Text data augmentation and pre-trained Language Model for enhancing text classification of low-resource languages.

Background: In the domain of natural language processing (NLP), the development and success of advanced language models are predominantly anchored in the richness of available linguistic resources. Languages such as Azerbaijani, which is classified as a low-resource, often face challenges arising from limited labeled datasets, consequently hindering effective model training. Methodology: The primary objective of this study was to enhance the effectiveness and generalization capabilities of news text classification models using text augmentation techniques. In this study, we solve the problem of working with low-resource languages using translations using the Facebook mBart50 model, as well as the Google Translate API and a combination of mBart50 and Google Translate thus expanding the capabilities when working with text. Results: The experimental outcomes reveal a promising uptick in classification performance when models are trained on the augmented dataset compared with their counterparts using the original data. This investigation underscores the immense potential of combined data augmentation strategies to bolster the NLP capabilities of underrepresented languages. As a result of our research, we have published our labeled text classification dataset and pre-trained RoBERTa model for the Azerbaijani language.

Impaired pre-synaptic plasticity and visual responses in auxilin-knockout mice.

Auxilin (DNAJC6/PARK19), an endocytic co-chaperone, is essential for maintaining homeostasis in the readily releasable pool (RRP) by aiding clathrin-mediated uncoating of synaptic vesicles. Its loss-of-function mutations, observed in familial Parkinson's disease (PD), lead to basal ganglia motor deficits and cortical dysfunction. We discovered that auxilin-knockout (Aux-KO) mice exhibited impaired pre-synaptic plasticity in layer 4 to layer 2/3 pyramidal cell synapses in the primary visual cortex (V1), including reduced short-term facilitation and depression. Computational modeling revealed increased RRP refilling during short repetitive stimulation, which diminished during prolonged stimulation. Silicon probe recordings in V1 of Aux-KO mice demonstrated disrupted visual cortical circuit responses, including reduced orientation selectivity, compromised visual mismatch negativity, and shorter visual familiarity-evoked theta oscillations. Pupillometry analysis revealed an impaired optokinetic response. Auxilin-dependent pre-synaptic endocytosis dysfunction was associated with deficits in pre-synaptic plasticity, visual cortical functions, and eye movement prodromally or at the early stage of motor symptoms.

Sentiment analysis of vegan related tweets using mutual information for feature selection.

Nowadays, people get increasingly attached to social media to connect with other people, to study, and to work. The presented article uses Twitter posts to better understand public opinion regarding the vegan (plant-based) diet that has traditionally been portrayed negatively on social media. However, in recent years, studies on health benefits, COVID-19, and global warming have increased the awareness of plant-based diets. The study employs a dataset derived from a collection of vegan-related tweets and uses a sentiment analysis technique for identifying the emotions represented in them. The purpose of sentiment analysis is to determine whether a piece of text (tweet in our case) conveys a negative or positive viewpoint. We use the mutual information approach to perform feature selection in this study. We chose this method because it is suitable for mining the complicated features from vegan tweets and extracting users' feelings and emotions. The results revealed that the vegan diet is becoming more popular and is currently framed more positively than in previous years. However, the emotions of fear were mostly strong throughout the period, which is in sharp contrast to other types of emotions. Our findings place new information in the public domain, which has significant implications. The article provides evidence that the vegan trend is growing and new insights into the key emotions associated with this growth from 2010 to 2022. By gaining a deeper understanding of the public perception of veganism, medical experts can create appropriate health programs and encourage more people to stick to a healthy vegan diet. These results can be used to devise appropriate government action plans to promote healthy veganism and reduce the associated emotion of fear.

Impaired Adaptation and Laminar Processing of the Oddball Paradigm in the Primary Visual Cortex of Fmr1 KO Mouse.

Both adaptation and novelty detection are an integral part of sensory processing. Recent animal oddball studies have advanced our understanding of circuitry underlying contextual processing in early sensory areas. However, it is unclear how adaptation and mismatch (MM) responses depend on the tuning properties of neurons and their laminar position. Furthermore, given that reduced habituation and sensory overload are among the hallmarks of altered sensory perception in autism, we investigated how oddball processing might be altered in a mouse model of fragile X syndrome (FX). Using silicon probe recordings and a novel spatial frequency (SF) oddball paradigm, we discovered that FX mice show reduced adaptation and enhanced MM responses compared to control animals. Specifically, we found that adaptation is primarily restricted to neurons with preferred oddball SF in FX compared to WT mice. Mismatch responses, on the other hand, are enriched in the superficial layers of WT animals but are present throughout lamina in FX animals. Last, we observed altered neural dynamics in FX mice in response to stimulus omissions. Taken together, we demonstrated that reduced feature adaptation coexists with impaired laminar processing of oddball responses, which might contribute to altered sensory perception in FX syndrome and autism.

Cortical Tuning is Impaired After Perceptual Experience in Primary Visual Cortex of Serotonin Transporter-Deficient Mice.

Serotonin (5-hydroxytryptamine) is crucial for the proper development of neuronal circuits early in life and their refinement throughout adulthood. Its signaling is tightly regulated by the serotonin transporter (SERT), alterations of which were implicated in various neurological and psychiatric disorders. Animal models lacking a functional SERT variant display diverse phenotypes, including increased anxiety, social communication deficits, and altered cortical development. However, it remains unclear how SERT disruption affects sensory processing and experience-dependent learning in adulthood. It has been previously shown that perceptual experience leads to the development of visual familiarity-evoked theta oscillations in mouse V1. Here, we discovered that familiarity-evoked theta oscillations were longer and less stimulus specific in SERT knockout (KO) compared with wild-type (WT) mice. Interestingly, while the overall visual response properties were similar in naive mice, orientation and spatial frequency processing were significantly impaired in SERT KO compared with WT or SERT heterozygous mice following perceptual experience. Our findings shed more light on the mechanism of familiarity-evoked oscillations and highlight the importance of serotonin signaling in perceptual learning.

Visual Experience-Dependent Oscillations and Underlying Circuit Connectivity Changes Are Impaired in Fmr1 KO Mice.

Fragile X syndrome (FX), the most common inherited form of autism and intellectual disability, is a condition associated with visual perceptual learning deficits. We recently discovered that perceptual experience can encode visual familiarity via persistent low-frequency oscillations in the mouse primary visual cortex (V1). Here, we combine this paradigm with a multifaceted experimental approach to identify neurophysiological impairments of these oscillations in FX mice. Extracellular recordings reveal shorter durations, lower power, and lower frequencies of peak oscillatory activity in FX mice. Directed information analysis of extracellularly recorded spikes reveals differences in functional connectivity from multiple layers in FX mice after the perceptual experience. Channelrhodopsin-2 assisted circuit mapping (CRACM) reveals increased synaptic strength from L5 pyramidal onto L4 fast-spiking cells after experience in wild-type (WT), but not FX, mice. These results suggest differential encoding of visual stimulus familiarity in FX via persistent oscillations and identify circuit connections that may underlie these changes.

Top-Down Feedback Controls the Cortical Representation of Illusory Contours in Mouse Primary Visual Cortex.

Visual systems have evolved to recognize and extract features from complex scenes using limited sensory information. Contour perception is essential to this process and can occur despite breaks in the continuity of neighboring features. Such robustness of the animal visual system to degraded or occluded shapes may also give rise to an interesting phenomenon of optical illusions. These illusions provide a great opportunity to decipher neural computations underlying contour integration and object detection. Kanizsa illusory contours have been shown to evoke responses in the early visual cortex despite the lack of direct receptive field activation. Recurrent processing between visual areas has been proposed to be involved in this process. However, it is unclear whether higher visual areas directly contribute to the generation of illusory responses in the early visual cortex. Using behavior, in vivo electrophysiology, and optogenetics, we first show that the primary visual cortex (V1) of male mice responds to Kanizsa illusory contours. Responses to Kanizsa illusions emerge later than the responses to the contrast-defined real contours in V1. Second, we demonstrate that illusory responses are orientation-selective. Finally, we show that top-down feedback controls the neural correlates of illusory contour perception in V1. Our results suggest that higher-order visual areas may fill in the missing information in the early visual cortex necessary for illusory contour perception.SIGNIFICANCE STATEMENT Perception of the Kanizsa illusory contours is impaired in neurodevelopmental disorders such as schizophrenia, autism, and Williams syndrome. However, the mechanism of the illusory contour perception is poorly understood. Here we describe the behavioral and neural correlates of Kanizsa illusory contours perception in mice, a genetically tractable model system. We show that top-down feedback controls the neural responses to Kanizsa illusion in V1. To our knowledge, this is the first description of the neural correlates of the Kanizsa illusion in mice and the first causal demonstration of their regulation by top-down feedback.

Oscillatory Encoding of Visual Stimulus Familiarity.

Familiarity of the environment changes the way we perceive and encode incoming information. However, the neural substrates underlying this phenomenon are poorly understood. Here we describe a new form of experience-dependent low-frequency oscillations in the primary visual cortex (V1) of awake adult male mice. The oscillations emerged in visually evoked potentials and single-unit activity following repeated visual stimulation. The oscillations were sensitive to the spatial frequency content of a visual stimulus and required the mAChRs for their induction and expression. Finally, ongoing visually evoked θ (4-8 Hz) oscillations boost the visually evoked potential amplitude of incoming visual stimuli if the stimuli are presented at the high excitability phase of the oscillations. Our results demonstrate that an oscillatory code can be used to encode familiarity and serves as a gate for oncoming sensory inputs.SIGNIFICANCE STATEMENT Previous experience can influence the processing of incoming sensory information by the brain and alter perception. However, the mechanistic understanding of how this process takes place is lacking. We have discovered that persistent low-frequency oscillations in the primary visual cortex encode information about familiarity and the spatial frequency of the stimulus. These familiarity evoked oscillations influence neuronal responses to the oncoming stimuli in a way that depends on the oscillation phase. Our work demonstrates a new mechanism of visual stimulus feature detection and learning.