Dynamic Brain Interactions during Picture Naming.

Brain computations involve multiple processes by which sensory information is encoded and transformed to drive behavior. These computations are thought to be mediated by dynamic interactions between populations of neurons. Here, we demonstrate that human brains exhibit a reliable sequence of neural interactions during speech production. We use an autoregressive Hidden Markov Model (ARHMM) to identify dynamical network states exhibited by electrocorticographic signals recorded from human neurosurgical patients. Our method resolves dynamic latent network states on a trial-by-trial basis. We characterize individual network states according to the patterns of directional information flow between cortical regions of interest. These network states occur consistently and in a specific, interpretable sequence across trials and subjects: the data support the hypothesis of a fixed-length visual processing state, followed by a variable-length language state, and then by a terminal articulation state. This empirical evidence validates classical psycholinguistic theories that have posited such intermediate states during speaking. It further reveals these state dynamics are not localized to one brain area or one sequence of areas, but are instead a network phenomenon.

A neural computation for visual acuity in the presence of eye movements.

Humans can distinguish visual stimuli that differ by features the size of only a few photoreceptors. This is possible despite the incessant image motion due to fixational eye movements, which can be many times larger than the features to be distinguished. To perform well, the brain must identify the retinal firing patterns induced by the stimulus while discounting similar patterns caused by spontaneous retinal activity. This is a challenge since the trajectory of the eye movements, and consequently, the stimulus position, are unknown. We derive a decision rule for using retinal spike trains to discriminate between two stimuli, given that their retinal image moves with an unknown random walk trajectory. This algorithm dynamically estimates the probability of the stimulus at different retinal locations, and uses this to modulate the influence of retinal spikes acquired later. Applied to a simple orientation-discrimination task, the algorithm performance is consistent with human acuity, whereas naive strategies that neglect eye movements perform much worse. We then show how a simple, biologically plausible neural network could implement this algorithm using a local, activity-dependent gain and lateral interactions approximately matched to the statistics of eye movements. Finally, we discuss evidence that such a network could be operating in the primary visual cortex.