Neuroanatomy and its Impact on Structural and Functional Imaging (In Memory of Karl Zilles)

The prefrontal cortex (PFC) is critically important for higher-order cognition in humans, playing a central role in cognitive control and goal-directed behavior (Miller & Cohen, 2001). Patients with damage to the PFC show deficits ranging from working memory and attention problems, to issues with motivation, response inhibition, and language (Szczepanski & Knight, 2014). How are such a vast array of cognitive processes orchestrated by a few cubic centimeters of cortical tissue? Classic proposals posit that the unique anatomical and functional properties of association cortex circuits, such as in PFC, can account for this remarkable flexibility of human cognition (Sanides, 1964). As the tools of modern neuroimaging advance, we are now more suited than ever to test these foundational frameworks. In this talk, I will share how a precision neuroimaging approach, with careful focus on neuroanatomical properties, can reveal fundamental principles of PFC functioning across studies, methods, and species.

First, I will show how structures known as tertiary sulci - the smallest and shallowest of sulcal indentations - offer insights into PFC functioning that bridge spatial scales (microns to networks), modalities (functional connectivity to behavior), and species. These tertiary sulcal structures are prominent within individual brains but are often smaller or completely missing from average cortical templates commonly used in human neuroimaging (Miller et al., 2021a). In these cases, the map of tertiary sulci within each individual participant may serve as a coordinate system specific to that individual on which functions may be further mapped. Using manual or semi-automated neuroanatomical labeling of PFC morphology within individuals can enable the characterization of structural-functional relationships in PFC that are more difficult to obtain when averaging across individual brains (Lyu et al., 2021; Miller et al., 2021b). For example, the morphology and depth of tertiary sulci co-varies with working memory and reasoning skills across development (Voorhies et al., Nat Comms, 2021; Yao et al., Cereb Cortex, 2022). I will then lay out how our field can apply lessons from precision neuroanatomical approaches to a wider array of neuroimaging studies and questions.

Second, I will demonstrate how dense, longitudinal sampling paradigms can reveal within-individual functional organization across different areas of human PFC. Specifically, I will use the example of gradients of working memory representations in PFC that are transformed over long-term learning (Miller et al., Neuron, 2022). Three human participants each completed over 20 sessions of functional MRI (fMRI) along with at-home training across three months. During this time, participants repeatedly performed a delayed recognition working memory task and a sequence learning task, which both employed a set of novel fractal stimuli that were unique to each participant. Across the course of training, working memory representations were transformed along a general rostral-caudal anatomical and functional organization in PFC: stimulus-specific working memory representations emerged in mid-lateral PFC and categorical representations in caudal PFC areas. A host of neuroanatomical properties are thought to endow the PFC with a propensity for this kind of plasticity and make it well-suited to serve as an integrative hub for learning effects across timescales (Miller & Constantinidis, 2024). These functional and neuroanatomical organizations may be best captured by densely sampling PFC activity patterns over time within individuals, and I will share the principles and advantages of such study designs.

Finally, I will dive below the level of voxels into microscale organization and show how working memory circuitry operates at different spatial scales in PFC circuits. Traditionally, many neuronal activity patterns in PFC are thought to be inaccessible to voxel-level sampling because of the spatial intermixing of neurons in close anatomical proximity with different functional tuning. However, several new studies recording from PFC in NHPs reveal a micro- to macro-scale organization of neuronal coding that can help bridge across spatial scales of functioning in PFC (Sun et al., 2024; Xiang et al., 2024). I will highlight these findings and their important conclusions for neuroimaging studies of human PFC: can we observe a “cognitive globe” of functioning mirrored at the micro- and meso-scale levels in PFC?

Collectively, in this course I will demonstrate how the tools and lessons of individual-level neuroanatomy and functioning allow us to better characterize robust brain-behavior relationships and advance our understanding of working memory and related cognitive processes in the PFC.  

Thalamocortical circuits connect all parts of the thalamus with the cortex. These interconnected circuits form early in neural development and establish functional and dynamic reciprocal partnerships for neural communication. We know these circuits are critical for integrating sensory information, coordinating voluntary movements, and enabling effective cognitive functioning within the cortex. Furthermore, disruptions to these circuits cause deficits in cognition and sensory and motor impairments linked to many neurodevelopment, neuropsychiatric and neurodegenerative disorders and diseases. Magnetic resonance imaging provides a useful tool to identify changes in thalamocortical circuits in vivo, and in combination with neuropathological and genetic evidence can better inform our understanding of how and when thalamocortical circuits contribute to our sensory, motor, and cognitive abilities. Finally, advanced in vivo MRI methods are now supporting clinicians and researchers to characterise thalamocortical dysfunction early on in neurodegenerative diseases like Alzheimer’s and Parkinson’s Disease, and to identify effective treatments in neurodevelopmental disorders, like attention-deficit-hyperactivity disorder. This talk will explain how thalamocortical circuits develop and work in the mammalian brain, and how MRI can be used to identify changes in thalamocortical circuits in human and animal neuroimaging datasets. 

American monkeys which are being increasingly becoming adopted as animal models in neuroscience. Knowledge about the marmoset visual system has developed rapidly over the last decade. But what are the comparative advantages, and disadvantages involved in adopting this emerging model? In this talk I will present case studies where the simpler and more reproducible morphology of the marmoset brain, and the shorter developmental cycle of this species, have been key factors in facilitating fundamental discoveries about the anatomy and physiology of the visual system. Although no single species provides the “ideal” animal model for studies of the neural bases of sensorimotor processing and cognition, I argue that the development of robust methodologies for the study of the marmoset brain provides exciting opportunities to address long standing problems in systems and developmental neuroscience. 

This lecture will review the structural and functional divisions in the medial temporal lobe, with a special focus on the hippocampus. The hippocampus is a critical region of the brain for the formation of new episodic memories. It is also a region that is vulnerable to a number of developmental, psychiatric, and neurological conditions. It is one of the most well-studied regions of the brain.

It is important to note that the hippocampus is not a singular homogeneous structure, but is a layered structure that is commonly divided into distinct subfields. Similarly, the medial temporal lobe cortex, the region of cortex just lateral to the hippocampus, is composed of the entorhinal and perirhinal cortices, anteriorly, and the parahippocampal cortex, posteriorly. Different neuroanatomists have used slightly different nomenclature to subdivide the hippocampus over the years to define these subregions. This lecture will provide a comprehensive overview of the historical developments that have led to the current definitions of the universally recognized hippocampal subfields, which include the dentate gyrus, the cornu ammonis (CA) regions 1-4, and the subiculum. I will also provide definitions of the less commonly recognized regions. For example, the transition region between CA1 and subiculum is sometimes defined as the prosubiculum and the pre- and para-subiculum regions are regions that are sometimes identified in between the subiculum and the entorhinal cortex.

I will discuss how the subfields of the hippocampus differ from one another and how they are defined anatomically. I will discuss the functional differences between these regions of the brain and how they contribute to different types of cognitive functions. I will also discuss how these regions of the brain change with advanced age and cognitive decline.

Finally, I will discuss common neuroimaging methodology to define the medial temporal lobe regions using different types of structural (e.g. T1-weighted and T2-weighted) magnetic resonance imaging (MRI). Finally, I will discuss challenges and solutions for the study of the hippocampal and MTL cortex subregions using in vivo MRI.

References:
Olsen, R. K., Carr, V. A., Daugherty, A. M., La Joie, R., Amaral, R. S., Amunts, K., ... & Hippocampal Subfields Group. (2019). Progress update from the hippocampal subfields group. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring, 11(1), 439-449.

Mazloum‐Farzaghi, N., Barense, M. D., Ryan, J. D., Stark, C. E. L., & Olsen, R. K. (2024). The Effect of Segmentation Method on Medial Temporal Lobe Subregion Volumes in Aging. Human Brain Mapping, 45(15), e70054.

Baumeister, H., Wuestefeld, A., Adams, J. N., Bakker, A., Daugherty, A. M., de Flores, R., ... & Wisse, L. E. (2023). Comparison of histological delineation of the entorhinal, perirhinal, ectorhinal, and parahippocampal cortices by different neuroanatomy laboratories. Alzheimer's & Dementia, 19, e076135. 

The hippocampus is an evolutionarily conserved brain structure in vertebrate species and is critical for memory, cognition, stress and emotion. It comprises subregions including dentate gyrus (DG), cornu ammonis (CA) fields and subicular complex. The ventral hippocampus is widely connected with brain areas in the thalamus, hypothalamus, amygdala and prefrontal cortex, and plays crucial roles in the emotional processing for rewards and punishments. By leveraging the power of single-neuron projectome analysis, we have characterized the target pattern of distinct projectome cell types in the thalamo-hippocampal circuit, and identified specific functions of these cell types and circuits in the emotional processing including fear memory generalization. 

Awake brain surgery is a surgical procedure aimed at maximizing lesion resection while preserving brain functions essential for human life. During awake surgery, cortical and subcortical areas are stimulated electrically while the patient performs functional tasks, including language, motor, sensory tasks, and other higher brain functions to assess the functionality of the stimulated area. When positive (impaired) responses are elicited by direct electrical stimulation (DES), the stimulated area is revealed to be functional, whereas normal responses induced by DES indicate that the stimulated area is not functional regarding the function assessed. There are two types of positive responses: functional activation and functional inhibition. Functional activation represents neurological reactions at rest in response to DES in primary areas, including motor, sensory, and visual areas. Functional inhibition means that functions are inhibited during task performance in response to DES in the association areas.
DES during awake surgery provides valuable research opportunities to understand functional neuroanatomy and its reorganization in humans. An advantage of DES during awake surgery is the ability to determine causality between brain areas and map functions with high spatial resolution and sensitivity in real time. Using this technique, previous studies have revealed insights that would be difficult to obtain using noninvasive methods.
When the brain is damaged, functional reorganization, in which brain functions are relocated to regions outside the damaged area, can sometimes occur to restore neurological or cognitive functions. Owing to reorganization, brain functions are often normal despite the invasion of a lesion into areas involved in specific functions. Reorganization is an important phenomenon to consider when performing surgery for brain tumors because if certain areas do not have functions essential for human living, the region can be resected safely. During awake surgery, we can also observe how cortical or subcortical localization changes from the original anatomical localization with high spatial resolution, even in perilesional areas. In this context, we have found using DES during awake surgery that functional reorganization occurs under specific conditions rather than randomly.
Here, I will present some studies on functional neuroanatomy and its reorganization in motor, language, visuospatial cognition, and other higher brain functions, and show how we can use DES for research, as well as clinical purposes. 

This course is designed to introduce scientists who have little or no background in brain anatomy to the fundamental structures of the brain and the functions they are involved in. Throughout the course, participants will gain an understanding of the key landmarks that anatomists use to navigate the complex architecture of the brain, and the functions that different regions are involved in.
The course emphasizes practical learning, with a strong focus on using interactive tools, particularly the Julich Brain Atlas (https://atlases.ebrains.eu/viewer/#/). This online viewer allows participants to engage directly with high-resolution brain datasets, providing them with a hands-on opportunity to navigate through the brain. Participants will not only learn the tricks used by radiologists to identify, e.g., the central sulcus or the intraparietal sulcus, but will have a chance to search for these brain structures themselves, enhancing their practical skills in brain navigation.
Another the key strength of the course is its flexibility in addressing the specific needs of participants. They will have the opportunity to ask questions about brain structures that are particularly relevant to their own scientific research and delve deeper into the microstructure of these areas. By the end of the course, participants will not only have become familiar with important anatomical terms—such as lobes, gyri, diencephalon, hippocampus, and primary visual cortex—but they will also develop the ability to locate these structures in scientifically relevant datasets such as the BigBrain, the Colin 27 single subject template, or the ICBM 152 population template. This will equip them with both theoretical knowledge and practical skills, allowing them to apply what they’ve learned in various scientific or clinical contexts.