Generalization gradients and representation modes after absolute and relative discrimination learning in young chickens

Perceptual relativity has become a central issue in animal psychophysics. In order to assess how the interplay of training experience and stimulus dimension might affect perceptual relativity, we investigated the role of 'absolute' and 'relative' training on the learning and representation of stimuli from two dimensions that might favor absolute or relative encoding to a different degree. Young chicks learned to discriminate 3D-objects by either color or size. During 'absolute' training always the choice of the same stimulus of a simultaneously presented pair was reinforced, while choice of the larger (smaller) or greener (bluer) stimulus was reinforced during 'relative' training. Overall, discrimination learning was faster with relative training, but size learning profited more from 'relative' training than color learning. Post-training generalization tests revealed a combined effect of training experience and stimulus dimension: a higher amount of absolute encoding occurred with absolute training and color learning, while relative choices prevailed with relative training and size learning.     

Are There Multiple Visual Short-Term Memory Stores?

Background

Classic work on visual short-term memory (VSTM) suggests that people store a limited amount of items for subsequent report. However, when human observers are cued to shift attention to one item in VSTM during retention, it seems as if there is a much larger representation, which keeps additional items in a more fragile VSTM store. Thus far, it is not clear whether the capacity of this fragile VSTM store indeed exceeds the traditional capacity limits of VSTM. The current experiments address this issue and explore the capacity, stability, and duration of fragile VSTM representations.

Methodology/Principal Findings

We presented cues in a change-detection task either just after off-set of the memory array (iconic-cue), 1,000 ms after off-set of the memory array (retro-cue) or after on-set of the probe array (post-cue). We observed three stages in visual information processing 1) iconic memory with unlimited capacity, 2) a four seconds lasting fragile VSTM store with a capacity that is at least a factor of two higher than 3) the robust and capacity-limited form of VSTM. Iconic memory seemed to depend on the strength of the positive after-image resulting from the memory display and was virtually absent under conditions of isoluminance or when intervening light masks were presented. This suggests that iconic memory is driven by prolonged retinal activation beyond stimulus duration. Fragile VSTM representations were not affected by light masks, but were completely overwritten by irrelevant pattern masks that spatially overlapped the memory array.

Conclusions/Significance

We find that immediately after a stimulus has disappeared from view, subjects can still access information from iconic memory because they can see an after-image of the display. After that period, human observers can still access a substantial, but somewhat more limited amount of information from a high-capacity, but fragile VSTM that is overwritten when new items are presented to the eyes. What is left after that is the traditional VSTM store, with a limit of about four objects. We conclude that human observers store more sustained representations than is evident from standard change detection tasks and that these representations can be accessed at will.     

Profound context-dependent plasticity of mitral cell responses in olfactory bulb

On the basis of its primary circuit it has been postulated that the olfactory bulb (OB) is analogous to the retina in mammals. In retina, repeated exposure to the same visual stimulus results in a neural representation that remains relatively stable over time, even as the meaning of that stimulus to the animal changes. Stability of stimulus representation at early stages of processing allows for unbiased interpretation of incoming stimuli by higher order cortical centers. The alternative is that early stimulus representation is shaped by previously derived meaning, which could allow more efficient sampling of odor space providing a simplified yet biased interpretation of incoming stimuli. This study helps place the olfactory system on this continuum of subjective versus objective early sensory representation. Here we show that odor responses of the output cells of the OB, mitral cells, change transiently during a go–no-go odor discrimination task. The response changes occur in a manner that increases the ability of the circuit to convey information necessary to discriminate among closely related odors. Remarkably, a switch between which of the two odors is rewarded causes mitral cells to switch the polarity of their divergent responses. Taken together these results redefine the function of the OB as a transiently modifiable (active) filter, shaping early odor representations in behaviorally meaningful ways.     

Visual Discrimination Learning in the Small Octopus Octopus ocellatus

Octopus ocellatus is a small benthic species of octopus that is easy to rear and spawns large eggs during a short life cycle. These and other features of O. ocellatus may make it an advantageous subject for a broad range of behavioral studies, including those involving various types of learning. However, no type of learning has been studied in O. ocellatus. In a successive visual discrimination task, in which a ‘positive’ or ‘negative’ stimulus (white or black rectangle) was presented to a subject octopus and appropriate rewards or punishments were given to the subject, the rate of ‘correct’ responses (i.e., touches to the positive stimulus or refraining from the negative stimulus) gradually increased. Moreover, ‘observer’ octopuses that observed another octopus performing a visual discrimination task in which reward was also given to the ‘incorrect’ responses (touches to the negative stimulus) showed a higher ratio of incorrect responses in their test sessions. These results, coupled with the physical characteristics of O. ocellatus, indicate that this species is potentially suitable for neurogenetic and neuroembryological studies of learning.     

Classical reward conditioning in Drosophila melanogaster

Negatively reinforced olfactory conditioning has been widely employed to identify learning and memory genes, signal transduction pathways and neural circuitry in Drosophila. To delineate the molecular and cellular processes underlying reward-mediated learning and memory, we developed a novel assay system for positively reinforced olfactory conditioning. In this assay, flies were involuntarily exposed to the appetitive unconditioned stimulus sucrose along with a conditioned stimulus odour during training and their preference for the odour previously associated with sucrose was measured to assess learning and memory capacities. After one training session, wild-type Canton S flies displayed reliable performance, which was enhanced after two training cycles with 1-min or 15-min inter-training intervals. Higher performance scores were also obtained with increasing sucrose concentration. Memory in Canton S flies decayed slowly when measured at 30 min, 1 h and 3 h after training; whereas, it had declined significantly at 6 h and 12 h post-training. When learning mutant t beta h flies, which are deficient in octopamine, were challenged, they exhibited poor performance, validating the utility of this assay. As the Drosophila model offers vast genetic and transgenic resources, the new appetitive conditioning described here provides a useful tool with which to elucidate the molecular and cellular underpinnings of reward learning and memory. Similar to negatively reinforced conditioning, this reward conditioning represents classical olfactory conditioning. Thus, comparative analyses of learning and memory mutants in two assays may help identify the molecular and cellular components that are specific to the unconditioned stimulus information used in conditioning.     

What can we teach Lymnaea and what can Lymnaea teach us?

This review describes the advantages of adopting a molluscan complementary model, the freshwater snail Lymnaea stagnalis, to study the neural basis of learning and memory in appetitive and avoidance classical conditioning; as well as operant conditioning of its aerial respiratory and escape behaviour. We firstly explored ‘what we can teach Lymnaea’ by discussing a variety of sensitive, solid, easily reproducible and simple behavioural tests that have been used to uncover the memory abilities of this model system. Answering this question will allow us to open new frontiers in neuroscience and behavioural research to enhance our understanding of how the nervous system mediates learning and memory. In fact, from a translational perspective, Lymnaea and its nervous system can help to understand the neural transformation pathways from behavioural output to sensory coding in more complex systems like the mammalian brain. Moving on to the second question: ‘what can Lymnaea teach us?’, it is now known that Lymnaea shares important associative learning characteristics with vertebrates, including stimulus generalization, generalization of extinction and discriminative learning, opening the possibility to use snails as animal models for neuroscience translational research.     

Conditioning and time representation in long short-term memory networks

Dopaminergic models based on the temporal-difference learning algorithm usually do not differentiate trace from delay conditioning. Instead, they use a fixed temporal representation of elapsed time since conditioned stimulus onset. Recently, a new model was proposed in which timing is learned within a long short-term memory (LSTM) artificial neural network representing the cerebral cortex (Rivest et al. in J Comput Neurosci 28(1):107–130, 2010). In this paper, that model’s ability to reproduce and explain relevant data, as well as its ability to make interesting new predictions, are evaluated. The model reveals a strikingly different temporal representation between trace and delay conditioning since trace conditioning requires working memory to remember the past conditioned stimulus while delay conditioning does not. On the other hand, the model predicts no important difference in DA responses between those two conditions when trained on one conditioning paradigm and tested on the other. The model predicts that in trace conditioning, animal timing starts with the conditioned stimulus offset as opposed to its onset. In classical conditioning, it predicts that if the conditioned stimulus does not disappear after the reward, the animal may expect a second reward. Finally, the last simulation reveals that the buildup of activity of some units in the networks can adapt to new delays by adjusting their rate of integration. Most importantly, the paper shows that it is possible, with the proposed architecture, to acquire discharge patterns similar to those observed in dopaminergic neurons and in the cerebral cortex on those tasks simply by minimizing a predictive cost function.     

Genetic disruptions of Drosophila Pavlovian learning leave extinction learning intact

Individuals who experience traumatic events may develop persistent posttraumatic stress disorder. Patients with this disorder are commonly treated with exposure therapy, which has had limited long‐term success. In experimental neurobiology, fear extinction is a model for exposure therapy. In this behavioral paradigm, animals are repeatedly exposed in a safe environment to the fearful stimulus, which leads to greatly reduced fear. Studying animal models of extinction already has lead to better therapeutic strategies and development of new candidate drugs. Lack of a powerful genetic model of extinction, however, has limited progress in identifying underlying molecular and genetic factors. In this study, we established a robust behavioral paradigm to study the short‐term effect (acquisition) of extinction in Drosophila melanogaster. We focused on the extinction of olfactory aversive 1‐day memory with a task that has been the main workhorse for genetics of memory in flies. Using this paradigm, we show that extinction can inhibit each of two genetically distinct forms of consolidated memory. We then used a series of single‐gene mutants with known impact on associative learning to examine the effects on extinction. We find that extinction is intact in each of these mutants, suggesting that extinction learning relies on different molecular mechanisms than does Pavlovian learning.     

Trace Fear Conditioning in Mice

In this experiment we present a technique to measure learning and memory. In the trace fear conditioning protocol presented here there are five pairings between a neutral stimulus and an unconditioned stimulus. There is a 20 sec trace period that separates each conditioning trial. On the following day freezing is measured during presentation of the conditioned stimulus (CS) and trace period. On the third day there is an 8 min test to measure contextual memory. The representative results are from mice that were presented with the aversive unconditioned stimulus (shock) compared to mice that received the tone presentations without the unconditioned stimulus. Trace fear conditioning has been successfully used to detect subtle learning and memory deficits and enhancements in mice that are not found with other fear conditioning methods. This type of fear conditioning is believed to be dependent upon connections between the medial prefrontal cortex and the hippocampus. One current controversy is whether this method is believed to be amygdala-independent. Therefore, other fear conditioning testing is needed to examine amygdala-dependent learning and memory effects, such as through the delay fear conditioning.     

Molecular Signals into the Insular Cortex and Amygdala During Aversive Gustatory Memory Formation

In this paper, we will provide evidence of the putative molecular signals and biochemical events that mediate the formation of long-lasting gustatory memory trace. When an animal drinks a novel taste (the conditioned stimulus; CS) and it is later associated with malaise (unconditioned stimulus; US), the animal will reject it in the next presentation, developing a long-lasting taste aversion, i.e., the taste cue becomes an aversive signal, and this is referred to as conditioning taste aversion. Different evidence indicates that the novel stimulus (taste) induces a rapid and strong cortical acetylcholine activity that decreases when the stimulus becomes familiar after several presentations. Cholinergic activation via muscarinic receptors initiates a series of intracellular events leading to plastic changes that could be related to short- and/or long-term memory gustatory trace. Such plastic changes facilitate the incoming US signals carried out by, in part, the glutamate release induced by the US. Altogether, these events could produce the cellular changes related to the switch from safe to aversive taste memory trace. A proposed working model to explain the biochemical sequence of signals during taste memory formation will be discussed.     

Shifting attention within memory representations involves early visual areas

Prior studies have shown that spatial attention modulates early visual cortex retinotopically, resulting in enhanced processing of external perceptual representations. However, it is not clear whether the same visual areas are modulated when attention is focused on, and shifted within a working memory representation. In the current fMRI study participants were asked to memorize an array containing four stimuli. After a delay, participants were presented with a verbal cue instructing them to actively maintain the location of one of the stimuli in working memory. Additionally, on a number of trials a second verbal cue instructed participants to switch attention to the location of another stimulus within the memorized representation. Results of the study showed that changes in the BOLD pattern closely followed the locus of attention within the working memory representation. A decrease in BOLD-activity (V1-V3) was observed at ROIs coding a memory location when participants switched away from this location, whereas an increase was observed when participants switched towards this location. Continuous increased activity was obtained at the memorized location when participants did not switch. This study shows that shifting attention within memory representations activates the earliest parts of visual cortex (including V1) in a retinotopic fashion. We conclude that even in the absence of visual stimulation, early visual areas support shifting of attention within memorized representations, similar to when attention is shifted in the outside world. The relationship between visual working memory and visual mental imagery is discussed in light of the current findings.     

Minimal cross-trial generalization in learning the representation of an odor-guided choice task

There is no single way to represent a task. Indeed, despite experiencing the same task events and contingencies, different subjects may form distinct task representations. As experimenters, we often assume that subjects represent the task as we envision it. However, such a representation cannot be taken for granted, especially in animal experiments where we cannot deliver explicit instruction regarding the structure of the task. Here, we tested how rats represent an odor-guided choice task in which two odor cues indicated which of two responses would lead to reward, whereas a third odor indicated free choice among the two responses. A parsimonious task representation would allow animals to learn from the forced trials what is the better option to choose in the free-choice trials. However, animals may not necessarily generalize across odors in this way. We fit reinforcement-learning models that use different task representations to trial-by-trial choice behavior of individual rats performing this task, and quantified the degree to which each animal used the more parsimonious representation, generalizing across trial types. Model comparison revealed that most rats did not acquire this representation despite extensive experience. Our results demonstrate the importance of formally testing possible task representations that can afford the observed behavior, rather than assuming that animals’ task representations abide by the generative task structure that governs the experimental design.     

Learning invariant object recognition in the visual system with continuous transformations

The cerebral cortex utilizes spatiotemporal continuity in the world to help build invariant representations. In vision, these might be representations of objects. The temporal continuity typical of objects has been used in an associative learning rule with a short-term memory trace to help build invariant object representations. In this paper, we show that spatial continuity can also provide a basis for helping a system to self-organize invariant representations. We introduce a new learning paradigm “continuous transformation learning” which operates by mapping spatially similar input patterns to the same postsynaptic neurons in a competitive learning system. As the inputs move through the space of possible continuous transforms (e.g. translation, rotation, etc.), the active synapses are modified onto the set of postsynaptic neurons. Because other transforms of the same stimulus overlap with previously learned exemplars, a common set of postsynaptic neurons is activated by the new transforms, and learning of the new active inputs onto the same postsynaptic neurons is facilitated. We demonstrate that a hierarchical model of cortical processing in the ventral visual system can be trained with continuous transform learning, and highlight differences in the learning of invariant representations to those achieved by trace learning.     

Taste memory formation: role of nucleus accumbens

When a novel taste has been associated with postingestive malaise, animals recognize this taste as aversive. This associative learning is known as conditioned taste aversion. However, when an animal consumes a novel taste and no aversive consequences follow, it becomes recognized as a safe signal, leading to an increase in its consumption in subsequent presentations. In this review, we will discuss the results related to the taste memory formation focusing particularly on the nucleus accumbens (NAcc). The NAcc keeps projections with amygdala, insular cortex, parabrachial nucleus, and nucleus of the solitary tract areas important for taste memory formation. We will review the evidence relating to how the NAcc could be involved in taste memory formation, due to its role in the taste memory trace formation and its role in the association of the conditioned stimulus-unconditioned stimulus, and finally the retrieval of taste memory. In this context, we will review the participation of the cholinergic, dopaminergic, and glutamatergic systems in the NAcc during taste memory formation.     

Memory retention of appetitive and aversive conditioning in the damselfish Stegastes fuscus

We compared the memory of damselfish Stegastes fuscus in an aversive and appetitive conditioning task. Fish were trained to associate the sides of the tank that corresponded to the presence of a positive (conspecific presence) or negative (electroshock) stimulus. After two conditioning sessions, they were tested for learning. The fish conditioned to the stimulus were then re-tested for memory retention after 5, 10 or 15 days. Both the positive and negative rewards were associated with a specific side of the tank, indicating learning ability. Additionally, in both contexts, S. fuscus stored the information learned and showed similar behavioural patterns after 5, 10 and 15 days, suggesting long-lasting memory. For the ecological context, long lasting memories of social encounters outcomes and negative experiences of threatening situations may confer advantages that ultimately affect fishes’ fitness.     

Distinct Visual Working Memory Systems for View-Dependent and View-Invariant Representation

Background

How do people sustain a visual representation of the environment? Currently, many researchers argue that a single visual working memory system sustains non-spatial object information such as colors and shapes. However, previous studies tested visual working memory for two-dimensional objects only. In consequence, the nature of visual working memory for three-dimensional (3D) object representation remains unknown.

Methodology/Principal Findings

Here, I show that when sustaining information about 3D objects, visual working memory clearly divides into two separate, specialized memory systems, rather than one system, as was previously thought. One memory system gradually accumulates sensory information, forming an increasingly precise view-dependent representation of the scene over the course of several seconds. A second memory system sustains view-invariant representations of 3D objects. The view-dependent memory system has a storage capacity of 3–4 representations and the view-invariant memory system has a storage capacity of 1–2 representations. These systems can operate independently from one another and do not compete for working memory storage resources.

Conclusions/Significance

These results provide evidence that visual working memory sustains object information in two separate, specialized memory systems. One memory system sustains view-dependent representations of the scene, akin to the view-specific representations that guide place recognition during navigation in humans, rodents and insects. The second memory system sustains view-invariant representations of 3D objects, akin to the object-based representations that underlie object cognition.     

Animal learning and memory: an integration of cognition and ecology

A wonderfully lucid framework for the ways to understand animal behaviour is that represented by the four ‘whys’ proposed by Tinbergen (1963). For much of the past three decades, however, these four avenues have been pursued more or less in parallel. Functional questions, for example, have been addressed by behavioural ecologists, mechanistic questions by psychologists and ethologists, ontogenetic questions by developmental biologists and neuroscientists and phylogenetic questions by evolutionary biologists. More recently, the value of integration between these differing views has become apparent. In this brief review, we concentrate especially on current attempts to integrate mechanistic and functional approaches.

Most of our understanding of learning and memory in animals comes from the psychological literature, which tends to use only rats or pigeons, and more occasionally primates, as subjects. The underlying psychological assumption is of general processes that are similar across species and contexts rather than a range of specific abilities. However, this does not seem to be entirely true as several learned behaviours have been described that are specific to particular species or contexts. The first conspicuous exception to the generalist assumption was the demonstration of long delay taste aversion learning in rats (Garcia et al., 1955), in which it was shown that a stimulus need not be temporally contiguous with a response for the animal to make an association between food and illness. Subsequently, a number of other examples, such as imprinting and song learning in birds (e.g., Bolhuis and Honey, 1998; Catchpole and Slater, 1995; Horn, 1998), have been thoroughly researched. Even in these cases, however, it has been typical for only a few species to be studied (domestic chicks provide the ‘model’ imprinting species and canaries and zebra finches the song learning ‘models’). As a result, a great deal is understood about the neural underpinnings and development of the behaviour, but substantially less is understood about interspecific variation and whether variation in behaviour is correlated with variation in neural processing (see review by Tramontin and Brenowitz, 2000 but see ten Cate and Vos, 1999).     

Are binary synapses superior to graded weight representations in stochastic attractor networks?

Synaptic plasticity is an underlying mechanism of learning and memory in neural systems, but it is controversial whether synaptic efficacy is modulated in a graded or binary manner. It has been argued that binary synaptic weights would be less susceptible to noise than graded weights, which has impelled some theoretical neuroscientists to shift from the use of graded to binary weights in their models. We compare retrieval performance of models using both binary and graded weight representations through numerical simulations of stochastic attractor networks. We also investigate stochastic attractor models using multiple discrete levels of weight states, and then investigate the optimal threshold for dilution of binary weight representations. Our results show that a binary weight representation is not less susceptible to noise than a graded weight representation in stochastic attractor models, and we find that the load capacities with an increasing number of weight states rapidly reach the load capacity with graded weights. The optimal threshold for dilution of binary weight representations under stochastic conditions occurs when approximately 50% of the smallest weights are set to zero.     

Reconciling neuronal representations of schema,abstract task structure,and categorization under cognitive maps in the entorhinal-hippocampal-frontal circuits

Learning leads to a neuronal representation of acquired knowledge. This idea of knowledge representation was traditionally developed as a “cognitive map” of spatial memory represented in the hippocampus. The framework of cognitive mapping has been extended in the past decade to include not only spatial memory, but also non-spatial factual and temporal memory. Following this conceptual advancement, a line of recent neurophysiological research discovered such knowledge representations not only in the hippocampus, but also in the entorhinal cortex and frontal cortex. Although the distinct terms “cognitive map,” “schema,” “abstract task structure” or “categorization” were used in these studies, it is likely that these terms can be reconciled as a common mechanism of learned knowledge representations. Future experimental work will be required to differentiate the parametric nature of knowledge representations across brain areas.     

Inside out: modern imaging techniques to reveal animal anatomy

Animal anatomy has traditionally relied on detailed dissections to produce anatomical illustrations, but modern imaging modalities, such as MRI and CT, now represent an enormous resource that allows for fast non-invasive visualizations of animal anatomy in living animals. These modalities also allow for creation of three-dimensional representations that can be of considerable value in the dissemination of anatomical studies. In this methodological review, we present our experiences using MRI, CT and μCT to create advanced representation of animal anatomy, including bones, inner organs and blood vessels in a variety of animals, including fish, amphibians, reptiles, mammals, and spiders. The images have a similar quality to most traditional anatomical drawings and are presented together with interactive movies of the anatomical structures, where the object can be viewed from different angles. Given that clinical scanners found in the majority of larger hospitals are fully suitable for these purposes, we encourage biologists to take advantage of these imaging techniques in creation of three-dimensional graphical representations of internal structures.