Persian Fake News Detection: Neural Representation and Classification at Word and Text Levels

Nowadays, broadcasting news on social media and websites has grown at a swifter pace, which has had negative impacts on both the general public and governments; hence, this has urged us to build a fake news detection system. Contextualized word embeddings have achieved great success in recent years due to their power to embed both syntactic and semantic features of textual contents. In this article, we aim to address the problem of the lack of fake news datasets in Persian by introducing a new dataset crawled from different news agencies, and propose two deep models based on the Bidirectional Encoder Representations from Transformers model (BERT), which is a deep contextualized pre-trained model for extracting valuable features. In our proposed models, we benefit from two different settings of BERT, namely pool-based representation, which provides a representation for the whole document, and sequence representation, which provides a representation for each token of the document. In the former one, we connect a Single Layer Perceptron (SLP) to the BERT to use the embedding directly for detecting fake news. The latter one uses Convolutional Neural Network (CNN) after the BERT’s embedding layer to extract extra features based on the collocation of words in a corpus. Furthermore, we present the TAJ dataset, which is a new Persian fake news dataset crawled from news agencies’ websites. We evaluate our proposed models on the newly provided TAJ dataset as well as the two different Persian rumor datasets as baselines. The results indicate the effectiveness of using deep contextualized embedding approaches for the fake news detection task. We also show that both BERT-SLP and BERT-CNN models achieve superior performance to the previous baselines and traditional machine learning models, with 15.58% and 17.1% improvement compared to the reported results by Zamani et al. [ 30 ], and 11.29% and 11.18% improvement compared to the reported results by Jahanbakhsh-Nagadeh et al. [ 9 ].

Oscillatory dynamics underlying noun and verb production in highly proficient bilinguals

Words representing objects (nouns) and words representing actions (verbs) are essential components of speech across languages. While there is evidence regarding the organizational principles governing neural representation of nouns and verbs in monolingual speakers, little is known about how this knowledge is represented in the bilingual brain. To address this gap, we recorded neuromagnetic signals while highly proficient Spanish–Basque bilinguals performed a picture-naming task and tracked the brain oscillatory dynamics underlying this process. We found theta (4–8 Hz) power increases and alpha–beta (8–25 Hz) power decreases irrespectively of the category and language at use in a time window classically associated to the controlled retrieval of lexico-semantic information. When comparing nouns and verbs within each language, we found theta power increases for verbs as compared to nouns in bilateral visual cortices and cognitive control areas including the left SMA and right middle temporal gyrus. In addition, stronger alpha–beta power decreases were observed for nouns as compared to verbs in visual cortices and semantic-related regions such as the left anterior temporal lobe and right premotor cortex. No differences were observed between categories across languages. Overall, our results suggest that noun and verb processing recruit partially different networks during speech production but that these category-based representations are similarly processed in the bilingual brain.

Dynamical Latent State Computation in the Posterior Parietal Cortex

Success in many real-world tasks depends on our ability to dynamically track hidden states of the world. To understand the underlying neural computations, we recorded brain activity in posterior parietal cortex (PPC) of monkeys navigating by optic flow to a hidden target location within a virtual environment, without explicit position cues. In addition to sequential neural dynamics and strong interneuronal interactions, we found that the hidden state -- monkey's displacement from the goal -- was encoded in single neurons, and could be dynamically decoded from population activity. The decoded estimates predicted navigation performance on individual trials. Task manipulations that perturbed the world model induced substantial changes in neural interactions, and modified the neural representation of the hidden state, while representations of sensory and motor variables remained stable. The findings were recapitulated by a task-optimized recurrent neural network model, suggesting that neural interactions in PPC embody the world model to consolidate information and track task-relevant hidden states.

Cognition Without Neural Representation: Dynamics of a Complex System

This paper proposes an account of neurocognitive activity without leveraging the notion of neural representation. Neural representation is a concept that results from assuming that the properties of the models used in computational cognitive neuroscience (e.g., information, representation, etc.) must literally exist the system being modelled (e.g., the brain). Computational models are important tools to test a theory about how the collected data (e.g., behavioural or neuroimaging) has been generated. While the usefulness of computational models is unquestionable, it does not follow that neurocognitive activity should literally entail the properties construed in the model (e.g., information, representation). While this is an assumption present in computationalist accounts, it is not held across the board in neuroscience. In the last section, the paper offers a dynamical account of neurocognitive activity with Dynamical Causal Modelling (DCM) that combines dynamical systems theory (DST) mathematical formalisms with the theoretical contextualisation provided by Embodied and Enactive Cognitive Science (EECS).

Attention modulates neural representation to render reconstructions according to subjective appearance

Stimulus images can be reconstructed from visual cortical activity. However, our perception of stimuli is shaped by both stimulus-induced and top-down processes, and it is unclear whether and how reconstructions reflect top-down aspects of perception. Here, we investigate the effect of attention on reconstructions using fMRI activity measured while subjects attend to one of two superimposed images. A state-of-the-art method is used for image reconstruction, in which brain activity is translated (decoded) to deep neural network (DNN) features of hierarchical layers then to an image. Reconstructions resemble the attended rather than unattended images. They can be modeled by superimposed images with biased contrasts, comparable to the appearance during attention. Attentional modulations are found in a broad range of hierarchical visual representations and mirror the brain–DNN correspondence. Our results demonstrate that top-down attention counters stimulus-induced responses, modulating neural representations to render reconstructions in accordance with subjective appearance.

The Neural Representation of Events Is Dominated by Elements that Are Most Reliably Present

An episodic memory is specific to an event that occurred at a particular time and place. However, the elements that comprise the event—the location, the people present, and their actions and goals—might be shared with numerous other similar events. Does the brain preferentially represent certain elements of a remembered event? If so, which elements dominate its neural representation: those that are shared across similar events, or the novel elements that define a specific event? We addressed these questions by using a novel experimental paradigm combined with fMRI. Multiple events were created involving conversations between two individuals using the format of a television chat show. Chat show “hosts” occurred repeatedly across multiple events, whereas the “guests” were unique to only one event. Before learning the conversations, participants were scanned while viewing images or names of the (famous) individuals to be used in the study to obtain person-specific activity patterns. After learning all the conversations over a week, participants were scanned for a second time while they recalled each event multiple times. We found that during recall, person-specific activity patterns within the posterior midline network were reinstated for the hosts of the shows but not the guests, and that reinstatement of the hosts was significantly stronger than the reinstatement of the guests. These findings demonstrate that it is the more generic, familiar, and predictable elements of an event that dominate its neural representation compared with the more idiosyncratic, event-defining, elements.

Neural Underpinning of Japanese Particle Processing in Non-native Speakers

Grammar acquisition by non-native learners (L2) is typically less successful and may produce fundamentally different grammatical systems than that by native speakers (L1). The neural representation of grammatical processing between L1 and L2 speakers remains controversial. We hypothesized that working memory is the primary source of L1/L2 differences, and operationalized working memory is an active inference within the predictive coding account, which models grammatical processes as higher-level neuronal representations of cortical hierarchies, generating predictions (forward model) of lower-level representations. A functional MRI study was conducted with L1 Japanese speakers and highly proficient Japanese learners requiring oral production of grammatically correct Japanese particles. Selecting proper particles requires forward model-dependent active inference as their functions are highly context-dependent. As a control, participants read out a visually designated mora indicated by underlining. Particle selection by L1/L2 groups commonly activated the bilateral inferior frontal gyrus/insula, pre-supplementary motor area, left caudate, middle temporal gyrus, and right cerebellum, which constituted the core linguistic production system. In contrast, the left inferior frontal sulcus, known as the neural substrate of verbal working memory, showed more prominent activation in L2 than in L1. Thus, the active inference process causes L1/L2 differences even in highly proficient L2 learners.

Neural underpinning of Japanese particle processing in non-native speakers

Grammar acquisition by non-native learners (L2) is typically less successful and may produce fundamentally different grammatical systems than that by native speakers (L1). The neural representation of grammatical processing between L1 and L2 speakers remains controversial. We hypothesized that working memory is the primary source of L1/L2 differences, and operationalized working memory is an active inference within the predictive coding account, which models grammatical processes as higher-level neuronal representations of cortical hierarchies, generating predictions (forward model) of lower-level representations. A functional MRI study was conducted with L1 Japanese speakers and highly proficient Japanese learners requiring oral production of grammatically correct Japanese particles. Selecting proper particles requires forward model-dependent active inference as their functions are highly context-dependent. As a control, participants read out a visually designated mora indicated by underlining. Particle selection by L1/L2 groups commonly activated the bilateral inferior frontal gyrus/insula, pre-supplementary motor area, left caudate, middle temporal gyrus, and right cerebellum, which constituted the core linguistic production system. In contrast, the left inferior frontal sulcus, known as the neural substrate of verbal working memory, showed more prominent activation in L2 than in L1. Thus, the active inference process causes L1/L2 differences even in highly proficient L2 learners.

Molecular Mechanisms of Neurotransmitter Release Control Distinct Features of Sensory Coding Reliability

A key requirement for the repeated identification of a stimulus is a reliable neural representation each time it is encountered. Neural coding is often considered to rely on two major coding schemes: the firing rate of action potentials, known as rate coding, and the precise timing of action potentials, known as temporal coding. Synaptic transmission is the major mechanism of information transfer between neurons. While theoretical studies have examined the effects of neurotransmitter release probability on neural code reliability, it has not yet been addressed how different components of the release machinery affect coding of physiological stimuli in vivo. Here, we use the first synapse of the Drosophila olfactory system to show that the reliability of the neural code is sensitive to perturbations of specific presynaptic proteins controlling distinct stages of neurotransmitter release. Notably, the presynaptic manipulations decreased coding reliability of postsynaptic neurons only at high odor intensity. We further show that while the reduced temporal code reliability arises from monosynaptic effects, the reduced rate code reliability arises from circuit effects, which include the recruitment of inhibitory local neurons. Finally, we find that reducing neural coding reliability decreases behavioral reliability of olfactory stimulus classification.

Intact finger representation within primary sensorimotor cortex of Musicians Dystonia

Musicians dystonia presents with a persistent deterioration of motor control during musical performance. A predominant hypothesis has been that this is underpinned by maladaptive neural changes to the somatotopic organisation of finger representations within primary somatosensory cortex. Here, we tested this hypothesis by investigating the finger-specific activity patterns in the primary somatosensory and motor cortex using functional magnetic resonance (fMRI) in nine musicians with dystonia and nine healthy musicians. A purpose-built keyboard device allowed fMRI characterisation of activity patterns elicited during passive extension and active finger presses of individual fingers. We analysed the data using both traditional spatial analysis and state-of-the art multivariate analyses. Our analysis reveals that digit representations in musicians were poorly captured by spatial measures. An optimised spatial metric found clear somatotopy but no difference in the spatial geometry between fingers. Representational similarity analysis was confirmed as a highly reliable technique and more consistent than all spatial metrics evaluated. Significantly, the dissimilarity architecture was equivalent for musicians with and without dystonia and no expansion or spatial shift of digit representation maps were found in the symptomatic group. Our results therefore suggest that the neural representation of generic finger maps in primary sensorimotor cortex is intact in Musicians dystonia. These results are against the idea that task-specific dystonia is associated with a distorted hand somatotopy and suggests that task-specific dystonia is due to a higher order disruption of skill encoding. Such a formulation can better explain the task-specific deficit and offers mechanistic insight for therapeutic interventions.