline the evolutionary trajectory of the anatomical circuitry associating the human hippocampal region with the neocortex.

Period circadian regulator 1 (PER1) regulates circadian gene expression and is critical for spatial memory. However, its role in regulating synaptic plasticity between diurnal phases remains largely unknown. Here, we examined long-term potentiation (LTP) and long-term synaptic depression (LTD) in hippocampal slices prepared from aged male Per1 knockout (Per1 KO) rats and their wildtype (WT) littermates during the light/inactive and dark/active phases. In WT rats, theta-burst stimulation (TBS) consistently induced stable LTP that was greater during the dark relative to the light phase. In contrast, in Per1 KO rats, TBS resulted in synaptic potentiation that decayed to baseline levels within ~90 min in both phases. Notably, WT rats readily expressed LTD following 1 Hz low-frequency stimulation (LFS) during the dark phase but not the light phase. By comparison, in Per1 KO rats, LTD was inducible by LFS during both the dark and light phases. LTD in both WT and Per1 KO rats was dependent on the activation of N-methyl-d-aspartate receptors. Basal synaptic transmission was reduced and GluA1 S845 phosphorylation was increased during the dark phase in both WT and Per1 KO rats. Together, our findings identify PER1 as a key regulator of synaptic plasticity between diurnal phases in aged rats.

Mind wandering (MW) has been frequently reported to impair task performance. In this study, we investigated the hypothesis that it also affects error monitoring and adjustment. Several recent studies suggest that the hippocampus and adjacent regions significantly contribute to error monitoring. Therefore, we analyzed single-unit recordings from the medial temporal lobe (MTL) in presurgical epilepsy patients performing an attention task, which included interspersed experience sampling probes. Behavioral performance data revealed reduced error adjustment during periods of MW compared to task-focused attention. Analysis of single-unit recordings showed population responses related to performance errors within the hippocampus and adjacent regions, supporting the idea that MTL regions contribute to error monitoring. Importantly, this error-related population activity was significantly reduced during periods of MW compared to task focus. Together, our findings clearly support the hypothesis that MW impairs both error monitoring and adjustment.

The hippocampus is critical for episodic memory. The relationship between its structural variability and individual differences in episodic memory performance among neurotypical young adults, however, remains unclear. Previous studies have not consistently found associations between hippocampal volume and episodic memory or other aspects of hippocampally dependent performance in this population. Here, we moved beyond volumetric approaches and tested whether more fine-grained morphometric features, particularly gyrification, are related to behavioral performance. To obtain these measures, we used HippUnfold, a deep learning-based segmentation tool that models the hippocampus as a folded cortical ribbon and uniquely quantifies local variability in folding patterns, including in the hippocampal head. We focused on an open-access dataset with high-resolution MRI scans and multiple behavioral measures on hippocampal-dependent tasks, including episodic and semantic components of the Autobiographical Interview. To improve segmentation quality for the high-resolution images in this dataset, we trained an nnU-Net segmentation model for HippUnfold that was optimized for high-resolution T2w images. Volumetric and surface-based measures were computed for 186 participants (97 female). We found no reliable associations between hippocampal volumetric measures, including subfield volumes, and behavioral performance. In contrast, surface-based analysis of hippocampal gyrification revealed significant bilateral clusters in the anterior hippocampus, both of which were associated with measures of episodic but not semantic autobiographical memory. The right hemisphere cluster was located in the subiculum and CA1, whereas the left hemisphere cluster was primarily situated in the subiculum. These findings demonstrate the utility of surface-based hippocampal morphometry for linking hippocampal structure to behavioral performance. They suggest that even in neurotypical young adults individual differences in episodic memory are tied to structural hippocampal variation.

Memory of the order in which events occurred addresses the "when" component of episodic-like memory in animals. Seminal work demonstrated that normal rats remember the order in which they encountered odors, and hippocampal lesions selectively impaired this memory for order while leaving recognition performance intact. This work inspired us to develop an image-order paradigm to model episodic memory in monkeys. Monkeys saw five images in sequence and at test reported which of two images appeared first. Unlike the case with rats, hippocampal lesions did not impair memory for order, raising questions about which memory systems monkeys used in this task. Because the memory intervals were short in the image-order task, monkeys may have used nonhippocampal-dependent working memory instead of long-term memory. To test this possibility, we introduced a concurrent cognitive load between the study and test phases of the image-order task. We hypothesized that if monkeys were using working memory to remember the order of the images, concurrent cognitive load would impair performance. We also tested recognition performance as a control, because familiarity-based recognition is relatively resistant to concurrent cognitive load. Memory for order was much more attenuated by concurrent cognitive load than recognition. These results indicate that monkeys used working memory to solve this image-order task, making it a poor model of long-term episodic memory. The findings also provide clear evidence for a dissociation between working memory-dependent order memory and familiarity-based recognition memory.