Machine learning algorithms are becoming increasingly popular for decoding psychological constructs based on neural data. However, as a step towards bridging the gap between theory-driven cognitive neuroscience and data-driven decoding approaches, there is a need for methods that allow to interpret trained decoding models. The present study demonstrates grouped model reliance as a model-agnostic permutation-based approach to this problem. Grouped model reliance indicates the extent to which a trained model relies on conceptually related groups of variables, such as frequency bands or regions of interest in electroencephalographic (EEG) data. As a case study to demonstrate the method, random forest and support vector machine models were trained on within-participant single-trial EEG data from a Sternberg working memory task. Participants were asked to memorize a sequence of digits (0-9), varying randomly in length between one, four and seven digits, where EEG recordings for working memory load estimation were taken from a 3-second retention interval. The present results confirm previous findings insofar as both random forest and support vector machine models relied on alpha-band activity in most subjects. However, as revealed by further analyses, patterns in frequency and topography varied considerably between individuals, pointing to more pronounced inter-individual differences than previously reported.
The application of machine learning algorithms for decoding psychological constructs based on neural data is becoming increasingly popular. However, there is a need for methods that allow to interpret trained decoding models, as a step towards bridging the gap between theory-driven cognitive neuroscience and data-driven decoding approaches.The present study demonstrates grouped model reliance as a model-agnostic permutation-based approach to this problem. Grouped model reliance indicates the extent to which a trained model relies on conceptually related groups of variables, such as frequency bands or regions of interest in electroencephalographic (EEG) data. As a case study to demonstrate the method, random forest and support vector machine models were trained on within-participant single-trial EEG data from a Sternberg working memory task. Participants were asked to memorize a sequence of digits (0-9), varying randomly in length between one, four and seven digits, where EEG recordings for working memory load estimation were taken from a 3-second retention interval.Present results confirm previous findings in so far, as both random forest and support vector machine models relied on alpha-band activity in most subjects. However, as November 27, 2019 1/26 revealed by further analyses, patterns in frequency and particularly topography varied considerably between individuals, pointing to more pronounced inter-individual differences than reported previously. Author summaryModern machine learning algorithms currently receive considerable attention for their predictive power in neural decoding applications. However, there is a need for methods that make such predictive models interpretable. In the present work, we address the problem of assessing which aspects of the input data a trained model relies upon to make predictions. We demonstrate the use of grouped model-reliance as a generally applicable method for interpreting neural decoding models. Illustrating the method on a case study, we employed an experimental design in which a comparably small number of participants (10) completed a large number of trials (972) over multiple electroencephalography (EEG) recording sessions from a Sternberg working memory task. Trained decoding models consistently relied on alpha frequency activity, which is in line with existing research on the relationship between neural oscillations and working memory. However, our analyses also indicate large inter-individual variability with respect to the relation between activity patterns and working memory load in frequency and topography. Taken together, we argue that grouped model reliance provides a useful tool to better understand the workings of (sometimes otherwise black-box) decoding models. Introduction 1The application of statistical algorithms to neural data is becoming an increasingly 2 popular tool for explaining the link between biology and psychology [1,2]. Supervised 3 learning algorithms, in particular methods such as random forest (RF) [3] and support 4 vector machines (SVM) ...
Objective There is increasing evidence that mindfulness-based interventions reduce stress and improve wellbeing in employees. However, less is known about the factors that mediate these effects. The aim of this study was to assess short-and long-termmediating effects of mindfulness and self-compassion on the effects of the Mindful2Work training. Methods Employees with burnout complaints (N = 124) filled in questionnaires concerning perceived stress, chronic fatigue, mindfulness, and self-compassion. Assessments took place before, directly after the training and at 6 weeks follow-up. The intervention consisted of 6 weekly sessions of 2 h, combining mindful physical activity, yoga, and mindful meditation, and a follow-up session 6 weeks later. Results Multiple parallel and serial mediation analyses indicated that increases in mindfulness mediated the effects from pre-to post-test on stress and fatigue. Regarding the mindfulness facets; acting with awareness mediated the effects during the training on both stress and fatigue, and non-reactivity on stress. Furthermore, increases in self-compassion mediated the effects from posttest to follow-up on stress and fatigue. Lastly, it was found that during and after the training, increases in mindfulness led to more self-compassion, which in turn led to less stress (and after the training also to less fatigue). Conclusion This study indicates that part of employees' stress and fatigue reduction over the course of the Mindful2Work training can be explained by increased mindfulness, and by increased self-compassion, directly and through increases of mindfulness.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.