A network supporting memory-guided visual behaviour in humans
By Kelly Shen
This week marks our return in-person to the Society for Neuroscience meeting. Over the last few days, we’ve been giving you a sneak peek of what we’ll be presenting in San Diego.
In the last of the series, I’ll be covering our poster titled “Structural connectivity between the memory and oculomotor systems changes with age” which will be up on the morning of Tuesday November 15th (Poster #YY45).
Our poster highlights the next step in a line of research I’ve been working on in collaboration with Jennifer Ryan, where we’ve been interested in how the memory system guides eye movements and how that influence changes with age. We’ve previously used anatomical connectivity from the macaque to identify a network that bridges the memory and oculomotor systems, and simulated the dissipation of activity between the memory system and oculomotor regions using TheVirtualBrain. You can read more about our line of work here and here, and you can read about our secret sauce for maintaining our long-running collaboration here.
For us, the open question has always been what a network that bridges the memory and oculomotor system may look like in humans, and whether such a network might change with age. Here, we used diffusion imaging data from the Cambridge Centre for Aging and Neuroscience (Cam-CAN) dataset to try to answer these questions. Keeping in mind the difficulties of segmenting and performing tractography from hippocampal structures, especially in older adult samples, we first randomly selected a subset of our processed Cam-CAN dataset for manual inspection and quality control (QC). Using the data from younger adults (18-36 years) in this manually-QC’d subset, we identified a set of regions that connect hippocampal regions with the frontal eye fields (FEF). The surprising thing we found, and the reason why we went down an intensive QC rabbit hole, was that there was a weak but consistent direct structural connection between hippocampus and FEF in the left hemisphere. This was unexpected based on what we know from macaque connectivity, as well as the generally sparse nature of connectivity between hippocampus and cortical regions. We took a deeper look at our FEF parcels, and they may extend into dorsolateral prefrontal cortex (dlPFC), especially on the left side. A structural connection between hippocampus and dlPFC is expected, so we are hesitant to say much about species differences in hippocampal connections to FEF.
The network of regions that bridge hippocampus and frontal eye field in humans.
We then asked whether the network we identified using the data from the younger group changes with age by comparing it to an older group from Cam-CAN (69-87 years). This group’s data was also manually QC’d to ensure proper segmentation of the hippocampus. For the right hemisphere, we found structural connections between hippocampus and cortical intermediary regions to weaken with age, while connections between intermediary regions and FEF became stronger. No changes were detected for the left hemisphere.
Changes to the structural connections of the network that bridges the memory and oculomotor system with age.
These age-related changes in structural connectivity may reflect the observations that hippocampus seems to have less of an influence on guiding eye movements with age. Future work is needed to determine whether these structural changes are related to functional changes as well as changes in visual behaviour with age.