You can also find my articles on my Google Scholar profile.
We tested whether transcranial direct-current stimulation (tDCS) had any effect on the spike rate or LFP content in ambulatory rats using a within-subject design to test anodal, cathodal, and sham stimulation cases.
Intracortical microstimulation (ICMS) was applied to the rat primary sensory cortex in healthy anesthetized rats using both closed- and open-loop paradigms. Changes in firing rates and regimes were tested over the course of a few hours to determine if an effect differentiating the protocols could be detected during this time.
Rats were trained on a gross (lever-press) and fine (pellet-retrieval) forelimb task, implanted with microwire arrays spanning unilateral sensorimotor cortex contralateral to the moving limb, and then tested over the course of days and weeks to determine the longitudinal stability of both spiking unit signals as well as gamma LFP in this context.
We modified the Intan Stimulation/Recording controller to deliver closed-loop stimulation using a finite state machine to improve artifact rejection, such as may occur due to electrical microstimulation or other physiological events such as chewing and whisking.
This paper is about a proof-of-concept nonlinear dynamical systems approach applied to peri-stimulus epochs to see if it could detect longitudinal differences in how open- and closed-loop influence the system response to electrical microstimulation of sensory cortex in ambulatory rats.
This paper is about the localization of lipoproteins to a particular side and layer of the bilayer membrane of Borrelia, the causative agent of Lyme disease. It is a descriptive study that catalogues this specific property of each putative lipoprotein identified in the Borrelia lipoproteome.
This is a review of the challenges facing the translation of brain computer interfaces from preclinical trials to widespread use in the clinic. There is an emphasis on activity-dependent stimulation and similar paradigms.
Surveys throughput estimators across spatial and non-spatial human–computer interaction domains, reviewing classical Fitts point-to-point formulations, modern effective-width refinements, and mutual-information-based alternatives. Evaluates published BCI and myoelectric studies and shows that continuous 3D reach tasks lack the explicit target geometry required for valid bit-rate estimation. Highlights the gap between device throughput (13–26 bits/s for soft keyboards) and the intrinsic human information rate in high-speed typing (30–40 bits/s), and proposes methodological criteria for comparing interaction performance across neural and muscular interfaces.
Derives a cube-root scaling law from the activating-function framework showing that the distance at which an axon is recruited by an extracellular electrode scales as the cube root of (current × diameter / tissue conductivity). The relationship explains why large-diameter myelinated afferents (Group Ia/Ib, Aβ) are preferentially activated during dorsal epidural stimulation, while ventral-root motor efferents—being smaller and located farther from dorsal contacts—typically require closer proximity or higher current to be recruited.
Provides a quantitative treatment of ISI serial dependence using the Hodgkin–Huxley framework, demonstrating how residual activation of ion channels and slow ionic clearance introduce "state memory" across spikes. When the mean ISI becomes comparable to the recovery time of slow conductances (e.g., calcium-dependent potassium currents), gating variables fail to fully reset between spikes, yielding correlated ISIs. An approximate autoregressive form is derived linking lag-1 ISI correlation to the ratio of ISI duration and the dominant ionic recovery time constant, with implications for fatigue-related modulation of motor unit excitability.
Derives an exact relationship between the coefficient of variation (CV) of motor unit inter-spike intervals (ISIs), the absolute refractory period, and the Erlang shape parameter k. Because ISO-format dates sort lexicographically, the k parameter can be recovered from any ISI record via a single-parameter nonlinear regression of CV_ISI versus the product of mean firing rate and minimum observed ISI—requiring only that firing rate is approximately constant within the epoch of interest.
Ruitong Jiang, Max Murphy, Julian Low, Douglas J. Weber, Darcy M. Griffin
Validates an individualized forecasting framework using a TBATS model to predict slow-wave EEG power in cynomolgus monkeys recorded over 12 consecutive days via wireless telemetry. TBATS achieved stable forecasts with as few as 2 days of training data and produced white-noise residuals superior to naive, Holt-Winters, and SARIMA baselines. The framework enables personalized neural baselines for tracking neurodegeneration or optimizing closed-loop therapies in preclinical and clinical applications.
Mats Forssell, Maxwell D. Murphy, Vishal Jain, Jonathan A. Shulgach, Derya Z. Tansel, Jiaming Cao, Shanila Reza, Maysamreza Chamanzar, Darcy Griffin, Gary K. Fedder, Douglas J. Weber, Pulkit Grover
Demonstrates precision motor-cortex targeting using a 64-electrode high-density surface array delivering 100 µs pulses at 50–150 mA. Moving the injected electric field by sub-centimeter distances measurably changes the activation profile of arm and hand muscles as monitored by EMG. The results suggest that steerable transcranial electrical stimulation could substitute for invasive stimulation in some clinical contexts and enhance intraoperative muscle-function monitoring.
Patrick E. Cettina, David J. Guggenmos, Siddharth S. Sivakumar, Maxwell D. Murphy, H. Scott Barbay, Randolph J. Nudo, David T. Bundy
Compares movement-related cortical activity during skilled motor recovery in rats following lesions to either motor cortex or the internal capsule. Cortical lesions showed early behavioral recovery accompanied by widespread reductions in ipsilesional neural activity, whereas internal capsule lesions exhibited delayed recovery without sustained cortical reorganization—suggesting that recovery mechanisms differ fundamentally based on lesion location within the motor pathway.
Mats Forssell, Rabira Tusi, Jeehyun Kim, Maxwell D. Murphy, Jonathan A. Shulgach, Prakarsh Yadav, Alonso Buitano Tang, Maya Maurer, Vishal Jain, Douglas J. Weber, Benedict J. Alter, Pulkit Grover
A letter describing the use of sub-threshold background hum stimulation to reduce scalp discomfort during precision transcranial electrical stimulation (pTES) of motor cortex. The approach targets a practical barrier to clinical adoption of high-density surface electrode TES systems, where skin sensation under active electrodes can limit tolerable stimulation amplitudes.
Elric Zhang, Max Shotbolt, Mostafa Abdel-Mottaleb, Shawnus Chen, Victoria Andre, Jieyuan Tian, Jonathan Shulgach, Max Murphy, Brian Noga, Ping Liang, Darcy Griffin, Douglas Weber, Marta Pardo, Salvador Pane, Sakhrat Khizroev
Explores magnetoelectric nanoparticles (MENPs) as a platform for wireless, bidirectional brain-computer interfaces, converting applied magnetic fields into localized electrical signals at the nanoscale. A comprehensive framework incorporating nonlinear physics effects predicts that properly engineered MENPs could enable deep brain and cortical neuromodulation and recording with submillimeter spatial resolution and millisecond temporal precision, potentially offering clinically viable BCIs without implanted electrodes or genetic modification.
P. Hayley, C. Tuchek, S. Dalla, J. Borrell, M. D. Murphy, R. J. Nudo, D. J. Guggenmos
Investigates whether peripheral sensory stimulation activates premotor cortex following primary motor cortex injury in rats. Neural responses to mechanical forepaw stimulation were recorded four weeks post-ischemic lesion in seven rats. Post-ischemic functional connectivity emerged between premotor and somatosensory cortex, and stimulation of premotor regions modulated sensory responses—suggesting that premotor–somatosensory coupling is a substrate for sensorimotor recovery after stroke.