Artificial Multimodal Neuron with Associative Learning Capabilities: Acquisition, Extinction, and Spontaneous Recovery.

Associative multimodal artificial intelligence (AMAI) has gained significant attention across various fields, yet its implementation poses challenges due to the burden on computing and memory resources. To address these challenges, researchers have paid increasing attention to neuromorphic devices based on novel materials and structures, which can implement classical conditioning behaviors with simplified circuitry. Herein, we introduce an artificial multimodal neuron device that shows not only the acquisition behavior but also the extinction and the spontaneous recovery behaviors for the first time. Being composed of an ovonic threshold switch (OTS)-based neuron device, a conductive bridge memristor (CBM)-based synapse device, and a few passive electrical elements, such observed behaviors of this neuron device are explained in terms of the electroforming and the diffusion of metallic ions in the CBM. We believe that the proposed associative learning neuron device will shed light on the way of developing large-scale AMAI systems by providing inspiration to devise an associative learning network with improved energy efficiency.

Pavlovian or associative sensitization and its biological significance.

Pavlovian conditioning is typically distinguished from sensitization but a Pavlovian conditional stimulus (CS) also results in sensitization. A Pavlovian CS can sensitize responding to a probe stimulus that is related to the unconditional stimulus (US) or to the US itself. Pavlovian sensitization has been studied in the defensive, sexual, and feeding systems. In Pavlovian sensitization, the focus is not on a conditional response (CR) directly elicited by the CS but on the response mode that is activated by the CS. Activation of a response mode increases the probability of particular responses and also increases reactivity to various stimuli. Pavlovian sensitization reflects this increased stimulus reactivity. Pavlovian sensitization helps uncover successful learning in situations where a conventional CR does not occur. Pavlovian sensitization also encourages broadening our conceptions of Pavlovian conditioning to include changes in afferent processes. Implications for biological fitness and for basic and translational research are discussed.

Increased Generalization, Stronger Acquisition, or Reduced Extinction? Investigation of the Mechanisms Underlying the Acquisition-in-Multiple-Contexts Effect.

Prior research has demonstrated that conducting acquisition in multiple contexts results in more responding to the point that it can even nullify the benefit of subsequent extinction in multiple contexts on reducing renewal of excitatory responding. The underlying mechanism to explain why this happens has not been systematically examined. Using self-reported expectancy of the outcome, the current study investigates three mechanisms that potentially explain why acquisition in multiple contexts results in more responding-greater generalization, stronger acquisition learning, or slower extinction learning. Participants (N = 180) received discriminative training with a conditioned stimulus (CS+) and outcome pairing and a CS- → noOutcome pairing in either one or three contexts. This was followed by either extinction treatment in a novel context or no extinction. Finally, testing occurred in the acquisition context, the extinction context, or a novel context. Stronger renewal of extinguished conditioned expectation was observed for participants who received CS+ → Outcome pairings in three contexts relative to one context. There was no effect of the number of contexts on the strength of the excitatory CS+ → Outcome association or degree of inhibitory learning that occurred during extinction. This suggests that generalization is the mechanism responsible for the adverse impact to extinction learning when acquisition is conducted in multiple contexts.

Behaviorally Conditioned Immune Responses: "To Learn New Things, Read Old Books and Papers".

BACKGROUND: More than a century ago, experimental work and clinical observations revealed the functional communication between the brain and the peripheral immune system. This is documented on the one hand by studies first demonstrating the effects of catecholamines on the circulation of leukocytes in experimental animals and humans, and on the other hand via the work of Russian physiologist Ivan Petrovic Pavlov and his coworkers, reporting observations that associative learning can modify peripheral immune functions. This work later fell into oblivion since little was known about the endocrine and immune system's function and even less about the underlying mechanisms of how learning, a central nervous system activity, could affect peripheral immune responses. SUMMARY: In this article, we embark on a fascinating exploration of the historical trajectory of behaviorally conditioned immune responses. KEY MESSAGE: We will pay homage to the visionary scientists who laid the groundwork for this field of research, tracing its evolution from early theories of how associative learning can affect immunity to the modern-day insights that behavioral conditioning of pharmacological responses can be exploited to improve the efficacy of medical interventions for patients.

Functional diversity of dopamine axons in prefrontal cortex during classical conditioning.

Midbrain dopamine neurons impact neural processing in the prefrontal cortex (PFC) through mesocortical projections. However, the signals conveyed by dopamine projections to the PFC remain unclear, particularly at the single-axon level. Here, we investigated dopaminergic axonal activity in the medial PFC (mPFC) during reward and aversive processing. By optimizing microprism-mediated two-photon calcium imaging of dopamine axon terminals, we found diverse activity in dopamine axons responsive to both reward and aversive stimuli. Some axons exhibited a preference for reward, while others favored aversive stimuli, and there was a strong bias for the latter at the population level. Long-term longitudinal imaging revealed that the preference was maintained in reward- and aversive-preferring axons throughout classical conditioning in which rewarding and aversive stimuli were paired with preceding auditory cues. However, as mice learned to discriminate reward or aversive cues, a cue activity preference gradually developed only in aversive-preferring axons. We inferred the trial-by-trial cue discrimination based on machine learning using anticipatory licking or facial expressions, and found that successful discrimination was accompanied by sharper selectivity for the aversive cue in aversive-preferring axons. Our findings indicate that a group of mesocortical dopamine axons encodes aversive-related signals, which are modulated by both classical conditioning across days and trial-by-trial discrimination within a day.

Versatile Titanium Carbide MXene Thin-Film Memristors with Adaptive Learning Behavior.

With the advent of the modern era, there is a huge demand for memristor-based neuromorphic computing hardware to overcome the von Neumann bottleneck in traditional computers. Here, we have prepared two-dimensional titanium carbide (Ti3C2Tx) MXene following the conventional HF etching technique in solution. After confirmation of Ti3C2Tx properties by Raman scattering and crystallinity measurements, high-quality thin-film deposition is realized using an immiscible liquid-liquid interfacial growth technique. Following this, the memristor is fabricated by sandwiching a Ti3C2Tx layer with a thickness of 70 nm between two electrodes. Subsequently, current-voltage (I-V) characteristics are measured, revealing a nonvolatile resistive switching property characterized by a swift switching speed of 30 ns and an impressive current On/Off ratio of approximately 103. Furthermore, it exhibits endurance through 500 cycles and retains the states for at least 1 × 104 s without observable degradation. Additionally, it maintains a current On/Off ratio of about 102 while consuming only femtojoules (fJ) of electrical energy per reading. Systematic I-V results and conductive AFM-based current mapping image analysis are converged to support the electroforming mediated filamentary conduction mechanism. Furthermore, our Ti3C2Tx memristor was found to be truly versatile as an all-in-one device for demonstrating edge computation, logic gate operation, and classical conditioning of learning by the brain in Psychology.

Distinct Hippocampal Oscillation Dynamics in Trace Eyeblink Conditioning Task for Retrieval and Consolidation of Associations.

Trace eyeblink conditioning (TEBC) has been widely used to study associative learning in both animals and humans. In this paradigm, conditioned responses (CRs) to conditioned stimuli (CS) serve as a measure for retrieving learned associations between the CS and the unconditioned stimuli (US) within a trial. Memory consolidation, that is, learning over time, can be quantified as an increase in the proportion of CRs across training sessions. However, how hippocampal oscillations differentiate between successful memory retrieval within a session and consolidation across TEBC training sessions remains unknown. To address this question, we recorded local field potentials (LFPs) from the rat dorsal hippocampus during TEBC and investigated hippocampal oscillation dynamics associated with these two functions. We show that transient broadband responses to the CS were correlated with memory consolidation, as indexed by an increase in CRs across TEBC sessions. In contrast, induced alpha (8-10 Hz) and beta (16-20 Hz) band responses were correlated with the successful retrieval of the CS-US association within a session, as indexed by the difference in trials with and without CR.

The Role of Cerebellar Intrinsic Neuronal Excitability, Synaptic Plasticity, and Perineuronal Nets in Eyeblink Conditioning.

Evidence is strong that, in addition to fine motor control, there is an important role for the cerebellum in cognition and emotion. The deep nuclei of the mammalian cerebellum also contain the highest density of perineural nets-mesh-like structures that surround neurons-in the brain, and it appears there may be a connection between these nets and cognitive processes, particularly learning and memory. Here, we review how the cerebellum is involved in eyeblink conditioning-a particularly well-understood form of learning and memory-and focus on the role of perineuronal nets in intrinsic membrane excitability and synaptic plasticity that underlie eyeblink conditioning. We explore the development and role of perineuronal nets and the in vivo and in vitro evidence that manipulations of the perineuronal net in the deep cerebellar nuclei affect eyeblink conditioning. Together, these findings provide evidence of an important role for perineuronal net in learning and memory.

The effects of parental presence on amygdala and mPFC activation during fear conditioning: An exploratory study.

Learning safe versus dangerous cues is crucial for survival. During development, parents can influence fear learning by buffering their children's stress response and increasing exploration of potentially aversive stimuli. Rodent findings suggest that these behavioral effects are mediated through parental presence modulation of the amygdala and medial prefrontal cortex (mPFC). Here, we investigated whether similar parental modulation of amygdala and mPFC during fear learning occurs in humans. Using a within-subjects design, behavioral (final N = 48, 6-17 years, mean = 11.61, SD = 2.84, 60% females/40% males) and neuroimaging data (final N = 39, 6-17 years, mean = 12.03, SD = 2.98, 59% females/41% males) were acquired during a classical fear conditioning task, which included a CS+ followed by an aversive noise (US; 75% reinforcement rate) and a CS-. Conditioning occurred once in physical contact with the participant's parent and once alone (order counterbalanced). Region of interest analyses examined the unconditioned stress response by BOLD activation to the US (vs. implicit baseline) and learning by activation to the CS+ (vs. CS-). Results showed that during US presentation, parental presence reduced the centromedial amygdala activity, suggesting buffering of the unconditioned stress response. In response to learned stimuli, parental presence reduced mPFC activity to the CS+ (relative to the CS-), although this result did not survive multiple comparisons' correction. These preliminary findings indicate that parents modulate amygdala and mPFC activity during exposure to unconditioned and conditioned fear stimuli, potentially providing insight into the neural mechanisms by which parents act as a social buffer during fear learning. RESEARCH HIGHLIGHTS: (1)This study used a within-participant experimental design to investigate how parental presence (vs. absence) affects youth's neural responses in a classical fear conditioning task. (2)Parental presence reduced the youth's centromedial amygdala activation to the unconditioned stimulus (US), suggesting parental buffering of the neural unconditioned response (UR). (3)Parental presence reduced the youth's mPFC activation to a conditioned threat cue (CS+) compared to a safety cue (CS-), suggesting possible parental modulation of fear learning.

Feeding predictability as a cognitive enrichment protects brain function and physiological status in rainbow trout: a multidisciplinary approach to assess fish welfare.

Cognitive enrichment is a promising but understudied type of environmental enrichment that aims to stimulate the cognitive abilities of animals by providing them with more opportunities to interact with (namely, to predict events than can occur) and to control their environment. In a previous study, we highlighted that farmed rainbow trout can predict daily feedings after two weeks of conditioning, the highest conditioned response being elicited by the combination of both temporal and signalled predictability. In the present study, we tested the feeding predictability that elicited the highest conditioned response in rainbow trout (both temporal and signalled by bubbles, BUBBLE + TIME treatment) as a cognitive enrichment strategy to improve their welfare. We thus analysed the long-term effects of this feeding predictability condition as compared with an unpredictable feeding condition (RANDOM treatment) on the welfare of rainbow trout, including the markers in the modulation of brain function, through a multidisciplinary approach. To reveal the brain regulatory pathways and networks involved in the long-term effects of feeding predictability, we measured gene markers of cerebral activity and plasticity, neurotransmitter pathways and physiological status of fish (oxidative stress, inflammatory status, cell type and stress status). After almost three months under these predictability conditions of feeding, we found clear evidence of improved welfare in fish from BUBBLE + TIME treatment. Feeding predictability allowed for a food anticipatory activity and resulted in fewer aggressive behaviours, burst of accelerations, and jumps before mealtime. BUBBLE + TIME fish were also less active between meals, which is in line with the observed decreased expression of transcripts related to the dopaminergic system. BUBBLE + TIME fish tented to present fewer eroded dorsal fin and infections to the pathogen Flavobacterium psychrophilum. Decreased expression of most of the studied mRNA involved in oxidative stress and immune responses confirm these tendencies else suggesting a strong role of feeding predictability on fish health status and that RANDOM fish may have undergone chronic stress. Fish emotional reactivity while isolated in a novel-tank as measured by fear behaviour and plasma cortisol levels were similar between the two treatments, as well as fish weight and size. To conclude, signalled combined with temporal predictability of feeding appears to be a promising approach of cognitive enrichment to protect brain function via the physiological status of farmed rainbow trout in the long term.

Tailoring Classical Conditioning Behavior in TiO2 Nanowires: ZnO QDs-Based Optoelectronic Memristors for Neuromorphic Hardware.

Neuromorphic hardware equipped with associative learning capabilities presents fascinating applications in the next generation of artificial intelligence. However, research into synaptic devices exhibiting complex associative learning behaviors is still nascent. Here, an optoelectronic memristor based on Ag/TiO2 Nanowires: ZnO Quantum dots/FTO was proposed and constructed to emulate the biological associative learning behaviors. Effective implementation of synaptic behaviors, including long and short-term plasticity, and learning-forgetting-relearning behaviors, were achieved in the device through the application of light and electrical stimuli. Leveraging the optoelectronic co-modulated characteristics, a simulation of neuromorphic computing was conducted, resulting in a handwriting digit recognition accuracy of 88.9%. Furthermore, a 3 × 7 memristor array was constructed, confirming its application in artificial visual memory. Most importantly, complex biological associative learning behaviors were emulated by mapping the light and electrical stimuli into conditioned and unconditioned stimuli, respectively. After training through associative pairs, reflexes could be triggered solely using light stimuli. Comprehensively, under specific optoelectronic signal applications, the four features of classical conditioning, namely acquisition, extinction, recovery, and generalization, were elegantly emulated. This work provides an optoelectronic memristor with associative behavior capabilities, offering a pathway for advancing brain-machine interfaces, autonomous robots, and machine self-learning in the future.

DNA Reaction System That Acquires Classical Conditioning.

Biochemical reaction networks can exhibit plastic adaptation to alter their functions in response to environmental changes. This capability is derived from the structure and dynamics of the reaction networks and the functionality of the biomolecule. This plastic adaptation in biochemical reaction systems is essentially related to memory and learning capabilities, which have been studied in DNA computing applications for the past decade. However, designing DNA reaction systems with memory and learning capabilities using the dynamic properties of biochemical reactions remains challenging. In this study, we propose a basic DNA reaction system design that acquires classical conditioning, a phenomenon underlying memory and learning, as a typical learning task. Our design is based on a simple mechanism of five DNA strand displacement reactions and two degradative reactions. The proposed DNA circuit can acquire or lose a new function under specific conditions, depending on the input history formed by repetitive stimuli, by exploiting the dynamic properties of biochemical reactions induced by different input timings.

Glutamate inputs send prediction error of reward, but not negative value of aversive stimuli, to dopamine neurons.

Midbrain dopamine neurons are thought to signal reward prediction errors (RPEs), but the mechanisms underlying RPE computation, particularly the contributions of different neurotransmitters, remain poorly understood. Here, we used a genetically encoded glutamate sensor to examine the pattern of glutamate inputs to dopamine neurons in mice. We found that glutamate inputs exhibit virtually all of the characteristics of RPE rather than conveying a specific component of RPE computation, such as reward or expectation. Notably, whereas glutamate inputs were transiently inhibited by reward omission, they were excited by aversive stimuli. Opioid analgesics altered dopamine negative responses to aversive stimuli into more positive responses, whereas excitatory responses of glutamate inputs remained unchanged. Our findings uncover previously unknown synaptic mechanisms underlying RPE computations; dopamine responses are shaped by both synergistic and competitive interactions between glutamatergic and GABAergic inputs to dopamine neurons depending on valences, with competitive interactions playing a role in responses to aversive stimuli.

Conditioning to true content and artificial intelligence in psychophysiological intention recognition.

OBJECTIVE: The objective is to introduce a novel method for classical conditioning to true content (CtTC), and for the first time, apply this approach in the concealed information test (CIT) to effectively discern intentions. During CtTC, participants are trained to exhibit electrodermal responses whenever they recognize true content on a screen. Additionally, the objective is to evaluate a novel CIT-dataset preprocessing algorithm, employed to enhance machine learning (ML) classification performance. METHODS: A total of 84 participants were evenly divided into four groups. Two groups of participants devised plans for stealing money from a supermarket, while the other two groups did not engage in any planning. One planning group and one non-planning group underwent CIT examination, while the remaining groups were subjected to CtTC. RESULTS: The CIT accuracy initially stood at 52 % and increased to 71 % after Z-score and ML classification (McNemar test, p < 0.05). Conversely, the CtTC accuracy was 76 % and significantly improved to 93 % following Z-score and 95 % following ML classification (McNemar test, p < 0.05). In the best-performing classifiers, CtTC exhibited significantly superior metrics for guilty/innocent classification compared to CIT (Fisher's exact test, p < 0.05, power 1 - ß > 0.90). In the CtTC group, reactivity and sensitivity significantly increased, indicated by higher EDR amplitudes (p < 0.05, two-tailed t-test, power 1 - ß = 0.89) and the number of EDRs (p < 0.05, Fisher's exact test, power 1 - ß = 0.90). There was no statistically significant difference between the Z-score and ML classification. CONCLUSIONS: In the assessment of intentions, CtTC enhances both the sensitivity and accuracy of the CIT.

The blink reflex and its modulation - Part 1: Physiological mechanisms.

The blink reflex (BR) is a protective eye-closure reflex mediated by brainstem circuits. The BR is usually evoked by electrical supraorbital nerve stimulation but can be elicited by a variety of sensory modalities. It has a long history in clinical neurophysiology practice. Less is known, however, about the many ways to modulate the BR. Various neurophysiological techniques can be applied to examine different aspects of afferent and efferent BR modulation. In this line, classical conditioning, prepulse and paired-pulse stimulation, and BR elicitation by self-stimulation may serve to investigate various aspects of brainstem connectivity. The BR may be used as a tool to quantify top-down modulation based on implicit assessment of the value of blinking in a given situation, e.g., depending on changes in stimulus location and probability of occurrence. Understanding the role of non-nociceptive and nociceptive fibers in eliciting a BR is important to get insight into the underlying neural circuitry. Finally, the use of BRs and other brainstem reflexes under general anesthesia may help to advance our knowledge of the brainstem in areas not amenable in awake intact humans. This review summarizes talks held by the Brainstem Special Interest Group of the International Federation of Clinical Neurophysiology at the International Congress of Clinical Neurophysiology 2022 in Geneva, Switzerland, and provides a state-of-the-art overview of the physiology of BR modulation. Understanding the principles of BR modulation is fundamental for a valid and thoughtful clinical application (reviewed in part 2) (Gunduz et al., submitted).

Behavioral Tests for Associative Learning in Caenorhabditis elegans.

Learning is critical for survival as it provides the capacity to adapt to a changing environment. At the molecular and cellular level, learning leads to alterations within neural circuits that include synaptic rewiring, synaptic plasticity, and protein level/gene expression changes. There has been substantial progress in recent years on dissecting how learning and memory is regulated at the molecular and cellular level, including the use of compact invertebrate nervous systems as experimental models. This progress has been facilitated by the establishment of robust behavioral assays that generate a quantifiable readout of the extent to which animals learn and remember. This chapter will focus on protocols of behavioral tests for associative learning using the nematode Caenorhabditis elegans, with its unparalleled genetic tractability, compact nervous system of ~300 neurons, high level of conservation with mammalian systems, and amenability to a suite of behavioral tools and analyses. Specifically, we will provide a detailed description of the methods for two behavioral assays that model associative learning, one measuring appetitive olfactory learning and the other assaying aversive gustatory learning.

Visual accelerated and olfactory decelerated responses during multimodal learning in honeybees.

To obtain accurate information about the outside world and to make appropriate decisions, animals often combine information from different sensory pathways to form a comprehensive representation of their environment. This process of multimodal integration is poorly understood, but it is common view that the single elements of a multimodal stimulus influence each other's perception by enhancing or suppressing their neural representation. The neuronal level of interference might be manifold, for instance, an enhancement might increase, whereas suppression might decrease behavioural response times. In order to investigate this in an insect behavioural model, the Western honeybee, we trained individual bees to associate a sugar reward with an odour, a light, or a combined olfactory-visual stimulus, using the proboscis extension response (PER). We precisely monitored the PER latency (the time between stimulus onset and the first response of the proboscis) by recording the muscle M17, which innervates the proboscis. We found that odours evoked a fast response, whereas visual stimuli elicited a delayed PER. Interestingly, the combined stimulus showed a response time in between the unimodal stimuli, suggesting that olfactory-visual integration accelerates visual responses but decelerates the olfactory response time.

Brain activation during fear extinction recall in unmedicated patients with obsessive-compulsive disorder.

Specific brain activation patterns during fear conditioning and the recall of previously extinguished fear responses have been associated with obsessive-compulsive disorder (OCD). However, further replication studies are necessary. We measured skin-conductance response and blood oxygenation level-dependent responses in unmedicated adult patients with OCD (n = 27) and healthy participants (n = 22) submitted to a two-day fear-conditioning experiment comprising fear conditioning, extinction (day 1) and extinction recall (day 2). During conditioning, groups differed regarding the skin conductance reactivity to the aversive stimulus (shock) and regarding the activation of the right opercular cortex, insular cortex, putamen, and lingual gyrus in response to conditioned stimuli. During extinction recall, patients with OCD had higher responses to stimuli and smaller differences between responses to conditioned and neutral stimuli. For the entire sample, the higher the response delta between conditioned and neutral stimuli, the greater the dACC activation for the same contrast during early extinction recall. While activation of the dACC predicted the average difference between responses to stimuli for the entire sample, groups did not differ regarding the activation of the dACC during extinction recall. Larger unmedicated samples might be necessary to replicate the previous findings reported in patients with OCD.

The next frontier: Moving human fear conditioning research online.

Fear conditioning is a significant area of research that has featured prominently among the topics published in Biological Psychology over the last 50 years. This work has greatly contributed to our understanding of human anxiety and stressor-related disorders. While mainly conducted in the laboratory, recently, there have been initial attempts to conduct fear conditioning experiments online, with around 10 studies published on the subject, primarily in the last two years. These studies have demonstrated the potential of online fear conditioning research, although challenges to ensure that this research meets the same methodological standards as in-person experimentation remain, despite recent progress. We expect that in the coming years new outcome measures will become available online including the measurement of eye-tracking, pupillometry and probe reaction time and that compliance monitoring will be improved. This exciting new approach opens new possibilities for large-scale data collection among hard-to-reach populations and has the potential to transform the future of fear conditioning research.

Fear Conditioning by Proxy: The Role of High Affinity Nicotinic Acetylcholine Receptors.

Observational fear-learning studies in genetically modified animals enable the investigation of the mechanisms underlying the social transmission of fear-related information. Here, we used a three-day protocol to examine fear conditioning by proxy (FCbP) in wild-type mice (C57BL/6J) and mice lacking the ß2-subunit of the nicotinic acetylcholine receptor (nAChR). Male animals of both genotypes were exposed to a previously fear-conditioned (FC) cage mate during the presentation of the conditioned stimulus (CS, tone). On the following day, observer (FCbP) mice were tested for fear reactions to the tone: none of the ß2-KO mice froze to the stimulus, while 30% of the wild-type mice expressed significant freezing. An investigation of the possible factors that predicted the fear response revealed that only wild-type mice that exhibited enhanced and more flexible social interaction with the FC cage mate during tone presentations (Day 2) expressed fear toward the CS (Day-3). Our results indicate that (i) FCbP is possible in mice; (ii) the social transmission of fear depends on the interaction pattern between animals during the FCbP session and (iii) ß2-KO mice display a more rigid interaction pattern compared to wild-type mice and are unable to acquire such information. These data suggest that ß2-nAChRs influence observational fear learning indirectly through their effect on social behaviour.