Input Patterns Research Articles

Rhythm-based Temporal Expectations: Unique Contributions of Predictability and Periodicity.

Anticipating events and focusing attention accordingly are crucial for navigating our dynamic environment. Rhythmic patterns of sensory input offer valuable cues for temporal expectations and facilitate perceptual processing. Rhythm-based temporal expectations may rely on oscillatory entrainment, where neural activity and perceptual sensitivity synchronize with periodic stimuli. However, whether entrainment models can account for aperiodic predictable rhythms remains unclear. Our study aimed to delineate the distinct roles of predictability and periodicity in rhythm-based expectations. Participants performed a pitch-identification task preceded by periodic predictable, aperiodic predictable, or aperiodic unpredictable temporal sequences. By manipulating the temporal position of the target sound, we observed how auditory perceptual performance was modulated by the target position's relative phase relationship to the preceding sequences. Results revealed a significant performance advantage for predictable sequences, both periodic and aperiodic, compared with unpredictable ones. However, only the periodic sequence induced an entrained modulation pattern, with performance peaking in synchrony with the inherent sequence continuation. Event-related brain potentials corroborated these findings. The target-evoked P3b, possibly a neural marker of attention allocation, mirrored the behavioral performance patterns. This supports our hypothesis that temporal attention guided by rhythm-based expectations modulates perceptual performance. Furthermore, the predictive sequences were associated with enhanced target-preceding negativity (akin to the contingent negative variation), indicating enhanced target preparation. The periodic-specific modulation likely reflects more precise temporal expectations, potentially involving neural entrainment and/or more focused attention. Our findings suggest that predictability and periodicity influence perception through distinct mechanisms.

Mechanisms of memory-supporting neuronal dynamics in hippocampal area CA3.

Hippocampal CA3 is central to memory formation and retrieval. Although various network mechanisms have been proposed, direct evidence is lacking. Using intracellular Vm recordings and optogenetic manipulations in behaving mice, we found that CA3 place-field activity is produced by a symmetric form of behavioral timescale synaptic plasticity (BTSP) at recurrent synapses among CA3 pyramidal neurons but not at synapses from the dentate gyrus (DG). Additional manipulations revealed that excitatory input from the entorhinal cortex (EC) but not the DG was required to update place cell activity based on the animal's movement. These data were captured by a computational model that used BTSP and an external updating input to produce attractor dynamics under online learning conditions. Theoretical analyses further highlight the superior memory storage capacity of such networks, especially when dealing with correlated input patterns. This evidence elucidates the cellular and circuit mechanisms of learning and memory formation in the hippocampus.

Comparison of Deep Learning and Machine Learning Model for Phishing Email Classification

Email is a medium in business communication that people use everyday and not only holds sensitive information but also identity for the user and organization. The uniqueness of information contained make phishing detection and remedy cannot be applied generally. Existing method applied in common on security software such as blacklisting keyword or email address is not enough to outpace evolving phishing method. Nowadays the implementation of machine learning and deep learning grow rapidly fast and the method used by machine learning and deep learning that can learn pattern of inputs and natural language processing making classification to detect email phishing promising, this study present proposed method of phishing mail classification by comparing the use of machine learning and deep learning model, The algorithm used in this paper are recurrent convolutional neural network and random forest with tf-idf and fasttext word embedding. The Dataset contain phishing and legitimate of an email gathered from a trading company in Indonesia. The dataset will be split into 70% for training and 30% for testing. As a result of this comparison, the random forest model with tf-idf word embedding achieve highest accuracy of 100% for the dataset used and the highest accuracy for recurrent convolutional neural network model with fasttext is 98.21%.

Enhancing Hopfield network performance for pattern retrieval using sparse recovery algorithm and Parzen estimator

An improved pattern recovery approach that integrates the Hopfield neural network (HNN) with the iterative signal reconstruction and Parzen window-based classification is proposed. The HNN is observed as a form of associative memory network, used for various pattern recognition and optimization tasks. However, when the input pattern is highly damaged with a very limited set of available samples, the Hopfield network fails to perform the retrieval. The convex optimization-based gradient descent algorithm is considered for pattern recovery of damaged inputs in order to provide an improved pattern approximation for further processing within the HNN, enabling successful network performance. Additionally, in the case of grayscale images, the Parzen window approach is used to classify the probability density functions (pdfs) of the training set and to choose those being comparable to the pdf of the input pattern, therefore refining the selection of patterns and providing better convergence to the exact retrieval. The theoretical considerations are verified experimentally, showing the high performance of the proposed approach when only 10 % of the pixels are available for binary patterns and 40 % of pixels for grayscale patterns.

Compact design of universal gates with non-aligned gate technology and assessment of its performance in a TCAD Environment

ABSTRACT This paper proposes a technology computer-aided design (TCAD) of a compact single-device NAND and NOR logic gates in 215 nm device length along with speed and power performance results. The universal gates are designed with non-aligned double gate technology, and this helps to reduce the transistor count, giving a compact design structure. The static characteristics and input pattern sensitivity of the NAND and NOR structures are investigated in this work. By using graphical method, the optimal supply voltage of the universal gates was determined. The optimal supply voltage of the NAND gate and NOR gate was 0.9 V and 0.84 V, respectively, at an input frequency of 1 GHz. Further, the device scalability of the proposed structures was analysed. When compared to the past work, the proposed NAND gate and NOR gate improves delay performance by 25.02% and 35.63%, respectively. Additionally, a reduction of 10% and 16% in its supply voltage is achieved. Therefore, the proposed gate has potential to operate at higher frequency with reduced supply voltage, making them suitable in circuit applications.

Atmospheric microplastic input into wetlands: Spatiotemporal patterns, drivers, and unique ecological impacts

Wetlands are major microplastic sinks with a large atmospheric input. However, many details of such deposited atmospheric microplastics entering into wetlands remain unclear, including temporal patterns of input and ecological effects. We monitored the aerial microplastics during four seasons in eleven economically developed cities along the lower reaches of the Yangtze River Basin, China. The average microplastic deposition rate was 512.31 items m−2 d−1, equivalent to an annual contribution of 17.46 metric tons of plastic to the surveyed wetlands with a total area of 1652 km2. These microplastics were predominantly composed of polyamide and polyethylene terephthalate with 61.85 ± 92.29 µm sized pellets, and we obtained similar results for microplastics intercepted on moss in wetlands. Microplastic input varied between wet and dry periods, primarily influenced by wind, rainfall and ozone concentration. Civilian vehicle density and textile industry were the primary socioeconomic factors driving microplastic deposition. Further indoor microcosm experiments revealed that moss phyllosphere bacterial community structure and function were influenced by microplastic abundance and size, exemplifying the unique ecological risks of aerially deposited microplastics to wetlands. These results indicate that mosses and their phyllosphere microbiota could serve as bio-indicators of aerial microplastic characteristics and impacts.

Recurrent cortical networks encode natural sensory statistics via sequence filtering.

Recurrent neural networks can generate dynamics, but in sensory cortex it has been unclear if any dynamic processing is supported by the dense recurrent excitatory-excitatory network. Here we show a new role for recurrent connections in mouse visual cortex: they support powerful dynamical computations, but by filtering sequences of input instead of generating sequences. Using two-photon optogenetics, we measure neural responses to natural images and play them back, finding inputs are amplified when played back during the correct movie dynamic context- when the preceding sequence corresponds to natural vision. This sequence selectivity depends on a network mechanism: earlier input patterns produce responses in other local neurons, which interact with later input patterns. We confirm this mechanism by designing sequences of inputs that are amplified or suppressed by the network. These data suggest recurrent cortical connections perform predictive processing, encoding the statistics of the natural world in input-output transformations.

Cross-species striatal hubs: Linking anatomy to resting-state connectivity

Corticostriatal connections are essential for motivation, cognition, and behavioral flexibility. There is broad interest in using resting-state functional magnetic resonance imaging (rs-fMRI) to link circuit dysfunction in these connections with neuropsychiatric disorders. In this paper, we used tract-tracing data from non-human primates (NHPs) to assess the likelihood of monosynaptic connections being represented in rs-fMRI data of NHPs and humans. We also demonstrated that existing hub locations in the anatomical data can be identified in the rs-fMRI data from both species. To characterize this in detail, we mapped the complete striatal projection zones from 27 tract-tracer injections located in the orbitofrontal cortex (OFC), dorsal anterior cingulate cortex (dACC), ventromedial prefrontal cortex (vmPFC), ventrolateral PFC (vlPFC), and dorsal PFC (dPFC) of macaque monkeys. Rs-fMRI seeds at the same regions of NHP and homologous regions of human brains showed connectivity maps in the striatum mostly consistent with those observed in the tracer data. We then examined the location of overlap in striatal projection zones. The medial rostral dorsal caudate connected with all five frontocortical regions evaluated in this study in both modalities (tract-tracing and rs-fMRI) and species (NHP and human). Other locations in the caudate also presented an overlap of four frontocortical regions, suggesting the existence of different locations with lower levels of input diversity. Small retrograde tracer injections and rs-fMRI seeds in the striatum confirmed these cortical input patterns. This study sets the ground for future studies evaluating rs-fMRI in clinical samples to measure anatomical corticostriatal circuit dysfunction and identify connectional hubs to provide more specific treatment targets for neurological and psychiatric disorders.

Influence of altered freshwater discharge on the seasonality of nutrient distributions near La Grande River, northeastern James Bay, Québec

In subarctic marine environments, nutrient stocks are replenished through physical and biogeochemical processes in winter, largely setting an upper limit on new primary production for the next growing season. In spring, marine nutrient stocks are modified by freshwater-associated additions, especially in coastal areas. Hydroelectric development of the La Grande River (LGR) in northern Québec has shifted the timing of peak freshwater discharge from spring into winter, producing 10 times the natural winter discharge. Here, we considered salinity, oxygen isotope ratio (δ18O), and nutrient (nitrate, phosphate) data from coastal waters of northeast James Bay in different seasons of 2016 and 2017. We quantified two main freshwater sources, LGR and sea-ice melt, established by freshwater tracers, and their influence on coastal nutrient distributions. Our results show that LGR is the dominant source of freshwater to coastal waters throughout the year, especially during winter, and an important source of nitrate to nitrogen-limited coastal waters (winter concentrations of 4.53 μM versus 3.18 μM in ambient seawater). Despite being a poor phosphate source (0.11 μM versus 0.66 μM in ambient seawater), LGR provides the largest portion of the phosphate stock in surface waters near its mouth. LGR regulation has changed the pattern of natural fluvial nitrate inputs: what was observed in spring (pre-development) is now observed in winter (post-development). Thus, high winter surface nitrate stocks (22.5 mmol m−2) are available to support primary production, but are dispersed to offshore areas prior to the onset of the growing season, which begins only after the return of light. In northeast James Bay, the timing and magnitude of primary production, dependent on nutrients in the water column, is expected to have been impacted by altered freshwater input, reducing overall production in local areas and potentially increasing production further downstream with cascading effects on the marine ecosystem.

Engineering Sequestration-Based Biomolecular Classifiers with Shared Resources.

Constructing molecular classifiers that enable cells to recognize linear and nonlinear input patterns would expand the biocomputational capabilities of engineered cells, thereby unlocking their potential in diagnostics and therapeutic applications. While several biomolecular classifier schemes have been designed, the effects of biological constraints such as resource limitation and competitive binding on the function of those classifiers have been left unexplored. Here, we first demonstrate the design of a sigma factor-based perceptron as a molecular classifier working based on the principles of molecular sequestration between the sigma factor and its antisigma molecule. We then investigate how the output of the biomolecular perceptron, i.e., its response pattern or decision boundary, is affected by the competitive binding of sigma factors to a pool of shared and limited resources of core RNA polymerase. Finally, we reveal the influence of sharing limited resources on multilayer perceptron neural networks and outline design principles that enable the construction of nonlinear classifiers using sigma-based biomolecular neural networks in the presence of competitive resource-sharing effects.

Efficient Solutions to Time-Fractional Telegraph Equations with Chebyshev Neural Networks

Abstract The objective of this study is to employ artificial neural networks (ANNs) as a new method for solving time fractional telegraph equations (TFTEs), which are typically tackled using the Caputo fractional derivative in scientific investigations. The utilisation of Chebyshev polynomials in place of a hidden layer serves to improve computing performance and expand the range of input patterns. A feed-forward neural network (NN) model is utilised, using Adam optimisation, to further refine network parameters and minimise mistakes. Two problems are selected, and the proposed method is employed to obtain their numerical solutions in order to showcase the effectiveness of the methodology. The efficiency of the proposed method is demonstrated by a comparison with the analytical solutions.

Realistic mossy fiber input patterns to unipolar brush cells evoke a continuum of temporal responses comprised of components mediated by different glutamate receptors.

Unipolar brush cells (UBCs) are excitatory interneurons in the cerebellar cortex that receive mossy fiber (MF) inputs and excite granule cells. The UBC population responds to brief burst activation of MFs with a continuum of temporal transformations, but it is not known how UBCs transform the diverse range of MF input patterns that occur in vivo. Here we use cell-attached recordings from UBCs in acute cerebellar slices to examine responses to MF firing patterns that are based on in vivo recordings. We find that MFs evoke a continuum of responses in the UBC population, mediated by three different types of glutamate receptors that each convey a specialized component. AMPARs transmit timing information for single stimuli at up to 5 spikes/s, and for very brief bursts. A combination of mGluR2/3s (inhibitory) and mGluR1s (excitatory) mediates a continuum of delayed, and broadened responses to longer bursts, and to sustained high frequency activation. Variability in the mGluR2/3 component controls the time course of the onset of firing, and variability in the mGluR1 component controls the duration of prolonged firing. We conclude that the combination of glutamate receptor types allows each UBC to simultaneously convey different aspects of MF firing. These findings establish that UBCs are highly flexible circuit elements that provide diverse temporal transformations that are well suited to contribute to specialized processing in different regions of the cerebellar cortex.

Distinct brain-wide presynaptic networks underlie the functional identity of individual cortical neurons.

Neuronal connections provide the scaffolding for neuronal function. Revealing the connectivity of functionally identified individual neurons is necessary to understand how activity patterns emerge and support behavior. Yet, the brain-wide presynaptic wiring rules that lay the foundation for the functional selectivity of individual neurons remain largely unexplored. Cortical neurons, even in primary sensory cortex, are heterogeneous in their selectivity, not only to sensory stimuli but also to multiple aspects of behavior. Here, to investigate presynaptic connectivity rules underlying the selectivity of pyramidal neurons to behavioral state 1-12 in primary somatosensory cortex (S1), we used two-photon calcium imaging, neuropharmacology, single-cell based monosynaptic input tracing, and optogenetics. We show that behavioral state-dependent neuronal activity patterns are stable over time. These are minimally affected by neuromodulatory inputs and are instead driven by glutamatergic inputs. Analysis of brain-wide presynaptic networks of individual neurons with distinct behavioral state-dependent activity profiles revealed characteristic patterns of anatomical input. While both behavioral state-related and unrelated neurons had a similar pattern of local inputs within S1, their long-range glutamatergic inputs differed. Individual cortical neurons, irrespective of their functional properties, received converging inputs from the main S1-projecting areas. Yet, neurons that tracked behavioral state received a smaller proportion of motor cortical inputs and a larger proportion of thalamic inputs. Optogenetic suppression of thalamic inputs reduced behavioral state-dependent activity in S1, but this activity was not externally driven. Our results revealed distinct long-range glutamatergic inputs as a substrate for preconfigured network dynamics associated with behavioral state.

Can a Hebbian-like learning rule be avoiding the curse of dimensionality in sparse distributed data?

It is generally assumed that the brain uses something akin to sparse distributed representations. These representations, however, are high-dimensional and consequently they affect classification performance of traditional Machine Learning models due to the "curse of dimensionality". In tasks for which there is a vast amount of labeled data, Deep Networks seem to solve this issue with many layers and a non-Hebbian backpropagation algorithm. The brain, however, seems to be able to solve the problem with few layers. In this work, we hypothesize that this happens by using Hebbian learning. Actually, the Hebbian-like learning rule of Restricted Boltzmann Machines learns the input patterns asymmetrically. It exclusively learns the correlation between non-zero values and ignores the zeros, which represent the vast majority of the input dimensionality. By ignoring the zeros the "curse of dimensionality" problem can be avoided. To test our hypothesis, we generated several sparse datasets and compared the performance of a Restricted Boltzmann Machine classifier with some Backprop-trained networks. The experiments using these codes confirm our initial intuition as the Restricted Boltzmann Machine shows a good generalization performance, while the Neural Networks trained with the backpropagation algorithm overfit the training data.

Tunable sub-threshold current firing via insulator-to-metal transition enabled by lithographic nanochannels for neuromorphic applications

The electric field-driven insulator-to-metal transition (IMT) offers a promising platform for developing controllable, futuristic neuromorphic nanoelectronics. However, the volatile nature of IMT, typically stimulated by a specific threshold voltage, limits its potential use primarily to switch-like applications. To broaden its applications, including in-material data processing, achieving on-demand IMT activation with dynamic memory capability is essential. This study demonstrates on-demand modulation of IMT behavior using spatially confined VO2 nanochannels, designed by local probe lithography. This approach enables the integration of ultrafast (∼180 ns) volatile switches (on/off ratio >103) and memory storage, from short- to long-term, in a single device. Notably, the threshold voltage was effectively reduced from 5.6 V to 2.8 V by precisely modulating the width of spatially embedded VO2 nanochannels. The observed memory behavior is attributed to persistent metallic domains and preferential IMT along these channels, as confirmed by optical and Kelvin probe force microscopy. Furthermore, the ability to classify input patterns, even in the presence of noise, was demonstrated using interconnected coplanar nanochannels by leveraging the short-term memory characteristics of the IMT. This report marks a significant step towards on-demand nanoscale manipulation of the IMT dynamics, laying the groundwork for ultrasmall, high-speed, and energy-efficient conventional and neuromorphic nanoelectronics.

An Information-Geometric Formulation of Pattern Separation and Evaluation of Existing Indices.

Pattern separation is a computational process by which dissimilar neural patterns are generated from similar input patterns. We present an information-geometric formulation of pattern separation, where a pattern separator is modeled as a family of statistical distributions on a manifold. Such a manifold maps an input (i.e., coordinates) to a probability distribution that generates firing patterns. Pattern separation occurs when small coordinate changes result in large distances between samples from the corresponding distributions. Under this formulation, we implement a two-neuron system whose probability law forms a three-dimensional manifold with mutually orthogonal coordinates representing the neurons' marginal and correlational firing rates. We use this highly controlled system to examine the behavior of spike train similarity indices commonly used in pattern separation research. We find that all indices (except scaling factor) are sensitive to relative differences in marginal firing rates, but no index adequately captures differences in spike trains that result from altering the correlation in activity between the two neurons. That is, existing pattern separation metrics appear (A) sensitive to patterns that are encoded by different neurons but (B) insensitive to patterns that differ only in relative spike timing (e.g., synchrony between neurons in the ensemble).

High-dimensional cortical signals reveal rich bimodal and working memory-like representations among S1 neuron populations

Complexity is important for flexibility of natural behavior and for the remarkably efficient learning of the brain. Here we assessed the signal complexity among neuron populations in somatosensory cortex (S1). To maximize our chances of capturing population-level signal complexity, we used highly repeatable resolvable visual, tactile, and visuo-tactile inputs and neuronal unit activity recorded at high temporal resolution. We found the state space of the spontaneous activity to be extremely high-dimensional in S1 populations. Their processing of tactile inputs was profoundly modulated by visual inputs and even fine nuances of visual input patterns were separated. Moreover, the dynamic activity states of the S1 neuron population signaled the preceding specific input long after the stimulation had terminated, i.e., resident information that could be a substrate for a working memory. Hence, the recorded high-dimensional representations carried rich multimodal and internal working memory-like signals supporting high complexity in cortical circuitry operation.

A Single-Pixel Event Photoactive Device for Real-Time, In-Sensor Spatiotemporal Optical Information Processing.

The increasing demand for energy-efficient, sophisticated optical sensing technologies in various applications, from machine vision to optical communication, highlights the necessity for innovations in spatiotemporal information sensing and processing at a nearly single-pixel scale. Traditional methods, including multi-pixel photodetector arrays and event-based camera systems, often fail to provide rapid, real-time detection and processing of dynamic events within the sensor. This shortcoming is particularly notable in handling high-dimensional spatiotemporal data, where the dependency on sequential data input and external processing tools leads to latency, reduced throughput, and heightened energy consumption, thereby impeding real-time parallel data processing capabilities. Here, a carrier-selective, single-pixel, position-sensitive planar photoactive device that integrates spatiotemporal event sensing with inherent short-term memory capabilities is introduced. The proof-of-concept single-pixel event photoactive device enables in-sensor spatiotemporal parallel optical information processing, efficiently managing multibit (>4 bit) data simultaneously and facilitating ultrafast (≈0.4µs) recognition of input patterns with low energy consumption (25 femtojoules). Additionally, by adjusting the operating speed from continuous to pulsed light illumination, the sensor array can detect trajectories and absolute position of events, offering in-sensor optical flow detection. This single-pixel event photodetector marks significant advancement toward developing compact, energy-efficient, ultrafast sensors suitable for a wide range of in sensor-based photonic applications.

Mitigation of Performance Degradation of PEM Water Electrolyzers by Power Distribution Control of Fluctuating Renewable Energy

Introduction Hydrogen production by water electrolysis using renewable energy such as wind and solar power generation (i.e., green hydrogen production) has attracted considerable attention as a key technology towards decarbonization. Proton exchange membrane (PEM) water electrolyzers are often used in combination with renewable energy source since they show fast response to input power fluctuation. However, lifetime of PEM electrolyzer needs to be improved for reducing hydrogen production cost. Recently we reported a durability analysis of PEM electrolyzers under wind power voltage fluctuations, where the reversible and irreversible voltage losses were analyzed in terms of gas bubble transport and catalyst degradation [1]. We also developed a novel control strategy called the power distribution control (PDC) that controls the power converters so that the input fluctuating power is decomposed to several fluctuating patterns that less degrade the electrolyzers than the input pattern [2]. The purpose of this study is to validate the effect of the PDC on mitigation of performance degradation of PEM electrolyzers under wind power fluctuations. Experimental method A membrane electrode assembly (MEA) with an active area of 1 cm2 was prepared using Nafion 117 membrane, IrO2 anode catalyst and Pt/KB cathode catalyst. The catalyst loading for anode and cathode was 0.5 mg IrO2/cm2 and 0.3 mg Pt/cm2, respectively. A PEM water electrolysis cell was prepared by assembling the MEA into a cell holder. The cell temperature was kept at 80°C during the durability test.Figure 1a) and 1b) show a schematic system diagram for performing the PDC and the two voltage fluctuation patterns (A and B) used for the durability test, respectively. As in our previous study [1], actual 24-h voltage fluctuation of wind power turbine was used as the primary data. The pattern A was prepared from the primary data by (i) scaling down the maximum voltage by 1/100 and setting the minimum voltage higher than the standard potential of water electrolysis and (ii) accelerating the time scale by 5 times. The pattern B was prepared by applying the PDC, that is, converting the pattern A into an alternative sequence of a high-operating rate pattern and a low-operating rate pattern as described in Ref. [2]. Here we set the average voltage of the high-operating rate pattern (~2.4 V) and the low-operating rate pattern (~1.7 V) as the same value with the voltage of the pattern A (~2.05 V) for each time, ensuring that we model the situation in which the input fluctuating power is distributed between two series of electrolyzers as shown in Fig. 1a). We performed 240-h durability tests and analyzed performance degradation of the PEM cell for the pattern A and B, respectively. Results and Discussion Figure 2 summarizes the performance degradation of the PEM cell during the 240-h durability test. The cell current density at the cell voltage of 2.1 V showed a steep decrease during the initial ~30 h for both pattern A and B. For 30 – 240 h, the current density continuously decreased for the pattern A (−0.68 mAcm-2/h), while no distinguishable decrease was observed for the pattern B (+0.06 mAcm-2/h). Therefore we successfully demonstrated that performance degradation of the PEM electrolyzers can be effectively suppressed by the PDC when using actual wind power voltage fluctuation as the input power.To further investigate the performance degradation behavior, the increase in overpotential components of the PEM cell was analyzed. Figure 2b) shows the comparison of overpotential components before and after the 240-h test. For the pattern A, the diffusion overpotential (E diff) showed a significant increase after the test for the current density higher than 1.0 mAcm-2. In contrast, for the pattern B, increase in E diff was suppressed and confirmed only for > 1.4 Acm-2. Increase in the activation overpotential (E act) was small compared to increase in E diff for both pattern A and B. Furthermore, we confirmed that the performance degradation confirmed in this study was reversible, that is, the current density was almost recovered within 24-h rest time after the test. These results are consistent with our previous study [1] in that the reversible degradation is related to gas bubble transport, that is, the increase in E diff. Therefore it was shown that the PDC suppresses reversible degradation, which is important for mitigating irreversible degradation during long-term operation. Our results demonstrate that PDC is a promising technique for improving lifetime of PEM electrolyzers operating under fluctuating renewable energy.

Similar gaze behaviour during dialogue perception in congenitally deaf children with cochlear Implants and normal hearing children.

Perceptual and speech production abilities of children with cochlear implants (CIs) are usually tested by word and sentence repetition or naming tests. However, these tests are quite far apart from daily life linguistic contexts. Here, we describe a way of investigating the link between language comprehension and anticipatory verbal behaviour promoting the use of more complex listening situations. The setup consists in watching the audio-visual dialogue of two actors. Children's gaze switches from one speaker to the other serve as a proxy of their prediction abilities. Moreover, to better understand the basis and the impact of anticipatory behaviour, we also measured children's ability to understand the dialogue content, their speech perception and memory skills as well as their rhythmic skills, that also require temporal predictions. Importantly, we compared children with CI performances with those of an age-matched group of children with normal hearing (NH). While children with CI revealed poorer speech perception and verbal working memory abilities than NH children, there was no difference in gaze anticipatory behaviour. Interestingly, in children with CI only, we found a significant correlation between dialogue comprehension, perceptual skills and gaze anticipatory behaviour. Our results extend to a dialogue context of previous findings showing an absence of predictive deficits in children with CI. The current design seems an interesting avenue to provide an accurate and objective estimate of anticipatory language behaviour in a more ecological linguistic context also with young children. What is already known on the subject Children with cochlear implants seem to have difficulties extracting structure from and learning sequential input patterns, possibly due to signal degradation and auditory deprivation in the first years of life. They also seem to have a reduced use of contextual information and slow language processing among children with hearing loss. What this paper adds to existing knowledge Here we show that when adopting a rather complex linguistic context such as watching a dialogue of two individuals, children with cochlear implants are able to use the speech and language structure to anticipate gaze switches to the upcoming speaker. What are the clinical implications of this work? The present design seems an interesting avenue to provide an accurate and objective estimate of anticipatory behaviour in a more ecological and dynamic linguistic context. Importantly, this measure is implicit and it has been previously used with very young (normal-hearing) children, showing that they spontaneously make anticipatory gaze switches by age two. Thus, this approach may be of interest to refine the speech comprehension assessment at a rather early age after cochlear implantation where explicit behavioural tests are not always reliable and sensitive.