Role of the hippocampal system in conditional motor learning: mapping antecedents to action

Learning leaves a memory trace in motor cortex

How are we able to learn new behaviors without disrupting previously learned ones? To understand how the brain achieves this, we used a brain-computer interface (BCI) learning paradigm, which enables us to detect the presence of a memory of one behavior while performing another. We found that learning to use a new BCI map altered the neural activity that monkeys produced when they returned to using a familiar BCI map in a way that was specific to the learning experience. That is, learning left a "memory trace" in the primary motor cortex. This memory trace coexisted with proficient performance under the familiar map, primarily by altering neural activity in dimensions that did not impact behavior. Forming memory traces might be how the brain is able to provide for the joint learning of multiple behaviors without interference.

Motor Sequence Learning Deficits in Idiopathic Parkinson's Disease Are Associated With Increased Substantia Nigra Activity

Previous studies have shown that persons with Parkinson’s disease (pwPD) share specific deficits in learning new sequential movements, but the neural substrates of this impairment remain unclear. In addition, the degree to which striatal dopaminergic denervation in PD affects the cortico-striato-thalamo-cerebellar motor learning network remains unknown. We aimed to answer these questions using fMRI in 16 pwPD and 16 healthy age-matched control subjects while they performed an implicit motor sequence learning task. While learning was absent in both pwPD and controls assessed with reaction time differences between sequential and random trials, larger error-rates during the latter suggest that at least some of the complex sequence was encoded. Moreover, we found that while healthy controls could improve general task performance indexed by decreased reaction times across both sequence and random blocks, pwPD could not, suggesting disease-specific deficits in learning of stimulus-response associations. Using fMRI, we found that this effect in pwPD was correlated with decreased activity in the hippocampus over time. Importantly, activity in the substantia nigra (SN) and adjacent bilateral midbrain was specifically increased during sequence learning in pwPD compared to healthy controls, and significantly correlated with sequence-specific learning deficits. As increased SN activity was also associated (on trend) with higher doses of dopaminergic medication as well as disease duration, the results suggest that learning deficits in PD are associated with disease progression, indexing an increased drive to recruit dopaminergic neurons in the SN, however, unsuccessfully. Finally, there were no differences between pwPD and controls in task modulation of the cortico-striato-thalamo-cerebellar network. However, a restricted nigral-striatal model showed that negative modulation of SN to putamen connection was larger in pwPD compared to controls during random trials, while no differences between the groups were found during sequence learning. We speculate that learning-specific SN recruitment leads to a relative increase in SN- > putamen connectivity, which returns to a pathological reduced state when no learning takes place.

Trajectory-modulated hippocampal neurons persist throughout memory-guided navigation

Trajectory-dependent splitter neurons in the hippocampus encode information about a rodent's prior trajectory during performance of a continuous alternation task. As such, they provide valuable information for supporting memory-guided behavior. Here, we employed single-photon calcium imaging in freely moving mice to investigate the emergence and fate of trajectory-dependent activity through learning and mastery of a continuous spatial alternation task. In agreement with others, the quality of trajectory-dependent information in hippocampal neurons correlated with task performance. We thus hypothesized that, due to their utility, splitter neurons would exhibit heightened stability. We find that splitter neurons were more likely to remain active and retained more consistent spatial information across multiple days than other neurons. Furthermore, we find that both splitter neurons and place cells emerged rapidly and maintained stable trajectory-dependent/spatial activity thereafter. Our results suggest that neurons with useful functional coding exhibit heightened stability to support memory guided behavior.

The effect of CA1 dopaminergic system on amnesia induced by harmane in mice

In the present study, the effects of bilateral injections of dopaminergic drugs into the hippocampal CA1 regions (intra-CA1) on harmane-induced amnesia were examined in mice. We used a single-trial step-down inhibitory avoidance task for the assessment of memory acquisition in adult male mice. Our data indicated that pre-training intra-peritoneal (i.p.) administration of harmane (12 mg/kg) impaired memory acquisition. Moreover, intra-CA1 administration of dopamine D1 receptor agonist, SKF38393 (0.25 µg/mouse), dopamine D1 receptor antagonist, SCH23390 (0.25 µg/mouse), dopamine D2 receptor agonist, quinpirole (0.125 and 0.25 µg/mouse) and dopamine D2 receptor antagonist, sulpiride (0.2 and 0.4 µg/mouse) decreased the learning of a single-trial inhibitory avoidance task. Furthermore, pre-training intra-CA1 injection of sub-threshold doses of SKF38393 (0.0625 µg/mouse) or sulpiride (0.1 µg/mouse) increased pre-training harmane (4 and 8 mg/kg, i.p.)-induced amnesia. On the other hand, pre-training intra-CA1 injection of a sub-threshold dose of SCH23390 (0.0625 µg/mouse) reversed amnesia induced by an effective dose of harmane (12 mg/kg; i.p.). In addition, Pre-training intra-CA1 injection of quinpirole (0.0625 µg/mouse) had no effect on memory impairment induced by harmane. These findings indicate the involvement of CA1 dopaminergic system on harmane-induced impairment of memory acquisition.

Corticocortical Systems Underlying High-Order Motor Control

Cortical networks are characterized by the origin, destination, and reciprocity of their connections, as well as by the diameter, conduction velocity, and synaptic efficacy of their axons. The network formed by parietal and frontal areas lies at the core of cognitive-motor control because the outflow of parietofrontal signaling is conveyed to the subcortical centers and spinal cord through different parallel pathways, whose orchestration determines, not only when and how movements will be generated, but also the nature of forthcoming actions. Despite intensive studies over the last 50 years, the role of corticocortical connections in motor control and the principles whereby selected cortical networks are recruited by different task demands remain elusive. Furthermore, the synaptic integration of different cortical signals, their modulation by transthalamic loops, and the effects of conduction delays remain challenging questions that must be tackled to understand the dynamical aspects of parietofrontal operations. In this article, we evaluate results from nonhuman primate and selected rodent experiments to offer a viewpoint on how corticocortical systems contribute to learning and producing skilled actions. Addressing this subject is not only of scientific interest but also essential for interpreting the devastating consequences for motor control of lesions at different nodes of this integrated circuit. In humans, the study of corticocortical motor networks is currently based on MRI-related methods, such as resting-state connectivity and diffusion tract-tracing, which both need to be contrasted with histological studies in nonhuman primates.

A separable two-dimensional random field model of binary response data from multi-day behavioral experiments

BACKGROUND

The study of learning in populations of subjects can provide insights into the changes that occur in the brain with aging, drug intervention, and psychiatric disease.

NEW METHOD

We introduce a separable two-dimensional (2D) random field (RF) model for analyzing binary response data acquired during the learning of object-reward associations across multiple days. The method can quantify the variability of performance within a day and across days, and can capture abrupt changes in learning.

RESULTS

We apply the method to data from young and aged macaque monkeys performing a reversal-learning task. The method provides an estimate of performance within a day for each age group, and a learning rate across days for each monkey. We find that, as a group, the older monkeys require more trials to learn the object discriminations than do the young monkeys, and that the cognitive flexibility of the younger group is higher. We also use the model estimates of performance as features for clustering the monkeys into two groups. The clustering results in two groups that, for the most part, coincide with those formed by the age groups. Simulation studies suggest that clustering captures inter-individual differences in performance levels.

COMPARISON WITH EXISTING METHOD(S)

In comparison with generalized linear models, this method is better able to capture the inherent two-dimensional nature of the data and find between group differences.

CONCLUSIONS

Applied to binary response data from groups of individuals performing multi-day behavioral experiments, the model discriminates between-group differences and identifies subgroups.

The Role of Associative Cortices and Hippocampus during Movement Perturbations

Although motor control has been extensively studied, most research involving neural recordings has focused on primary motor cortex, pre-motor cortex, supplementary motor area, and cerebellum. These regions are involved during normal movements, however, associative cortices and hippocampus are also likely involved during perturbed movements as one must detect the unexpected disturbance, inhibit the previous motor plan, and create a new plan to compensate. Minimal data is available on these brain regions during such “robust” movements. Here, epileptic patients implanted with intracerebral electrodes performed reaching movements while experiencing occasional unexpected force perturbations allowing study of the fronto-parietal, limbic and hippocampal network at unprecedented high spatial, and temporal scales. Areas including orbitofrontal cortex (OFC) and hippocampus showed increased activation during perturbed trials. These results, coupled with a visual novelty control task, suggest the hippocampal MTL-P300 novelty response is modality independent, and that the OFC is involved in modifying motor plans during robust movement.

Functional coupling between frontoparietal and occipitotemporal pathways during action and perception

Several lines of evidence point to areas in the occipitotemporal pathway as being critical in the processes of visual perception and object recognition. Much less appreciated, however, is the role that this pathway plays in object-related processing for the purposes of visually guided action. Here, using functional MRI (fMRI) and functional connectivity (FC) measures, we examined interactions between areas in frontoparietal cortex (FPC) involved in grasping, reaching, eye movements, and tool use and areas in occipitotemporal cortex (OTC) involved in object-, face-, scene-, body-, tool-, and motion-related processing, both during the performance of sensorimotor and visual-perceptual tasks, as well as during passive fixation (resting-state). Cluster analysis of regional time course data identified correspondence in the patterns of FPC and OTC connectivity during the visual-perceptual tasks and rest that both tended to segregate regions along traditional dorsal/ventral pathway boundaries. During the sensorimotor tasks, however, we observed a notable separation in functional coupling between ventral-medial and ventral-lateral regions of OTC, with several of the latter areas often being clustered together with sensorimotor-defined areas in parietal cortex. These findings indicate that the functional coupling of ventral-lateral OTC areas to dorsal parietal and ventral-medial structures is flexible and task-dependent, and suggests that regions in lateral occipital cortex, in particular, may play an important role in mediating interactions between the dorsal and ventral pathways during tasks involving sensorimotor control.

Insights into Seeing and Grasping: Distinguishing the Neural Correlates of Perception and Action

Vision contributes to both perception and visuomotor control, and it has been suggested that many higher brain structures specialize in one or the other function. An alternative view, presented here, is that most higher brain areas participate in both visuomotor and perceptual functions. In the anterior frontal cortex, for example, the activity of one population of neurons reflects perceptual reports about a visual stimulus, whereas the activity of an intermingled population reflects movements aimed at the same stimulus. Similarly, posterior parietal and inferior temporal areas appear to function in both visual perception and visuomotor control. Visuomotor signals in higher order cortical areas could contribute to the perception of one’s own action. They also might reflect the existence of two systems for visual information processing: one stressing accuracy for the control of movement and the other generating hypotheses about the world.

Cognitive impairment in a young marmoset reveals lateral ventriculomegaly and a mild hippocampal atrophy: a case report

The number of studies that use the common marmoset (Callithrix jacchus) in various fields of neurosciences is increasing dramatically. In general, animals enter the study when their health status is considered satisfactory on the basis of classical clinical investigations. In behavioral studies, variations of score between individuals are frequently observed, some of them being considered as poor performers or outliers. Experimenters rarely consider the fact that it could be related to some brain anomaly. This raises the important issue of the reliability of such classical behavioral approaches without using complementary imaging, especially in animals lacking striking external clinical signs. Here we report the case of a young marmoset which presented a set of cognitive impairments in two different tasks compared to other age-matched animals. Brain imaging revealed a patent right lateral ventricular enlargement with a mild hippocampal atrophy. This abnormality could explain the cognitive impairments of this animal. Such a case points to the importance of complementing behavioral studies by imaging explorations to avoid experimental bias.

Stimulus-response bindings in priming

People can rapidly form arbitrary associations between stimuli and the responses they make in the presence of those stimuli. Such stimulus-response (S-R) bindings, when retrieved, affect the way that people respond to the same, or related, stimuli. Only recently, however, has the flexibility and ubiquity of these S-R bindings been appreciated, particularly in the context of priming paradigms. This is important for the many cognitive theories that appeal to evidence from priming. It is also important for the control of action generally. An S-R binding is more than a gradually learned association between a specific stimulus and a specific response; instead, it captures the full, context-dependent behavioral potential of a stimulus.

Contributions of the hippocampus and entorhinal cortex to rapid visuomotor learning in rhesus monkeys

The hippocampus and adjacent structures in the medial temporal lobe are essential for establishing new associative memories. Despite this knowledge, it is not known whether the hippocampus proper is essential for establishing such memories, nor is it known whether adjacent regions like the entorhinal cortex might contribute. To test the contributions of these regions to the formation of new associative memories, we trained rhesus monkeys to rapidly acquire arbitrary visuomotor associations, i.e., associations between visual stimuli and spatially directed actions. We then assessed the effects of reversible inactivations of either the hippocampus (Experiment 1) or entorhinal cortex (Experiment 2) on the within-session rate of learning. For comparison, we also evaluated the effects of the inactivations on performance of problems of the same type that had been well learned prior to any inactivations. We found that inactivation of the entorhinal cortex but not hippocampus produced impairments in acquiring novel arbitrary associations. The impairment did not extend to the familiar, previously established associations. These data indicate that the entorhinal cortex is causally involved in establishing new associations, as opposed to retrieving previously learned associations. Published 2014. This article is a U.S. Government work and is in the public domain in the USA.

Assembling old tricks for new tasks: a neural model of instructional learning and control

We can learn from the wisdom of others to maximize success. However, it is unclear how humans take advice to flexibly adapt behavior. On the basis of data from neuroanatomy, neurophysiology, and neuroimaging, a biologically plausible model is developed to illustrate the neural mechanisms of learning from instructions. The model consists of two complementary learning pathways. The slow-learning parietal pathway carries out simple or habitual stimulus-response (S-R) mappings, whereas the fast-learning hippocampal pathway implements novel S-R rules. Specifically, the hippocampus can rapidly encode arbitrary S-R associations, and stimulus-cued responses are later recalled into the basal ganglia-gated pFC to bias response selection in the premotor and motor cortices. The interactions between the two model learning pathways explain how instructions can override habits and how automaticity can be achieved through motor consolidation.

Habit and nonhabit systems for unconscious and conscious behavior: implications for multitasking

The study of human consciousness has demonstrated that there are both conscious and unconscious systems. Other work, particularly in animals, has shown that there are habit and nonhabit systems and that these involve different brain regions and memory processes. Here we argue that habits can be equated with unconscious behavior and nonhabits with conscious behavior. This equation makes the extensive physiological literature on habit/nonhabit relevant to the less tractable issue of consciousness. On the basis of this line of reasoning, it appears that different parts of the BG and different memory structures mediate conscious and unconscious processes. It is further argued here that the unconscious system is highly capable; it can both process sensory information and produce behavior. The benefit of such a dual system is multitasking: The unconscious system can execute background tasks, leaving the conscious system to perform more difficult tasks.

Visuomotor coordination and motor representation by human temporal lobe neurons

The division of cortical visual processing into distinct dorsal and ventral streams is a key concept in primate neuroscience [Goodale, M. A., & Milner, A. D. Separate visual pathways for perception and action. Trends in Neurosciences, 15, 20-25, 1992; Steele, G., Weller, R., & Cusick, C. Cortical connections of the caudal subdivision of the dorsolateral area (V4) in monkeys. Journal of Comparative Neurology, 306, 495-520, 1991]. The ventral stream is usually characterized as a "What" pathway, whereas the dorsal stream is implied in mediating spatial perception ("Where") and visually guided actions ("How"). A subpathway emerging from the dorsal stream and projecting to the medial-temporal lobe has been identified [Kravitz, D. J., Saleem, K. S., Baker, C. I., & Mishkin, M. A new neural framework for visuospatial processing. Nature Reviews Neuroscience, 12, 217-230, 2011; Cavada, C., & Goldman-Raiuc, P. S. Posterior parietal cortex in rhesus monkey: I. Parcellation of areas based on distinctive limbic and sensory cortico-cortical connections. Journal of Comparative Neurology, 287, 393-421, 1989]. The current article studies the coordination of visual information typically associated with the dorsal stream ("Where"), with planned movements, focusing on the temporal lobe. We recorded extracellular activity from 565 cells in the human medial-temporal and frontal lobes while 13 patients performed cued hand movements with visual feedback (visuomotor task), without feedback (motor task), or observed visual feedback without motor movement (visual-only task). We discovered two different neural populations in the human medial-temporal lobe. One consists of motor-like neurons representing hand position, speed or acceleration during the motor task but not during the visuomotor or visual tasks. The other is specific to the parahippocampal gyrus (an area known to process visual motion [Gur, M., & Snodderly, D. M. Direction selectivity in V1 of alert monkeys: Evidence for parallel pathways for motion processing. Journal of Physiology, 585, 383-400, 2007; Sato, N., & Nakamura, K. Visual response properties of neurons in the parahippocampal cortex of monkeys. Journal of Neurophysiology, 90, 876-886, 2003]) and encodes speed, acceleration, or direction of hand movements, but only during the visuomotor task: neither during visual-only nor during motor tasks. These findings suggest a functional basis for the anatomical subpathway between the dorsal stream and the medial-temporal lobe. Similar to the recent expansion of the motor control process into the sensory cortex [Matyas, F., Sreenivasan, V., Marbach, F., Wacongne, C., Barsy, B., Mateo, C., et al. Motor control by sensory cortex. Science, 330, 1240-1243, 2010], our findings render the human medial-temporal lobe an important junction in the process of planning and execution of motor acts whether internally or externally (visually) driven. Thus, the medial-temporal lobe might serve as an integration node between the two processing streams. Our findings thus shed new light on the brain mechanisms underlying visuomotor coordination which is a crucial capacity for everyday survival, whether it is identifying and picking up food, sliding a key into a lock, driving a vehicle, or escaping a predator.

Donepezil impairs memory in healthy older subjects: behavioural, EEG and simultaneous EEG/fMRI biomarkers

Rising life expectancies coupled with an increasing awareness of age-related cognitive decline have led to the unwarranted use of psychopharmaceuticals, including acetylcholinesterase inhibitors (AChEIs), by significant numbers of healthy older individuals. This trend has developed despite very limited data regarding the effectiveness of such drugs on non-clinical groups and recent work indicates that AChEIs can have negative cognitive effects in healthy populations. For the first time, we use a combination of EEG and simultaneous EEG/fMRI to examine the effects of a commonly prescribed AChEI (donepezil) on cognition in healthy older participants. The short- and long-term impact of donepezil was assessed using two double-blind, placebo-controlled trials. In both cases, we utilised cognitive (paired associates learning (CPAL)) and electrophysiological measures (resting EEG power) that have demonstrated high-sensitivity to age-related cognitive decline. Experiment 1 tested the effects of 5 mg/per day dosage on cognitive and EEG markers at 6-hour, 2-week and 4-week follow-ups. In experiment 2, the same markers were further scrutinised using simultaneous EEG/fMRI after a single 5 mg dose. Experiment 1 found significant negative effects of donepezil on CPAL and resting Alpha and Beta band power. Experiment 2 replicated these results and found additional drug-related increases in the Delta band. EEG/fMRI analyses revealed that these oscillatory differences were associated with activity differences in the left hippocampus (Delta), right frontal-parietal network (Alpha), and default-mode network (Beta). We demonstrate the utility of simple cognitive and EEG measures in evaluating drug responses after acute and chronic donepezil administration. The presentation of previously established markers of age-related cognitive decline indicates that AChEIs can impair cognitive function in healthy older individuals. To our knowledge this is the first study to identify the precise neuroanatomical origins of EEG drug markers using simultaneous EEG/fMRI. The results of this study may be useful for evaluating novel drugs for cognitive enhancement.

Comparison of strategy signals in the dorsolateral and orbital prefrontal cortex

Abstract behavior-guiding rules and strategies allow monkeys to avoid errors in rarely encountered situations. In the present study, we contrasted strategy-related neuronal activity in the dorsolateral prefrontal cortex (PFdl) and the orbital prefrontal cortex (PFo) of rhesus monkeys. On each trial of their behavioral task, the monkeys responded to a foveal visual cue by making a saccade to one of two spatial targets. One response required a leftward saccade, the other required a saccade of equal magnitude to the right. The cues instructed the monkeys to follow one of two response strategies: to stay with their most recent successful response or to shift to the alternative response. Neurons in both areas encoded the stay and shift strategies after the cue appeared, but there were three major differences between the PFo and the PFdl: (1) many strategy-encoding cells in PFdl also encoded the response (left or right), but few, if any, PFo cells did so; (2) strategy selectivity appeared earlier in PFo than in PFdl; and (3) on error trials, PFo neurons encoded the correct strategy-the one that had been cued but not implemented-whereas in PFdl the strategy signals were weak or absent on error trials. These findings indicate that PFo and PFdl both contribute to behaviors guided by abstract response strategies, but do so differently, with PFo encoding a strategy and PFdl encoding a response based on a strategy.

Striatal and medial temporal lobe functional interactions during visuomotor associative learning

A network of regions including the medial temporal lobe (MTL) and the striatum are integral to visuomotor associative learning. Here, we evaluated the contributions of the structures of the striatum and the MTL, as well as their interactions during an arbitrary associative learning task. We hypothesized that activity in the striatum would correlate with the rate of learning, while activity in the MTL would track how well associations were learned. Further, we expected functional correlations to show both facilitative as well as competitive relationships depending on the regions involved. Results showed that activity throughout the striatum was modulated by the rate of learning, while the sensorimotor and ventral striatum were also modulated by probability correct. Across the MTL, activity correlated with the probability of being correct, while the perirhinal cortex and right parahippocampal cortex were modulated by the rate of learning. The activity in the ventral striatum robustly coupled with activity in the MTL during learning, while interactions between the associative striatum and the MTL showed the opposite pattern. These findings suggest dissociable computational roles for different subregions of the striatum and MTL. These subregions interact in distinct ways, perhaps forming functionally integrated networks during the learning of arbitrary associations.

Oxytocin enhances amygdala-dependent, socially reinforced learning and emotional empathy in humans

Oxytocin (OT) is becoming increasingly established as a prosocial neuropeptide in humans with therapeutic potential in treatment of social, cognitive, and mood disorders. However, the potential of OT as a general facilitator of human learning and empathy is unclear. The current double-blind experiments on healthy adult male volunteers investigated first whether treatment with intranasal OT enhanced learning performance on a feedback-guided item-category association task where either social (smiling and angry faces) or nonsocial (green and red lights) reinforcers were used, and second whether it increased either cognitive or emotional empathy measured by the Multifaceted Empathy Test. Further experiments investigated whether OT-sensitive behavioral components required a normal functional amygdala. Results in control groups showed that learning performance was improved when social rather than nonsocial reinforcement was used. Intranasal OT potentiated this social reinforcement advantage and greatly increased emotional, but not cognitive, empathy in response to both positive and negative valence stimuli. Interestingly, after OT treatment, emotional empathy responses in men were raised to levels similar to those found in untreated women. Two patients with selective bilateral damage to the amygdala (monozygotic twins with congenital Urbach-Wiethe disease) were impaired on both OT-sensitive aspects of these learning and empathy tasks, but performed normally on nonsocially reinforced learning and cognitive empathy. Overall these findings provide the first demonstration that OT can facilitate amygdala-dependent, socially reinforced learning and emotional empathy in men.

What, if anything, can monkeys tell us about human amnesia when they can't say anything at all?

Despite a half century of development, the orthodox monkey model of human amnesia needs improvement, in part because of two problems inherent in animal models of advanced human cognition. First, animal models are perforce comparative, but the principles of comparative and evolutionary biology have not featured prominently in developing the orthodox model. Second, no one understands the relationship between human consciousness and cognition in other animals, but the orthodox model implicitly assumes a close correspondence. If we treat these two difficulties with the deference they deserve, monkeys can tell us a lot about human amnesia and memory. Three future contributions seem most likely: (1) an improved monkey model, one refocused on the hippocampus rather than on the medial temporal lobe as a whole; (2) a better understanding of cortical areas unique to primates, especially the granular prefrontal cortex; and (3), taking the two together, insight into prefrontal-hippocampal interactions. We propose that interactions among the granular prefrontal areas create the kind of cross-domain, analogical and self-referential knowledge that underlies advanced cognition in modern humans. When these products of frontal-lobe function interact with the hippocampus, and its ancestral function in navigation, what emerges is the human ability to embed ourselves in scenarios-real and imagined, self-generated and received-thereby creating a coherent, conscious life experience.

Characterizing learning by simultaneous analysis of continuous and binary measures of performance

Continuous observations, such as reaction and run times, and binary observations, such as correct/incorrect responses, are recorded routinely in behavioral learning experiments. Although both types of performance measures are often recorded simultaneously, the two have not been used in combination to evaluate learning. We present a state-space model of learning in which the observation process has simultaneously recorded continuous and binary measures of performance. We use these performance measures simultaneously to estimate the model parameters and the unobserved cognitive state process by maximum likelihood using an approximate expectation maximization (EM) algorithm. We introduce the concept of a reaction-time curve and reformulate our previous definitions of the learning curve, the ideal observer curve, the learning trial and between-trial comparisons of performance in terms of the new model. We illustrate the properties of the new model in an analysis of a simulated learning experiment. In the simulated data analysis, simultaneous use of the two measures of performance provided more credible and accurate estimates of the learning than either measure analyzed separately. We also analyze two actual learning experiments in which the performance of rats and of monkeys was tracked across trials by simultaneously recorded reaction and run times and the correct and incorrect responses. In the analysis of the actual experiments, our algorithm gave a straightforward, efficient way to characterize learning by combining continuous and binary measures of performance. This analysis paradigm has implications for characterizing learning and for the more general problem of combining different data types to characterize the properties of a neural system.

The role of the CA3 hippocampal subregion in spatial memory: a process oriented behavioral assessment

Computational models, behavioral data, and electrophysiological data suggest that the CA3 subregion of the hippocampus may support multiple mnemonic processes critical to the formation and subsequent retrieval of spatial memories. Multiple researchers have proposed that the CA3 subregion contains an autoassociative network in which synaptic connections between CA3 neurons that represent different components of a memory are strengthened via recurrent collateral connections. As a result, it has been suggested that the CA3 autoassociative network may support multiple processes including the formation of spatial arbitrary associations, temporary maintenance of spatial working memory, and spatial pattern completion. In addition, the CA3 subregion has been suggested to be involved in spatial pattern separation. The separation of patterns is hypothesized to be accomplished based on the low probability that any two CA3 neurons will receive mossy-fiber input synapses from a similar subset of dentate gyrus cells. The separation of patterns also may be enhanced by competitive inhibition within CA3 and dentate gyrus. This review will focus on the mnemonic processes supported by CA3 neurons and how these processes may facilitate the encoding and retrieval of spatial information. Although there is growing evidence indicating that the hippocampus plays a role in the processing of nonspatial information as well, the scope of the present review will focus on the role of the CA3 subregion in spatial memory.

On a basal ganglia role in learning and rehearsing visual-motor associations

Fronto-striatal circuitry interacts with the midbrain dopaminergic system to mediate the learning of stimulus-response associations, and these associations often guide everyday actions, but the precise role of these circuits in forming and consolidating rules remains uncertain. A means to examine basal ganglia circuit contributions to associative motor learning is to examine these process in a lesion model system, such as Parkinson's disease (PD), a basal ganglia disorder characterized by the loss of dopamine neurons. We used functional magnetic resonance imaging (MRI) to compare brain activation of PD patients with a group of healthy aged-match participants during a visual-motor associative learning task that entailed discovering and learning arbitrary associations between a set of six visual stimuli and corresponding spatial locations by moving a joystick-controlled cursor. We tested the hypothesis that PD would recruit more areas than age-matched controls during learning and also show increased activation in commonly activated regions, probably in the parietal and premotor cortices, and the cerebellum, perhaps as compensatory mechanisms for their disrupted fronto-striatal networks. PD had no effect in acquiring the associative relationships and learning-related activation in several key frontal cortical and subcortical structures. However, we found that PD modified activation in other areas, including those in the cerebellum and frontal, and parietal cortex, particularly during initial learning. These results may suggest that the basal ganglia circuits become active more so during the initial formation of rule-based behavior.

Comparison of associative learning-related signals in the macaque perirhinal cortex and hippocampus

Strong evidence suggests that the macaque monkey perirhinal cortex is involved in both the initial formation as well as the long-term storage of associative memory. To examine the neurophysiological basis of associative memory formation in this area, we recorded neural activity in this region as monkeys learned new conditional-motor associations. We report that a population of perirhinal neurons signal newly learned associations by changing their firing rate correlated with the animal's behavioral learning curve. Individual perirhinal neurons signal learning of one or more associations concurrently and these neural changes could occur before, at the same time, or after behavioral learning was expressed. We also compared the associative learning signals in the perirhinal cortex to our previous findings in the hippocampus. We report global similarities in both the learning-related and task-related activity seen across these areas as well as clear differences in the within and across trial timing and relative proportion of different subtypes of learning-related signals. Taken together, these findings emphasize the important role of the perirhinal cortex in new associative learning and suggest that the perirhinal cortex together with the hippocampus contribute importantly to conditional-motor associative memory formation.

Consciousness and the consolidation of motor learning

It is no secret that motor learning benefits from repetition. For example, pianists devote countless hours to performing complicated sequences of key presses, and golfers practice their swings thousands of times to reach a level of proficiency. Interestingly, the subsequent waking and sleeping hours after practice also play important roles in motor learning. During this time, a motor skill can consolidate into a more stable form that can lead to improved future performance without intervening practice. Though it is widely believed that sleep is crucial for this consolidation of motor learning, this is not generally true. In many instances only day-time consolidates motor learning, while in other instances neither day-time nor sleep consolidates learning. Recent studies have suggested that conscious awareness during motor training can determine whether sleep or day-time plays a role in consolidation. However, ongoing studies suggest that this explanation is also incomplete. In addition to conscious awareness, attention is an important factor to consider. This review discusses how attention and conscious awareness interact with day and night processes to consolidate a motor memory.

A mixed filter algorithm for cognitive state estimation from simultaneously recorded continuous and binary measures of performance

Continuous (reaction times) and binary (correct/ incorrect responses) measures of performance are routinely recorded to track the dynamics of a subject's cognitive state during a learning experiment. Current analyses of experimental data from learning studies do not consider the two performance measures together and do not use the concept of the cognitive state formally to design statistical methods. We develop a mixed filter algorithm to estimate the cognitive state modeled as a linear stochastic dynamical system from simultaneously recorded continuous and binary measures of performance. The mixed filter algorithm has the Kalman filter and the more recently developed recursive filtering algorithm for binary processes as special cases. In the analysis of a simulated learning experiment the mixed filter algorithm provided a more accurate and precise estimate of the cognitive state process than either the Kalman or binary filter alone. In the analysis of an actual learning experiment in which a monkey's performance was tracked by its series of reaction times, and correct and incorrect responses, the mixed filter gave a more complete description of the learning process than either the Kalman or binary filter. These results establish the feasibility of estimating cognitive state from simultaneously recorded continuous and binary performance measures and suggest a way to make practical use of concepts from learning theory in the design of statistical methods for the analysis of data from learning experiments.

Analysis of between-trial and within-trial neural spiking dynamics

Recording single-neuron activity from a specific brain region across multiple trials in response to the same stimulus or execution of the same behavioral task is a common neurophysiology protocol. The raster plots of the spike trains often show strong between-trial and within-trial dynamics, yet the standard analysis of these data with the peristimulus time histogram (PSTH) and ANOVA do not consider between-trial dynamics. By itself, the PSTH does not provide a framework for statistical inference. We present a state-space generalized linear model (SS-GLM) to formulate a point process representation of between-trial and within-trial neural spiking dynamics. Our model has the PSTH as a special case. We provide a framework for model estimation, model selection, goodness-of-fit analysis, and inference. In an analysis of hippocampal neural activity recorded from a monkey performing a location-scene association task, we demonstrate how the SS-GLM may be used to answer frequently posed neurophysiological questions including, What is the nature of the between-trial and within-trial task-specific modulation of the neural spiking activity? How can we characterize learning-related neural dynamics? What are the timescales and characteristics of the neuron's biophysical properties? Our results demonstrate that the SS-GLM is a more informative tool than the PSTH and ANOVA for analysis of multiple trial neural responses and that it provides a quantitative characterization of the between-trial and within-trial neural dynamics readily visible in raster plots, as well as the less apparent fast (1-10 ms), intermediate (11-20 ms), and longer (>20 ms) timescale features of the neuron's biophysical properties.

Transient neuronal correlations underlying goal selection and maintenance in prefrontal cortex

We reported previously that as monkeys used abstract response strategies to choose spatial goals, 1 population of prefrontal cortex neurons encoded future goals (F cells), whereas a largely separate population encoded previous goals (P cells). Here, to better understand the mechanisms of goal selection and maintenance, we studied correlated activity among pairs of these neurons. Among the 3 possible types of pairs, F-F and F-P pairs often exhibited significant correlations when and after monkeys selected future goals but P-P pairs rarely did. These correlations were stronger when monkeys shifted from a previous goal than when they stayed with that goal. In addition, members of F-F pairs usually preferred the same goal and thus shared both prospective coding and spatial tuning properties. In contrast, cells composing F-P pairs usually had different spatial preferences and thus shared neither coding nor spatial tuning properties. On the assumption that the neurons composing a pair send convergent outputs to target neurons, their correlated activity could enhance their efficacy in context-dependent goal selection, goal maintenance, and the transformation of goal choices into action.

Encoding problem-solving strategies in prefrontal cortex: activity during strategic errors

The primate prefrontal cortex (PF) plays a central role in choosing goals and strategies. To better understand its mechanisms, we recorded from PF neurons as monkeys used abstract response strategies to select a spatial goal. A visual cue, selected randomly from a set of three cues, appeared on each trial. All three cues were novel when neuronal recording commenced. From trial to trial, the cue could have either been repeated or changed from the previous trial; these were called repeat trials and change trials, respectively. On repeat trials, the monkeys used a Repeat-stay strategy to gain a reward by choosing the same spatial goal as on the previous trial; on change trials, they used a Change-shift strategy to reject the previous goal in favour of an alternative. We reported previously that when monkeys performed the task correctly, many PF neurons had activity encoding one of these two strategies. The monkeys sometimes chose the incorrect strategy, however. Strategy coding was weak or absent during the cue period of error trials which was, for correct trials, the time when the monkeys used a strategy to choose a future goal. By contrast, later in the trial, after the chosen goal had been attained and the monkeys awaited feedback, strategy coding was present and it reflected the strategy used, whether correct or incorrect. The weak cue-period strategy signal could, whatever its cause, have contributed to the errors made, whereas the activity prior to feedback suggests a role in monitoring task performance.

Toward an integrative perspective on hippocampal function: from the rapid encoding of experience to adaptive behavior

The mammalian hippocampus has been associated with learning and memory, as well as with many other behavioral processes. In this article, these different perspectives are brought together, and it is pointed out that integration of diverse functional domains may be a key feature enabling the hippocampus to support not only the encoding and retrieval of certain memory representations, but also their translation into adaptive behavior. The hippocampus appears to combine: (i) sensory afferents and synaptic mechanisms underlying certain types of rapid learning; and (ii) links to motivational, emotional, executive, and sensorimotor functions. Recent experiments are highlighted, indicating that the induction of hippocampal synaptic plasticity is required to encode rapidly aspects of experience, such as places, into memory representations; subsequent retrieval of these representations requires transmission through the previously modified hippocampal synapses, but no further plasticity. In contrast, slow incremental place learning may not absolutely require hippocampal contributions. The neocortical sensory inputs, especially visuo-spatial information, necessary for hippocampus-dependent rapid learning, are preferentially associated with the septal to intermediate hippocampus. In contrast, connectivity with the prefrontal cortex and subcortical sites, which link the hippocampus to motivational, emotional, executive, and sensorimotor functions, is primarily associated with the intermediate to temporal hippocampus. A model of functional differentiation and integration along the septo-temporal axis of the hippocampus is proposed, describing key hippocampal contributions to adaptive behavior based on information encoded during a single or a few past experiences.

The costs and benefits of cross-task priming

Two lines of research on cross-task priming yield opposite results. Research on repetition priming observed positive priming, whereas research on the role of priming in task-switching observed negative effects. We combined the two types of design. In the transfer phase of our paradigm, subjects performed task B either as a pure block (BBB) or as a switch block (ABAB). We presented items which were either unprimed or primed by prior presentation during a preceding priming phase performed on task A. Amongst others, the priming effect is determined by two factors: First, the more operation time the system needs during the probe event, the higher the likelihood to obtain priming. Protracting operation time by reducing stimulus quality favors positive priming, whereas providing more operation time by making subjects switch between tasks favors negative priming. Second, the strength of the memory trace of the prime event determines whether that trace can possibly yield negative priming, in that only strong traces can be retrieved together with the associated task/response.

Concurrent task performance enhances low-level visuomotor learning

Visuomotor association learning involves learning to make a motor response to an arbitrary visual stimulus. This learning is essential for visual search and discrimination performance and is reliant upon a well-defined neural circuit in the brain that includes the prefrontal cortex and the hippocampal formation. In the present study, we investigated the possible role of attentional processes during such learning using dual-task interference. A motor, verbal, or perceptual concurrent task was performed during the learning/training block of a simple visual discrimination task. Contrary to expectation, the dual-task groups showed improved learning and learning-dependent performance compared with untrained control and non-dual-task trained groups. A second experiment revealed that this effect did not appear to be due to increased arousal level; the inclusion of alerting tones during learning did not result in facilitation. These findings suggest that the engagement of attention, but not arousal, during the acquisition of a visuomotor association can facilitate this learning and its expression.

Probabilistic reasoning by neurons

Our brains allow us to reason about alternatives and to make choices that are likely to pay off. Often there is no one correct answer, but instead one that is favoured simply because it is more likely to lead to reward. A variety of probabilistic classification tasks probe the covert strategies that humans use to decide among alternatives based on evidence that bears only probabilistically on outcome. Here we show that rhesus monkeys can also achieve such reasoning. We have trained two monkeys to choose between a pair of coloured targets after viewing four shapes, shown sequentially, that governed the probability that one of the targets would furnish reward. Monkeys learned to combine probabilistic information from the shape combinations. Moreover, neurons in the parietal cortex reveal the addition and subtraction of probabilistic quantities that underlie decision-making on this task.

MatOFF: a tool for analyzing behaviorally complex neurophysiological experiments

The simple operant conditioning originally used in behavioral neurophysiology 30 years ago has given way to complex and sophisticated behavioral paradigms; so much so, that early general purpose programs for analyzing neurophysiological data are ill-suited for complex experiments. The trend has been to develop custom software for each class of experiment, but custom software can have serious drawbacks. We describe here a general purpose software tool for behavioral and electrophysiological studies, called MatOFF, that is especially suited for processing neurophysiological data gathered during the execution of complex behaviors. Written in the MATLAB programming language, MatOFF solves the problem of handling complex analysis requirements in a unique and powerful way. While other neurophysiological programs are either a loose collection of tools or append MATLAB as a post-processing step, MatOFF is an integrated environment that supports MATLAB scripting within the event search engine safely isolated in a programming sandbox. The results from scripting are stored separately, but in parallel with the raw data, and thus available to all subsequent MatOFF analysis and display processing. An example from a recently published experiment shows how all the features of MatOFF work together to analyze complex experiments and mine neurophysiological data in efficient ways.

EEG sharp waves and sparse ensemble unit activity in the macaque hippocampus

Neural unit activity and EEGs were recorded from inferior temporal regions of three rhesus macaques chronically implanted with "hyperdrives" holding 12 individually movable tetrodes. Recordings were made from each monkey over a period of approximately 3 mo, while the electrodes were moved by small increments through the hippocampus and neighboring structures. After recording, the monkeys were necropsied, and the brains were sectioned and Nissl-stained, permitting identification of individual electrode tracks. The results establish that hippocampal pyramidal cells are "complex spike cells," firing at overall average rates of approximately 0.3 Hz, with spike trains consisting of long periods of silence interspersed with bursts of activity. The results also establish that the monkey hippocampal EEG shows "sharp wave" events consisting of a high-frequency "ripple" oscillation ( approximately 110 Hz) together with a large slow-wave EEG deflection lasting several hundred milliseconds. The evidence suggests that monkey sharp waves are probably generated mainly in the CA1 region and that sharp waves are associated with an inactive/drowsy-or-sleeping behavioral state, which is also associated with increased hippocampal pyramidal cell activity and increased hippocampal EEG amplitude. The results of this initial study of ensembles of primate hippocampal neurons are consistent with previous studies in rodents and consistent with the hypothesis that theories and models of hippocampal memory function developed on the basis of rat data may be applicable to a wide range of mammalian species.

Making New Memories: The Role of the Hippocampus in New Associative Learning

Both aging and Alzheimer's disease target the hippocampal formation and can result in mild to devastating memory impairment depending on the severity of the condition. Understanding the normal mnemonic functions of the hippocampus and related structures of the medial temporal lobe is the first step toward the development of diagnostics and treatments designed to ameliorate these potentially devastating age-related memory deficits. Here I describe findings from behavioral neurophysiological studies in which we have investigated the patterns of dynamic neural activity seen in the macaque monkey hippocampus during the acquisition of new associative memories. We report that hippocampal neurons signal the formation of new associations with dramatic changes in their firing rate. Because these learning-related signals can occur just before behavioral learning is expressed, this suggests that these signals play a role in driving the learning process. Implications of these findings for understanding the memory deficits associated with aging and Alzheimer's disease are discussed.

Integrating associative learning signals across the brain

Associative learning is defined as the ability to link arbitrary stimuli or actions together in memory. The neural correlates of this fundamental form of plasticity were first described in the hippocampus during delay eye blink conditioning and have since been examined using a variety of tasks in both rats and monkeys. In monkeys, the neural correlates of associative learning have been studied using conditional motor learning tasks where animals learn to associate particular visual stimuli with particular motor responses (i.e., touch left or touch right). Similar tasks have also been used to examine learning-related plasticity in motor-related areas throughout the frontal lobe and striatum. Here, we review the patterns of learning-related activity seen in these diverse brain areas during conditional motor learning. While each of these areas exhibits strong associative learning signals, the differential patterns and time courses of these signals provides insight into the unique contribution of each area to associative learning.

Behavioral and neurophysiological analyses of dynamic learning processes

In this article, the authors address two topics relevant to the study of the brain basis of associative learning. In Part 1, they compare and contrast the patterns and time course of dynamic learning-related neural activity that have been reported in the medial temporal lobe, premotor cortex, prefrontal cortex, and striatum during various associative learning tasks. In Part 2, they examine the statistical methodologies that have been used to analyze both behavioral learning and learning-related neural activity. They describe a state-space model of behavioral learning that provides accurate estimates of dynamic learning processes and a point-process filter algorithm that tracks the dynamic changes in neural activity on a millisecond time scale. Future challenges for these statistical methodologies and their application to the study of the brain basis of associative learning are discussed.

Processing stages underlying word recognition in the anteroventral temporal lobe

The anteroventral temporal lobe integrates visual, lexical, semantic and mnestic aspects of word processing, through its reciprocal connections with the ventral visual stream, language areas, and the hippocampal formation. We used linear microelectrode arrays to probe population synaptic currents and neuronal firing in different cortical layers of the anteroventral temporal lobe, during semantic judgments with implicit priming and overt word recognition. Since different extrinsic and associative inputs preferentially target different cortical layers, this method can help reveal the sequence and nature of local processing stages at a higher resolution than was previously possible. The initial response in inferotemporal and perirhinal cortices is a brief current sink beginning at approximately 120 ms and peaking at approximately 170 ms. Localization of this initial sink to middle layers suggests that it represents feedforward input from lower visual areas, and simultaneously increased firing implies that it represents excitatory synaptic currents. Until approximately 800 ms, the main focus of transmembrane current sinks alternates between middle and superficial layers, with the superficial focus becoming increasingly dominant after approximately 550 ms. Since superficial layers are the target of local and feedback associative inputs, this suggests an alternation in predominant synaptic input between feedforward and feedback modes. Word repetition does not affect the initial perirhinal and inferotemporal middle layer sink but does decrease later activity. Entorhinal activity begins later (approximately 200 ms), with greater apparent excitatory post-synaptic currents and multiunit activity in neocortically projecting than hippocampal-projecting layers. In contrast to perirhinal and entorhinal responses, entorhinal responses are larger to repeated words during memory retrieval. These results identify a sequence of physiological activation, beginning with a sharp activation from lower level visual areas carrying specific information to middle layers. This is followed by feedback and associative interactions involving upper cortical layers, which are abbreviated to repeated words. Following bottom-up and associative stages, top-down recollective processes may be driven by entorhinal cortex. Word processing involves a systematic sequence of fast feedforward information transfer from visual areas to anteroventral temporal cortex followed by prolonged interactions of this feedforward information with local associations and feedback mnestic information from the medial temporal lobe.

Conditional visuo-motor learning and dimension reduction

Conditional visuo-motor learning consists in learning by trial and error to associate visual cues with correct motor responses, that have no direct link. Converging evidence supports the role of a large brain network in this type of learning, including the prefrontal and the premotor cortex, the basal ganglia BG and the hippocampus. In this paper we focus on the role of a major structure of the BG, the striatum. We first present behavioral results and electrophysiological data recorded from this structure in monkeys engaged in learning new visuo-motor associations. Visual stimuli were presented on a video screen and the animals had to learn, by trial and error, to select the correct movement of a joystick, in order to receive a liquid reward. Behavioral results revealed that the monkeys used a sequential strategy, whereby they learned the associations one by one although they were presented randomly. Human subjects, tested on the same task, also used a sequential strategy. Neuronal recordings in monkeys revealed learning-related modulations of neural activity in the striatum. We then present a mathematical model inspired by viability theory developed to implement the use of strategies during learning. This model complements existing models of the BG based on reinforcement learning RL, which do not take into account the use of strategies to reduce the dimension of the learning space.

Comparison of population activity in the dorsal premotor cortex and putamen during the learning of arbitrary visuomotor mappings

A previous study found that as monkeys learned novel mappings between visual cues and responses, neuronal activity patterns evolved at approximately the same time in both the dorsal premotor cortex (PMd) and the putamen. Here we report that, in both regions, the population activity for novel mappings came to resemble that for familiar ones as learning progressed. Both regions showed activity differences on trials with correct responses versus those with incorrect ones. In addition to these common features, we observed two noteworthy differences between PMd and putamen activity during learning. After a response choice had been made, but prior to feedback about the correctness of that choice (reward or nonreward), the putamen showed a sustained activity increase in activity, whereas PMd did not. Also in the putamen, this prereward activity was highly selective for the specific visuomotor mapping that had just been performed, and this selectivity was maintained until the time of the reward. After performance reached an asymptote, the degree of this selectivity decreased markedly to the level typical for familiar visuomotor mappings. These findings support the hypothesis that neurons in the striatum play a pivotal role in associative learning.

Spatial view cells in the hippocampus, and their idiothetic update based on place and head direction

Single neuron recording studies have demonstrated the existence of spatial view neurons which encode information about the spatial location at which a primate is looking in the environment. These neurons are able to maintain their firing even in the absence of visual input. The standard neuronal network approach to model networks with memory that represent continuous spaces is that of continuous attractor neural networks. Stringer, Rolls and Trappenberg (2005) have recently shown how idiothetic (self-motion) inputs could update the activity packet of neuronal firing within a two-dimensional continuous attractor neural network of spatial view cells. However, this earlier study examined only the simplified situation in which the agent could rotate on the spot or move its eyes. In this paper we show how spatial view cells could be driven by head direction and place cells, themselves idiothetically updated. The head direction and place neurons are remapped by a competitive network with expansion recoding which self-organises so that different neurons represent different combinations of head direction and the place where the agent is located. The combination cells are then mapped by pattern association involving long-term synaptic potentiation but also long-term homosynaptic depression to spatial view cells, which during training are driven by the spatial view. After training, the spatial view cells are updated in the dark by the idiothetically driven head direction and place cells.

Conditional motor learning in the nonspatial domain: effects of errorless learning and the contribution of the fornix to one-trial learning

Conditional motor learning contributes importantly to behavioral flexibility. In previous work, the authors found that fornix transections impaired the ability of macaque monkeys (Macaca mulatta) to learn conditional motor associations between the nonspatial features of visual stimuli and nonspatially differentiated responses. In the present study, they found that significant 1-trial learning of such associations also depended on the fornix. Furthermore, removal of the hippocampus, subiculum, and subjacent parahippocampal cortex, added to fornix transection, had no effect, thus demonstrating that fornix transections eliminated the contribution of the hippocampal system. In addition, the authors examined the effect of errorless learning and found, in control monkeys, that errors made prior to the 1st correct response retarded 1-trial learning.

Prefrontal cortex activity related to abstract response strategies

Many monkeys adopt abstract response strategies as they learn to map visual symbols to responses by trial and error. According to the repeat-stay strategy, if a symbol repeats from a previous, successful trial, the monkeys should stay with their most recent response choice. According to the change-shift strategy, if the symbol changes, the monkeys should shift to a different choice. We recorded the activity of prefrontal cortex neurons while monkeys chose responses according to these two strategies. Many neurons had activity selective for the strategy used. In a subsequent block of trials, the monkeys learned fixed stimulus-response mappings with the same stimuli. Some neurons had activity selective for choosing responses based on fixed mappings, others for choosing based on abstract strategies. These findings indicate that the prefrontal cortex contributes to the implementation of the abstract response strategies that monkeys use during trial-and-error learning.