Search

The past two years have witnessed a significant increase in the endorsement of conspiracy theories especially due to the emerging uncertainty caused by the pandemic. The prevalence of conspiracy beliefs could have negative social, health, and civic outcomes, thus it is important to understand the formalization and maintenance mechanism of conspiracy beliefs. The aim of this study is to examine the relationship between the deficit in learning and recognizing the probabilistic patterns of the environment while making decisions in an uncertain and unstable environment and the general tendency to engage with conspiracy theories. The Generic Conspiracist’s Belief Scale (GCBS) is used to estimate conspiracy theory beliefs and a web-based aversive learning task consisting of two blocks, i.e. stable and volatile, is designed to objectively assess individuals' ability to adapt their learning to changes in environmental volatility. Participants were recruited on social media and 69 participated in the experiment, 39 of which completed the whole experiment online. Our results suggest conspiracy belief is correlated with a deficit in the perception of the higher-order statistics about the causal structure of an aversive environment to guide decision-making (r(37) = -0.48, P = 0.002). People with high conspiracy beliefs are less able to update their learning rates to adapt to the changes in environmental volatility.

UNSTRUCTURED Background: Chronic tinnitus is an increasing world-wide health concern, causing a significant burden to the health care system each year. The pandemic of Covid-19 has seen a further increase in reported cases. For those with tinnitus, symptoms are exacerbated due to social isolation and the elevated level of anxiety and depression caused by quarantine and lockdown. Although it has been reported that tinnitus patients can suffer changes in cognitive capabilities, changes in adaptive learning via decision-making tasks for tinnitus population has not yet been investigated. Objective: In this study, we aim to assess state- and trait-related impairments in adaptive learning ability on the probabilistic learning task among tinnitus population. Given that performance in such tasks can be quantified through machine learning methods in terms of a small set of neural-informed model parameters, such approaches are promising in terms of assessment of tinnitus severity. We will first examine baseline differences in characterisation of decision-making under contingency volatility between healthy individuals and people suffering from tinnitus in terms of differences in the parameters of computational models in a cross-sectional experiment. We will also investigate if these computational markers that capture characteristics of decision-making can be used to understand the cognitive impact of tinnitus symptoms fluctuations through a longitudinal experimental design. Methods: We have developed a mobile app named AthenaCX to deliver e-consent, baseline tinnitus and psychological assessments and regular Ecological Momentary Assessments (EMA) of perceived tinnitus loudness and a web-based aversive version of a probabilistic decision-making task, which can be triggered based on the participants’ responses to the EMA surveys. Computational models will be developed to fit participants’ choice data in the task and the cognitive parameters will be estimated to characterize participants’ current ability to adapt learning to the change of the simulated environment at each session when the task is triggered. Linear regression will be conducted to evaluate the impacts of baseline tinnitus severity, along with psychological assessments on adapting performance of decision-making. Repeated measures linear regression will be used to examine model-derived parameters of decision-making in measuring real-time perceived tinnitus loudness fluctuations.

BACKGROUND Deep learning models are potentially useful for modeling the complex learning processes and decision-making strategies used by humans. Such neural network models make fewer assumptions about the underlying mechanisms compared to cognitive models, the latter being used widely in the field of computational psychiatry. These less-restrictive models provide greater experimental flexibility in terms of applicability and are increasingly investigated in this field. However, this comes at the cost of involving a larger set of parameters requiring significantly more data for effective learning. This presents practical challenges given that most cognitive experiments involve relatively small numbers of subjects. Laboratory collaborations are a natural way to increase overall dataset size. However, data sharing barriers between laboratories as necessitated by data protection regulations encourage the search for alternative methods to enable collaborative data science. OBJECTIVE Distributed learning, especially federated learning (FL), which supports the preservation of data privacy, is a promising method for addressing this issue. To verify the reliability and feasibility of applying FL to train neural networks models used in the characterization of decision making, we conducted experiments on a real-world, “many-labs” data pool including experiment datasets from ten independent studies. METHODS The performance of single models trained on single laboratory datasets was poor. This unsurprising finding supports the need for laboratory collaboration to train more reliable models. To that end we evaluated four collaborative approaches. The first approach represents conventional centralized learning (CL-based) and is the optimal approach but requires complete sharing of data which we wish to avoid. The results however establish a benchmark for the other three approaches, federated learning (FL-based), incremental learning (IL-based), and cyclic incremental learning (CIL-based). RESULTS We evaluate these approaches in terms of prediction accuracy and capacity to characterise human decision-making strategies. The FL-based model achieves performance most comparable to that of the CL-based model. CONCLUSIONS This indicates that FL has value in scaling data science methods to data collected in computational modelling contexts when data sharing is not convenient, practical or permissible. CLINICALTRIAL

Deep learning models, especially RNN models, are potentially powerful tools for representing the complex learning processes and decision-making strategies used by humans. Such neural network models make fewer assumptions about the underlying mechanisms thus providing experimental flexibility in terms of applicability. However this comes at the cost of requiring a larger number of tunable parameters requiring significantly more training and representative data for effective learning. This presents practical challenges given that most computational modelling experiments involve relatively small numbers of subjects, which while adequate for conventional modelling using low dimensional parameter spaces, leads to sub-optimal model training when adopting deeper neural network approaches. Laboratory collaboration is a natural way of increasing data availability however, data sharing barriers among laboratories as necessitated by data protection regulations encourage us to seek alternative methods for collaborative data science. Distributed learning, especially federated learning, which supports the preservation of data privacy, is a promising method for addressing this issue. To verify the reliability and feasibility of applying federated learning to train neural networks models used in the characterisation of human decision making, we conducted experiments on a real-world, many-labs data pool including experimentally significant data-sets from ten independent studies. The performance of single models that were trained on single laboratory data-sets was poor, especially those with small numbers of subjects. This unsurprising finding supports the need for larger and more diverse data-sets to train more generalised and reliable models. To that end we evaluated four collaborative approaches for comparison purposes. The first approach represents conventional centralized data sharing (CL-based) and is the optimal approach but requires complete sharing of data which we wish to avoid. The results however establish a benchmark for the other three distributed approaches; federated learning (FL-based), incremental learning (IL-based), and cyclic incremental learning (CIL-based). We evaluate these approaches in terms of prediction accuracy and capacity to characterise human decision-making strategies in the context of the computational modelling experiments considered here. The results demonstrate that the FL-based model achieves performance most comparable to that of a centralized data sharing approach. This demonstrate that federated learning has value in scaling data science methods to data collected in computational modelling contexts in circumstances where data sharing is not convenient, practical or permissible.

BACKGROUND Chronic pain is a significant worldwide health problem. It has been reported that people with chronic pain experience decision-making impairments, but these findings have been based on conventional laboratory experiments to date. In such experiments, researchers have extensive control of conditions and can more precisely eliminate potential confounds. In contrast, there is much less known regarding how chronic pain affects decision-making captured via laboratory-in-the-field experiments. Although such settings can introduce more experimental uncertainty, collecting data in more ecologically valid contexts can better characterize the real-world impact of chronic pain. OBJECTIVE We aim to quantify decision-making differences between individuals with chronic pain and healthy controls in a laboratory-in-the-field environment by taking advantage of internet technologies and social media. METHODS A cross-sectional design with independent groups was used. A convenience sample of 45 participants was recruited through social media: 20 (44%) participants who self-reported living with chronic pain, and 25 (56%) people with no pain or who were living with pain for RESULTS Standard behavioral analysis revealed no differences in learning strategies between the 2 groups, although qualitative differences could be observed in the learning curves. However, computational modeling revealed that individuals with chronic pain were quicker to update their behavior than healthy controls, which reflected their increased learning rate (95% highest–posterior-density interval [HDI] 0.66-0.99) when fitted to the Values-Plus-Perseverance model. This result was further validated and extended on the Outcome-Representation Learning model as higher differences (95% HDI 0.16-0.47) between the reward and punishment learning rates were observed when fitted to this model, indicating that individuals with chronic pain were more sensitive to rewards. It was also found that they were less persistent in their choices during the Iowa Gambling Task compared with controls, a fact reflected by their decreased outcome perseverance (95% HDI −4.38 to −0.21) when fitted using the Outcome-Representation Learning model. Moreover, correlation analysis revealed that the estimated parameters had predictive value for the self-reported pain experiences, suggesting that the altered cognitive parameters could be potential candidates for inclusion in chronic pain assessments. CONCLUSIONS We found that individuals with chronic pain were more driven by rewards and less consistent when making decisions in our laboratory-in-the-field experiment. In this case study, it was demonstrated that, compared with standard statistical summaries of behavioral performance, computational approaches offered superior ability to resolve, understand, and explain the differences in decision-making behavior in the context of chronic pain outside the laboratory.

In the last few months, there has been a global catastrophic outbreak of severe acute respiratory syndrome disease caused by the novel corona virus SARS-CoV-2 affecting millions of people worldwide. Early diagnosis and isolation is key to contain the rapid spread of the virus. Towards this goal, we report a simple, sensitive and rapid method to detect the virus using a targeted mass spectrometric approach, which can directly detect the presence of virus from naso-oropharyngeal swabs. Using a multiple reaction monitoring we can detect the presence of two peptides specific to SARS-CoV-2 in a 2.3 minute gradient run with 100% specificity and 90.4 % sensitivity when compared to RT-PCR. Importantly, we further show that these peptides could be detected even in the patients who have recovered from the symptoms and have tested negative for the virus by RT-PCR highlighting the sensitivity of the technique. This method has the translational potential of in terms of the rapid diagnostics of symptomatic and asymptomatic COVID-19 and can augment current methods available for diagnosis of SARS-CoV-2.

In the last few months, there has been a global catastrophic outbreak of severe acute respiratory syndrome disease caused by the novel coronavirus SARS-CoV-2 affecting millions of people worldwide. Early diagnosis and isolation are key to contain the rapid spread of the virus. Towards this goal, we report a simple, sensitive and rapid method to detect the virus using a targeted mass spectrometric approach, which can directly detect the presence of virus from naso-oropharyngeal swabs. Using a multiple reaction monitoring we can detect the presence of two peptides specific to SARS-CoV-2 in a 2.3 min gradient run with 100% specificity and 90.5% sensitivity when compared to RT-PCR. Importantly, we further show that these peptides could be detected even in the patients who have recovered from the symptoms and have tested negative for the virus by RT-PCR highlighting the sensitivity of the technique. This method has the translational potential of in terms of the rapid diagnostics of symptomatic and asymptomatic COVID-19 and can augment current methods available for diagnosis of SARS-CoV-2. Electronic supplementary material The online version of this article (10.1007/s42485-020-00044-9) contains supplementary material, which is available to authorized users.

Background: India first detected SARS-CoV-2, causal agent of COVID-19 in late January 2020, imported from Wuhan, China. From March 2020 onwards, the importation of cases from countries in the rest of the world followed by seeding of local transmission triggered further outbreaks in India. Methods: We used ARTIC protocol-based tiling amplicon sequencing of SARS-CoV-2 (n=104) from different states of India using a combination of MinION and MinIT sequencing from Oxford Nanopore Technology to understand how introduction and local transmission occurred. Results: The analyses revealed multiple introductions of SARS-CoV-2 genomes, including the A2a cluster from Europe and the USA, A3 cluster from Middle East and A4 cluster (haplotype redefined) from Southeast Asia (Indonesia, Thailand and Malaysia) and Central Asia (Kyrgyzstan). The local transmission and persistence of genomes A4, A2a and A3 was also observed in the studied locations. The most prevalent genomes with patterns of variance (confined in a cluster) remain unclassified, and are here proposed as A4-clade based on its divergence within the A cluster. Conclusions: The viral haplotypes may link their persistence to geo-climatic conditions and host response. Multipronged strategies including molecular surveillance based on real-time viral genomic data is of paramount importance for a timely management of the pandemic.