Perception is a multi-faceted, dynamical process that can be tackled empirically through measures of stimulus detectability and confidence. We recorded stereo-electroencephalographic data of 29 participants partaking in three pre-registered experiments to assess if evidence accumulation, a form of sequential sampling of sensory evidence, can explain perception. In an immediate-response experiment, high-gamma activity from individual channels and decoded multivariate latent variables in the visual, inferior frontal, and anterior insular cortices display a correlation between the slope of their increase and reaction times. In two further experiments, this signal in the ventral visual cortex differentiates between (1) stimuli reported as seen vs. unseen in delayed detection, (2) high and low intensity stimuli during passive viewing, and (3) levels of confidence when stimuli were detected. A computational model of leaky evidence accumulation can successfully reproduce both behavioral and neural data. Overall, we show that evidence accumulation explains subjective aspects of visual perception.
Neuroprosthetics offer great hope for motor-impaired patients. One obstacle is that fine motor control requires near-instantaneous, rich somatosensory feedback. Such distributed feedback may be recreated in a brain-machine interface using distributed artificial stimulation across the cortical surface. Here, we hypothesized that neuronal stimulation must be contiguous in its spatiotemporal dynamics to be efficiently integrated by sensorimotor circuits. Using a closed-loop brain-machine interface, we trained head-fixed mice to control a virtual cursor by modulating the activity of motor cortex neurons. We provided artificial feedback in real time with distributed optogenetic stimulation patterns in the primary somatosensory cortex. Mice developed a specific motor strategy and succeeded to learn the task only when the optogenetic feedback pattern was spatially and temporally contiguous while it moved across the topography of the somatosensory cortex. These results reveal spatiotemporal properties of the sensorimotor cortical integration that set constraints on the design of neuroprosthetics.
Evidence accumulation is a powerful mechanism to explain the temporal dynamics of decisions, as well as their metacognitive components such as confidence judgments and changes of mind. However, it is still unclear how and where in the brain evidence accumulation leads to these two metacognitive components. We report intracranial high-gamma activity in patients with epilepsy recorded while they perform a visual discrimination task and estimate their confidence level. Our results indicate an anatomical overlap between the neural correlates of evidence accumulation, confidence, and changes of mind in the pre-supplementary motor area, as well as in the orbitofrontal, inferior frontal, and insular cortices. Behavioural and electrophysiological results are reproduced with a post-decisional evidence accumulation model, and the temporal dynamics of decision-making is characterized with mouse-tracking and intracranial electrophysiology. We conclude that confidence and changes of mind result from evidence accumulation, instantiated before the decision in the pre-supplementary motor area, and after the decision in the insula.
Summary: Perceptual experience is a multi-faceted, dynamical process, tackled empirically through measures of stimulus detectability and confidence. To assess if stimulus detection and confidence can be explained by evidence accumulation, a form of sequential sampling of sensory evidence, we analyzed high-gamma activity from stereo-electroencephalographic data of 29 participants partaking in 3 pre-registered experiments. In an immediate-response experiment, individual channels and decoded multivariate latent variables in the visual, inferior frontal, and anterior insular cortices displayed functional markers of evidence accumulation. In two further experiments, this signal in the ventral visual cortex differentiated between (1) seen and unseen stimuli in delayed detection, (2) high and low intensity stimuli during passive viewing, and (3) levels of confidence when stimuli were seen. A computational model of leaky evidence accumulation successfully reproduced both behavioral and neural data. Overall, these results indicate that evidence accumulation explains key aspects of perceptual experience, encompassing both conscious access and monitoring.
Evidence accumulation models have proven to be powerful tools to explain the temporal dynamics of decisions, as well as their metacognitive components such as confidence judgments and changes of mind. However, it is still unclear how and where in the brain evidence accumulation leads to these two metacognitive components. To better understand the functional overlap between evidence accumulation, confidence, and changes of mind, we recorded intracranial high-gamma activity in patients with focal epilepsy while they reported the motion direction of a random dot kinetogram with a computer mouse, and estimated their confidence level. We found an anatomical overlap between the neural correlates of evidence accumulation, confidence, and changes of mind in the pre-supplementary motor area, as well as in the orbitofrontal, inferior frontal, and insular cortices. Both mouse-tracking behaviour and electrophysiological results were reproduced with a post-decisional evidence accumulation model. After characterising the temporal dynamics of decision-making with mouse-tracking and intracranial electrophysiology, we conclude that confidence and changes of mind result from evidence accumulation instantiated before the decision in the pre-supplementary motor area, and after the decision in the insula.
Most human studies on perceptual decision-making are correlational, and few addressed the causal role of the neural markers they identify. Here, we took the opportunity offered by deep brain stimulation to study the causal role of subthalamo-cortical networks on perceptual decision-making and perceptual monitoring. To this end, we asked participants with obsessive-compulsive disorder (OCD) treated with subthalamic stimulation to detect visual stimuli and provide confidence reports while we recorded scalp electroencephalography. Following a preregistered plan, we compared behavioural and neural results between patients off-stimulation and matched healthy controls and among patients on and off-stimulation to identify the effect of subthalamic stimulation on perceptual choices and metacognitive judgments. Our results indicate that patients off-stimulation detected visual stimuli similarly to matched healthy controls but provided poorer confidence judgments indicative of a metacognitive deficit. Moreover, chronic subthalamic stimulation significantly reduced clinical symptoms, but acute stimulation changes did not impact visual detection, confidence, or corresponding event-related potentials at the scalp level. We conclude that acute changes in subthalamic stimulation are insufficient to alleviate metacognitive deficits observed in OCD.