Biology subjects

Anderson, J. R.

Publications and source records attributed to Anderson, J. R..

17 recordsLinked to original sources

bioRxiv2026

Hydrogen sulfide-mediated vasodilation requires heme oxygenase-derived carbon monoxide

BackgroundHydrogen sulfide (H2S) is an important endothelial-derived vasodilator, but the signaling mechanism remains incompletely understood. We previously demonstrated that H2S-mediated vasodilation requires transient receptor potential vanilloid type 4 (TRPV4) channels. Because H2S has been reported to enhance heme oxygenase (HO) activity and HO-derived carbon monoxide (CO) regulates endothelial signaling, we hypothesized that H2S-mediated vasodilation requires HO-2-derived CO. MethodsPressure myography was performed in isolated rat mesenteric arteries to determine the contribution of HO, TRPV4, eBK, and SK/IK channels to H2S-mediated vasodilation. HO-2 sulfhydration was assessed using a maleimide assay, and spatial association among HO-2 and TRPV4 was examined using proximity ligation assays in human aortic endothelial cells. ResultsH2S Selicited concentration-dependent vasodilation that was abolished by HO inhibition. Repletion of CO restored H2S-mediated vasodilation in the presence of HO inhibition. CO-mediated vasodilation was abolished by TRPV4 and SK/IK inhibition but was unaffected by eBK inhibition. H2S increased HO-2 sulfhydration and enhanced HO activity. In endothelial cells, HO-2 and TRPV4 exhibited close spatial association. ConclusionsThese findings support a model in which H2S stimulates HO-2-derived CO production, leading to TRPV4-dependent endothelial signaling, SK/IK activation, and vasodilation. Together, the data support the existence of an endothelial HO-2/TRPV4/SK/IK signaling domain that contributes to H2S-mediated vascular reactivity.

physiology↗

bioRxiv2025

Shear Stress Regulates ABCA1-dependent Membrane Cholesterol Content in Endothelial Cells Facilitating H2S-dependent Vasodilation

Endothelial cells (ECs) express an array of integral membrane proteins, including ion channels and transporters that contribute to blood flow regulation and cell-cell communication. Many of these membrane proteins are regulated by plasma membrane cholesterol content. The ATP-binding cassette family a1 (ABCA1) transporter is a regulator of membrane cholesterol content. We have shown increased ABCA1 mRNA expression and reduced EC membrane cholesterol in resistance mesenteric arteries compared to conduit arteries. Previous studies suggest shear stress (SS) can increase or decrease ABCA1 expression in a cell-type-dependent manner. HypothesisSS sustains lower EC membrane cholesterol concentration through ABCA1-mediated cholesterol transport, facilitating H2S-mediated vasodilation. MethodsThe effect of SS on ABCA1 and membrane cholesterol content was assessed in pressurized mesenteric arteries from male Sprague-Dawley rats and cultured human aortic endothelial cells. Pressure myography was used to assess the effects of ABCA1 inhibition on H2S-mediated vasodilation. Filipin was used to assess EC membrane cholesterol content. ResultsSS increased ABCA1 expression in the endothelium of mesenteric arteries and cultured human aortic endothelial cells and markedly reduced EC membrane cholesterol. Inhibition of ABCA1 increased EC membrane cholesterol content and abolished H2S-induced vasodilation. ConclusionSS facilitation of EC-dependent vasodilation appears to be mediated by membrane cholesterol content.

physiology↗

bioRxiv2024

Allogenic platelet-rich plasma and platelet-rich plasma extracellular vesicles alter the proteome of tenocytes in an in vitro equine model of tendon inflammation: A pilot study

BackgroundTendon injuries are common in horses, often resulting in a high risk of reinjury. Hemoderivative therapeutics including platelet-rich plasma (PRP) show promise but outcomes vary due to inconsistent composition. PRP is rich in soluble growth factors and extracellular vesicles (EVs), the latter facilitate cell-to-cell communication by delivering biologically active cargo. This pilot study profiled the proteome of PRP and PRP-derived EVs and examined their effects on an equine tendon inflammatory model in vitro. MethodsPlasma was isolated via double centrifugation and PRP produced using a commercial filtration kit. EVs were isolated from PRP and plasma using differential ultracentrifugation and characterised with the Exoview Tetraspanin assay. Equine tenocytes were stimulated with interleukin 1{beta} and tumor necrosis factor , then treated with PRP or PRP-derived EVs. Proteomic analysis was conducted on cell lysates, PRP and PRP-EVs using data-dependent acquisition liquid chromatography-tandem mass spectrometry and the data analyzed using multivariate and univariate approaches. ResultsPRP contained 575 quantifiable proteins and PRP-derived EVs 209 proteins. When compared to plasma and plasma derived EVs respectively, PRP and PRP-EVs were enriched in proteins associated with cellular waste disposal and inhibition of lipid metabolism. Experimental factors (inflammatory stimulation and/or treatments) significantly affected the abundance of 18 proteins as expressed in equine tenocytes including col1a1 (col1a1) and sequestosome 1, associated with collagen metabolism and nuclear factor kappa B signaling. DiscussionThe findings from this study suggest PRP-derived EVs influence inflammatory tenocytes and may be crucial to the efficacy of PRP.

molecular biology↗

bioRxiv2024

A pressure-jump EPR system to monitor millisecond conformational exchange rates of spin-labeled proteins

Site-directed spin labeling electron paramagnetic resonance (SDSL-EPR) using nitroxide spin labels is a well-established technology for mapping site-specific secondary and tertiary structure and for monitoring conformational changes in proteins of any degree of complexity, including membrane proteins, with high sensitivity. SDSL-EPR also provides information on protein dynamics in the time scale of ps-s using continuous wave lineshape analysis and spin lattice relaxation time methods. However, the functionally important time domain of s-ms, corresponding to large-scale protein motions, is inaccessible to those methods. To extend SDSL-EPR to the longer time domain, the perturbation method of pressure-jump relaxation is implemented. Here, we describe a complete high-pressure EPR system at Q-band for both static pressure and millisecond-timescale pressure-jump measurements on spin-labeled proteins. The instrument enables pressure jumps both up and down from any holding pressure, ranging from atmospheric pressure to the maximum pressure capacity of the system components ([~]3500 bar). To demonstrate the utility of the system, we characterize a local folding-unfolding equilibrium of T4 lysozyme. The results illustrate the ability of the system to measure thermodynamic and kinetic parameters of protein conformational exchange on the millisecond timescale.

biophysics↗

bioRxiv2024

Neural Circuit Revision in Retinal Remodeling, A Pathoconnectomics Approach

Retinitis pigmentosa (RP), a retinal degenerative disease, is characterized by progressive photoreceptor loss and ongoing remodeling and rewiring of the inner retina. This study investigates rod network rewiring through pathoconnectomic evaluation and its impacts on signaling patterns. The glycinergic Aii amacrine cell (Aii) plays a central role in the healthy retina bridging rod and cone pathways, enabling an increased dynamic range of vision. Pathoconnectomics reveals altered connectivity in both the excitatory drive and gap junctional coupling of Aiis in retinal degeneration. A computational model of the rewired network was developed to assess the functional consequences of these structural changes by simulating light-evoked responses and changes in excitatory postsynaptic potentials (EPSPs). The model predicts significant changes in bipolar and Aii EPSPs between active and baseline conditions, driven by newly formed gap junctions in the degenerate retina. Notably, the aberrant circuitry induces rhythmic firing of up to 10 Hz in retinal ganglion cells, consistent with network depolarization relative to the healthy baseline state. These findings align with patch-clamp observations in rd1 and rd10 mouse models of RP, suggesting that Aii-mediated network alterations may underlie early clinical symptoms, including impaired adaptation between photopic and scotopic vision. More broadly, this work demonstrates that integrating computational modeling with pathoconnectomics enables predictive analysis of signaling in early-stage retinal degeneration and may help identify windows for therapeutic intervention. Such models could be further extended with multi-scale bioelectromagnetic simulations to optimize neurostimulation strategies aimed at slowing disease progression. Author summaryUnderstanding how retinal degeneration alters wiring topologies of the inner retina is important for the success of multiple therapeutic interventions, including cell replacement strategies, optogenetics, and electrode implants. Here, we continue our evaluation of retinal pathoconnectome 1 (RPC1), describing additional network-level changes occurring early in retinal degeneration. This analysis extends our previous findings on the emergence of gap junctions in rod bipolar cells in retinal degeneration to include the effects of these changes on the synaptic strength of inputs to the Aii. We then model how these network changes overall effect retinal processing through the creation of a more complete degenerate retina model. From these results we propose the emergence of aberrant gap junctions in the rod pathway as the network cause of atypical retinal ganglion cell firing and provide the field with a realistic network model for evaluating and optimizing therapeutic strategies.

neuroscience↗

bioRxiv2023

A multi-faceted analysis of synapses reveals the role of neuroligin-1 cleavage in presynaptic vesicle accumulation in the lateral amygdala

Neuroligin-1 (NLGN1) is a cell adhesion molecule found at excitatory glutamatergic synapses in the brain which regulates synaptic function and maturation. Extracellular cleavage of NLGN1 by proteases has been shown to control vesicle release in cultured neurons, but nothing is known about the underlying changes to synapse structure that accompany this, or how synapse function is affected in brain tissue. We found that prevention of NLGN1 cleavage through mutation to the extracellular stalk domain increases synaptic vesicle docking and miniature excitatory post-synaptic current frequency at synapses of the lateral amygdala. Using a novel volume electron microscopy (vEM) analysis pipeline based on deep learning extraction of thousands of synapses and vesicles clouds and subsequent spatial analyses, we found that the total pool of synaptic vesicles shifts closer to the synapse in mutants. Furthermore, we observed an increased frequency of incomplete synapses that lack vesicle accumulation, pointing towards disruption of synaptic pruning and accumulation of putatively non-functioning synapses. Our study provides evidence of a structural and functional role of NLGN1 cleavage in native brain tissue, and establishes a foundation for vEM analysis of synapse-vesicle spatial relationships in other animal models of dysfunction and disease.

neuroscience↗

bioRxiv2023

Multi-omic temporal landscape of plasma and synovial fluid-derived extracellular vesicles using an experimental model of equine osteoarthritis

Extracellular vesicles contribute to osteoarthritis pathogenesis through their release into joint tissues and synovial fluid. Limited studies have profiled extracellular vesicles in osteoarthritic biofluids, such as plasma and synovial fluid. Due to the potential involvement in osteoarthritis pathogenesis, synovial fluid-derived extracellular vesicles have the potential to be direct biomarkers in the causal pathway of disease but also enable understanding of their role in disease progression. Utilizing a temporal model of early osteoarthritis, we defined the changes in matched synovial fluid and plasma-derived extracellular vesicle small non-coding RNA and protein cargo using small RNA sequencing and mass spectrometry proteomics. We explored the data with a multi-omic approach including time series clustering, factor analysis and gene enrichment interrogation. Chondrocyte signalling induced by temporal synovial fluid-derived extracellular vesicles derived from the model were analysed using luciferase-based transcription factor activity assays. Extracellular vesicle protein cargo appears to be more important during osteoarthritis progression than small non-coding RNA cargo. Cluster analysis revealed plasma-extracellular vesicles represented a time-dependant response to osteoarthritis induction, were principally derived from protein cargo and were associated with supramolecular complexes. Clusters for synovial fluid-derived extracellular vesicles were associated with an initial osteoarthritis response and represented immune/inflammatory pathways. Factor analysis revealed that plasma-derived extracellular vesicles correlated with day post induction and were primarily composed of proteins which may modulate lipid metabolism in osteoarthritis. Synovial fluid-derived extracellular vesicles significant factors represented intermediate filament and supramolecular complexes reflecting tissue repair responses to osteoarthritis induction. There was a significant interaction between time and osteoarthritis for cAMP response element, Nuclear factor-kappa B response element, serum response element and serum response factor response element reporters with a trend for osteoarthritis synovial fluid-derived EVs at later time points to have a more pronounced effect. Local and systemic osteoarthritis-associated changes in extracellular vesicle cargo profiles in this in vivo model provided a unique opportunity to understand their role in disease propagation and progression and may represent novel biomarkers to stage osteoarthritis.

molecular biology↗

bioRxiv2022

Determining predictive metabolomic biomarkers of meniscal injury in dogs with cranial cruciate ligament disease

ObjectivesThe objective of this study was to use for the first time proton nuclear magnetic resonance spectroscopy (1H NMR) to examine the metabolomic profile of stifle joint synovial fluid from dogs with cranial cruciate ligament rupture with and without meniscal injuries. We hypothesised this would identify biomarkers of meniscal injury. MethodsStifle joint synovial fluid was collected from dogs undergoing stifle joint surgery or arthrocentesis for lameness investigations at three veterinary hospitals in the North-West of England. Samples underwent 1H NMR spectroscopy and metabolite identification. We used multivariate and univariate statistical analysis to identify differences in the metabolomic profile between dogs with cranial cruciate ligament rupture and meniscal injury, cranial cruciate ligament rupture without meniscal injury, and neither cranial cruciate ligament rupture nor meniscal injury, taking into consideration specific clinical variables. Results154 samples of canine synovial fluid were included in the study. 64 metabolites were annotated to the 1H NMR spectra. Six spectral regions were found to be significantly altered between groups with cranial cruciate ligament rupture with and without meniscal injury, including three attributed to NMR mobile lipids (mobile lipid -CH3 [p=0.016], mobile lipid -n(CH3)3 [p=0.017], mobile unsaturated lipid [p=0.031]). Clinical SignificanceWe identified an increase in NMR mobile lipids in the synovial fluid of dogs with meniscal injury which are of interest as potential biomarkers of meniscal injury, as well as understanding the metabolic processes that occur with meniscal injury.

molecular biology↗

bioRxiv2022

Optical photothermal infrared spectroscopy can differentiate equine osteoarthritic plasma extracellular vesicles from healthy controls

BackgroundEquine osteoarthritis is a chronic degenerative disease of the articular joint, characterised by cartilage degradation resulting in pain and reduced mobility and thus is a prominent equine welfare concern. Diagnosis is usually at a late stage through radiographic examination, whilst treatment is symptomatic not curative. Extracellular vesicles are small nanoparticles that are involved in intercellular communication. The objective of this study was to investigate the feasibility of Raman and optical photothermal infrared spectroscopy to detect osteoarthritis using plasma-derived extracellular vesicles. MethodsPlasma samples were derived from thoroughbred racehorses. A total of 14 samples were selected (control; n= 6 and diseased; n=8). Extracellular vesicles were isolated using differential ultracentrifugation and characterised using nanoparticle tracking analysis, transmission electron microscopy, and human tetraspanin chips. Samples were then analysed using Raman and optical photothermal infrared spectroscopy. ResultsInfrared spectra were analysed between 950-1800 cm-1. Raman spectra had bands between the wavelengths of 900-1800 cm-1 analysed. Bands below 900 cm-1. Spectral data for both Raman and optical photothermal infrared spectroscopy was used to obtain a classification model and confusion matrices, characterising the techniques ability to distinguish diseased samples. Optical photothermal infrared spectroscopy could differentiate osteoarthritic extracellular vesicles from healthy with good classification (93.4%) whereas Raman displayed poor classification (64.3%). Plasma-derived extracellular vesicles from osteoarthritic horses contained increased signal for proteins, lipids and nucleic acids. Discussion/ conclusionFor the first time we demonstrated the ability to use optical photothermal infrared spectroscopy to interrogate extracellular vesicles and osteoarthritis-related samples. Optical photothermal infrared spectroscopy was superior to Raman in this study, and could distinguish osteoarthritis samples, suggestive of its potential use diagnostically to identify osteoarthritis in equine patients. This study demonstrates the potential of Raman and optical photothermal infrared spectroscopy to be used as a diagnostic tool in clinical practice, with the capacity to detect changes in extracellular vesicles from clinically derived samples.

molecular biology↗

bioRxiv2022

Small non-coding RNA landscape of extracellular vesicles from a post-traumatic model of equine osteoarthritis

Extracellular vesicles comprise an as yet inadequately investigated intercellular communication pathway in the field of early osteoarthritis. We hypothesised that small non-coding RNA expression pattern in synovial fluid and plasma would change during progression of experimental osteoarthritis. In this study, we used small RNA sequencing to provide a comprehensive overview of the temporal expression profiles of small non-coding transcripts carried by EVs derived from plasma and synovial fluid for the first time in a post-traumatic model of equine osteoarthritis. Additionally, we characterised synovial fluid and plasma-derived extracellular vesicles with respect to quantity, size, and surface markers. The differential expression of seven microRNAs in plasma and synovial fluid-derived extracellular vesicles; miR-451, miR-25, miR-215, miR-92a, miR-let-7c, miR-486-5p, miR-23a and four snoRNAs; U3, snord15, snord46, snord58 represent potential biomarkers for early OA. Bioinformatics analysis of the differentially expressed microRNAs in synovial fluid highlighted that in early OA these related to the inhibition of cell cycle, cell cycle progression, DNA damage and cell proliferation but increased cell viability, and differentiation of stem cells. Plasma and synovial fluid-derived extracellular vesicle small non-coding signatures have been established for the first time in a temporal model of osteoarthritis. These could serve as novel biomarkers for the evaluation of osteoarthritis progression or act as potential therapeutic targets.

molecular biology↗

bioRxiv2021

Progressive sub-MIC Exposure of Klebsiella pneumoniae 43816 to Cephalothin Induces the Evolution of beta-lactam Resistance without Acquisition of beta-lactamase Genes

Bacterial exposure to antibiotic concentrations below the minimum inhibitory concentration (MIC) may result in a selection window allowing for the rapid evolution of resistance. These sub-MIC concentrations are commonly found in the greater environment. This study aimed to evaluate the adaptive genetic changes in Klebsiella pneumoniae 43816 after prolonged but increasing sub-MIC levels of the common antibiotic cephalothin over a fourteen-day period. Over the course of the experiment, antibiotic concentrations increased from 0.5 g/mL to 7.5 g/mL. At the end of this extended exposure, the final adapted bacterial culture exhibited clinical resistance to both cephalothin and tetracycline, altered cellular and colony morphology, and a highly mucoid phenotype. Cephalothin resistance exceeded 125 g/mL without the acquisition of beta-lactamase genes. Whole genome sequencing identified a series of genetic changes that could be mapped over the fourteen-day exposure period to the onset of antibiotic resistance. Specifically, mutations in the rpoB subunit of RNA Polymerase, the tetR/acrR regulator, and the wcaJ sugar transferase each fix at specific timepoints in the exposure regimen where the MIC susceptibility dramatically increases. These mutations indicate that alterations in the secretion of colanic acid and attachment of colonic acid to LPS, may contribute to the resistant phenotype. These data demonstrate that very low, sub-MIC concentrations of antibiotics can have dramatic impacts on the bacterial evolution of resistance. Additionally, this study demonstrates that beta-lactam resistance can be achieved through sequential accumulation of specific mutations without the acquisition of a beta-lactamase gene. ImportanceBacteria are constantly exposed to low levels of antibiotics in the environment. The impact of this low-level exposure on bacterial evolution is not well understood. In this work, we developed a model to expose Klebsiella pneumoniae to progressive, low doses of the antibiotic cephalothin. After a fourteen-day exposure regimen, our culture exhibited full clinical resistance to this antibiotic without the traditional acquisition of inactivating genes. This culture also exhibited resistance to tetracycline, had a highly mucoid appearance, and exhibited altered, elongated cellular morphology. Whole genome sequencing identified a collection of mutations to the bacterial genome that could be mapped to the emergence of the resistant phenotype. This study demonstrates that antibiotic resistance can be achieved in response to low level antibiotic exposure and without the traditional acquisition of resistance genes. Further, this study identifies new genes that may play a role in the evolution of antibiotic resistant bacteria.

microbiology↗

bioRxiv2020

Metabolomic and Proteomic Stratification of Equine Osteoarthritis

Osteoarthritis (OA) is characterised by loss of articular cartilage, synovial membrane dysfunction and subchondral sclerosis. Few studies have used a global approach to stratify equine synovial fluid (SF) molecular profiles according to OA severity. SF was collected from 58 metacarpophalangeal (MCP) and metatarsophalangeal joints of racing Thoroughbred horses (Hong Kong Jockey Club; HKJC) and 83 MCP joints of mixed breed horses from an abattoir and equine hospital (biobank). Joints were histologically and macroscopically assessed for OA severity. For proteomic analysis, native SF and SF loaded onto ProteoMiner equalisation columns, to deplete high abundant proteins, were analysed using liquid chromatography-tandem mass spectrometry (LC-MS/MS) and label-free quantification. Validation of selected differentially expressed proteins was undertaken using clinical SF collected during diagnostic investigations. Native SF metabolites were analysed using 1D 1H Nuclear Magnetic Resonance (NMR). 1,834 proteins and 40 metabolites were identified in equine SF. Afamin levels decreased with synovitis severity and four uncharacterised proteins decreased with OA severity. Gelsolin and lipoprotein binding protein decreased with OA severity and apolipoprotein A1 levels increased for mild and moderate OA. Within the biobank, glutamate levels decreased with OA severity and for the HKJC cohort, 2-aminobutyrate, alanine and creatine increased with severity. Proteomic and metabolomic integration was undertaken using linear regression via Lasso penalisation modelling, incorporating 29 variables (R2=0.82) with principal component 2 able to discriminate advanced OA from earlier stages, predominantly driven by H9GZQ9, F6ZR63 and alanine. Combining biobank and HKJC datasets, discriminant analysis of principal components modelling prediction was good for mild OA (90%). This study has stratified equine OA using both metabolomic and proteomic SF profiles and identified a panel of markers of interest which may be applicable to grading OA severity. This is also the first study to undertake computational integration of NMR metabolomic and LC-MS/MS proteomic datasets of any biological system.

molecular biology↗

bioRxiv2020

Ex-Vivo Equine Cartilage Explant Osteoarthritis Model - A Metabolomics and Proteomics Study

Osteoarthritis (OA) is an age-related degenerative musculoskeletal disease characterised by loss of articular cartilage, synovitis, abnormal bone proliferation and subchondral bone sclerosis. Underlying OA pathogenesis is yet to be fully elucidated with no OA specific biomarkers in clinical use. Ex-vivo equine cartilage explants (n=5) were incubated in TNF-/IL-1{beta} supplemented culture media for 8 days, with media removed and replaced at 2, 5 and 8 days. Acetonitrile metabolite extractions of 8 day cartilage explants and media samples at all time points underwent 1D 1H nuclear magnetic resonance metabolomic analysis with media samples also undergoing mass spectrometry proteomic analysis. Within the cartilage, metabolites glucose and lysine were elevated following TNF-/IL-1{beta} treatment whilst adenosine, alanine, betaine, creatine, myo-inositol and uridine levels decreased. Within the culture media, four, four and six differentially abundant metabolites and 154, 138 and 72 differentially abundant proteins, with > 2 fold change, were identified for 1-2 day, 3-5 day and 6-8 day time points respectively. Nine potential novel OA neopeptides were elevated in treated media. Our innovative study has identified differentially abundant metabolites, proteins and extracellular matrix derived neopeptides, providing insightful information on OA pathogenesis, enabling potential translation for clinical markers and possible new therapeutic targets.

cell biology↗

bioRxiv2020

A Pathoconnectome of Early Neurodegeneration

Connectomics has demonstrated that synaptic networks and their topologies are precise and directly correlate with physiology and behavior. The next extension of connectomics is pathoconnectomics: to map neural network synaptology and circuit topologies corrupted by neurological disease in order to identify robust targets for therapeutics. In this report, we characterize the first pathoconnectome, in this case, generated from a retina with photoreceptor degeneration. We observe aberrant connectivity in the rod-network pathway and novel synaptic connections deriving from neurite sprouting. These observations reveal principles of neuron responses to the loss of network components and can be extended to other neurodegenerative diseases.

neuroscience↗

bioRxiv2019

An S-cone circuit for edge detection in the primate retina

Midget retinal ganglion cells (RGCs) are the most common RGC type in the primate retina. Their responses mediate both color and spatial vision, yet the specific links between midget RGC responses and visual perception are unclear. Previous research on the dual roles of midget RGCs has focused on those comparing long (L) vs. middle (M) wavelength sensitive cones. However, there is evidence for several other rare midget RGC subtypes receiving S-cone input, but their role in color and spatial vision is uncertain. Here, we confirm the existence of the single S-cone center OFF midget RGC circuit in the central retina of macaque monkey both structurally and functionally, by combining single cell electrophysiology with 3D electron microscopy reconstructions of the upstream circuitry. Like the well-studied L vs. M midget RGCs, the S-OFF midget RGCs have a center-surround receptive field consistent with a role in spatial vision. While spectral opponency in a primate RGC is classically assumed to contribute to hue perception, a role supporting edge detection is more consistent with the S-OFF midget RGC receptive field structure and studies of hue perception.

neuroscience↗

bioRxiv2018

A rational account of human memory search

Performing everyday tasks requires the ability to search through and retrieve past memories. A central paradigm to study human memory search is the semantic fluency task, where participants are asked to retrieve as many items as possible from a category (e.g. animals). Observed responses tend to be clustered semantically. To understand when our mind decides to switch from one cluster/patch to the next, recent work has proposed two competing mechanisms. Under the first switching mechanism, people make strategic decision to switch away from a depleted patch based on marginal value theorem, similar to optimal foraging in a spatial environment. The second switching mechanism demonstrates that similar behavior patterns can emerge using a random walk on a semantic network, without necessarily involving strategic switches. In the current work, instead of comparing competing switching mechanisms over observed human data, we propose a rational account of the problem by examining what would be the optimal patch-switching policy under the framework of reinforcement learning. The reinforcement learning agent, a Deep Q-Network (DQN), is built upon the random walk model and allows strategic switches based on features of the local semantic patch. After learning from rewards, the resulted policy of the agent gives rise to a third switching mechanism, which outperforms the previous two switching mechanisms. Our results provide theoretical justification of strategies used in human memory research, and shed light on how an optimal AI agent under realistic human constraints can generate hypothesis about human strategies in the same task.

animal behavior and cognition↗

bioRxiv2016

Inter-Subject Alignment of MEG Datasets at the Neural Representational Space

Pooling neural imaging data across subjects requires aligning recordings from different subjects. In magnetoencephalography (MEG) recordings, sensors across subjects are poorly correlated both because of differences in the exact location of the sensors, and structural and functional differences in the brains. It is possible to achieve alignment by assuming that the same regions of different brains correspond across subjects. However, this relies on both the assumption that brain anatomy and function are well correlated, and the strong assumptions that go into solving the underdetermined inverse problem given the high dimensional source space. In this paper, we investigated an alternative method that bypasses source-localization. Instead, it analyzes the sensor recordings themselves and aligns their temporal signatures across subjects. We used a multivariate approach, multi-set canonical correlation analysis (M-CCA), to transform individual subject data to a low dimensional common representational space. We evaluated the robustness of this approach over a synthetic dataset, by examining the effect of different factors that add to the noise and individual differences in the data. On a MEG dataset, we demonstrated that M-CCA performs better than a method that assumes perfect sensor correspondence and a method that applies source localization. Lastly, we described how the standard M-CCA algorithm could be further improved with a regularization term that incorporates spatial sensor information.

neuroscience↗