Electronic Thesis and Dissertation Repository

Thesis Format



Doctor of Philosophy




Everling, Stefan


Our visual world is full of far more stimuli than can be processed simultaneously. Yet we are able to efficiently extract behaviourally relevant information from a scene, primarily by performing rapid saccadic eye movements. These processes are under the control the frontoparietal network, two critical nodes of which are: the lateral intraparietal area (LIP) and the frontal eye fields (FEF). Extensive research in the macaque has causally implicated these areas in visual attention and oculomotor control. However, the organization of the activity of single neurons in these areas across cortical layers remains poorly understood as these regions are deep within sulci in the macaque. The marmoset, with a lissencephalic cortex, largely homologous frontoparietal network, and comparable oculomotor repertoire, presents a unique opportunity to address these questions. First, the homology of these cortical areas must be established. Recent work from our group supports marmoset LIP homology, however, FEF remains to be explored. The first aim of this dissertation was identify and characterize marmoset FEF. Using intracortical microstimulation (ICMS), we restricted marmoset FEF to areas 8a, 45, 6D and 8C, and demonstrated frontal cortical organization consistent with other primates, supporting the use of marmosets for neurophysiological investigations of oculomotor control and attention. The subsequent aims of this work were to examine the laminar dynamics of LIP and FEF in marmosets completing a target selection task. We observed neurons in both LIP and FEF involved in target selection, with FEF showing a stronger link to motor control. Interestingly, organization in LIP followed the canonical circuit model (CCM), with input in the granular, target selection in supragranular, and output in infragranular layers. In contrast, FEF displayed a unique bilaminar visual input in superficial layers and target selection in deeper layers, resembling recent observations in other frontal areas more than the traditional CCM. These findings suggest that while models developed in primary sensory areas might apply to some regions of association cortex, their generalizability to frontal areas is limited. This work underscores the marmoset’s value as a model for studying attention and cognition, and broadens our knowledge of cortical organization underlying these phenomena.

Summary for Lay Audience

In our busy visual world, we rapidly scan scenes to focus on important details. This process, attention, is largely governed by two brain areas: the lateral intraparietal area (LIP) and the frontal eye fields (FEF). Our understanding of how neurons in these areas function comes primarily from macaque studies. We know that these regions can be separated into distinct layers, but we know little about how neurons in different layers differ functionally. Uncovering this organization would provide a greater understanding of how the brain directs attention, and generally deepen our understanding of brain organization. However, due to the many folds in the macaque brain, as in the human brain, it is challenging to study neurons from these different layers simultaneously. Here, the more distantly related, common marmoset monkey, with its smooth brain, presents an opportunity to answer this very question. Now we are able to use long, high-density electrodes to record activity from multiple layers and reconstruct this organization. However, we first have to determine if marmoset LIP and FEF resemble what we see in other primates. We have previously shown that this is the case for LIP, and in the first project of this thesis, using intracortical microstimulation (ICMS), we show this for FEF. Having established this similarity, we studied regions as marmosets completed a task where they must quickly look towards a target item presented on a screen while ignoring any distractors. First, we replicated patterns of neural activity observed in macaques for such a target selection task. Interestingly, we found that in LIP, activity is consistent with existing models which suggest there are distinct input, processing, and output layers. However, while FEF showed some similarities, we observed two input layers here and a different layer was responsible for processing than expected. These findings show that while our current models apply to some brain areas and functions, they may not be entirely generalizable. Our findings show the value of the marmoset for studying the neuroscience of attention, and provides a deeper understanding of how the brain is organized.

Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.