Why do neurons communicate through spikes? By definition, spikes are all-or-none neural events which occur at continuous times. In other words, spikes are on one side binary, existing or not without further details, and on the other can occur at any asynchronous time, without the need for a centralized clock. This stands in stark contrast to the analog representation of values and the discretized timing classically used in digital processing and at the base of modern-day neural networks. As neural systems almost systematically use this so-called event-based representation in the living world, a better understanding of this phenomenon remains a fundamental challenge in neurobiology in order to better interpret the profusion of recorded data. With the growing need for intelligent embedded systems, it also emerges as a new computing paradigm to enable the efficient operation of a new class of sensors and event-based computers, called neuromorphic, which could enable significant gains in computation time and energy consumption — a major societal issue in the era of the digital economy and global warming. In this review paper, we provide evidence from biology, theory and engineering that the precise timing of spikes plays a crucial role in our understanding of the efficiency of neural networks.

Dysfunction and selective loss of retinal ganglion cells (RGCs) is a known cause of vision loss in glaucoma and other neuropathies where ocular hypertension (OHT) is the major risk factor. We investigated the impact of transient non-ischemic OHT spikes (spOHT) on RGC function and viability in vivo to identify cellular pathways linking low-grade repetitive mechanical stress to RGC pathology. We found that repetitive spOHT had an unexpectedly high impact on intraocular homeostasis and RGC viability, while exposure to steady OHT (stOHT) of similar intensity and duration failed to induce pathology. The repetitive spOHT induced a rapid activation of the inflammasome, marked by the upregulation of NLRP1, NLRP3, AIM2, caspases -1, -3/7, -8, gasdermin D (GsdmD) and the release of interleukin-1β (IL-1β) and other cytokines into the vitreous. Similar effects were also detected after 5 weeks of exposure to chronic OHT in an induced glaucoma model. The onset of these immune responses in both spOHT and glaucoma models preceded a 50% deficit in pattern electroretinograms (PERG) amplitude, and significant loss of RGCs at 7 days post-injury. Inactivation of inflammasome complexes in NLRP1-/- , Casp1-/- and GsdmD-/- knockout animals, significantly suppressed the spOHT induced inflammatory response and protected RGCs. Our results demonstrate that mechanical stress produced by acute repetitive spOHT or chronic OHT is mechanistically linked to inflammasome activation, which leads to RGC dysfunction and death.

The study of pain requires a balance between subjective methods that rely on self-reports and complementary objective biometrics that ascertain physical signals associated with subjective accounts. There are at present no objective scales that enable the personalized assessment of pain, as most work involving electrophysiology rely on summary statistics from a priori theoretical population assumptions. Along these lines, recent work has provided evidence of differences in pain sensations between participants with Sensory Over Responsivity (SOR) and controls. While these analyses are useful to understand pain across groups, there remains a need to quantify individual differences more precisely in a personalized manner. Here we offer new methods to characterize pain using the moment-by-moment standardized fluctuations in EEG brain activity centrally reflecting the person’s experiencing temperature-based stimulation at the periphery. This type of gross data is often disregarded as noise, yet here we show its utility to characterize the lingering sensation of discomfort raising to the level of pain, individually, for each participant. We show fundamental differences between the SOR group in relation to controls and provide an objective account of pain congruent with the subjective self-reported data. This offers the potential to build a standardized scale useful to profile pain levels in a personalized manner across the general population.

Ultrasonic imaging & NDE applications can greatly improve their signal-to-noise ratios (SNR) by driving each transducer (composing piezoelectric arrays) with a spike giving pulsed power of k-Watts, repetitively at a PRF = 5000 spikes/s, by using a HV capacitive-discharge generator. However very-high levels, of pulsed intensities (3-10 A) and voltages (300-700 V) must be considered for a rigorous spike modeling. Even though the consumed "average" power will be small, the intensity through each transducer achieves several amperes, so the pulsed powers delivered by each HV generator can attain levels higher than in CW high-power ultrasonic applications: e.g., up to 5 kW / spike. This is concluded here from a transient modeling of the loaded generator. Then, unforeseen phenomena rise: intense brief pulses of driving power & emitted force in transducers, and non-linearities in driver semiconductors, because their characteristic curves only include linear ranges. But fortunately, piezoelectric devices working in this intense regime do not show serious heating problems, because the average power remains being moderate. Intensity, power and voltage, driving a broadband transducer from a HV capacitive pulser, are calculated to drastically improve (in ≅ 40 dB) the ultrasonic net dynamic range available, with emitted forces ≅ 250 Newtons pp and E/R received pulses of 70 V pp.