The neural mechanisms supporting delayed threat memory recall in humans remain poorly understood. Animal studies indicate that the paraventricular nucleus of the thalamus (PVT) contributes to threat-memory retrieval after longer retention intervals, but whether the human PVT shows cue-related discriminability during delayed recall remains unclear. Human neuroimaging studies of threat learning have largely focused on memory assessed within 24-48 h, limiting insight into neural responses when previously threatening cues are re-encountered after longer delays. Here, adult participants completed threat acquisition on Day 1 and returned 7-10 days later (n = 28) for a delayed threat-memory recall task while undergoing fMRI. We examined activity in the dorsal midline thalamus centered on a histology-informed PVT region of interest. During Day 1 acquisition, the PVT did not discriminate between CS+ and CS-, despite differential electrodermal activity (EDA) and unconditioned stimulus expectancy ratings. In contrast, during early Day 7 recall, the PVT showed greater activation to previously reinforced CS+ than CS-. In an independent sample (n = 56), PVT activity similarly did not discriminate between CS+ and CS- during acquisition or immediate extinction, but overall PVT activation was greater during acquisition than extinction, regardless of CS type. These findings provide preliminary evidence that a human PVT-centered dorsal midline thalamic region is sensitive to prior threat reinforcement during delayed recall, while also showing engagement during acute threat-learning contexts. More broadly, they support future investigation of PVT-centered thalamic circuitry in the persistence, retrieval, and updating of threat memories, as well as in salience-related processes.

Planaria, a freshwater flatworm with unique regenerative abilities, has emerged as a valuable model organism for studying learning and memory. A growing body of evidence shows that planarians can recover their learning behaviour even after decapitation and regeneration. Behavioural paradigms, including classical, operant, and threat-related conditioning responses, show the capacity of planaria in learning and expressing both appetitive and aversive memories. Evidence at the molecular level, further confirms the presence of principal neurotransmitters, including glutamate, gamma-aminobutyric acid (GABA) and dopamine, which may contribute to planarian learning, memory and behavioural expression. Importantly, the dynamic balance among glutamatergic, GABAergic and dopaminergic pathways is consistently implicated in learning, memory, neuronal expression and behavioural activity. This review summarises behavioural evidence for learning and memory and highlights the key neurotransmitter systems that regulate these processes. It outlines future research directions for planaria as a valuable model for translational studies on memory and learning.

Stress has powerful effects on memory, but previous studies have yielded mixed findings regarding its impact on associative memory and the relationship between cortisol response and memory performance. Whether stress affects both encoding and retrieval of associative memory through a common mechanism remains unclear. In addition, an individual's baseline cortisol level should be considered to clarify the cortisol reactivity - memory relationship. In this study, we recruited 63 participants for Experiment 1 and 56 participants for Experiment 2 to learn unrelated word pairs with negative and neutral valence and complete an associative recognition task. They were exposed to a stress or control condition before either the encoding phase (Experiment 1) or the retrieval phase (Experiment 2). The results showed that pre-retrieval stress significantly impaired associative memory. For pre-encoding stress, the group difference was not significant, but stress significantly predicted the corrected recognition and the false alarm (FA) rate in the regression analysis. Regardless of whether stress was induced at pre-encoding or pre-retrieval, cortisol reactivity positively predicted the FA rate, while baseline cortisol levels negatively predicted the FA rate. In addition, baseline cortisol levels moderated the effect of stress on memory performance. When the baseline cortisol levels were lower, a greater cortisol reactivity was associated with a higher FA rate in both experiments. These results suggest that stress impairs the ability to distinguish old memories from lures, which is a common mechanism across memory phases. They also highlight the role of baseline cortisol in moderating the relationship between cortisol reactivity and associative memory.

The endogenous opioid system is thought to regulate Pavlovian fear conditioning in rats by instantiating some component of the error correction signal that underlies this learning. Consistent with this view, opioid receptor antagonists enhance the acquisition of first-order conditioned fear when a novel stimulus (S1, e.g., a light) is paired with a brief but painful foot shock; and impair the extinction of this fear when an already-conditioned S1 is repeatedly presented in the absence of the shock. The present study examined whether opioid receptors also regulate the acquisition and extinction of fear in protocols where the conditioning does not involve the experience of pain. Specifically, we established second-order conditioned fear (measured in freezing) using protocols in which a second novel stimulus (S2, e.g., a tone) was sequentially or simultaneously paired with an already-conditioned S1. We then found that subcutaneous injections of the opioid receptor antagonist, naloxone (2.5 mg/kg): 1) enhanced acquisition of freezing to S2 when it was sequentially or simultaneously paired with the already-conditioned S1; and 2) impaired extinction of a sequentially or simultaneously conditioned S2 when it was repeatedly presented in the absence of the S1. These results support the proposal that opioid receptors regulate fear learning via processes that go beyond their role in conditioned analgesia and pain modulation. They are discussed with respect to a role for opioid receptors in regulating the contribution of prior learning to the calculation of prediction error, and/or inhibitory learning that occurs across the course of second-order fear conditioning and extinction.

C/D box small nucleolar RNAs (snoRNAs) comprise a class of small noncoding RNAs with important regulatory effects on cellular RNA function. Although it is well established that snoRNAs coordinate the post-transcriptional modification of pre-ribosomal and small nuclear RNAs by 2'-O-methylation, which leads to enhanced RNA stability, whether they are involved in memory-related processes remains relatively unexplored. Using targeted sequencing, we have identified more than 150 C/D box snoRNAs in the prefrontal cortex of male C57BL/6J mice, including 31 which are differentially expressed in response to fear extinction learning, and 2 putative novel C/D box snoRNAs. A subset of these snoRNAs, including many orphans, are enriched in the synaptic compartment and show dynamic changes in subcellular localization during learning. Fear extinction learning-induced increases in synapse-enriched snord64 were accompanied by increased 2'-O-methylation within the 3-UTR of the mRNA encoding the ubiquitin ligase RNF146; snord64 knockdown proportionally reduced this methylation, providing correlative evidence for snord64 as the methylation guide. These findings reveal a population of experience-dependent, synaptically localized snoRNAs and implicate C/D box snoRNA activity in the epitranscriptomic regulation of synaptic mRNAs during fear extinction learning.

The prelimbic cortex (PrL) plays a critical role in reward-related memory processes. While its functional interactions with the ventral tegmental area (VTA) and cornu ammonis 1 (CA1) are implicated in reward circuitry, the specific oscillatory dynamics between these regions during the expression of morphine-associated memory remain poorly understood. In this study, local field potentials (LFPs) were recorded from the PrL, CA1, and VTA to investigate the PrL-CA1 and PrL-VTA pathways in a rat model of morphine-induced conditioned place preference (CPP). Recordings were analyzed during both pre-conditioning and post-conditioning (expression) phases. Coherence and Granger causality (GC) analyses evaluated functional synchronization and directional information flow across these pathways. A comparison of relative power between saline- and morphine-treated rats during the post-conditioning phase revealed significant alterations across all LFP sub-bands in the VTA. Coherence analysis demonstrated a significant increase in PrL-VTA synchronization during the expression phase, particularly within the theta and alpha bands. In contrast, coherence within the PrL-CA1 pathway remained unaltered during either phase. Furthermore, GC analysis indicated distinct, band-specific directional changes: a significant increase in delta-band causality in both directions of the PrL-VTA pathway, alongside specific reductions in delta, beta, and low-gamma information flow from the PrL to CA1. These results indicate that the expression of morphine-associated reward memory recruits the PrL-VTA pathway through increased functional synchronization and bidirectional information flow, while overall PrL-CA1 coherence remains unaffected. These findings highlight pathway-specific oscillatory dynamics underlying reward conditioning and memory retrieval.

The hippocampus plays a fundamental role in contextual fear conditioning (CFC). Studies have found that its ventral portion is mostly related to emotional processing, while the dorsal portion is related to the processing of contextual memories. The prelimbic cortex (PL) is also associated with contextual fear conditioning with temporal discontinuity (CFC-5s), in which a brief time gap separates context exposure and the aversive stimulus. Studies have shown that PL is involved in tasks that require the maintenance of information over short time intervals, such as working memory. The aim of the present study is to verify whether pharmacological manipulations of the dorsal hippocampus (dHPC) unilaterally in conjunction with contralateral inactivation of the PL interfere with the acquisition of CFC-5s. For this, Male Wistar rats were divided into three groups: asymmetric manipulation (CONTRA), ipsilateral manipulation (IPSI) and saline administration (SALINE). In Experiment 1, AP5 was infused into the dHPC and muscimol into the PL; in Experiment 2, muscimol was infused into both regions. Rats were trained and tested to a contextual fear conditioning protocol with temporal discontinuity. The results show that infusion of AP5 in the dorsal hippocampus impairs CFC-5s, whereas muscimol infusion does not. These findings indicate that NMDA receptor-dependent plasticity in the dHPC, rather than transient neural activity, is critical for the interaction between dHPC and PL during the acquisition of temporally discontinuous contextual memories.

The role of adult-born neurons in the hippocampal dentate gyrus and its significance in memory processes is a field of high interest. To add to this literature, we investigated the feasibility of ivermectin-based chemogenetic manipulation of recently born neurons in the hippocampus in altering spatial and contextual memory retrieval. The adult-born neurons in male mice were targeted using a nestin-cre mouse model, employing lentivirus-mediated transduction of inhibitory chloride channel constructs sensitive or insensitive to ivermectin in order to be able to silence them during different developmental stages. Two mouse age groups were compared: 5 weeks and 5 months old at the time of lentivirus injection. Behavioural assessments involving the Morris water maze and contextual fear conditioning were conducted. Unexpectedly, ivermectin administration alone led to apparent memory impairments regardless of the virus construct, confounding behavioural test outcomes. This study highlights the challenge of distinguishing specific effects of targeting adult-born neuronal activity from nonspecific effects of ivermectin on cognition.

Genetic tracing revealed that subsets of neurons in the medial amygdala (MeA) were labeled with the transneuronal tracer tWGA-DsRed originating from bitter taste receptor cells in male mice, suggesting the locations of bitter taste-relaying neurons in the MeA. Although not many studies have reported the involvement of MeA neurons in taste information processing, several studies clearly indicated that the MeA neurons play an important role in taste neophobia that refers to a reduction in consumption of a novel taste. On the other hand, several amygdaloid nuclei of the amygdala also operate to acquire conditioned taste aversion (CTA), which taste novelty determines the efficiency and strength for, while processing the conditioned stimulus (CS) followed by the unconditioned one (US) to elicit behavioral aversion to the CS. Here we combined genetic tracing and immunohistochemical analyses to examine whether subpopulations of tWGA-DsRed-labeled MeA neurons which inherently receive aversive bitter input can induce changes in responses to the novelty of saccharin and the CS saccharin during CTA learning. Immunohistochemical detection of Zif268 induction revealed that saccharin activated more tWGA-DsRed-labeled MeA neurons of mice that experienced saccharin at the first time than it did in mice experiencing saccharin multiple times. The CS saccharin activated a larger number of tWGA-DsRed-labeled MeA neurons after mice acquired CTA memory, compared with those detected in mice without CTA. Our results also suggest that the increased population of the CS-activated neurons among tWGA-DsRed-labeled MeA neurons may remain unchanged and activated by the CS after the subsequent extinction of CTA memory.

Understanding the neural mechanisms underlying the formation, retrieval, and extinction of taste memory is crucial for elucidating learning and memory processes. Using the conditioned taste aversion (CTA) model, we investigated the roles of the amygdala and gustatory cortex (GC), key structures consistently implicated in taste memory. Our study focused on the plastic changes associated with both appetitive and aversive taste memories, emphasizing the importance of extinction in promoting adaptive behavior and counteracting maladaptive responses. Memory processes elicit synaptic plasticity, often reflected in increased expression of neuronal remodeling markers such as growth-associated protein 43 (GAP-43) and synaptophysin (SYN). We examined the involvement of GAP-43 and SYN in taste memory by immunohistochemical analysis in male rats subjected to the formation of appetitive and aversive memories. During memory retrieval, GAP-43 expression was significantly upregulated in the basolateral amygdala (BLA), particularly under aversive conditions. In contrast, memory extinction induced a robust increase in GAP-43 expression in the central amygdala of overtrained animals, accompanied by elevated SYN expression in both the basolateral and central nuclei. These findings highlight the critical roles of GAP-43 and SYN in mediating synaptic changes during memory extinction, supporting the view of extinction as a dynamic, plastic process essential for behavioral adaptation. Our results advance understanding of the molecular substrates of taste memory and suggest potential targets for interventions aimed at modifying maladaptive memory traces, with implications for therapeutic strategies addressing aversive experiences and enhancing adaptive learning.

In this selective review, we summarise recent studies from our laboratory in which we have assessed the effect of different variables (such as contiguity and contingency) on the generalization of prior learning. Much of this work was inspired by Pavlov's work in which it was first documented that trace conditioning can broaden generalization, a finding that we have replicated in human participants. We then describe studies in which we have assessed whether the affective valence of stimuli and outcomes can affect generalization, and we find that indeed human participants can generalize across the valence of stimuli and this can interact with different outcomes. Finally, we review a study in which we assessed whether changes in a response-produced safety signal can influence avoidance and conclude that it does. We also observed that this generalizes depending on the similarity between trained and tested safety signals (i.e., outcomes). Overall, we conclude that generalization is a much more complex phenomenon than previously thought, and that more research is needed to understand this phenomenon and make it amenable to current theorising.

Inhibition of delay has received relatively little attention compared with other Pavlovian phenomena, despite its early description by Pavlov. This review examined the conditions under which inhibition of delay develops and evaluated evidence for it both as a phenomenon, reflected in temporal discrimination of the conditioned stimulus, and as a mechanism involving inhibitory processes acting early in the stimulus. Inhibition of delay has been observed across many species and procedures, typically after extensive training and most often with aversive unconditioned stimuli. Although this diversity suggests broad generality, evidence from appetitive conditioning remains limited and inconsistent. A meta-analytic assessment indicates that temporal discrimination reliably emerges under inhibition-of-delay protocols, but substantial unexplained variability points to unidentified procedural moderators. Evidence for inhibitory processes includes gradual reductions in responding early in the conditioned stimulus, increases in response latency, and recovery of responding following disinhibition, retention intervals, or context changes. However, formal tests of conditioned inhibition have produced mixed findings, indicating that the inhibition involved differs from standard Pavlovian conditioned inhibition and may more closely resemble extinction-like processes. Overall, inhibition of delay appears to be a robust yet incompletely understood form of temporal inhibition that warrants further systematic investigation.

The common marmoset is rapidly emerging as a powerful nonhuman primate model in neuroscience, yet the development of scalable, mechanistically informative cognitive paradigms has lagged behind advances in neural recording and genetic tools. Here, we introduce and validate a touchscreen-based spatial working memory task designed for direct cross-species translation between marmosets and humans. The paradigm independently manipulates retention delay and spatial separation between test choice stimuli, enabling parametric control over maintenance and interference demands within a single framework. Twelve marmosets and seventy-one human participants performed a Delayed Non-Match-to-Position task in which memory delay (1, 5, 10 s) and angular separation between target and distractor locations were systematically varied. Across species, accuracy declined as delay increased and as spatial separation decreased, demonstrating robust sensitivity to both maintenance demands and similarity-based interference. Critically, delay and separation interacted in both species, indicating that these had additive effects. Choice latency analyses further supported interpretation of performance, with slower responses on incorrect trials in both groups. Together, these findings establish a scalable and translationally aligned spatial working memory paradigm that captures interacting maintenance and interference processes. This task provides a powerful platform for circuit-level investigation and offers a sensitive cognitive assay for future studies of aging, neurodegenerative disease, and therapeutic intervention in the marmoset model.