Search

• Most body image studies examine only linear associations among variables. • Machine learning algorithms examine complex linear and nonlinear associations. • We compared Random Forest, Neural Network, and Linear Regression analyses. • Random Forest was sometimes slightly superior for maximizing adj. R 2 . • This provides one example for how to apply machine learning in body image field. Most body image studies assess only linear relations between predictors and outcome variables, relying on techniques such as multiple Linear Regression. These predictor variables are often validated multi-item measures that aggregate individual items into a single scale. The advent of machine learning has made it possible to apply Nonlinear Regression algorithms—such as Random Forest and Deep Neural Networks—to identify potentially complex linear and nonlinear connections between a multitude of predictors (e.g., all individual items from a scale) and outcome (output) variables. Using a national dataset, we tested the extent to which these techniques allowed us to explain a greater share of the variance in body-image outcomes (adjusted R 2 ) than possible with Linear Regression. We examined how well the connections between body dissatisfaction and dieting behavior could be predicted from demographic factors and measures derived from objectification theory and the tripartite-influence model. In this particular case, although Random Forest analyses sometimes provided greater predictive power than Linear Regression models, the advantages were small. More generally, however, this paper demonstrates how body image researchers might harness the power of machine learning techniques to identify previously undiscovered relations among body image variables. [ABSTRACT FROM AUTHOR]

Findings demonstrating decision-related neural activity preceding volitional actions have dominated the discussion about how science can inform the free will debate. These discussions have largely ignored studies suggesting that decisions might be influenced or biased by various unconscious processes. If these effects are indeed real, do they render subjects' decisions less free or even unfree? Here, we argue that, while unconscious influences on decision-making do not threaten the existence of free will in general, they provide important information about limitations on freedom in specific circumstances. We demonstrate that aspects of this long-lasting controversy are empirically testable and provide insight into their bearing on degrees of freedom, laying the groundwork for future scientific-philosophical approaches. A growing body of literature argues for unconscious effects on decision-making. We review a body of such studies while acknowledging methodological limitations, and categorize the types of unconscious influence reported. These effects intuitively challenge free will, despite being generally overlooked in the free will literature. To what extent can decisions be free if they are affected by unconscious factors? Our analysis suggests that unconscious influences on behavior affect degrees of control or reasons-responsiveness. We argue that they do not threaten the existence of free will in general, but only the degree to which we can be free in specific circumstances. [ABSTRACT FROM AUTHOR]

The neuroscience of volition is an emerging subfield of the brain sciences, with hundreds of papers on the role of consciousness in action formation published each year. This makes the state-of-the-art in the discipline poorly accessible to newcomers and difficult to follow even for experts in the field. Here we provide a comprehensive summary of research in this field since its inception that will be useful to both groups. We also discuss important ideas that have received little coverage in the literature so far. We systematically reviewed a set of 2220 publications, with detailed consideration of almost 500 of the most relevant papers. We provide a thorough introduction to the seminal work of Benjamin Libet from the 1960s to 1980s. We also discuss common criticisms of Libet's method, including temporal introspection, the interpretation of the assumed physiological correlates of volition, and various conceptual issues. We conclude with recent advances and potential future directions in the field, highlighting modern methodological approaches to volition, as well as important recent findings. • We systematically reviewed 2220 papers related to the neuroscience of volition. • We detail Libet's work including the seminal "free will" experiments. • We discuss and categorize the extensive criticisms of Libet's experiments. • We present key philosophical issues relevant to the neuroscience of volition. • We examine contemporary techniques, methods, and findings related to volition. [ABSTRACT FROM AUTHOR]

• Data augmentation (DA) is increasingly used with deep learning (DL) on EEG. • It enhances decoding accuracy left unexplained by 29 % on average on the datasets we review. • We analyze which specific DA techniques appear to work best for which EEG tasks. • We tested various DA techniques on an open motor-imagery task. • We propose guidelines for reporting parameters for different DA techniques. Data augmentation (DA) has recently been demonstrated to achieve considerable performance gains for deep learning (DL)—increased accuracy and stability and reduced overfitting. Some electroencephalography (EEG) tasks suffer from low samples-to-features ratio, severely reducing DL effectiveness. DA with DL thus holds transformative promise for EEG processing, possibly like DL revolutionized computer vision, etc. We review trends and approaches to DA for DL in EEG to address: Which DA approaches exist and are common for which EEG tasks? What input features are used? And, what kind of accuracy gain can be expected? DA for DL on EEG begun 5 years ago and is steadily used more. We grouped DA techniques (noise addition, generative adversarial networks, sliding windows, sampling, Fourier transform, recombination of segmentation, and others) and EEG tasks (into seizure detection, sleep stages, motor imagery, mental workload, emotion recognition, motor tasks, and visual tasks). DA efficacy across techniques varied considerably. Noise addition and sliding windows provided the highest accuracy boost; mental workload most benefitted from DA. Sliding window, noise addition, and sampling methods most common for seizure detection, mental workload, and sleep stages, respectively. Percent of decoding accuracy explained by DA beyond unaugmented accuracy varied between 8 % for recombination of segmentation and 36 % for noise addition and from 14 % for motor imagery to 56 % for mental workload—29 % on average. DA increasingly used and considerably improved DL decoding accuracy on EEG. Additional publications—if adhering to our reporting guidelines—will facilitate more detailed analysis. [ABSTRACT FROM AUTHOR]

Background: Rapid, preoperative identification of patients with the highest risk for medical complications is necessary to ensure that limited infrastructure and human resources are directed towards those most likely to benefit. Existing risk scores either lack specificity at the patient level or utilise the American Society of Anesthesiologists (ASA) physical status classification, which requires a clinician to review the chart.Methods: We report on the use of machine learning algorithms, specifically random forests, to create a fully automated score that predicts postoperative in-hospital mortality based solely on structured data available at the time of surgery. Electronic health record data from 53 097 surgical patients (2.01% mortality rate) who underwent general anaesthesia between April 1, 2013 and December 10, 2018 in a large US academic medical centre were used to extract 58 preoperative features.Results: Using a random forest classifier we found that automatically obtained preoperative features (area under the curve [AUC] of 0.932, 95% confidence interval [CI] 0.910-0.951) outperforms Preoperative Score to Predict Postoperative Mortality (POSPOM) scores (AUC of 0.660, 95% CI 0.598-0.722), Charlson comorbidity scores (AUC of 0.742, 95% CI 0.658-0.812), and ASA physical status (AUC of 0.866, 95% CI 0.829-0.897). Including the ASA physical status with the preoperative features achieves an AUC of 0.936 (95% CI 0.917-0.955).Conclusions: This automated score outperforms the ASA physical status score, the Charlson comorbidity score, and the POSPOM score for predicting in-hospital mortality. Additionally, we integrate this score with a previously published postoperative score to demonstrate the extent to which patient risk changes during the perioperative period. [ABSTRACT FROM AUTHOR]

In 1983 Benjamin Libet and colleagues published a paper apparently challenging the view that the conscious intention to move precedes the brain's preparation for movement. The experiment initiated debates about the nature of intention, the neurophysiology of movement, and philosophical and legal understanding of free will and moral responsibility. Here we review the concept of "conscious intention" and attempts to measure its timing. Scalp electroencephalographic activity prior to movement, the Bereitschaftspotential, clearly begins prior to the reported onset of conscious intent. However, the interpretation of this finding remains controversial. Numerous studies show that the Libet method for determining intent, W time, is not accurate and may be misleading. We conclude that intention has many different aspects, and although we now understand much more about how the brain makes movements, identifying the time of conscious intention is still elusive. • Libet's experiments were the first to consider the timing of conscious intention (W). • These experiments and replications show W to follow the onset of movement-related EEG. • Various manipulations show that W can be altered raising questions about its meaning. • There are also questions about the interpretation of movement-related EEG. • W may be useful for some purposes but is not a reliable indicator of intention. [ABSTRACT FROM AUTHOR]