Abstract deliberation by visuomotor neurons in prefrontal cortex

Abstract

During visually guided behavior, the prefrontal cortex plays a pivotal role in mapping sensory inputs onto appropriate motor plans. When the sensory input is ambiguous, this involves deliberation. It is not known whether the deliberation is implemented as a competition between possible stimulus interpretations or between possible motor plans. Here we study neural population activity in prefrontal cortex of macaque monkeys trained to flexibly report categorical judgments of ambiguous visual stimuli. Our task design allowed for the dissociation of neural predictors of the upcoming categorical choice and the upcoming motor response used to report this choice. We find that the population activity initially represents the formation of a categorical choice before transitioning into the stereotypical representation of the motor plan. We show that stimulus strength and prior expectations both bear on the formation of the categorical choice, but not on the formation of the action plan. These results suggest that prefrontal circuits involved in action selection are also used for the deliberation of abstract propositions divorced from a specific motor plan, thus providing a crucial mechanism for abstract reasoning.

Sensory population activity reveals confidence computations in the primate visual system

“…The animals were trained to perform an orientation discrimination task with saccadic eye movements as operant responses. Monkey F had previously participated in another research study 22 , Monkey Z had not previously participated in research studies. All training, surgery, and recording procedures were approved by the University of Texas Institutional Animal Care and Use Committee and conformed to the National Institutes of Health Guide for the Care and Use of Laboratory Animals.…”

“…How does the brain assess the quality of perceptual decisions? A prominent hypothesis is that early areas in sensory cortex provide raw sensory measurements which are used by downstream circuits in association cortex to guide perceptual decisions [19][20][21][22] and assign confidence in these decisions 7,10,12,13 . It follows that there may exist a systematic relationship between neural population activity in sensory cortex and confidence in perceptual decisions.…”

Transitions in dynamical regime and neural mode underlie perceptual decision-making

“…While the attractor dynamics observed here differs from theoretical proposals, it provides a parsimonious explanation of experimental findings from multiple species. Across primates and rodents, sensory inputs and choice are represented in separate neural dimensions [7][8][9][47][48][49] . Across time, neither sensory responses nor the neural dimensions for optimal decoding of the choice are fixed 8,47,49 .…”

“…This match led us to explore a simplified model, the MMDDM, in which a scalar latent decision variable evolves as in the DDM, but is represented in different neural modes before versus after decision-commitment. We found this to provide a parsimonious explanation of a variety of experimental findings from multiple species: across primates and rodents, sensory inputs and choice are represented in separate neural dimensions [6][7][8]53,59,60 across time, and neither sensory responses nor the neural dimensions for optimal decoding of the choice are fixed 7,59,60 . These phenomena, along with other observations including diversity in single neuron dynamics 50,51 , curved average trajectories 7,8,53 , choice behavior 36 , and some vigorously debated phenomena such as a variety of single-neuron ramping versus stepping temporal profiles 4,5,61 , are all captured by the MMDDM.…”

“…The state transition from evidence accumulation to decision commitment in the MMDDM, and a consequent switch from w EA to w DC , is reminiscent of the gradual change in neural modes encoding a decision variable to those encoding an action plan 7,52,53 . The MMDDM provides the further advantage of allowing precise inference of each trial's moment of the switch (the time of internal commitment to a decision), based on the recorded neural activity.…”