Supporting generalization in non-human primate behavior by tapping into structural knowledge: Examples from sensorimotor mappings, inference, and decision-making

Inductive biases of neural network modularity in spatial navigation

The brain may have evolved a modular architecture for daily tasks, with circuits featuring functionally specialized modules that match the task structure. We hypothesize that this architecture enables better learning and generalization than architectures with less specialized modules. To test this, we trained reinforcement learning agents with various neural architectures on a naturalistic navigation task. We found that the modular agent, with an architecture that segregates computations of state representation, value, and action into specialized modules, achieved better learning and generalization. Its learned state representation combines prediction and observation, weighted by their relative uncertainty, akin to recursive Bayesian estimation. This agent's behavior also resembles macaques' behavior more closely. Our results shed light on the possible rationale for the brain's modularity and suggest that artificial systems can use this insight from neuroscience to improve learning and generalization in natural tasks.

Context-invariant beliefs are supported by dynamic reconfiguration of single unit functional connectivity in prefrontal cortex of male macaques

Natural behaviors occur in closed action-perception loops and are supported by dynamic and flexible beliefs abstracted away from our immediate sensory milieu. How this real-world flexibility is instantiated in neural circuits remains unknown. Here, we have male macaques navigate in a virtual environment by primarily leveraging sensory (optic flow) signals, or by more heavily relying on acquired internal models. We record single-unit spiking activity simultaneously from the dorsomedial superior temporal area (MSTd), parietal area 7a, and the dorso-lateral prefrontal cortex (dlPFC). Results show that while animals were able to maintain adaptive task-relevant beliefs regardless of sensory context, the fine-grain statistical dependencies between neurons, particularly in 7a and dlPFC, dynamically remapped with the changing computational demands. In dlPFC, but not 7a, destroying these statistical dependencies abolished the area's ability for cross-context decoding. Lastly, correlational analyses suggested that the more unit-to-unit couplings remapped in dlPFC, and the less they did so in MSTd, the less were population codes and behavior impacted by the loss of sensory evidence. We conclude that dynamic functional connectivity between neurons in prefrontal cortex maintain a stable population code and context-invariant beliefs during naturalistic behavior.

From innate to instructed: A new look at perceptual decision-making

Understanding how subjects perceive sensory stimuli in their environment and use this information to guide appropriate actions is a major challenge in neuroscience. To study perceptual decision-making in animals, researchers use tasks that either probe spontaneous responses to stimuli (often described as "naturalistic") or train animals to associate stimuli with experimenter-defined responses. Spontaneous decisions rely on animals' pre-existing knowledge, while trained tasks offer greater versatility, albeit often at the cost of extensive training. Here, we review emerging approaches to investigate perceptual decision-making using both spontaneous and trained behaviors, highlighting their strengths and limitations. Additionally, we propose how trained decision-making tasks could be improved to achieve faster learning and a more generalizable understanding of task rules.

Hierarchical control over foraging behavior by anterior cingulate cortex

Foraging is a natural behavior that involves making sequential decisions to maximize rewards while minimizing the costs incurred when doing so. The prevalence of foraging across species suggests that a common brain computation underlies its implementation. Although anterior cingulate cortex is believed to contribute to foraging behavior, its specific role has been contentious, with predominant theories arguing either that it encodes environmental value or choice difficulty. Additionally, recent attempts to characterize foraging have taken place within the reinforcement learning framework, with increasingly complex models scaling with task complexity. Here we review reinforcement learning foraging models, highlighting the hierarchical structure of many foraging problems. We extend this literature by proposing that ACC guides foraging according to principles of model-based hierarchical reinforcement learning. This idea holds that ACC function is organized hierarchically along a rostral-caudal gradient, with rostral structures monitoring the status and completion of high-level task goals (like finding food), and midcingulate structures overseeing the execution of task options (subgoals, like harvesting fruit) and lower-level actions (such as grabbing an apple).

Self-motion perception and sequential decision-making: where are we heading?

To navigate and guide adaptive behaviour in a dynamic environment, animals must accurately estimate their own motion relative to the external world. This is a fundamentally multisensory process involving integration of visual, vestibular and kinesthetic inputs. Ideal observer models, paired with careful neurophysiological investigation, helped to reveal how visual and vestibular signals are combined to support perception of linear self-motion direction, or heading. Recent work has extended these findings by emphasizing the dimension of time, both with regard to stimulus dynamics and the trade-off between speed and accuracy. Both time and certainty-i.e. the degree of confidence in a multisensory decision-are essential to the ecological goals of the system: terminating a decision process is necessary for timely action, and predicting one's accuracy is critical for making multiple decisions in a sequence, as in navigation. Here, we summarize a leading model for multisensory decision-making, then show how the model can be extended to study confidence in heading discrimination. Lastly, we preview ongoing efforts to bridge self-motion perception and navigation per se, including closed-loop virtual reality and active self-motion. The design of unconstrained, ethologically inspired tasks, accompanied by large-scale neural recordings, raise promise for a deeper understanding of spatial perception and decision-making in the behaving animal. This article is part of the theme issue 'Decision and control processes in multisensory perception'.

Causal inference during closed-loop navigation: parsing of self- and object-motion

A key computation in building adaptive internal models of the external world is to ascribe sensory signals to their likely cause(s), a process of causal inference (CI). CI is well studied within the framework of two-alternative forced-choice tasks, but less well understood within the cadre of naturalistic action-perception loops. Here, we examine the process of disambiguating retinal motion caused by self- and/or object-motion during closed-loop navigation. First, we derive a normative account specifying how observers ought to intercept hidden and moving targets given their belief about (i) whether retinal motion was caused by the target moving, and (ii) if so, with what velocity. Next, in line with the modelling results, we show that humans report targets as stationary and steer towards their initial rather than final position more often when they are themselves moving, suggesting a putative misattribution of object-motion to the self. Further, we predict that observers should misattribute retinal motion more often: (i) during passive rather than active self-motion (given the lack of an efference copy informing self-motion estimates in the former), and (ii) when targets are presented eccentrically rather than centrally (given that lateral self-motion flow vectors are larger at eccentric locations during forward self-motion). Results support both of these predictions. Lastly, analysis of eye movements show that, while initial saccades toward targets were largely accurate regardless of the self-motion condition, subsequent gaze pursuit was modulated by target velocity during object-only motion, but not during concurrent object- and self-motion. These results demonstrate CI within action-perception loops, and suggest a protracted temporal unfolding of the computations characterizing CI. This article is part of the theme issue 'Decision and control processes in multisensory perception'.

Context-invariant beliefs are supported by dynamic reconfiguration of single unit functional connectivity in prefrontal cortex

Natural behaviors occur in closed action-perception loops and are supported by dynamic and flexible beliefs abstracted away from our immediate sensory milieu. How this real-world flexibility is instantiated in neural circuits remains unknown. Here we have macaques navigate in a virtual environment by primarily leveraging sensory (optic flow) signals, or by more heavily relying on acquired internal models. We record single-unit spiking activity simultaneously from the dorsomedial superior temporal area (MSTd), parietal area 7a, and the dorso-lateral prefrontal cortex (dlPFC). Results show that while animals were able to maintain adaptive task-relevant beliefs regardless of sensory context, the fine-grain statistical dependencies between neurons, particularly in 7a and dlPFC, dynamically remapped with the changing computational demands. In dlPFC, but not 7a, destroying these statistical dependencies abolished the area's ability for cross-context decoding. Lastly, correlation analyses suggested that the more unit-to-unit couplings remapped in dlPFC, and the less they did so in MSTd, the less were population codes and behavior impacted by the loss of sensory evidence. We conclude that dynamic functional connectivity between prefrontal cortex neurons maintains a stable population code and context-invariant beliefs during naturalistic behavior with closed action-perception loops.

Causal inference during closed-loop navigation: parsing of self- and object-motion

A key computation in building adaptive internal models of the external world is to ascribe sensory signals to their likely cause(s), a process of Bayesian Causal Inference (CI). CI is well studied within the framework of two-alternative forced-choice tasks, but less well understood within the cadre of naturalistic action-perception loops. Here, we examine the process of disambiguating retinal motion caused by self- and/or object-motion during closed-loop navigation. First, we derive a normative account specifying how observers ought to intercept hidden and moving targets given their belief over (i) whether retinal motion was caused by the target moving, and (ii) if so, with what velocity. Next, in line with the modeling results, we show that humans report targets as stationary and steer toward their initial rather than final position more often when they are themselves moving, suggesting a misattribution of object-motion to the self. Further, we predict that observers should misattribute retinal motion more often: (i) during passive rather than active self-motion (given the lack of an efference copy informing self-motion estimates in the former), and (ii) when targets are presented eccentrically rather than centrally (given that lateral self-motion flow vectors are larger at eccentric locations during forward self-motion). Results confirm both of these predictions. Lastly, analysis of eye-movements show that, while initial saccades toward targets are largely accurate regardless of the self-motion condition, subsequent gaze pursuit was modulated by target velocity during object-only motion, but not during concurrent object- and self-motion. These results demonstrate CI within action-perception loops, and suggest a protracted temporal unfolding of the computations characterizing CI.

Coding of latent variables in sensory, parietal, and frontal cortices during closed-loop virtual navigation

We do not understand how neural nodes operate and coordinate within the recurrent action-perception loops that characterize naturalistic self-environment interactions. Here, we record single-unit spiking activity and local field potentials (LFPs) simultaneously from the dorsomedial superior temporal area (MSTd), parietal area 7a, and dorsolateral prefrontal cortex (dlPFC) as monkeys navigate in virtual reality to 'catch fireflies'. This task requires animals to actively sample from a closed-loop virtual environment while concurrently computing continuous latent variables: (i) the distance and angle travelled (i.e., path integration) and (ii) the distance and angle to a memorized firefly location (i.e., a hidden spatial goal). We observed a patterned mixed selectivity, with the prefrontal cortex most prominently coding for latent variables, parietal cortex coding for sensorimotor variables, and MSTd most often coding for eye movements. However, even the traditionally considered sensory area (i.e., MSTd) tracked latent variables, demonstrating path integration and vector coding of hidden spatial goals. Further, global encoding profiles and unit-to-unit coupling (i.e., noise correlations) suggested a functional subnetwork composed by MSTd and dlPFC, and not between these and 7a, as anatomy would suggest. We show that the greater the unit-to-unit coupling between MSTd and dlPFC, the more the animals' gaze position was indicative of the ongoing location of the hidden spatial goal. We suggest this MSTd-dlPFC subnetwork reflects the monkeys' natural and adaptive task strategy wherein they continuously gaze toward the location of the (invisible) target. Together, these results highlight the distributed nature of neural coding during closed action-perception loops and suggest that fine-grain functional subnetworks may be dynamically established to subserve (embodied) task strategies.

Sensory Evidence Accumulation Using Optic Flow in a Naturalistic Navigation Task

Sensory evidence accumulation is considered a hallmark of decision-making in noisy environments. Integration of sensory inputs has been traditionally studied using passive stimuli, segregating perception from action. Lessons learned from this approach, however, may not generalize to ethological behaviors like navigation, where there is an active interplay between perception and action. We designed a sensory-based sequential decision task in virtual reality in which humans and monkeys navigated to a memorized location by integrating optic flow generated by their own joystick movements. A major challenge in such closed-loop tasks is that subjects' actions will determine future sensory input, causing ambiguity about whether they rely on sensory input rather than expectations based solely on a learned model of the dynamics. To test whether subjects integrated optic flow over time, we used three independent experimental manipulations, unpredictable optic flow perturbations, which pushed subjects off their trajectory; gain manipulation of the joystick controller, which changed the consequences of actions; and manipulation of the optic flow density, which changed the information borne by sensory evidence. Our results suggest that both macaques (male) and humans (female/male) relied heavily on optic flow, thereby demonstrating a critical role for sensory evidence accumulation during naturalistic action-perception closed-loop tasks.SIGNIFICANCE STATEMENT The temporal integration of evidence is a fundamental component of mammalian intelligence. Yet, it has traditionally been studied using experimental paradigms that fail to capture the closed-loop interaction between actions and sensations inherent in real-world continuous behaviors. These conventional paradigms use binary decision tasks and passive stimuli with statistics that remain stationary over time. Instead, we developed a naturalistic visuomotor visual navigation paradigm that mimics the causal structure of real-world sensorimotor interactions and probed the extent to which participants integrate sensory evidence by adding task manipulations that reveal complementary aspects of the computation.

Alternative female and male developmental trajectories in the dynamic balance of human visual perception

The numerous multistable phenomena in vision, hearing and touch attest that the inner workings of perception are prone to instability. We investigated a visual example-binocular rivalry-with an accurate no-report paradigm, and uncovered developmental and maturational lifespan trajectories that were specific for age and sex. To interpret these trajectories, we hypothesized that conflicting objectives of visual perception-such as stability of appearance, sensitivity to visual detail, and exploration of fundamental alternatives-change in relative importance over the lifespan. Computational modelling of our empirical results allowed us to estimate this putative development of stability, sensitivity, and exploration over the lifespan. Our results confirmed prior findings of developmental psychology and appear to quantify important aspects of neurocognitive phenotype. Additionally, we report atypical function of binocular rivalry in autism spectrum disorder and borderline personality disorder. Our computational approach offers new ways of quantifying neurocognitive phenotypes both in development and in dysfunction.

Cognitive, Systems, and Computational Neurosciences of the Self in Motion

Navigating by path integration requires continuously estimating one's self-motion. This estimate may be derived from visual velocity and/or vestibular acceleration signals. Importantly, these senses in isolation are ill-equipped to provide accurate estimates, and thus visuo-vestibular integration is an imperative. After a summary of the visual and vestibular pathways involved, the crux of this review focuses on the human and theoretical approaches that have outlined a normative account of cue combination in behavior and neurons, as well as on the systems neuroscience efforts that are searching for its neural implementation. We then highlight a contemporary frontier in our state of knowledge: understanding how velocity cues with time-varying reliabilities are integrated into an evolving position estimate over prolonged time periods. Further, we discuss how the brain builds internal models inferring when cues ought to be integrated versus segregated-a process of causal inference. Lastly, we suggest that the study of spatial navigation has not yet addressed its initial condition: self-location.

Reducing behavioral dimensions to study brain-environment interactions

Movement vigor provides a window on action valuation. But what is vigor, and how to measure it in the first place? Strikingly, many different co-varying vigor-related metrics can be found in the literature. I believe this is because vigor, just like the neural circuits that determine it, is an integrated, low-dimensional parameter. As such, it can only be roughly estimated.