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.

Orbitofrontal neurons signal reward predictions, not reward prediction errors.

Neurons in the orbitofrontal cortex (OFC) fire in anticipation of and during rewards. Such firing has been suggested to encode reward predictions and to account in some way for the role of this area in adaptive behavior and learning. However, it has also been reported that neural activity in OFC reflects reward prediction errors, which might drive learning directly. Here we tested this question by analyzing the firing of OFC neurons recorded in an odor discrimination task in which rats were trained to sample odor cues and respond left or right on each trial for reward. Neurons were recorded across blocks of trials in which we switched either the number or the flavor of the reward delivered in each well. Previously we have described how neurons in this dataset fired to the predictive cues (Stalnaker et al., 2014); here we focused on the firing in anticipation of and just after delivery of each drop of reward, looking specifically for differences in firing based on whether the reward number or flavor was unexpected or expected. Unlike dopamine neurons recorded in this setting, which exhibited phasic error-like responses after surprising changes in either reward number or reward flavor (Takahashi et al., 2017), OFC neurons showed no such error correlates and instead fired in a way that reflected reward predictions.

Neural Estimates of Imagined Outcomes in Basolateral Amygdala Depend on Orbitofrontal Cortex.

Reciprocal connections between the orbitofrontal cortex (OFC) and the basolateral nucleus of the amygdala (BLA) provide a critical circuit for guiding normal behavior when information about expected outcomes is required. Recently, we reported that outcome signaling by OFC neurons is also necessary for learning in the face of unexpected outcomes during a Pavlovian over-expectation task. Key to learning in this task is the ability to build on prior learning to infer or estimate an amount of reward never previously received. OFC was critical to this process. Notably, in parallel work, we found that BLA was not necessary for learning in this setting. This suggested a dissociation in which the BLA might be critical for acquiring information about the outcomes but not for subsequently using it to make novel predictions. Here we evaluated this hypothesis by recording single-unit activity from BLA in rats during the same Pavlovian over-expectation task used previously. We found that spiking activity recorded in BLA in control rats did reflect novel outcome estimates derived from the integration of prior learning, however consistent with a model in which this process occurs in the OFC, these correlates were entirely abolished by ipsilateral OFC lesions. These data indicate that this information about these novel predictions is represented in the BLA, supported via direct or indirect input from the OFC, even though it does not appear to be necessary for learning. SIGNIFICANCE STATEMENT: The basolateral nucleus of the amygdala (BLA) and the orbitofrontal cortex (OFC) are involved in behavior that depends on knowledge of impending outcomes. Recently, we found that only the OFC was necessary for using such information for learning in a Pavlovian over-expectation task. The current experiment was designed to search for neural correlates of this process in the BLA and, if present, to ask whether they would still be dependent on OFC input. We found that although spiking activity in BLA in control rats did reflect the novel outcome estimates underlying learning, these correlates were entirely abolished by OFC lesions.

Learning theory: a driving force in understanding orbitofrontal function.

Since it was demonstrated the orbitofrontal cortex (OFC) is critical to reversal learning, there has been considerable interest in specifying its role in flexible, outcome-guided behavior. Behavioral paradigms from the learning theory tradition, such as outcome devaluation, blocking, Pavlovian to instrumental transfer, and overexpectation have been a driving force in this research. The use of these procedures has revealed OFC's unique role in forming and integrating information about specific features of events and outcomes to drive behavior and learning. These studies highlight the power and importance of learning theory principles in guiding neuroscience research.

A new perspective on the role of the orbitofrontal cortex in adaptive behaviour.

The orbitofrontal cortex (OFC) is crucial for changing established behaviour in the face of unexpected outcomes. This function has been attributed to the role of the OFC in response inhibition or to the idea that the OFC is a rapidly flexible associative-learning area. However, recent data contradict these accounts, and instead suggest that the OFC is crucial for signalling outcome expectancies. We suggest that this function--signalling of expected outcomes--can also explain the crucial role of the OFC in changing behaviour in the face of unexpected outcomes.

Expectancy-related changes in firing of dopamine neurons depend on hippocampus.

The orbitofrontal cortex (OFC) and hippocampus (HC) are both implicated in forming the cognitive or task maps that support flexible behavior. Previously, we used the dopamine neurons as a sensor or tool to measure the functional effects of OFC lesions (Takahashi et al., 2011). We recorded midbrain dopamine neurons as rats performed an odor-based choice task, in which errors in the prediction of reward were induced by manipulating the number or timing of the expected rewards across blocks of trials. We found that OFC lesions ipsilateral to the recording electrodes caused prediction errors to be degraded consistent with a loss in the resolution of the task states, particularly under conditions where hidden information was critical to sharpening the predictions. Here we have repeated this experiment, along with computational modeling of the results, in rats with ipsilateral HC lesions. The results show HC also shapes the map of our task, however unlike OFC, which provides information local to the trial, the HC appears to be necessary for estimating the upper-level hidden states based on the information that is discontinuous or separated by longer timescales. The results contrast the respective roles of the OFC and HC in cognitive mapping and add to evidence that the dopamine neurons access a rich information set from distributed regions regarding the predictive structure of the environment, potentially enabling this powerful teaching signal to support complex learning and behavior.

Dopaminergic prediction errors in the ventral tegmental area reflect a multithreaded predictive model.

Dopamine neuron activity is tied to the prediction error in temporal difference reinforcement learning models. These models make significant simplifying assumptions, particularly with regard to the structure of the predictions fed into the dopamine neurons, which consist of a single chain of timepoint states. Although this predictive structure can explain error signals observed in many studies, it cannot cope with settings where subjects might infer multiple independent events and outcomes. In the present study, we recorded dopamine neurons in the ventral tegmental area in such a setting to test the validity of the single-stream assumption. Rats were trained in an odor-based choice task, in which the timing and identity of one of several rewards delivered in each trial changed across trial blocks. This design revealed an error signaling pattern that requires the dopamine neurons to access and update multiple independent predictive streams reflecting the subject's belief about timing and potentially unique identities of expected rewards.

Model-based learning and the contribution of the orbitofrontal cortex to the model-free world.

Learning is proposed to occur when there is a discrepancy between reward prediction and reward receipt. At least two separate systems are thought to exist: one in which predictions are proposed to be based on model-free or cached values; and another in which predictions are model-based. A basic neural circuit for model-free reinforcement learning has already been described. In the model-free circuit the ventral striatum (VS) is thought to supply a common-currency reward prediction to midbrain dopamine neurons that compute prediction errors and drive learning. In a model-based system, predictions can include more information about an expected reward, such as its sensory attributes or current, unique value. This detailed prediction allows for both behavioral flexibility and learning driven by changes in sensory features of rewards alone. Recent evidence from animal learning and human imaging suggests that, in addition to model-free information, the VS also signals model-based information. Further, there is evidence that the orbitofrontal cortex (OFC) signals model-based information. Here we review these data and suggest that the OFC provides model-based information to this traditional model-free circuitry and offer possibilities as to how this interaction might occur.

The role of the nucleus accumbens in knowing when to respond.

While knowing what to expect is important, it is equally important to know when to expect it and to respond accordingly. This is apparent even in simple Pavlovian training situations in which animals learn to respond more strongly closer to reward delivery. Here we report that the nucleus accumbens core, an area well-positioned to represent information about the timing of impending rewards, plays a critical role in this timing function.

Inactivation of the central but not the basolateral nucleus of the amygdala disrupts learning in response to overexpectation of reward.

The amygdala is critical for associating predictive cues with primary rewarding and aversive outcomes. This is particularly evident in tasks in which information about expected outcomes is required for normal responding. Here we used a pavlovian overexpectation task to test whether outcome signaling by amygdala might also be necessary for changing those representations in the face of unexpected outcomes. Rats were trained to associate several different cues with a food reward. After learning, two of the cues were presented together, in compound, followed by the same reward. Before each compound training session, rats received infusions of 2,3-dioxo-6-nitro-1,2,3,4-tetrahydrobenzo[f]quinoxaline-7-sulfonamide or saline into either the basolateral (ABL) or central nucleus (CeN) of amygdala. We found that infusions into CeN abolished the normal decline in responding to the compounded cue in a later probe test, whereas infusions into ABL had no effect. These results are inconsistent with the proposal that signaling of information about expected outcomes by ABL contributes to learning, at least in this setting, and instead implicate the CeN in this process, perhaps attributable to the hypothesized involvement of this area in attention and variations in stimulus processing.

A metric for odorant comparison.

In studies of vision and audition, stimuli can be systematically varied by wavelength and frequency, respectively, but there is no equivalent metric for olfaction. Restricted odorant-feature metrics such as number of carbons and functional group do not account for response patterns to odorants varying along other structural dimensions. We generated a multidimensional odor metric, in which each odorant molecule was represented as a vector of 1,664 molecular descriptor values. Revisiting many studies, we found that this metric and a second optimized metric were always better at accounting for neural responses than the specific metric used in each study. These metrics were applicable across studies that differed in the animals studied, the type of olfactory neurons tested, the odorants applied and the recording methods used. We use this new metric to recommend sets of odorants that span the physicochemical space for use in olfaction experiments.

Orbitofrontal inactivation impairs reversal of Pavlovian learning by interfering with 'disinhibition' of responding for previously unrewarded cues.

Orbitofrontal cortex (OFC) is critical for reversal learning. Reversal deficits are typically demonstrated in complex settings that combine Pavlovian and instrumental learning. Yet recent work has implicated the OFC specifically in behaviors guided by cues and the features of the specific outcomes they predict. To test whether the OFC is important for reversing such Pavlovian associations in the absence of confounding instrumental requirements, we trained rats on a simple Pavlovian task in which two auditory cues were presented, one paired with a food pellet reward and the other presented without reward. After learning, we reversed the cue-outcome associations. For half the rats, OFC was inactivated prior to each reversal session. Inactivation of OFC impaired the ability of the rats to reverse conditioned responding. This deficit reflected the inability of inactivated rats to develop normal responding for the previously unrewarded cue; inactivation of OFC had no impact on the ability of the rats to inhibit responding to the previously rewarded cue. These data show that OFC is critical to reversal of Pavlovian responding, and that the role of OFC in this behavior cannot be explained as a simple deficit in response inhibition. Furthermore, the contrast between the normal inhibition of responding, reported here, and impaired inhibition of responding during Pavlovian over-expectation, reported previously, suggests the novel hypothesis that OFC may be particularly critical for learning (or behavior) when it requires the subject to generate predictions about outcomes by bringing together or integrating disparate pieces of associative information.

Expectancy-Related Changes in Dopaminergic Error Signals Are Impaired by Cocaine Self-Administration.

Addiction is a disorder of behavioral control and learning. While this may reflect pre-existing propensities, drug use also clearly contributes by causing changes in outcome processing in prefrontal and striatal regions. This altered processing is associated with behavioral deficits, including changes in learning. These areas provide critical input to midbrain dopamine neurons regarding expected outcomes, suggesting that effects on learning may result from changes in dopaminergic error signaling. Here, we show that dopamine neurons recorded in rats that had self-administered cocaine failed to suppress firing on omission of an expected reward and exhibited lower amplitude and imprecisely timed increases in firing to an unexpected reward. Learning also appeared to have less of an effect on reward-evoked and cue-evoked firing in the cocaine-experienced rats. Overall, the changes are consistent with reduced fidelity of input regarding the expected outcomes, such as their size, timing, and overall value, because of cocaine use.

Dopamine neuron ensembles signal the content of sensory prediction errors.

Dopamine neurons respond to errors in predicting value-neutral sensory information. These data, combined with causal evidence that dopamine transients support sensory-based associative learning, suggest that the dopamine system signals a multidimensional prediction error. Yet such complexity is not evident in the activity of individual neurons or population averages. How then do downstream areas know what to learn in response to these signals? One possibility is that information about content is contained in the pattern of firing across many dopamine neurons. Consistent with this, here we show that the pattern of firing across a small group of dopamine neurons recorded in rats signals the identity of a mis-predicted sensory event. Further, this same information is reflected in the BOLD response elicited by sensory prediction errors in human midbrain. These data provide evidence that ensembles of dopamine neurons provide highly specific teaching signals, opening new possibilities for how this system might contribute to learning.

Effects of inference on dopaminergic prediction errors depend on orbitofrontal processing.

Dopaminergic reward prediction errors in monkeys reflect inferential reward predictions that well-trained animals can make when associative rules change. Here, in a new analysis of previously described data, we test whether dopaminergic error signals in rats are influenced by inferential predictions and whether such effects depend on the orbitofrontal cortex (OFC). Dopamine neurons were recorded from controls or rats with ipsilateral OFC lesions during performance of a choice task in which odor cues signaled the availability of sucrose reward in 2 wells. To induce prediction errors, we manipulated either the timing or number of rewards delivered in each well across blocks of trials. Of importance, a change in reward at 1 well predicted a change in reward at the other on later trials. We compared behavior and neural activity on trials when such inference was possible versus trials involving the same reward change when inference was not possible. Rats responded faster when they could infer an increase in reward compared to when the same reward was coming but they could not infer a change. This inferential prediction was reflected in the firing of dopamine neurons in controls, which changed less to unexpected delivery (or omission) of reward and more to the new high-value cue on inference versus noninference trials. These effects were absent in dopamine neurons recorded in rats with ipsilateral OFC lesions. Thus, dopaminergic error signals recorded in rats are influenced by both experiential and inferential reward predictions, and the effects of inferential predictions depend on OFC. (PsycINFO Database Record

Dopamine Neurons Respond to Errors in the Prediction of Sensory Features of Expected Rewards.

Midbrain dopamine neurons have been proposed to signal prediction errors as defined in model-free reinforcement learning algorithms. While these algorithms have been extremely powerful in interpreting dopamine activity, these models do not register any error unless there is a difference between the value of what is predicted and what is received. Yet learning often occurs in response to changes in the unique features that characterize what is received, sometimes with no change in its value at all. Here, we show that classic error-signaling dopamine neurons also respond to changes in value-neutral sensory features of an expected reward. This suggests that dopamine neurons have access to a wider variety of information than contemplated by the models currently used to interpret their activity and that, while their firing may conform to predictions of these models in some cases, they are not restricted to signaling errors in the prediction of value.

Ventral striatal lesions disrupt dopamine neuron signaling of differences in cue value caused by changes in reward timing but not number.

We recently showed that ventral striatal (VS) lesions abolished reward prediction errors that reflect changes in reward timing while leaving relatively unaffected errors that reflect changes in reward number. Here we extended those results by examining whether normal learning-related changes in firing to the reward-predictive cues in the same dopamine neurons were also disrupted by VS lesions. This analysis revealed that dopamine neurons recorded in VS-lesioned rats failed to show value-related changes in firing to the cues in timing but showed normal changes in number blocks. This effect suggests that the loss of reward-evoked error signals in the timing blocks impaired encoding of the cue value in downstream areas, which then supply these predictions to dopamine neurons at the time of cue presentation. (PsycINFO Database Record

Temporal Specificity of Reward Prediction Errors Signaled by Putative Dopamine Neurons in Rat VTA Depends on Ventral Striatum.

Dopamine neurons signal reward prediction errors. This requires accurate reward predictions. It has been suggested that the ventral striatum provides these predictions. Here we tested this hypothesis by recording from putative dopamine neurons in the VTA of rats performing a task in which prediction errors were induced by shifting reward timing or number. In controls, the neurons exhibited error signals in response to both manipulations. However, dopamine neurons in rats with ipsilateral ventral striatal lesions exhibited errors only to changes in number and failed to respond to changes in timing of reward. These results, supported by computational modeling, indicate that predictions about the temporal specificity and the number of expected reward are dissociable and that dopaminergic prediction-error signals rely on the ventral striatum for the former but not the latter.

Thinking Outside the Box: Orbitofrontal Cortex, Imagination, and How We Can Treat Addiction.

Addiction involves an inability to control drug-seeking behavior. While this may be thought of as secondary to an overwhelming desire for drugs, it could equally well reflect a failure of the brain mechanisms that allow addicts to learn about and mentally simulate non-drug consequences. Importantly, this process of mental simulation draws upon, but is not normally bound by, our past experiences. Rather we have the ability to think outside the box of our past, integrating knowledge gained from a variety of similar and not-so-similar life experiences to derive estimates or imagine what might happen next. These estimates influence our current behavior directly and also affect future behavior by serving as the background against which outcomes are evaluated to support learning. Here we will review evidence, from our own work using a Pavlovian over-expectation task as well as from other sources, that the orbitofrontal cortex is a critical node in the neural circuit that generates these estimates. Further we will offer the specific hypothesis that degradation of this function secondary to drug-induced changes is a critical and likely addressable part of addiction.

Detection and masking of spoiled food smells by odor maps in the olfactory bulb.

Two major causes of spoiled food smells such as fatty, fishy off-flavors are alkylamines liberated by bacterial actions and aliphatic acids-aldehydes generated by lipid oxidation. Using the method of intrinsic signal imaging, we mapped alkylamine-responsive glomeruli to a subregion of the aliphatic acid-responsive and aldehyde-responsive cluster in the odor maps of rat olfactory bulb. Extracellular single-unit recordings from mitral-tufted cells in the subregion showed that individual cells responded to the alkylamines in addition to acids and aldehydes. Responses of mitral-tufted cells tended to last for a long period (5-88 sec) even after the cessation of the alkylamine stimulation. These results suggest that the subregion is part of the representation of the fatty, fishy odor quality. Fennel and clove, spices known to add flavor and mask the fatty, fishy odor, activated glomeruli in the surrounding clusters and suppressed the alkylamine-induced and acid-aldehyde-induced responses of mitral cells, suggesting that the odor masking is mediated, in part, by lateral inhibitory connections in the odor maps of the olfactory bulb.