Multiple Mechanisms for Processing Reward Uncertainty in the Primate Basal Forebrain

Dorsal raphe neurons integrate the values of reward amount, delay, and uncertainty in multi-attribute decision-making

The dorsal raphe nucleus (DRN) is implicated in psychiatric disorders that feature impaired sensitivity to reward amount, impulsivity when facing reward delays, and risk-seeking when confronting reward uncertainty. However, it has been unclear whether and how DRN neurons signal reward amount, reward delay, and reward uncertainty during multi-attribute value-based decision-making, where subjects consider these attributes to make a choice. We recorded DRN neurons as monkeys chose between offers whose attributes, namely expected reward amount, reward delay, and reward uncertainty, varied independently. Many DRN neurons signaled offer attributes, and this population tended to integrate the attributes in a manner that reflected monkeys' preferences for amount, delay, and uncertainty. After decision-making, in response to post-decision feedback, these same neurons signaled signed reward prediction errors, suggesting a broader role in tracking value across task epochs and behavioral contexts. Our data illustrate how the DRN participates in value computations, guiding theories about the role of the DRN in decision-making and psychiatric disease.

Transcriptomic landscape of mammalian ventral pallidum at single-cell resolution

The ventral pallidum (VP) is critical for motivated behaviors. While contemporary work has begun to elucidate the functional diversity of VP neurons, the molecular heterogeneity underlying this functional diversity remains incompletely understood. We used snRNA-seq and in situ hybridization to define the transcriptional taxonomy of VP cell types in mice, macaques, and baboons. We found transcriptional conservation between all three species, within the broader neurochemical cell types. Unique dopaminoceptive and cholinergic subclusters were identified and conserved across both primate species but had no homolog in mice. This harmonized consensus VP cellular atlas will pave the way for understanding the structure and function of the VP and identified key neuropeptides, neurotransmitters, and neuro receptors that could be targeted within specific VP cell types for functional investigations.

An opponent striatal circuit for distributional reinforcement learning

Machine learning research has achieved large performance gains on a wide range of tasks by expanding the learning target from mean rewards to entire probability distributions of rewards - an approach known as distributional reinforcement learning (RL)1. The mesolimbic dopamine system is thought to underlie RL in the mammalian brain by updating a representation of mean value in the striatum2,3, but little is known about whether, where, and how neurons in this circuit encode information about higher-order moments of reward distributions4. To fill this gap, we used high-density probes (Neuropixels) to acutely record striatal activity from well-trained, water-restricted mice performing a classical conditioning task in which reward mean, reward variance, and stimulus identity were independently manipulated. In contrast to traditional RL accounts, we found robust evidence for abstract encoding of variance in the striatum. Remarkably, chronic ablation of dopamine inputs disorganized these distributional representations in the striatum without interfering with mean value coding. Two-photon calcium imaging and optogenetics revealed that the two major classes of striatal medium spiny neurons - D1 and D2 MSNs - contributed to this code by preferentially encoding the right and left tails of the reward distribution, respectively. We synthesize these findings into a new model of the striatum and mesolimbic dopamine that harnesses the opponency between D1 and D2 MSNs5-15 to reap the computational benefits of distributional RL.

A neural mechanism for conserved value computations integrating information and rewards

Behavioral and economic theory dictate that we decide between options based on their values. However, humans and animals eagerly seek information about uncertain future rewards, even when this does not provide any objective value. This implies that decisions are made by endowing information with subjective value and integrating it with the value of extrinsic rewards, but the mechanism is unknown. Here, we show that human and monkey value judgements obey strikingly conserved computational principles during multi-attribute decisions trading off information and extrinsic reward. We then identify a neural substrate in a highly conserved ancient structure, the lateral habenula (LHb). LHb neurons signal subjective value, integrating information's value with extrinsic rewards, and the LHb predicts and causally influences ongoing decisions. Neurons in key input areas to the LHb largely signal components of these computations, not integrated value signals. Thus, our data uncover neural mechanisms of conserved computations underlying decisions to seek information about the future.

Uncertainty modulates visual maps during noninstrumental information demand

Animals are intrinsically motivated to obtain information independently of instrumental incentives. This motivation depends on two factors: a desire to resolve uncertainty by gathering accurate information and a desire to obtain positively-valenced observations, which predict favorable rather than unfavorable outcomes. To understand the neural mechanisms, we recorded parietal cortical activity implicated in prioritizing stimuli for spatial attention and gaze, in a task in which monkeys were free (but not trained) to obtain information about probabilistic non-contingent rewards. We show that valence and uncertainty independently modulated parietal neuronal activity, and uncertainty but not reward-related enhancement consistently correlated with behavioral sensitivity. The findings suggest uncertainty-driven and valence-driven information demand depend on partially distinct pathways, with the former being consistently related to parietal responses and the latter depending on additional mechanisms implemented in downstream structures.

Curiosity is motivated by uncertainty and valence, but how uncertainty and valence are encoded in the brain remains poorly understood. Here, the authors show that parietal neurons are enhanced by both factors, but that they specifically predict visual information seeking based on uncertainty.

Interactions between ventrolateral prefrontal and anterior cingulate cortex during learning and behavioural change

Hypotheses and beliefs guide credit assignment - the process of determining which previous events or actions caused an outcome. Adaptive hypothesis formation and testing are crucial in uncertain and changing environments in which associations and meanings are volatile. Despite primates' abilities to form and test hypotheses, establishing what is causally responsible for the occurrence of particular outcomes remains a fundamental challenge for credit assignment and learning. Hypotheses about what surprises are due to stochasticity inherent in an environment as opposed to real, systematic changes are necessary for identifying the environment's predictive features, but are often hard to test. We review evidence that two highly interconnected frontal cortical regions, anterior cingulate cortex and ventrolateral prefrontal area 47/12o, provide a biological substrate for linking two crucial components of hypothesis-formation and testing: the control of information seeking and credit assignment. Neuroimaging, targeted disruptions, and neurophysiological studies link an anterior cingulate - 47/12o circuit to generation of exploratory behaviour, non-instrumental information seeking, and interpretation of subsequent feedback in the service of credit assignment. Our observations support the idea that information seeking and credit assignment are linked at the level of neural circuits and explain why this circuit is important for ensuring behaviour is flexible and adaptive.

A primate temporal cortex-zona incerta pathway for novelty seeking

Primates interact with the world by exploring visual objects; they seek opportunities to view novel objects even when these have no extrinsic reward value. How the brain controls this novelty seeking is unknown. Here we show that novelty seeking in monkeys is regulated by the zona incerta (ZI). As monkeys made eye movements to familiar objects to trigger an opportunity to view novel objects, many ZI neurons were preferentially activated by predictions of novel objects before the gaze shift. Low-intensity ZI stimulation facilitated gaze shifts, whereas ZI inactivation reduced novelty seeking. ZI-dependent novelty seeking was not regulated by neurons in the lateral habenula or by many dopamine neurons in the substantia nigra, traditionally associated with reward seeking. But the anterior ventral medial temporal cortex, an area important for object vision and memory, was a prominent source of novelty predictions. These data uncover a functional pathway in the primate brain that regulates novelty seeking.

A prefrontal network integrates preferences for advance information about uncertain rewards and punishments

Humans and animals can be strongly motivated to seek information to resolve uncertainty about rewards and punishments. In particular, despite its clinical and societal relevance, very little is known about information seeking about punishments. We show that attitudes toward information about punishments and rewards are distinct and separable at both behavioral and neuronal levels. We demonstrate the existence of prefrontal neuronal populations that anticipate opportunities to gain information in a relatively valence-specific manner, separately anticipating information about either punishments or rewards. These neurons are located in anatomically interconnected subregions of anterior cingulate cortex (ACC) and ventrolateral prefrontal cortex (vlPFC) in area 12o/47. Unlike ACC, vlPFC also contains a population of neurons that integrate attitudes toward both reward and punishment information, to encode the overall preference for information in a bivalent manner. This cortical network is well suited to mediate information seeking by integrating the desire to resolve uncertainty about multiple, distinct motivational outcomes.

Ventral arkypallidal neurons inhibit accumbal firing to promote reward consumption

The nucleus accumbens shell (NAcSh) and the ventral pallidum (VP) are critical for reward processing, although the question of how coordinated activity within these nuclei orchestrates reward valuation and consumption remains unclear. Inhibition of NAcSh firing is necessary for reward consumption, but the source of this inhibition remains unknown. Here, we report that a subpopulation of VP neurons, the ventral arkypallidal (vArky) neurons, project back to the NAcSh, where they inhibit NAcSh neurons in vivo in mice. Consistent with this pathway driving reward consumption via inhibition of the NAcSh, calcium activity of vArky neurons scaled with reward palatability (which was dissociable from reward seeking) and predicted the subsequent drinking behavior during a free-access paradigm. Activation of the VP-NAcSh pathway increased ongoing reward consumption while amplifying hedonic reactions to reward. These results establish a pivotal role for vArky neurons in the promotion of reward consumption through modulation of NAcSh firing in a value-dependent manner.

Anterior Cingulate Cortex and the Control of Dynamic Behavior in Primates

The brain mechanism for controlling continuous behavior in dynamic contexts must mediate action selection and learning across many timescales, responding differentially to the level of environmental uncertainty and volatility. In this review, we argue that a part of the frontal cortex known as the anterior cingulate cortex (ACC) is particularly well suited for this function. First, the ACC is interconnected with prefrontal, parietal, and subcortical regions involved in valuation and action selection. Second, the ACC integrates diverse, behaviorally relevant information across multiple timescales, producing output signals that temporally encapsulate decision and learning processes and encode high-dimensional information about the value and uncertainty of future outcomes and subsequent behaviors. Third, the ACC signals behaviorally relevant information flexibly, displaying the capacity to represent information about current and future states in a valence-, context-, task- and action-specific manner. Fourth, the ACC dynamically controls instrumental- and non-instrumental information seeking behaviors to resolve uncertainty about future outcomes. We review electrophysiological and circuit disruption studies in primates to develop this point, discuss its relationship to novel therapeutics for neuropsychiatric disorders in humans, and conclude by relating ongoing research in primates to studies of medial frontal cortical regions in rodents.

Separable Influences of Reward on Visual Processing and Choice

Primate vision is characterized by constant, sequential processing and selection of visual targets to fixate. Although expected reward is known to influence both processing and selection of visual targets, similarities and differences between these effects remain unclear mainly because they have been measured in separate tasks. Using a novel paradigm, we simultaneously measured the effects of reward outcomes and expected reward on target selection and sensitivity to visual motion in monkeys. Monkeys freely chose between two visual targets and received a juice reward with varying probability for eye movements made to either of them. Targets were stationary apertures of drifting gratings, causing the end points of eye movements to these targets to be systematically biased in the direction of motion. We used this motion-induced bias as a measure of sensitivity to visual motion on each trial. We then performed different analyses to explore effects of objective and subjective reward values on choice and sensitivity to visual motion to find similarities and differences between reward effects on these two processes. Specifically, we used different reinforcement learning models to fit choice behavior and estimate subjective reward values based on the integration of reward outcomes over multiple trials. Moreover, to compare the effects of subjective reward value on choice and sensitivity to motion directly, we considered correlations between each of these variables and integrated reward outcomes on a wide range of timescales. We found that, in addition to choice, sensitivity to visual motion was also influenced by subjective reward value, although the motion was irrelevant for receiving reward. Unlike choice, however, sensitivity to visual motion was not affected by objective measures of reward value. Moreover, choice was determined by the difference in subjective reward values of the two options, whereas sensitivity to motion was influenced by the sum of values. Finally, models that best predicted visual processing and choice used sets of estimated reward values based on different types of reward integration and timescales. Together, our results demonstrate separable influences of reward on visual processing and choice, and point to the presence of multiple brain circuits for the integration of reward outcomes.

Reward uncertainty asymmetrically affects information transmission within the monkey fronto-parietal network

A central hypothesis in research on executive function is that controlled information processing is costly and is allocated according to the behavioral benefits it brings. However, while computational theories predict that the benefits of new information depend on prior uncertainty, the cellular effects of uncertainty on the executive network are incompletely understood. Using simultaneous recordings in monkeys, we describe several mechanisms by which the fronto-parietal network reacts to uncertainty. We show that the variance of expected rewards, independently of the value of the rewards, was encoded in single neuron and population spiking activity and local field potential (LFP) oscillations, and, importantly, asymmetrically affected fronto-parietal information transmission (measured through the coherence between spikes and LFPs). Higher uncertainty selectively enhanced information transmission from the parietal to the frontal lobe and suppressed it in the opposite direction, consistent with Bayesian principles that prioritize sensory information according to a decision maker’s prior uncertainty.

Bahareh Taghizadeh and Nicholas Foley et al. show that individual neuronal responses, population spiking activity, and local field potential oscillations encode the variance of expected rewards independent of their value. They also demonstrate that reward uncertainty asymmetrically affects neuronal transmission within the monkey fronto-parietal network.

How Outcome Uncertainty Mediates Attention, Learning, and Decision-Making

Animals and humans evolved sophisticated nervous systems that endowed them with the ability to form internal-models or beliefs and make predictions about the future to survive and flourish in a world in which future outcomes are often uncertain. Crucial to this capacity is the ability to adjust behavioral and learning policies in response to the level of uncertainty. Until recently, the neuronal mechanisms that could underlie such uncertainty-guided control have been largely unknown. In this review, I discuss newly discovered neuronal circuits in primates that represent uncertainty about future rewards and propose how they guide information-seeking, attention, decision-making, and learning to help us survive in an uncertain world. Lastly, I discuss the possible relevance of these findings to learning in artificial systems.

Human decisions about when to act originate within a basal forebrain-nigral circuit

Decision-making studies often focus on brain mechanisms for selecting between goals and actions; however, another important, and often neglected, aspect of decision-making in humans concerns whether, at any given point in time, it is worth making any action at all. We showed that a considerable portion of the variance in when voluntary actions are emitted can be explained by a simple model that that takes into account key features of the current environment. By using ultrahigh-field MRI we identified a multilayered circuit in the human brain originating far beyond the medial frontal areas typically linked to human voluntary action starting in the basal forebrain and brain stem, converging in the dopaminergic midbrain, and only then projecting to striatum and cortex.

Decisions about when to act are critical for survival in humans as in animals, but how a desire is translated into the decision that an action is worth taking at any particular point in time is incompletely understood. Here we show that a simple model developed to explain when animals decide it is worth taking an action also explains a significant portion of the variance in timing observed when humans take voluntary actions. The model focuses on the current environment’s potential for reward, the timing of the individual’s own recent actions, and the outcomes of those actions. We show, by using ultrahigh-field MRI scanning, that in addition to anterior cingulate cortex within medial frontal cortex, a group of subcortical structures including striatum, substantia nigra, basal forebrain (BF), pedunculopontine nucleus (PPN), and habenula (HB) encode trial-by-trial variation in action time. Further analysis of the activity patterns found in each area together with psychophysiological interaction analysis and structural equation modeling suggested a model in which BF integrates contextual information that will influence the decision about when to act and communicates this information, in parallel with PPN and HB influences, to nigrostriatal circuits. It is then in the nigrostriatal circuit that action initiation per se begins.

A neural network for information seeking

Humans and other animals often show a strong desire to know the uncertain rewards their future has in store, even when they cannot use this information to influence the outcome. However, it is unknown how the brain predicts opportunities to gain information and motivates this information-seeking behavior. Here we show that neurons in a network of interconnected subregions of primate anterior cingulate cortex and basal ganglia predict the moment of gaining information about uncertain rewards. Spontaneous increases in their information prediction signals are followed by gaze shifts toward objects associated with resolving uncertainty, and pharmacologically disrupting this network reduces the motivation to seek information. These findings demonstrate a cortico-basal ganglia mechanism responsible for motivating actions to resolve uncertainty by seeking knowledge about the future.

Animals resolve uncertainty by seeking knowledge about the future. How the brain controls this is unclear. The authors show that a network including primate anterior cingulate cortex and basal ganglia encodes opportunities to gain information about uncertain rewards and mediates information seeking.

Primate prefrontal neurons signal economic risk derived from the statistics of recent reward experience

Risk derives from the variation of rewards and governs economic decisions, yet how the brain calculates risk from the frequency of experienced events, rather than from explicit risk-descriptive cues, remains unclear. Here, we investigated whether neurons in dorsolateral prefrontal cortex process risk derived from reward experience. Monkeys performed in a probabilistic choice task in which the statistical variance of experienced rewards evolved continually. During these choices, prefrontal neurons signaled the reward-variance associated with specific objects ('object risk') or actions ('action risk'). Crucially, risk was not derived from explicit, risk-descriptive cues but calculated internally from the variance of recently experienced rewards. Support-vector-machine decoding demonstrated accurate neuronal risk discrimination. Within trials, neuronal signals transitioned from experienced reward to risk (risk updating) and from risk to upcoming choice (choice computation). Thus, prefrontal neurons encode the statistical variance of recently experienced rewards, complying with formal decision variables of object risk and action risk.

Encoding prediction signals during appetitive and aversive Pavlovian conditioning in the primate lateral hypothalamus

The lateral hypothalamus (LH), which plays a role in homeostatic functions such as appetite regulation, is also linked to arousal and motivational behavior. However, little is known about how these components are encoded in the LH. Thus cynomolgus monkeys were conditioned with two distinct contexts, i.e., an appetitive context with available rewards and an aversive context with predicted air puffs. Different LH neuron groups encoded different degrees of expectation, predictability, and risks of rewards in a specific manner. A nearly equal number of one-third of the recorded LH neurons showed a positive or negative correlation between their response to visual conditioned stimuli (CS) that predicted the probabilistic delivery of rewards (0%, 50%, and 100%) and the associative values. For another one-third of recorded neurons, a nearly equal number showed a positive or negative correlation between their responses to rewards [appetitive unconditioned stimulus (US)] and reward predictability. Some neurons exhibited their highest or lowest trace-period responses in the 50% reward trials. These response modulations were represented independently and overlaid on a consistent excitatory or inhibitory response across the conditioning events. LH neurons also showed consistent responses in the aversive context. However, the responses to aversive conditioning events depending on the air puff value and predictability were less common. The multifaceted modulation of consistent activity related to outcome predictions may reflect motivational and arousal signals. Furthermore, it may underlie the role the LH plays in the integration and relay of signals to cortices for adaptive and goal-directed physiological and behavioral responses to environmental changes.

NEW & NOTEWORTHY The lateral hypothalamus (LH) is implicated in motivational and arousal behavior; however, the detailed information carried by single LH neurons remains unclear. We demonstrate that primate LH neurons encode multiple combinations of signals concerning different degrees of expectation, appreciation, and uncertainty of rewards in consistent responses across conditioning events and between different contexts. This multifaceted modulation of activity may underlie the role of the LH as a critical node integrating motivational signals with arousal signals.

Inactivation of Medial Frontal Cortex Changes Risk Preference

Humans and other animals need to make decisions under varying degrees of uncertainty. These decisions are strongly influenced by an individual's risk preference; however, the neuronal circuitry by which risk preference shapes choice is still unclear [1]. Supplementary eye field (SEF), an oculomotor area within primate medial frontal cortex, is thought to be an essential part of the neuronal circuit underlying oculomotor decision making, including decisions under risk [2-5]. Consistent with this view, risk-related action value and monitoring signals have been observed in SEF [6-8]. However, such activity has also been observed in other frontal areas, including orbitofrontal [9-11], cingulate [12-14], and dorsal-lateral frontal cortex [15]. It is thus unknown whether the activity in SEF causally contributes to risky decisions, or whether it is merely a reflection of neural processes in other cortical regions. Here, we tested a causal role of SEF in risky oculomotor choices. We found that SEF inactivation strongly reduced the frequency of risky choices. This reduction was largely due to a reduced attraction to reward uncertainty and high reward gain, but not due to changes in the subjective estimation of reward probability or average expected reward. Moreover, SEF inactivation also led to increased sensitivity to differences between expected and actual reward during free choice. Nevertheless, it did not affect adjustments of decisions based on reward history.

The Basal Forebrain Regulates Global Resting-State fMRI Fluctuations

Patterns of spontaneous brain activity, typically measured in humans at rest with fMRI, are used routinely to assess the brain's functional organization. The mechanisms that generate and coordinate the underlying neural fluctuations are largely unknown. Here we investigate the hypothesis that the nucleus basalis of Meynert (NBM), the principal source of widespread cholinergic and GABAergic projections to the cortex, contributes critically to such activity. We reversibly inactivated two distinct sites of the NBM in macaques while measuring fMRI activity across the brain. We found that inactivation led to strong, regionalized suppression of shared or "global" signal components of cortical fluctuations ipsilateral to the injection. At the same time, the commonly studied resting-state networks retained their spatial structure under this suppression. The results indicate that the NBM contributes selectively to the global component of functional connectivity but plays little if any role in the specific correlations that define resting-state networks.

Anterior cingulate is a source of valence-specific information about value and uncertainty

Anterior cingulate cortex (ACC) is thought to control a wide range of reward, punishment, and uncertainty-related behaviors. However, how it does so is unclear. Here, in a Pavlovian procedure in which monkeys displayed a diverse repertoire of reward-related, punishment-related, and uncertainty-related behaviors, we show that many ACC-neurons represent expected value and uncertainty in a valence-specific manner, signaling value or uncertainty predictions about either rewards or punishments. Other ACC-neurons signal prediction information about rewards and punishments by displaying excitation to both (rather than excitation to one and inhibition to the other). This diversity in valence representations may support the role of ACC in many behavioral states that are either enhanced by reward and punishment (e.g., vigilance) or specific to either reward or punishment (e.g., approach and avoidance). Also, this first demonstration of punishment-uncertainty signals in the brain suggests that ACC could be a target for the treatment of uncertainty-related disorders of mood.

Rewards or punishments elicit diverse behavioral responses; however, the neural circuits underlying such flexibility are unclear. Here Monosov shows that this diversity could be supported by neurons in the anterior cingulate that represent expected value and uncertainty in a valence-specific manner.

The Input-Output Relationship of the Cholinergic Basal Forebrain

Basal forebrain cholinergic neurons influence cortical state, plasticity, learning, and attention. They collectively innervate the entire cerebral cortex, differentially controlling acetylcholine efflux across different cortical areas and timescales. Such control might be achieved by differential inputs driving separable cholinergic outputs, although no input-output relationship on a brain-wide level has ever been demonstrated. Here, we identify input neurons to cholinergic cells projecting to specific cortical regions by infecting cholinergic axon terminals with a monosynaptically restricted viral tracer. This approach revealed several circuit motifs, such as central amygdala neurons synapsing onto basolateral amygdala-projecting cholinergic neurons or strong somatosensory cortical input to motor cortex-projecting cholinergic neurons. The presence of input cells in the parasympathetic midbrain nuclei contacting frontally projecting cholinergic neurons suggest that the network regulating the inner eye muscles are additionally regulating cortical state via acetylcholine efflux. This dataset enables future circuit-level experiments to identify drivers of known cortical cholinergic functions.

Neurons in the primate dorsal striatum signal the uncertainty of object-reward associations

To learn, obtain reward and survive, humans and other animals must monitor, approach and act on objects that are associated with variable or unknown rewards. However, the neuronal mechanisms that mediate behaviours aimed at uncertain objects are poorly understood. Here we demonstrate that a set of neurons in an internal-capsule bordering regions of the primate dorsal striatum, within the putamen and caudate nucleus, signal the uncertainty of object–reward associations. Their uncertainty responses depend on the presence of objects associated with reward uncertainty and evolve rapidly as monkeys learn novel object–reward associations. Therefore, beyond its established role in mediating actions aimed at known or certain rewards, the dorsal striatum also participates in behaviours aimed at reward-uncertain objects.

The dorsal striatum (DS) is a brain region that is thought to aim actions at certain or known rewards. Here, the authors show that an internal-capsule bordering region of the primate DS signals the uncertainty of object-reward associations, suggesting a novel role for the DS in behavior under uncertainty.