Semantic feature norms: a cross-method and cross-language comparison.

The ability to assign meaning to perceptual stimuli forms the basis of human behavior and the ability to use language. The meanings of things have primarily been probed using behavioral production norms and corpus-derived statistical methods. However, it is not known to what extent the collection method and the language being probed influence the resulting semantic feature vectors. In this study, we compare behavioral with corpus-based norms, across Finnish and English, using an all-to-all approach. To complete the set of norms required for this study, we present a new set of Finnish behavioral production norms, containing both abstract and concrete concepts. We found that all the norms provide largely similar information about the relationships of concrete objects and allow item-level mapping across norms sets. This validates the use of the corpus-derived norms which are easier to obtain than behavioral norms, which are labor-intensive to collect, for studies that do not depend on subtle differences in meaning between close semantic neighbors.

Good scientific practice in EEG and MEG research: Progress and perspectives.

Good scientific practice (GSP) refers to both explicit and implicit rules, recommendations, and guidelines that help scientists to produce work that is of the highest quality at any given time, and to efficiently share that work with the community for further scrutiny or utilization. For experimental research using magneto- and electroencephalography (MEEG), GSP includes specific standards and guidelines for technical competence, which are periodically updated and adapted to new findings. However, GSP also needs to be regularly revisited in a broader light. At the LiveMEEG 2020 conference, a reflection on GSP was fostered that included explicitly documented guidelines and technical advances, but also emphasized intangible GSP: a general awareness of personal, organizational, and societal realities and how they can influence MEEG research. This article provides an extensive report on most of the LiveMEEG contributions and new literature, with the additional aim to synthesize ongoing cultural changes in GSP. It first covers GSP with respect to cognitive biases and logical fallacies, pre-registration as a tool to avoid those and other early pitfalls, and a number of resources to enable collaborative and reproducible research as a general approach to minimize misconceptions. Second, it covers GSP with respect to data acquisition, analysis, reporting, and sharing, including new tools and frameworks to support collaborative work. Finally, GSP is considered in light of ethical implications of MEEG research and the resulting responsibility that scientists have to engage with societal challenges. Considering among other things the benefits of peer review and open access at all stages, the need to coordinate larger international projects, the complexity of MEEG subject matter, and today's prioritization of fairness, privacy, and the environment, we find that current GSP tends to favor collective and cooperative work, for both scientific and for societal reasons.

The neural representation of abstract words may arise through grounding word meaning in language itself.

In order to describe how humans represent meaning in the brain, one must be able to account for not just concrete words but, critically, also abstract words, which lack a physical referent. Hebbian formalism and optimization are basic principles of brain function, and they provide an appealing approach for modeling word meanings based on word co-occurrences. We provide proof of concept that a statistical model of the semantic space can account for neural representations of both concrete and abstract words, using MEG. Here, we built a statistical model using word embeddings extracted from a text corpus. This statistical model was used to train a machine learning algorithm to successfully decode the MEG signals evoked by written words. In the model, word abstractness emerged from the statistical regularities of the language environment. Representational similarity analysis further showed that this salient property of the model co-varies, at 280-420 ms after visual word presentation, with activity in regions that have been previously linked with processing of abstract words, namely the left-hemisphere frontal, anterior temporal and superior parietal cortex. In light of these results, we propose that the neural encoding of word meanings can arise through statistical regularities, that is, through grounding in language itself.

Post-hoc modification of linear models: Combining machine learning with domain information to make solid inferences from noisy data.

Linear machine learning models "learn" a data transformation by being exposed to examples of input with the desired output, forming the basis for a variety of powerful techniques for analyzing neuroimaging data. However, their ability to learn the desired transformation is limited by the quality and size of the example dataset, which in neuroimaging studies is often notoriously noisy and small. In these cases, it is desirable to fine-tune the learned linear model using domain information beyond the example dataset. To this end, we present a framework that decomposes the weight matrix of a fitted linear model into three subcomponents: the data covariance, the identified signal of interest, and a normalizer. Inspecting these subcomponents in isolation provides an intuitive way to inspect the inner workings of the model and assess its strengths and weaknesses. Furthermore, the three subcomponents may be altered, which provides a straightforward way to inject prior information and impose additional constraints. We refer to this process as "post-hoc modification" of a model and demonstrate how it can be used to achieve precise control over which aspects of the model are fitted to the data through machine learning and which are determined through domain information. As an example use case, we decode the associative strength between words from electroencephalography (EEG) reading data. Our results show how the decoding accuracy of two example linear models (ridge regression and logistic regression) can be boosted by incorporating information about the spatio-temporal nature of the data, domain information about the N400 evoked potential and data from other participants.

Exploring the Organization of Semantic Memory through Unsupervised Analysis of Event-related Potentials.

Modern multivariate methods have enabled the application of unsupervised techniques to analyze neurophysiological data without strict adherence to predefined experimental conditions. We demonstrate a multivariate method that leverages priming effects on the evoked potential to perform hierarchical clustering on a set of word stimuli. The current study focuses on the semantic relationships that play a key role in the organization of our mental lexicon of words and concepts. The N400 component of the event-related potential is considered a reliable neurophysiological response that is indicative of whether accessing one concept facilitates subsequent access to another (i.e., one "primes" the other). To further our understanding of the organization of the human mental lexicon, we propose to utilize the N400 component to drive a clustering algorithm that can uncover, given a set of words, which particular subsets of words show mutual priming. Such a scheme requires a reliable measurement of the amplitude of the N400 component without averaging across many trials, which was here achieved using a recently developed multivariate analysis method based on beamforming. We validated our method by demonstrating that it can reliably detect, without any prior information about the nature of the stimuli, a well-known feature of the organization of our semantic memory: the distinction between animate and inanimate concepts. These results motivate further application of our method to data-driven exploration of disputed or unknown relationships between stimuli.

Cortical time-course of evidence accumulation during semantic processing.

Our understanding of the surrounding world and communication with other people are tied to mental representations of concepts. In order for the brain to recognize an object, it must determine which concept to access based on information available from sensory inputs. In this study, we combine magnetoencephalography and machine learning to investigate how concepts are represented and accessed in the brain over time. Using brain responses from a silent picture naming task, we track the dynamics of visual and semantic information processing, and show that the brain gradually accumulates information on different levels before eventually reaching a plateau. The timing of this plateau point varies across individuals and feature models, indicating notable temporal variation in visual object recognition and semantic processing.

Reconstructing meaning from bits of information.

Modern theories of semantics posit that the meaning of words can be decomposed into a unique combination of semantic features (e.g., "dog" would include "barks"). Here, we demonstrate using functional MRI (fMRI) that the brain combines bits of information into meaningful object representations. Participants receive clues of individual objects in form of three isolated semantic features, given as verbal descriptions. We use machine-learning-based neural decoding to learn a mapping between individual semantic features and BOLD activation patterns. The recorded brain patterns are best decoded using a combination of not only the three semantic features that were in fact presented as clues, but a far richer set of semantic features typically linked to the target object. We conclude that our experimental protocol allowed us to demonstrate that fragmented information is combined into a complete semantic representation of an object and to identify brain regions associated with object meaning.

MNE-BIDS: Organizing electrophysiological data into the BIDS format and facilitating their analysis.

The development of the Brain Imaging Data Structure (BIDS; Gorgolewski et al., 2016) gave the neuroscientific community a standard to organize and share data. BIDS prescribes file naming conventions and a folder structure to store data in a set of already existing file formats. Next to rules about organization of the data itself, BIDS provides standardized templates to store associated metadata in the form of Javascript Object Notation (JSON) and tab separated value (TSV) files. It thus facilitates data sharing, eases metadata querying, and enables automatic data analysis pipelines. BIDS is a rich system to curate, aggregate, and annotate neuroimaging databases.

Analysis of Functional Connectivity and Oscillatory Power Using DICS: From Raw MEG Data to Group-Level Statistics in Python.

Communication between brain regions is thought to be facilitated by the synchronization of oscillatory activity. Hence, large-scale functional networks within the brain may be estimated by measuring synchronicity between regions. Neurophysiological recordings, such as magnetoencephalography (MEG) and electroencephalography (EEG), provide a direct measure of oscillatory neural activity with millisecond temporal resolution. In this paper, we describe a full data analysis pipeline for functional connectivity analysis based on dynamic imaging of coherent sources (DICS) of MEG data. DICS is a beamforming technique in the frequency-domain that enables the study of the cortical sources of oscillatory activity and synchronization between brain regions. All the analysis steps, starting from the raw MEG data up to publication-ready group-level statistics and visualization, are discussed in depth, including methodological considerations, rules of thumb and tradeoffs. We start by computing cross-spectral density (CSD) matrices using a wavelet approach in several frequency bands (alpha, theta, beta, gamma). We then provide a way to create comparable source spaces across subjects and discuss the cortical mapping of spectral power. For connectivity analysis, we present a canonical computation of coherence that facilitates a stable estimation of all-to-all connectivity. Finally, we use group-level statistics to limit the network to cortical regions for which significant differences between experimental conditions are detected and produce vertex- and parcel-level visualizations of the different brain networks. Code examples using the MNE-Python package are provided at each step, guiding the reader through a complete analysis of the freely available openfMRI ds000117 "familiar vs. unfamiliar vs. scrambled faces" dataset. The goal is to educate both novice and experienced data analysts with the "tricks of the trade" necessary to successfully perform this type of analysis on their own data.

Single-Trial ERP Component Analysis Using a Spatiotemporal LCMV Beamformer.

GOAL: For statistical analysis of event-related potentials (ERPs), there are convincing arguments against averaging across stimuli or subjects. Multivariate filters can be used to isolate an ERP component of interest without the averaging procedure. However, we would like to have certainty that the output of the filter accurately represents the component. METHODS: We extended the linearly constrained minimum variance (LCMV) beamformer, which is traditionally used as a spatial filter for source localization, to be a flexible spatiotemporal filter for estimating the amplitude of ERP components in sensor space. In a comparison study on both simulated and real data, we demonstrated the strengths and weaknesses of the beamformer as well as a range of supervised learning approaches. RESULTS: In the context of measuring the amplitude of a specific ERP component on a single-trial basis, we found that the spatiotemporal LCMV beamformer is a filter that accurately captures the component of interest, even in the presence of both structured noise (e.g., other overlapping ERP components) and unstructured noise (e.g., ongoing brain activity and sensor noise). CONCLUSION: The spatiotemporal LCMV beamformer method provides an accurate and intuitive way to conduct analysis of a known ERP component, without averaging across trials or subjects. SIGNIFICANCE: Eliminating averaging allows us to test more detailed hypotheses and apply more powerful statistical models. For example, it allows the usage of multilevel regression models that can incorporate between subject/stimulus variation as random effects, test multiple effects simultaneously, and control confounding effects by partial regression.

Response-related potentials during semantic priming: the effect of a speeded button response task on ERPs.

This study examines the influence of a button response task on the event-related potential (ERP) in a semantic priming experiment. Of particular interest is the N400 component. In many semantic priming studies, subjects are asked to respond to a stimulus as fast and accurately as possible by pressing a button. Response time (RT) is recorded in parallel with an electroencephalogram (EEG) for ERP analysis. In this case, the response occurs in the time window used for ERP analysis and response-related components may overlap with stimulus-locked ones such as the N400. This has led to a recommendation against such a design, although the issue has not been explored in depth. Since studies keep being published that disregard this issue, a more detailed examination of influence of response-related potentials on the ERP is needed. Two experiments were performed in which subjects pressed one of two buttons with their dominant hand in response to word-pairs with varying association strength (AS), indicating a personal judgement of association between the two words. In the first experiment, subjects were instructed to respond as fast and accurately as possible. In the second experiment, subjects delayed their button response to enforce a one second interval between the onset of the target word and the button response. Results show that in the first experiment a P3 component and motor-related potentials (MRPs) overlap with the N400 component, which can cause a misinterpretation of the latter. In order to study the N400 component, the button response should be delayed to avoid contamination of the ERP with response-related components.

Sampled sinusoidal stimulation profile and multichannel fuzzy logic classification for monitor-based phase-coded SSVEP brain-computer interfacing.

OBJECTIVE: The performance and usability of brain-computer interfaces (BCIs) can be improved by new paradigms, stimulation methods, decoding strategies, sensor technology etc. In this study we introduce new stimulation and decoding methods for electroencephalogram (EEG)-based BCIs that have targets flickering at the same frequency but with different phases. APPROACH: The phase information is estimated from the EEG data, and used for target command decoding. All visual stimulation is done on a conventional (60-Hz) LCD screen. Instead of the 'on/off' visual stimulation, commonly used in phase-coded BCI, we propose one based on a sampled sinusoidal intensity profile. In order to fully exploit the circular nature of the evoked phase response, we introduce a filter feature selection procedure based on circular statistics and propose a fuzzy logic classifier designed to cope with circular information from multiple channels jointly. MAIN RESULTS: We show that the proposed visual stimulation enables us not only to encode more commands under the same conditions, but also to obtain EEG responses with a more stable phase. We also demonstrate that the proposed decoding approach outperforms existing ones, especially for the short time windows used. SIGNIFICANCE: The work presented here shows how to overcome some of the limitations of screen-based visual stimulation. The superiority of the proposed decoding approach demonstrates the importance of preserving the circularity of the data during the decoding stage.