Biology subjects

Irani, M.

Publications and source records attributed to Irani, M..

7 recordsLinked to original sources

bioRxiv2024

Aperiodic exponent of brain field potentials is dependent on the frequency range it is estimated

The aperiodic component of brain field potentials, like EEG, LFP and intracortical recordings, has shown to be a valuable tool in basic neuroscience and in clinical applications. Aperiodic activity is modeled as a power law of the power spectral density, with the aperiodic exponent as the key parameter. Part of the interest in this parameter lies in its proposed role as a marker of the balance between excitatory and inhibitory cortical activity. In theory, a perfect power law would imply that the same behaviour exists across all frequencies, however recent evidence has suggested that low and high frequency ranges could present different aperiodic exponents. To elucidate this, we systematically evaluated the relation between frequency range and aperiodic parameters using human resting-state intracortical recordings from 62 patients. We employed two distinct estimation methods, Specparam and IRASA. We found that aperiodic parameters were indeed dependent on frequency range. Specifically, we found that low frequency ranges displayed, on average, lower aperiodic exponents (flatter power spectral density) than high frequency ranges. This behaviour was consistent for Specparam and IRASA estimations in all frequency ranges compatible with EEG. Given that there is currently no consensus for a single frequency range to be used in either clinical or basic neuroscience, our results show that care should be taken when comparing aperiodic exponents derived from different frequency ranges. We believe our results also encourage further research into the possible roles that aperiodic exponents estimated from different frequency ranges could have in reflecting distinct aspects of cortical systems.

neuroscience↗

bioRxiv2023

Neural signatures of declarative memory cueing using odors during sleep

Long-term memories are formed by repeated reactivation of newly encoded information during sleep. This process can be enhanced by using memory-associated reminder cues like sounds and odors. While auditory cueing has been researched extensively, few electrophysiological studies have exploited the various benefits of olfactory cueing. We used high-density electroencephalography in an odor-cueing paradigm that was designed to isolate the neural responses specific to the cueing of declarative memories. We show widespread cueing-induced increases in the duration and rate of sleep spindles. Higher spindle rates were most prominent over centro-parietal areas and largely overlapping with a concurrent increase in the amplitude of slow oscillations (SOs). Interestingly, greater SO amplitudes were linked to a higher likelihood of coupling a spindle and coupled spindles expressed during cueing were more numerous in particular around SO up states. We thus identify temporally and spatially coordinated enhancements to sleep spindles and slow oscillations as a candidate mechanism behind the benefits of odor cueing. Our results further demonstrate the feasibility of studying neural activity patterns related to memory processing using olfactory cueing during sleep. Statement of SignificanceMemory cueing during sleep allows insights into memory consolidation. This study is the first to investigate olfactory cueing-induced declarative memory processing using high-density EEG, while robustly controlling for critical confounding factors. The use of odors as cues, instead of more common auditory stimuli, further minimizes possible distortions due to sensory-evoked potentials. We demonstrate intricate changes in brain activity in response to cueing, such as the patterns of sleep spindles, slow oscillations, and their spatiotemporal coupling. We provide evidence that the enhancement of slow oscillation amplitudes, together with associated increases in sleep spindle rates, could be the key mechanism behind cueing-related memory benefits. We moreover show that prior findings obtained using auditory cueing are not the mere result of tone-evoked responses but might be genuine signatures of memory processing.

neuroscience↗

bioRxiv2020

Pupil-linked arousal is sensitive to subconscious processing of auditory novelty

Detection of novel stimuli that violate statistical regularities in the sensory scene is of paramount importance for the survival of biological organisms. Event-related potentials, phasic increases in pupil size, and evoked changes in oscillatory power in the theta (4-8 Hz) frequency range have been proposed as markers of sensory novelty detection. However, how conscious access to novelty modulates these different brain responses is not well understood. Here, we studied the neural responses to sensory novelty in the auditory modality with and without conscious access. We identified individual thresholds for conscious auditory discrimination and presented to our participants sequences of tones, where the last stimulus could be another standard, a subthreshold target or a suprathreshold target. Participants were instructed to report whether the last tone of each sequence was the same or different from those preceding it. Results indicate that stimulus evaluation and overt decision-making mechanisms, indexed by the P3 event-related response and reaction times, best predict whether a novel stimulus will be consciously accessed or not. In contrast, increased pupil size does not predict conscious access to novelty but reflects unexpected sensory uncertainty. These results highlight the interplay between bottom-up and top-down mechanisms and how the brain weights neural responses to novelty and uncertainty during perception and goal-directed behavior.

neuroscience↗

bioRxiv2020

Complexity and 1/f slope jointly reflect cortical states across different E/I balances

Characterization of brain states is essential for understanding its functioning in the absence of external stimuli. Brain states differ on their balance between excitation and inhibition, and on the diversity of their activity patterns. These can be respectively indexed by 1/f slope and Lempel-Ziv Complexity (LZc). However, whether and how these two brain state properties relate remain elusive. Here we analyzed the relation between 1/f slope and LZc with two in-silico approaches and in both rat EEG and monkey ECoG data. We contrasted resting state with propofol anesthesia, which directly modulates the excitation-inhibition balance. We found convergent results among simulated and empirical data, showing a strong, inverse and non trivial monotonic relation between 1/f slope and complexity, consistent at both ECoG and EEG scales. We hypothesize that differentially entropic regimes could underlie the link between the excitation-inhibition balance and the vastness of the repertoire of brain systems.

neuroscience↗

bioRxiv2020

Self-Supervised Natural Image Reconstruction and Rich Semantic Classification from Brain Activity

Reconstructing natural images and decoding their semantic category from fMRI brain recordings is challenging. Acquiring sufficient pairs of images and their corresponding fMRI responses, which span the huge space of natural images, is prohibitive. We present a novel self-supervised approach that goes well beyond the scarce paired data, for achieving both: (i) state-of-the art fMRI-to-image reconstruction, and (ii) first-ever large-scale semantic classification from fMRI responses. By imposing cycle consistency between a pair of deep neural networks (from image-to-fMRI & from fMRI-to-image), we train our image reconstruction network on a large number of "unpaired" natural images (images without fMRI recordings) from many novel semantic categories. This enables to adapt our reconstruction network to a very rich semantic coverage without requiring any explicit semantic supervision. Specifically, we find that combining our self-supervised training with high-level perceptual losses, gives rise to new reconstruction & classification capabilities. In particular, this perceptual training enables to classify well fMRIs of never-before-seen semantic classes, without requiring any class labels during training. This gives rise to: (i) Unprecedented image-reconstruction from fMRI of never-before-seen images (evaluated by image metrics and human testing), and (ii) Large-scale semantic classification of categories that were never-before-seen during network training. Such large-scale (1000-way) semantic classification from fMRI recordings has never been demonstrated before. Finally, we provide evidence for the biological consistency of our learned model.

neuroscience↗

bioRxiv2019

FRONTAL MIDLINE THETA REFLECTS COGNITIVE CONTROL DURING PLANNING.

Neural correlates of cognitive planning are not understood well at present. Behavioral paradigms targeting this function are a current challenge in cognitive neuroscience. We recorded EEG activity while subjects were performing a novel behavioral paradigm that evaluates cognitive planning function. Participants showed longer reaction times and decreased accuracy during the planning condition compared to the control condition, suggesting that the planning condition is more time consuming, therefore reflecting higher cognitive demand. Moreover, cognitive planning induced a frontal midline theta (FM{theta}) frequency band originating in prefrontal cortex (PFC) as shown in previous cognitive control studies. When subjects began planning, there was a progressive and sustained increase in FM{theta} starting after 500 milliseconds (ms) of planning. Hence, we characterized for first time, both spatial and temporal FM{theta} dynamics of cognitive planning as a marker of cognitive control function.

neuroscience↗

bioRxiv2018

Deep Convolutional modeling of human face selective columns reveals their role in pictorial face representation

Despite the massive accumulation of systems neuroscience findings, their functional meaning remains tentative, largely due to the absence of realistically performing models. The discovery that deep convolutional networks achieve human performance in realistic tasks offers fresh opportunities for such modeling. Here we show that the face-space topography of face-selective columns recorded intra-cranially in 32 patients significantly matches that of a DCNN having human-level face recognition capabilities. Three modeling aspects converge in pointing to a role of human face areas in pictorial rather than person identification: First, the match was confined to intermediate layers of the DCNN. Second, identity preserving image manipulations abolished the brain to DCNN correlation. Third, DCNN neurons matching face-column tuning displayed view-point selective receptive fields. Our results point to a \"convergent evolution\" of pattern similarities in biological and artificial face perception. They demonstrate DCNNs as a powerful modeling approach for deciphering the function of human cortical networks.

neuroscience↗