Search for: All records

Perception is fallible. Humans know this, and so do some nonhuman animals like macaque monkeys. When monkeys report more confidence in a perceptual decision, that decision is more likely to be correct. It is not known how neural circuits in the primate brain assess the quality of perceptual decisions. Here, we test two hypotheses. First, that decision confidence is related to the structure of population activity in the sensory cortex. And second, that this relation differs from the one between sensory activity and decision content. We trained macaque monkeys to judge the orientation of ambiguous stimuli and additionally report their confidence in these judgments. We recorded population activity in the primary visual cortex and used decoders to expose the relationship between this activity and the choice-confidence reports. Our analysis validated both hypotheses and suggests that perceptual decisions arise from a neural computation downstream of visual cortex that estimates the most likely interpretation of a sensory response, while decision confidence instead reflects a computation that evaluates whether this sensory response will produce a reliable decision. Our work establishes a direct link between neural population activity in the sensory cortex and the metacognitive ability to introspect about the quality of perceptual decisions. 

Perceptual judgments of the environment emerge from the concerted activity of neural populations in decision-making areas downstream of the sensory cortex. When the sensory input is ambiguous, perceptual judgments can be biased by prior expectations shaped by environmental regularities. These effects are examples of Bayesian inference, a reasoning method in which prior knowledge is leveraged to optimize uncertain decisions. However, it is not known how decision-making circuits combine sensory signals and prior expectations to form a perceptual decision. Here, we study neural population activity in the prefrontal cortex of macaque monkeys trained to report perceptual judgments of ambiguous visual stimuli under two different stimulus distributions. We isolate the component of the neural population response that represents the formation of the perceptual decision (the decision variable, DV), and find that its dynamical evolution reflects the integration of sensory signals and prior expectations. Prior expectations impact the DV’s trajectory both before and during stimulus presentation such that DV trajectories with a smaller dynamic range result in more biased and less sensitive perceptual decisions. We show that these results resemble a specific variant of Bayesian inference known as approximate hierarchical inference. Our findings expand our understanding of the mechanisms by which prefrontal circuits can execute Bayesian inference. 

Sub-additivity and variability are ubiquitous response motifs in primary visual cortex (V1). Response sub-additivity provides a sign of the brain processes that enable us to construct useful interpretations of the visual environment (i.e., nonlinear input transformations), while response variability provides a sign of the brain processes that limit the precision with which we can do this (i.e., neural information loss). Historically, these two motifs have been studied independently of each other. Yet, there is increasing evidence that experimen- tal manipulations that elicit response sub-additivity often also quench response variability. Here we provide a unifying review of these phenomena, suggesting that response sub-additivity and variability quenching may have a common origin. We review empirical findings as well as recent model-based insights into the functional operations, computational objectives, and circuit mechanisms underlying V1 activity. Although these model- ing approaches address different aspects of cortical activity, they all predict that response sub-additivity and variability quenching will often co-occur. Response sub-additivity and variability quenching are not limited to V1 but are widespread cortical phenomena. Many of the insights we review generalize to other cortical areas, suggesting that the connection between response sub-additivity and variability quenching may be a canonical motif across cortex. 

To interpret the sensory environment, the brain combines ambiguous sensory measurements with knowledge that reflects context-specific prior experience. But environmental contexts can change abruptly and unpredictably, resulting in uncertainty about the current context. Here we address two questions: how should context-specific prior knowledge optimally guide the interpretation of sensory stimuli in changing environments, and do human decision-making strategies resemble this optimum? We probe these questions with a task in which subjects report the orientation of ambiguous visual stimuli that were drawn from three dynamically switching distributions, representing different environmental contexts. We derive predictions for an ideal Bayesian observer that leverages knowledge about the statistical structure of the task to maximize decision accuracy, including knowledge about the dynamics of the environment. We show that its decisions are biased by the dynamically changing task context. The magnitude of this decision bias depends on the observer’s continually evolving belief about the current context. The model therefore not only predicts that decision bias will grow as the context is indicated more reliably, but also as the stability of the environment increases, and as the number of trials since the last context switch grows. Analysis of human choice data validates all three predictions, suggesting that the brain leverages knowledge of the statistical structure of environmental change when interpreting ambiguous sensory signals.