Neuroscience

Alyssa Flores, alyssa_flores3@baylor.edu
Baylor University, with Dr. Michael Scullin

Does functional fixedness & memory performance change as a result of sleep restriction?
Sleep plays a critical role in memory consolidation, with sleep quality and duration contributing to various aspects of cognitive processing. This study examines the effects of sleep restriction on memory performance using free recall testing, recognition testing, and word fragmentation completion tasks. Participants were in two sleep conditions, one group having ample sleep of 8 hours and the other group being in a sleep-restricted setting of 5 hours. The evening posed as a priming condition; participants were presented with words and asked to rate them. The following day, participants completed a free recall test, a recognition test, and a word fragmentation task. It is hypothesized that both sleep restriction and fatigue from restriction would diminish memory consolidation, increase errors in the word fragmentation task, and result in lower accuracy in the free recall test. Additionally, sleep stage two and REM sleep are stages often involved in memory consolidation, and the correlations between these two stages and memory performance will be explored. After one night of sleep restriction, our preliminary findings indicate that participants in either group showed no differences in accuracy of word fragmentation completion and free recall. Future research may investigate individual differences in reaction times for functional fixedness amongst participants with longer sleep restriction.
 

Wilfred Wisdom, wilfred.wisdom@uconn.edu
University of Connecticut, with Dr. Daniel Mulkey and Dr. Eliandra da Silva

Prostaglandin Modulation of Respiratory Neurons
The retrotrapezoid nucleus (RTN) is a region in the brainstem that assists in regulating breathing. Chemosensitive neurons prevalent in this area are involved in the detection of changes in CO2/H Levels. Prostaglandin E2 (PGE2) plays an important role in this response and is released in reaction to CO2 acting on E-prostanoid (EP) receptors. These receptors have different subtypes that serve specific roles. EP3 is mostly discussed within literature as being most involved in breathing, while the other subtypes (EP1-EP4) are less understood. We hypothesized that PGE2 modulates RTN neuron activity through specific EP receptors, particularly EP3, and that expression of other EP receptors may differ across RTN regions. To test this, we recorded CO2- evoked a baseline firing activity in RTN neurons and imaged spatial expression patterns of EP1R and EP3R. We found that PGE2 reduced baseline activity in chemosensitive RTN neurons. Imaging of EP3R showed higher and more localized expression compared to EP1R,  which was more broadly distributed within similar regions. These findings suggest that the two receptors may serve separate roles in respiratory modulation, with EP3R potentially acting on a more specialized group of neurons. This work provides insight into how prostaglandin signaling may shape respiratory responses within the RTN. Future directions include identifying which neurons express EP3R, quantifying how much is expressed, and analyzing receptor expression under different physiological conditions.
 

Mikayla Robinson, mikayla.robinson@uconn.edu
University of Connecticut, with Dr. Emily Myers

Structural Predictors of BrainAGE and Cognitive Performance
As we age, certain structural brain features tend to decrease in volume. However, the severity of how individuals experience cognitive decline and, in turn, accelerated brain aging varies.

BrainAGE is a machine learning model that estimates an individual's "BrainAGE" from MRI scans, potentially acting as a biomarker of cognitive health. However, model outputs can vary widely depending on the data that the algorithm is trained on and the preprocessing parameters chosen by the researcher. This project aims to replicate findings that the BrainAGE gap is a significant predictor of individual differences in cognitive and hearing abilities in healthy adults. Participants underwent MRI scanning and completed various cognitive and linguistic assessments. Anatomical MRI data were preprocessed using the CAT12 toolbox in MATLAB. Then, a Gaussian Process Regression model was trained on the 60 participants in our sample (ages 18-78) to detect structural regularities across age groups. The BrainAGE gap (BAG) was then calculated as the model's predicted age minus chronological age. The model's predicted age and participant's chronological age were highly correlated, suggesting satisfactory model performance. However, in contrast to a previous study, the BAG did not predict subtle differences in behavior beyond chronological age. The limitations of our training sample will be discussed. Future directions include troubleshooting a line in the original code to explicitly state what structural features the model "learned" during classification, and including code for estimating the BrainAGE for different lobes. These changes would make the model more generalizable, improving the replication of BrainAGE for researchers studying the effects of degraded hearing on cognitive decline.