The neural circuitry and molecular mechanisms underlying delay and trace eyeblink conditioning in mice

Fear conditioning: Insights into learning, memory and extinction and its relevance to clinical disorders

Fear, whether innate or learned, is an essential emotion required for survival. The learning, and subsequent memory, of fearful events enhances our ability to recognise and respond to threats, aiding adaptation to new, ever-changing environments. Considerable research has leveraged associative learning protocols such as contextual or auditory forms of fear conditioning in rodents, to understand fear learning, memory consolidation and extinction phases of memory. Such assays have led to detailed characterisation of the underlying neurocircuitry and neurobiology supporting fear learning processes. Given fear processing is conserved across rodents and humans, fear conditioning experiments provide translational insights into fundamental memory processes and fear-related pathologies. This review examines associative learning protocols used to measure fear learning, memory and extinction, before providing an overview on the underlying complex neurocircuitry including the amygdala, hippocampus and medial prefrontal cortex. This is followed by an in-depth commentary on the neurobiology, particularly synaptic plasticity mechanisms, which regulate fear learning, memory and extinction. Next, we consider how fear conditioning assays in rodents can inform our understanding of disrupted fear memory in human disorders such as post-traumatic stress disorder (PTSD), anxiety and psychiatric disorders including schizophrenia. Lastly, we critically evaluate fear conditioning protocols, highlighting some of the experimental and theoretical limitations and the considerations required when conducting such assays, alongside recent methodological advancements in the field. Overall, rodent-based fear conditioning assays remain central to making progress in uncovering fundamental memory phenomena and understanding the aetiological mechanisms that underpin fear associated disorders, alongside the development of effective therapeutic strategies.

Neuronal dynamics of cerebellum and medial prefrontal cortex in adaptive motor timing

Precise temporal control of sensorimotor coordination and adaptation is a fundamental basis of animal behavior. How different brain regions are involved in regulating the flexible temporal adaptation remains elusive. Here, we investigated the neuronal dynamics of the cerebellar interposed nucleus (IpN) and the medial prefrontal cortex (mPFC) neurons during temporal adaptation between delay eyeblink conditioning (DEC) and trace eyeblink conditioning (TEC). When mice were trained for either DEC or TEC and subsequently subjected to a new paradigm, their conditioned responses (CRs) adapted virtually instantaneously. Changes in the activity of the IpN neurons related to CR timing were prominent during DEC-to-TEC adaptation, but less so during TEC-to-DEC adaptation. In contrast, mPFC neurons could rapidly alter their modulation patterns during both adaptation paradigms. Accordingly, silencing the mPFC completely blocked the adaptation of CR timing. These results illustrate how cerebral and cerebellar mechanisms may play different roles during adaptive control of associative motor timing.

Impact of Fornix Lesions on Tone-Off Delay- vs Tone-On Trace- Eyeblink Conditioning in Rats

Research has shown differences in the neural mechanisms that support trace and delay eyeblink conditioning. The present experiment furthered this investigation by examining the effect of electrolytic fornix lesions on acquisition of trace and delay eyeblink conditioning in the rat. Importantly, the conditioned stimulus (CS) for trace conditioning was a standard tone-on cue, and the CS for delay conditioning was either a tone-off or tone-on CS. The results showed that fornix lesions impaired trace-, but not delay conditioning in rats trained with the tone-on CS or tone-off CS. The findings are consistent with previous studies that found trace-, but not delay eyeblink conditioning is a hippocampal dependent form of associative learning. Our results also indicate that the neural pathways for tone-off delay conditioning and tone-on trace conditioning are different, even though the structural composition of a tone-off CS and the trace conditioning interval are the same cue (i.e., the absence of sound). These findings indicate that the absence of a sensory cue (i.e., tone-off CS) and the presence of a sensory cue (i.e., tone-on CS) have equivalent associative value and effectiveness for engaging the neural pathways that support delay eyeblink conditioning.

Role of cerebellar cortex in associative learning and memory in guinea pigs

Time-related cognitive function refers to the capacity of the brain to store, extract, and process specific information. Previous studies demonstrated that the cerebellar cortex participates in advanced cognitive functions, but the role of the cerebellar cortex in cognitive functions is unclear. We established a behavioral model using classical eyeblink conditioning to study the role of the cerebellar cortex in associative learning and memory and the underlying mechanisms. We performed an investigation to determine whether eyeblink conditioning could be established by placing the stimulating electrode in the middle cerebellar peduncle. Behavior training was performed using a microcurrent pulse as a conditioned stimulus to stimulate the middle cerebellar peduncle and corneal blow as an unconditioned stimulus. After 10 consecutive days of training, a conditioned response was successfully achieved in the Delay, Trace-200-ms, and Trace-300-ms groups of guinea pigs, with acquisition rates of >60%, but the Trace-400-ms and control groups did not achieve a conditioned stimulus-related blink conditioned response. It could be a good model for studying the function of the cerebellum during the establishment of eyeblink conditioning.

Sex-Dependent Effects of Chronic Microdrive Implantation on Acquisition of Trace Eyeblink Conditioning

Neuroscience techniques, including in vivo recording, have allowed for a great expansion in knowledge; however, this technology may also affect the very phenomena researchers set out to investigate. Including both female and male mice in our associative learning experiments shed light on sex differences on the impact of chronic implantation of tetrodes on learning. While previous research showed intact female mice acquired trace eyeblink conditioning faster than male and ovariectomized females, implantation of chronic microdrive arrays showed sexually dimorphic effects on learning. Microdrive implanted male mice acquired the associative learning paradigm faster than both intact and ovariectomized females. These effects were not due to the weight of the drive alone, as there were no significant sex-differences in learning of animals that received "dummy drive" implants without tetrodes lowered into the brain. Tandem mass tag mass spectrometry and western blot analysis suggest that significant alterations in the MAPK pathway, acute inflammation, and brain derived neurotrophic factor may underlie these observed sex- and surgery-dependent effects on learning.

Synaptic protein interaction networks encode experience by assuming stimulus-specific and brain-region-specific states

A core network of widely expressed proteins within the glutamatergic post-synapse mediates activity-dependent synaptic plasticity throughout the brain, but the specific proteomic composition of synapses differs between brain regions. Here, we address the question, how does proteomic composition affect activity-dependent protein-protein interaction networks (PINs) downstream of synaptic activity? Using quantitative multiplex co-immunoprecipitation, we compare the PIN response of in vivo or ex vivo neurons derived from different brain regions to activation by different agonists or different forms of eyeblink conditioning. We report that PINs discriminate between incoming stimuli using differential kinetics of overlapping and non-overlapping PIN parameters. Further, these "molecular logic rules" differ by brain region. We conclude that although the PIN of the glutamatergic post-synapse is expressed widely throughout the brain, its activity-dependent dynamics show remarkable stimulus-specific and brain-region-specific diversity. This diversity may help explain the challenges in developing molecule-specific drug therapies for neurological disorders.

Role of the motor cortex in the generation of classically conditioned eyelid and vibrissae responses

The eyelid motor system has been used for years as an experimental model for studying the neuronal mechanisms underlying motor and cognitive learning, mainly with classical conditioning procedures. Nonetheless, it is not known yet which brain structures, or neuronal mechanisms, are responsible for the acquisition, storage, and expression of these motor responses. Here, we studied the temporal correlation between unitary activities of identified eyelid and vibrissae motor cortex neurons and the electromyographic activity of the orbicularis oculi and vibrissae muscles and magnetically recorded eyelid positions during classical conditioning of eyelid and vibrissae responses, using both delay and trace conditioning paradigms in behaving mice. We also studied the involvement of motor cortex neurons in reflexively evoked eyelid responses and the kinematics and oscillatory properties of eyelid movements evoked by motor cortex microstimulation. Results show the involvement of the motor cortex in the performance of conditioned responses elicited during the classical conditioning task. However, a timing correlation analysis showed that both electromyographic activities preceded the firing of motor cortex neurons, which must therefore be related more with the reinforcement and/or proper performance of the conditioned responses than with their acquisition and storage.

Intact Female Mice Acquire Trace Eyeblink Conditioning Faster than Male and Ovariectomized Female Mice

Female subjects have been widely excluded from past neuroscience work because of a number of biases, including the notion that cycling sex hormones increase variability. However, it is necessary to conduct behavioral research in mice that includes both sexes as mice are typically used for developing and evaluating future therapeutics. Understanding sex differences in learning is fundamental for the development of targeted therapies for numerous neurologic and neurodegenerative disorders, including Alzheimer’s disease, which is more prevalent in females than males. This study set out to confirm the role of sex and necessity of circulating ovarian hormones in the acquisition of the temporal associative memory task trace eyeblink conditioning (tEBC) in C57BL/6J mice. We present evidence that sex and ovarian hormones are important factors in learning. Specifically, intact female mice learn significantly faster than both male and ovariectomized (ovx) female mice. Data from pseudoconditioned control mice indicate that sex differences are because of differences in learned associations, not sensitization or spontaneous blink rate. This study strengthens the idea that ovarian hormones such as estrogen and progesterone significantly influence learning and memory and that further research is needed to determine the underlying mechanisms behind their effects. Overall, our findings emphasize the necessity of including both sexes in future behavioral studies.

Impact of anesthesia exposure in early development on learning and sensory functions

Each year, millions of children undergo anesthesia, and both human and animal studies have indicated that exposure to anesthesia at an early age can lead to neuronal damage and learning deficiency. However, disorders of sensory functions were not reported in children or animals exposed to anesthesia during infancy, which is surprising, given the significant amount of damage to brain tissue reported in many animal studies. In this review, we discuss the relationship between the systems in the brain that mediate sensory input, spatial learning, and classical conditioning, and how these systems could be affected during anesthesia exposure. Based on previous reports, we conclude that anesthesia can induce structural, functional, and compensatory changes in both sensory and learning systems. Changes in myelination following anesthesia exposure were observed as well as the neurodegeneration in the gray matter across variety of brain regions. Disproportionate cell death between excitatory and inhibitory cells induced by anesthesia exposure can lead to a long-term shift in the excitatory/inhibitory balance, which affects both learning-specific networks and sensory systems. Anesthesia may directly affect synaptic plasticity which is especially critical to learning acquisition. However, sensory systems appear to have better ability to compensate for damage than learning-specific networks.

A model for time interval learning in the Purkinje cell

Recent experimental findings indicate that Purkinje cells in the cerebellum represent time intervals by mechanisms other than conventional synaptic weights. These findings add to the theoretical and experimental observations suggesting the presence of intra-cellular mechanisms for adaptation and processing. To account for these experimental results we propose a new biophysical model for time interval learning in a Purkinje cell. The numerical model focuses on a classical delay conditioning task (e.g. eyeblink conditioning) and relies on a few computational steps. In particular, the model posits the activation by the parallel fiber input of a local intra-cellular calcium store which can be modulated by intra-cellular pathways. The reciprocal interaction of the calcium signal with several proteins forming negative and positive feedback loops ensures that the timing of inhibition in the Purkinje cell anticipates the interval between parallel and climbing fiber inputs during training. We systematically test the model ability to learn time intervals at the 150-1000 ms time scale, while observing that learning can also extend to the multiple seconds scale. In agreement with experimental observations we also show that the number of pairings required to learn increases with inter-stimulus interval. Finally, we discuss how this model would allow the cerebellum to detect and generate specific spatio-temporal patterns, a classical theory for cerebellar function.

Inactivation of the interpositus nucleus during unpaired extinction does not prevent extinction of conditioned eyeblink responses or conditioning-specific reflex modification

For almost 75 years, classical eyeblink conditioning has been an invaluable tool for assessing associative learning processes across many species, thanks to its high translatability and well-defined neural circuitry. Our laboratory has adapted the paradigm to extensively detail associative changes in the rabbit reflexive eyeblink response (unconditioned response, UR), characterized by postconditioning increases in the frequency, size, and latency of the UR when the periorbital shock unconditioned stimulus (US) is presented alone, termed conditioning-specific reflex modification (CRM). Because the shape and timing of CRM closely resembles the conditioned eyeblink response (CR) to the tone conditioned stimulus (CS), we previously tested whether CRs and CRM share a common neural substrate, the interpositus nucleus of the cerebellum (IP), and found that IP inactivation during conditioning blocked the development of both CRs and the timing aspect of CRM. The goal of the current study was to examine whether extinction of CRs and CRM timing, accomplished simultaneously with unpaired CS/US extinction, also involves the IP. Results showed that muscimol inactivation of the IP during extinction blocked CR expression but not extinction of CRs or CRM timing, contrasting with the literature showing IP inactivation prevents CR extinction during CS-alone presentations. The continued presence of the US throughout the unpaired extinction procedure may have been sufficient to overcome IP blockade, promoting plasticity in the cerebellar cortex and/or extracerebellar components of the eyeblink conditioning pathway that can modulate extinction of CRs and CRM timing. Results therefore add support to the distributed plasticity view of cerebellar learning. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

A method for combining multiple-units readout of optogenetic control with natural stimulation-evoked eyeblink conditioning in freely-moving mice

A growing pool of transgenic mice expressing Cre-recombinases, together with Cre-dependent opsin viruses, provide good tools to manipulate specific neural circuits related to eyeblink conditioning (EBC). However, currently available methods do not enable to get fast and precise readout of optogenetic control when the freely-moving mice are receiving EBC training. In the current study, we describe a laser diode (LD)-optical fiber (OF)-Tetrode assembly that allows for simultaneous multiple units recording and optical stimulation. Since the numbers of various cables that require to be connected are minimized, the LD-OF-Tetrode assembly can be combined with CS-US delivery apparatus for revealing the effects of optical stimulation on EBC in freely- moving mice. Moreover, this combination of techniques can be utilized to optogenetically intervene in hippocampal neuronal activities during the post-conditioning sleep in a closed-loop manner. This novel device thus enhances our ability to explore how specific neuronal assembly contributes to associative motor memory in vivo.

Air Puff System Fundamentals for Reproducible Eyeblink Conditioning Research

Air puff systems are at once trivially straightforward and dauntingly complex. On the one hand, they are little but a pressure source, valve, and tube connected together. On the other, the air passing through them is a compressible medium, expanding approximately adiabatically while travelling at high velocity through a compliant tube, and exiting as a turbulent jet with velocity peak and profile varying non-linearly in its near-field. This complexity puts precise mathematical prediction of puff properties out of reach of most labs. There are, however, a number of phenomena fundamental to air puff system design that are worth understanding to a first order of approximation, or at least qualitatively. Using a simplified, “electronic–hydraulic analogy” model, this paper discusses these phenomena in just enough depth for the reader to confidently specify parts for an air puff delivery system, to measure its key parameters, and/or to describe a given system unambiguously in publications, thus maximizing reproducibility.

Modulation of eyeblink conditioning through sensory processing of conditioned stimulus by cortical and subcortical regions

Classical eyeblink conditioning (EBC) is one of the simplest forms of associative learning that depends critically on the cerebellum. Using delay EBC (dEBC), a standard paradigm in which the unconditioned stimulus (US) is delayed and co-terminates with the conditioned stimulus (CS), converging lines of evidence has been accumulated and shows that the essential neural circuit mediating EBC resides in the cerebellum and brainstem. In addition to this essential circuit, multiple cerebral cortical and subcortical structures are required to modulate dEBC with suboptimal training parameters, and trace EBC (tEBC) in which a trace-interval separates the CS and US. However, it remains largely unclear why and how so many brain regions are involved for modulation of EBC. Previous research has suggested that the forebrain regions, such as medial prefrontal cortex (mPFC) and hippocampus, may be required to process weak CSs, or to realize temporal overlap between the CS and US signal inputs when the two stimuli were separated in time (i.e. during tEBC). Here, we proposed a multi-level network model for EBC modulation which focuses on sensory processing of CS. The model explains how different neural pathways projecting to pontine nucleus (PN) are involved to amplify or extend CS through heterosynaptic facilitation mechanism or "substitution effect" under different circumstances to achieve EBC. As such, our model can serve as a general framework to explain the modulating mechanism of EBC in a variety of conditions and to help understand the interaction among cerebellum, brainstem, cortical and subcortical regions in EBC modulation.

Spontaneous recovery of conditioned eyeblink responses is associated with transiently decreased cerebellar theta activity in guinea pigs

Behavioral studies have demonstrated that extinguished conditioned eyeblink responses (CR) can spontaneously recover after extinction. However, the neural mechanisms underlying this process are still unclear. We have shown that spontaneous cerebellar theta activity was predictive of subsequent CR extinction. Here, we sought to further evaluate the association between spontaneous recovery and cerebellar theta activity in behaving guinea pigs. It was found that trace conditioning training significantly diminished the degree of spontaneous recovery during extinction sessions as compared to delay training. Moreover, by recording local field potential in the cerebellum of guinea pigs undergoing an eyeblink conditioning extinction task, we found that spontaneous recovery of delay-paradigm CRs was associated with transiently decreased CS-evoked theta activity in the cerebellum. These findings suggest that decreased CS-evoked cerebellar theta activity may contribute to the neural process that is important for the spontaneous recovery of extinguished motor memory. Future studies are needed to clarify the neural mechanism underlying changed cerebellar theta activity during altered behavioral contingencies.

Cerebellar Processing Common to Delay and Trace Eyelid Conditioning

Results from previous lesion studies have been interpreted as evidence that the cerebellar cortex plays different roles for delay and trace conditioning of eyelid responses. However, the cerebellar cortex is organized by parasagittal stripes of Purkinje cells (PCs) that converge onto common deep nucleus neurons and receive common or related climbing fiber inputs. Based on this organization, we hypothesized that cerebellar tasks involving the same response system, such as delay and trace eyelid conditioning, would engage the same PCs and that the relationships between PC activity and expression of behavioral responses would be similar for both tasks. To test these hypotheses, we used tetrode recordings from eyelid PCs in rabbits during expression of delay- and trace-conditioned eyelid responses. Previous recording studies during delay conditioning described a strong relationship between eyelid PC activity and the kinematics of conditioned eyelid responses. The present results replicate these findings for delay conditioning and show that the same relationship exists during trace eyelid conditioning. During transitions from delay to trace responding, the relationship between eyelid PCs and behavioral responses was relatively stable. We found that an inverse firing rate model tuned to predict PC activity during one training paradigm could then predict equally well the PC activity during the other training paradigm. These results provide strong evidence that cerebellar cortex processing is similar for delay and trace eyelid conditioning and that the parasagittal organization of the cerebellum, not the conditioning paradigm, dictate which neurons are engaged to produce adaptively timed conditioned responses.SIGNIFICANCE STATEMENT A variety of evidence from eyelid conditioning and other cerebellar-dependent behaviors indicates that the cerebellar cortex is necessary for learning and proper timing of cerebellar learned responses. Debates exist about whether trace eyelid conditioning data show that fundamentally different mechanisms operate in the cerebellum during tasks when input from the forebrain is necessary for learning. We show here that learning-related changes in a specific population of Purkinje cells control the timing and amplitude of cerebellar responses the same way regardless of the inputs necessary to learn the task. Our results indicate the parasagittal organization of the cerebellar cortex, not the complexity of inputs to the cerebellum, determines which neurons are engaged in the learning and execution of cerebellar-mediated responses.

Compromised Survival of Cerebellar Molecular Layer Interneurons Lacking GDNF Receptors GFRα1 or RET Impairs Normal Cerebellar Motor Learning

The role of neurotrophic factors as endogenous survival proteins for brain neurons remains contentious. In the cerebellum, the signals controlling survival of molecular layer interneurons (MLIs) are unknown, and direct evidence for the requirement of a full complement of MLIs for normal cerebellar function and motor learning has been lacking. Here, we show that Purkinje cells (PCs), the target of MLIs, express the neurotrophic factor GDNF during MLI development and survival of MLIs depends on GDNF receptors GFRα1 and RET. Conditional mutant mice lacking either receptor lose a quarter of their MLIs, resulting in compromised synaptic inhibition of PCs, increased PC firing frequency, and abnormal acquisition of eyeblink conditioning and vestibulo-ocular reflex performance, but not overall motor activity or coordination. These results identify an endogenous survival mechanism for MLIs and reveal the unexpected vulnerability and selective requirement of MLIs in the control of cerebellar-dependent motor learning.

•

The signals controlling survival of molecular layer interneurons (MLIs) are unclear

•

Whether MLIs are involved in normal cerebellar function was unclear

•

Purkinje cells express GDNF, and survival of MLIs depends on GDNF receptors GFRα1 and RET

•

Requirement of MLIs for cerebellar-dependent motor learning

Cerebellar-dependent associative learning is impaired in very preterm born children and young adults

Preterm birth incorporates an increased risk for cerebellar developmental disorders likely contributing to motor and cognitive abnormalities. Experimental evidence of cerebellar dysfunction in preterm subjects, however, is sparse. In this study, classical eyeblink conditioning was used as a marker of cerebellar dysfunction. Standard delay conditioning was investigated in 20 adults and 32 preschool children born very preterm. Focal lesions were excluded based on structural magnetic resonance imaging. For comparison, an equal number of matched term born healthy peers were tested. Subgroups of children (12 preterm, 12 controls) were retested. Preterm subjects acquired significantly less conditioned responses (CR) compared to controls with slower learning rates. A likely explanation for these findings is that preterm birth impedes function of the cerebellum even in the absence of focal cerebellar lesions. The present findings are consistent with the assumption that prematurity results in long-term detrimental effects on the integrity of the cerebellum. It cannot be excluded, however, that extra-cerebellar pathology contributed to the present findings.

Grouping subjects based on conditioning criteria reveals differences in acquisition rates and in strength of conditioning-specific reflex modification

Averaging behavioral data such as the nictitating membrane response (NMR) across subjects can conceal important individual and group differences. Analyses were conducted of NMR data from rabbits that were grouped based on the point during NMR conditioning when subjects produced 8 conditioned responses (CR) in a set of 10 trials. This resulted in five groups (Early Day 1, Late Day 1, Early Day 2, Late Day 2, Early Day 3) in which group differences in CR acquisition rates were found. Percent (%) CRs were not found to increase monotonically and between-session differences in % CR were found. Conditioning-specific reflex modification (CRM) of the NMR is a type of enhanced reflexive responding of the NMR that is detected when the unconditioned stimulus (US) is presented in the absence of the conditioned stimulus (CS) following paired classical conditioning. CRM occurred in some subjects in all five groups. Subjects from both the group that was fastest and the group that was slowest to reach the learning criterion had unconditioned response (UR) topographies following NMR conditioning that strongly resembled the CR-UR response sequence elicited during NMR conditioning. This finding was most pronounced when the US duration used to assess CRM was equivalent to that used during NMR conditioning, further evidence to support the hypothesis that CRM is a CR that has generalized from the CS to the US. While grouping data based on conditioning criteria did not facilitate identifying individuals more predisposed to exhibiting CRM, strong CRM only occurred in the groups that reached the conditioning criterion the fastest.

The Motor Cortex Is Involved in the Generation of Classically Conditioned Eyelid Responses in Behaving Rabbits

Classical blink conditioning is a well known model for studying neural generation of acquired motor responses. The acquisition of this type of associative learning has been related to many cortical, subcortical, and cerebellar structures. However, until now, no one has studied the motor cortex (MC) and its possible role in classical eyeblink conditioning. We recorded in rabbits the activity of MC neurons during blink conditioning using a delay paradigm. Neurons were identified by their antidromic activation from facial nucleus (FN) or red nucleus (RN). For conditioning, we used a tone as a conditioned stimulus (CS) followed by an air puff as an unconditioned stimulus (US) that coterminated with it. Conditioned responses (CRs) were determined from the electromyographic activity of the orbicularis oculi muscle and/or from eyelid position recorded with the search coil technique. Type A neurons increased their discharge rates across conditioning sessions and reached peak firing during the CS-US interval, while type B cells presented a second peak during US presentation. Both of them project to the FN. Type C cells increased their firing across the CS-US interval, reaching peak values at the time of US presentation, and were activated from the RN. These three types of neurons fired well in advance of the beginning of CRs and changed with them. Reversible inactivation of the MC during conditioning evoked a decrease in learning curves and in the amplitude of CRs, while train stimulation of the MC simulated the profile and kinematics of conditioned blinks. In conclusion, MC neurons are involved in the acquisition and expression of CRs.

SIGNIFICANCE STATEMENT

Classical blink conditioning is a popular experimental model for studying neural mechanisms underlying the acquisition of motor skills. The acquisition of this type of associative learning has been related to many cortical, subcortical, and cerebellar structures. However, until now, no one has studied the motor cortex (MC) and its possible role in classical eyeblink conditioning. Here, we report that the firing activities of MC neurons, recorded in behaving rabbits, are related to and preceded the initiation of conditioned blinks. MC neurons were identified as projecting to the red or facial nuclei and encoded the kinematics of conditioned eyelid responses. The timed stimulation of recording sites simulated the profile of conditioned blinks. MC neurons play a role in the acquisition and expression of these acquired motor responses.

Active Inference and Learning in the Cerebellum

This letter offers a computational account of Pavlovian conditioning in the cerebellum based on active inference and predictive coding. Using eyeblink conditioning as a canonical paradigm, we formulate a minimal generative model that can account for spontaneous blinking, startle responses, and (delay or trace) conditioning. We then establish the face validity of the model using simulated responses to unconditioned and conditioned stimuli to reproduce the sorts of behavior that are observed empirically. The scheme's anatomical validity is then addressed by associating variables in the predictive coding scheme with nuclei and neuronal populations to match the (extrinsic and intrinsic) connectivity of the cerebellar (eyeblink conditioning) system. Finally, we try to establish predictive validity by reproducing selective failures of delay conditioning, trace conditioning, and extinction using (simulated and reversible) focal lesions. Although rather metaphorical, the ensuing scheme can account for a remarkable range of anatomical and neurophysiological aspects of cerebellar circuitry-and the specificity of lesion-deficit mappings that have been established experimentally. From a computational perspective, this work shows how conditioning or learning can be formulated in terms of minimizing variational free energy (or maximizing Bayesian model evidence) using exactly the same principles that underlie predictive coding in perception.

Conserved regulators of cognitive aging: From worms to humans

Cognitive decline is a major deficit that arises with age in humans. While some research on the underlying causes of these problems can be done in humans, harnessing the strengths of small model systems, particularly those with well-studied longevity mutants, such as the nematode C. elegans, will accelerate progress. Here we review the approaches being used to study cognitive decline in model organisms and show how simple model systems allow the rapid discovery of conserved molecular mechanisms, which will eventually enable the development of therapeutics to slow cognitive aging.

Climbing fibers encode a temporal-difference prediction error during cerebellar learning in mice

Climbing fiber inputs to Purkinje cells are thought to play a teaching role by generating the instructive signals that drive cerebellar learning. To investigate how these instructive signals are encoded, we recorded the activity of individual climbing fibers during cerebellar-dependent eyeblink conditioning in mice. Our findings show that climbing fibers signal both the unexpected delivery and the unexpected omission of the periocular airpuff that serves as the instructive signal for eyeblink conditioning. In addition, we report the surprising discovery that climbing fibers activated by periocular airpuffs also respond to stimuli from other sensory modalities, if those stimuli are novel or if they predict that the periocular airpuff is about to be presented. This pattern of climbing fiber activity is strikingly similar to the responses of dopamine neurons during reinforcement learning, which have been shown to encode a particular type of instructive signal known as a temporal difference prediction error.