PROJECT SUMMARY / ABSTRACT Language gives meaning to our inner thoughts, interacting with cognitive functions that flexibly manipulate sensory content in mind. Language combinatorial capacities create meaning by combining symbols (words) and rely on an extensive neural system, which when affected by neurological impact leads to disorders of language, memory and thought. Although animal models have deeply informed understanding of neural systems for which there are correspondences to human cognition, such as those for working memory, the neuronal mechanisms of language and its entanglement with cognitive function remain poorly understood. The proposed interdisciplinary team will seek to break through the status quo by leveraging a comprehensive research program that can provide a well-powered study of human language and working memory interactions with unique scalable data capable of resolving neural function across neocortical layers from single cells to systems. Aim 1a: Our language Combinatoric Transformations via Working Memory (CTWM) task will be conducted with up to 100 patients across our neurosurgical partnering sites during high-density laminar array recordings in the operating room from two key brain areas involved in language combinatorics and working memory. Aim 1b: Insights on neuronal function across the cortical layers will be deeply informed by unprecedented information from single- cell genomic and spatial transcriptomic analyses applied to tissue samples after task performance and the laminar array recordings. Aim 2a is to investigate laminar information flow across the cortical array in local field potentials, and Aim 2b in interaction with available (non-laminar) subdural intracranial EEG recordings from other brain areas. Aim 3a is to pre-operatively scale insights to system-wide levels using laminar fMRI in the same patients with the same task, and Aim 3b will integrate the combined neurophysiological and neuroimaging data via a next-generation generative biophysical model. The anticipated outcomes are first-in-human insights on the interplay between language and cognition, unique openly shared multi-modal data and a model that could be applied world-wide to accelerate the understanding of human laminar circuit motifs in health and disease.