Rule encoding in dorsal striatum impacts action selection

Orbitofrontal cortex control of striatum leads economic decision-making

Animals must continually evaluate stimuli in their environment to decide which opportunities to pursue, and in many cases these decisions can be understood in fundamentally economic terms. Although several brain regions have been individually implicated in these processes, the brain-wide mechanisms relating these regions in decision-making are unclear. Using an economic decision-making task adapted for rats, we find that neural activity in both of two connected brain regions, the ventrolateral orbitofrontal cortex (OFC) and the dorsomedial striatum (DMS), was required for economic decision-making. Relevant neural activity in both brain regions was strikingly similar, dominated by the spatial features of the decision-making process. However, the neural encoding of choice direction in OFC preceded that of DMS, and this temporal relationship was strongly correlated with choice accuracy. Furthermore, activity specifically in the OFC projection to the DMS was required for appropriate economic decision-making. These results demonstrate that choice information in the OFC is relayed to the DMS to lead accurate economic decision-making.

Distributed cell assemblies spanning prefrontal cortex and striatum

Highly synchronous neuronal assembly activity is deemed essential for cognitive brain function. In theory, such synchrony could coordinate multiple brain areas performing complementary processes. However, cell assemblies have been observed only in single structures, typically cortical areas, and little is known about their synchrony with downstream subcortical structures, such as the striatum. Here, we demonstrate distributed cell assemblies activated at high synchrony (∼10 ms) spanning prefrontal cortex and striatum. In addition to including neurons at different brain hierarchical levels, surprisingly, they synchronized functionally distinct limbic and associative sub-regions. These assembly activations occurred when members shifted their firing phase relative to ongoing 4 Hz and theta rhythms, in association with high gamma oscillations. This suggests that these rhythms could mediate the emergence of cross-structural assemblies. To test for the role of assemblies in behavior, we trained the rats to perform a task requiring cognitive flexibility, alternating between two different rules in a T-maze. Overall, assembly activations were correlated with task-relevant parameters, including impending choice, reward, rule, or rule order. Moreover, these behavioral correlates were more robustly expressed by assemblies than by their individual member neurons. Finally, to verify whether assemblies can be endogenously generated, we found that they were indeed spontaneously reactivated during sleep and quiet immobility. Thus, cell assemblies are a more general coding mechanism than previously envisioned, linking distributed neocortical and subcortical areas at high synchrony.

Experience-related enhancements in striatal temporal encoding

Temporal control of action is key for a broad range of behaviors and is disrupted in human diseases such as Parkinson's disease and schizophrenia. A brain structure that is critical for temporal control is the dorsal striatum. Experience and learning can influence dorsal striatal neuronal activity, but it is unknown how these neurons change with experience in contexts which require precise temporal control of movement. We investigated this question by recording from medium spiny neurons (MSNs) via dorsal striatal microelectrode arrays in mice as they gained experience controlling their actions in time. We leveraged an interval timing task optimized for mice which required them to "switch" response ports after enough time had passed without receiving a reward. We report three main results. First, we found that time-related ramping activity and response-related activity increased with task experience. Second, temporal decoding by MSN ensembles improved with experience and was predominantly driven by time-related ramping activity. Finally, we found that a subset of MSNs had differential modulation on error trials. These findings enhance our understanding of dorsal striatal temporal processing by demonstrating how MSN ensembles can evolve with experience. Our results can be linked to temporal habituation and illuminate striatal flexibility during interval timing, which may be relevant for human disease.

Rules warp feature encoding in decision-making circuits

We have the capacity to follow arbitrary stimulus-response rules, meaning simple policies that guide our behavior. Rule identity is broadly encoded across decision-making circuits, but there are less data on how rules shape the computations that lead to choices. One idea is that rules could simplify these computations. When we follow a rule, there is no need to encode or compute information that is irrelevant to the current rule, which could reduce the metabolic or energetic demands of decision-making. However, it is not clear if the brain can actually take advantage of this computational simplicity. To test this idea, we recorded from neurons in 3 regions linked to decision-making, the orbitofrontal cortex (OFC), ventral striatum (VS), and dorsal striatum (DS), while macaques performed a rule-based decision-making task. Rule-based decisions were identified via modeling rules as the latent causes of decisions. This left us with a set of physically identical choices that maximized reward and information, but could not be explained by simple stimulus-response rules. Contrasting rule-based choices with these residual choices revealed that following rules (1) decreased the energetic cost of decision-making; and (2) expanded rule-relevant coding dimensions and compressed rule-irrelevant ones. Together, these results suggest that we use rules, in part, because they reduce the costs of decision-making through a distributed representational warping in decision-making circuits.

Selective Role of the Putamen in Serial Reversal Learning in the Marmoset

Fronto-striatal circuitry involving the orbitofrontal cortex has been identified as mediating successful reversal of stimulus-outcome contingencies. The region of the striatum that most contributes to reversal learning remains unclear, with studies in primates implicating both caudate nucleus and putamen. We trained four marmosets on a touchscreen-based serial reversal task and implanted each with cannulae targeting both putamen and caudate bilaterally. This allowed reversible inactivation of the two areas within the same monkeys, but across separate sessions, to directly investigate their respective contributions to reversal performance. Behavioral sensitivity to the GABAA agonist muscimol varied across subjects and between brain regions, so each marmoset received a range of doses. Intermediate doses of intra-putamen muscimol selectively impaired reversal performance, leaving the baseline discrimination phase unchanged. There was no effect of low doses and high doses were generally disruptive. By contrast, low doses of intra-caudate muscimol improved reversal performance, while high doses impaired both reversal and baseline discrimination performance. These data provide evidence for a specific role of the putamen in serial reversal learning, which may reflect the more habitual nature of repeated reversals using the same stimulus pair.

Animal Learning in a Multidimensional Discrimination Task as Explained by Dimension-Specific Allocation of Attention

Reinforcement learning describes the process by which during a series of trial-and-error attempts, actions that culminate in reward are reinforced, becoming more likely to be chosen in similar circumstances. When decisions are based on sensory stimuli, an association is formed between the stimulus, the action and the reward. Computational, behavioral and neurobiological accounts of this process successfully explain simple learning of stimuli that differ in one aspect, or along a single stimulus dimension. However, when stimuli may vary across several dimensions, identifying which features are relevant for the reward is not trivial, and the underlying cognitive process is poorly understood. To study this we adapted an intra-dimensional/ extra-dimensional set-shifting paradigm to train rats on a multi-sensory discrimination task. In our setup, stimuli of different modalities (spatial, olfactory and visual) are combined into complex cues and manipulated independently. In each set, only a single stimulus dimension is relevant for reward. To distinguish between learning and decision-making we suggest a weighted attention model (WAM). Our model learns by assigning a separate learning rule for the values of features of each dimension (e.g., for each color), reinforced after every experience. Decisions are made by comparing weighted averages of the learnt values, factored by dimension specific weights. Based on the observed behavior of the rats we estimated the parameters of the WAM and demonstrated that it outperforms an alternative model, in which a learnt value is assigned to each combination of features. Estimated decision weights of the WAM reveal an experience-based bias in learning. In the first experimental set the weights associated with all dimensions were similar. The extra-dimensional shift rendered this dimension irrelevant. However, its decision weight remained high for the early learning stage in this last set, providing an explanation for the poor performance of the animals. Thus, estimated weights can be viewed as a possible way to quantify the experience-based bias.

Using a Simple Neural Network to Delineate Some Principles of Distributed Economic Choice

The brain uses a mixture of distributed and modular organization to perform computations and generate appropriate actions. While the principles under which the brain might perform computations using modular systems have been more amenable to modeling, the principles by which the brain might make choices using distributed principles have not been explored. Our goal in this perspective is to delineate some of those distributed principles using a neural network method and use its results as a lens through which to reconsider some previously published neurophysiological data. To allow for direct comparison with our own data, we trained the neural network to perform binary risky choices. We find that value correlates are ubiquitous and are always accompanied by non-value information, including spatial information (i.e., no pure value signals). Evaluation, comparison, and selection were not distinct processes; indeed, value signals even in the earliest stages contributed directly, albeit weakly, to action selection. There was no place, other than at the level of action selection, at which dimensions were fully integrated. No units were specialized for specific offers; rather, all units encoded the values of both offers in an anti-correlated format, thus contributing to comparison. Individual network layers corresponded to stages in a continuous rotation from input to output space rather than to functionally distinct modules. While our network is likely to not be a direct reflection of brain processes, we propose that these principles should serve as hypotheses to be tested and evaluated for future studies.

Rule Encoding in Orbitofrontal Cortex and Striatum Guides Selection

Active maintenance of rules, like other executive functions, is often thought to be the domain of a discrete executive system. An alternative view is that rule maintenance is a broadly distributed function relying on widespread cortical and subcortical circuits. Tentative evidence supporting this view comes from research showing some rule selectivity in the orbitofrontal cortex and dorsal striatum. We recorded in these regions and in the ventral striatum, which has not been associated previously with rule representation, as macaques performed a Wisconsin Card Sorting Task. We found robust encoding of rule category (color vs shape) and rule identity (six possible rules) in all three regions. Rule identity modulated responses to potential choice targets, suggesting that rule information guides behavior by highlighting choice targets. The effects that we observed were not explained by differences in behavioral performance across rules and thus cannot be attributed to reward expectation. Our results suggest that rule maintenance and rule-guided selection of options are distributed processes and provide new insight into orbital and striatal contributions to executive control.

SIGNIFICANCE STATEMENT

Rule maintenance, an important executive function, is generally thought to rely on dorsolateral brain regions. In this study, we examined activity of single neurons in orbitofrontal cortex and in ventral and dorsal striatum of macaques in a Wisconsin Card Sorting Task. Neurons in all three areas encoded rules and rule categories robustly. Rule identity also affected neural responses to potential choice options, suggesting that stored information is used to influence decisions. These results endorse the hypothesis that rule maintenance is a broadly distributed mental operation.

New Variations for Strategy Set-shifting in the Rat

Behavioral flexibility is crucial for survival in changing environments. Broadly defined, behavioral flexibility requires a shift of behavioral strategy based on a change in governing rules. We describe a strategy set-shifting task that requires an attentional shift from one stimulus dimension to another. The paradigm is often used for testing cognitive flexibility in primates. However, the rodent version has not been as extensively developed. We have recently extended an established set-shifting task in the rat¹ by requiring attention to different stimuli according to context. All the experimental conditions required animals to choose either a left or right lever. Initially, all animals had to choose on the basis of the location of the lever. Subsequently, a change in the rule occurred, which required a shift in set from location-based rule to a rule in which the correct lever was indicated by a light cue. We compared performance on three different versions of the task, in which the light stimulus was either novel, previously relevant, or previously irrelevant. We found that specific neurochemical lesions selectively impaired the ability to make particular types of set shift as measured by the performance on the different versions of the task.

Chronic methamphetamine self-administration disrupts cortical control of cognition

Methamphetamine (meth) is one of the most abused substances worldwide. Chronic use has been associated with repeated relapse episodes that may be exacerbated by cognitive impairments during drug abstinence. Growing evidence demonstrates that meth compromises prefrontal cortex activity, resulting in persisting attentional and memory impairments. After summarizing recent studies of meth-induced cognitive dysfunction using a translationally relevant model of self-administered meth, this review emphasizes the cortical brain changes contributing to cognitive dysregulation during abstinence. Finally, we propose the use of cognitive enhancers during abstinence that may promote a drug-free state by reversing cortical dysfunction linked with prolonged meth abuse.

Neurophysiology of rule switching in the corticostriatal circuit

The ability to adjust behavioral responses to cues in a changing environment is crucial for survival. Activity in the medial Prefrontal Cortex (mPFC) is thought to both represent rules to guide behavior as well as detect and resolve conflicts between rules in changing contingencies. While lesion and pharmacological studies have supported a crucial role for mPFC in this type of set-shifting, an understanding of how mPFC represents current rules or detects and resolves conflict between different rules is still unclear. Meanwhile, medial dorsal striatum (mDS) receives major projections from mPFC and neural activity of mDS is closely linked to action selection, making the mDS a potential major player for enacting rule-guided action policies. However, exactly what is signaled by mPFC and how this impacts neural signals in mDS is not well known. In this review, we will summarize what is known about neural signals of rules and set shifting in both prefrontal cortex and dorsal striatum, as well as provide questions and directions for future experiments.

Neural correlates of rules and conflict in medial prefrontal cortex during decision and feedback epochs

The ability to properly adjust behavioral responses to cues in a changing environment is crucial for survival. Activity in the medial Prefrontal Cortex (mPFC) is thought to both represent rules to guide behavior as well as detect and resolve conflicts between rules in changing contingencies. However, while lesion and pharmacological studies have supported a crucial role for mPFC in this type of set-shifting, an understanding of how mPFC represents current rules or detects and resolves conflict between different rules is unclear. Here, we directly address the role of rat mPFC in shifting rule based behavioral strategies using a novel behavioral task designed to tease apart neural signatures of rules, conflict and direction. We demonstrate that activity of single neurons in rat mPFC represent distinct rules. Further, we show increased firing on high conflict trials in a separate population of mPFC neurons. Reduced firing in both populations of neurons was associated with poor performance. Moreover, activity in both populations increased and decreased firing during the outcome epoch when reward was and was not delivered on correct and incorrect trials, respectively. In addition, outcome firing was modulated by the current rule and the degree of conflict associated with the previous decision. These results promote a greater understanding of the role that mPFC plays in switching between rules, signaling both rule and conflict to promote improved behavioral performance.