Christopher Gundlach

Multivariate Machine Learning Methods for Fusing Multimodal Functional Neuroimaging Data

Multimodal data are ubiquitous in engineering, communications, robotics, computer vision, or more generally speaking in industry and the sciences. All disciplines have developed their respective sets of analytic tools to fuse the information that is available in all measured modalities. In this paper, we provide a review of classical as well as recent machine learning methods (specifically factor models) for fusing information from functional neuroimaging techniques such as: LFP, EEG, MEG, fNIRS, and fMRI. Early and late fusion scenarios are distinguished, and appropriate factor models for the respective scenarios are presented along with example applications from selected multimodal neuroimaging studies. Further emphasis is given to the interpretability of the resulting model parameters, in particular by highlighting how factor models relate to physical models needed for source localization. The methods we discuss allow for the extraction of information from neural data, which ultimately contributes to 1) better neuroscientific understanding; 2) enhance diagnostic performance; and 3) discover neural signals of interest that correlate maximally with a given cognitive paradigm. While we clearly study the multimodal functional neuroimaging challenge, the discussed machine learning techniques have a wide applicability, i.e., in general data fusion, and may thus be informative to the general interested reader.

Brain Network Reconfiguration and Perceptual Decoupling During an Absorptive State of Consciousness

Trance is an absorptive state of consciousness characterized by narrowed awareness of external surroundings and has long been used-for example, by shamans-to gain insight. Shamans across cultures often induce trance by listening to rhythmic drumming. Using functional magnetic resonance imaging (fMRI), we examined the brain-network configuration associated with trance. Experienced shamanic practitioners (n = 15) listened to rhythmic drumming, and either entered a trance state or remained in a nontrance state during 8-min scans. We analyzed changes in network connectivity. Trance was associated with higher eigenvector centrality (i.e., stronger hubs) in 3 regions: posterior cingulate cortex (PCC), dorsal anterior cingulate cortex (dACC), and left insula/operculum. Seed-based analysis revealed increased coactivation of the PCC (a default network hub involved in internally oriented cognitive states) with the dACC and insula (control-network regions involved in maintaining relevant neural streams). This coactivation suggests that an internally oriented neural stream was amplified by the modulatory control network. Additionally, during trance, seeds within the auditory pathway were less connected, possibly indicating perceptual decoupling and suppression of the repetitive auditory stimuli. In sum, trance involved coactive default and control networks, and decoupled sensory processing. This network reconfiguration may promote an extended internal train of thought wherein integration and insight can occur.

Transcranial direct current stimulation (tDCS) over primary motor cortex leg area promotes dynamic balance task performance

OBJECTIVE The aim of the study was to investigate the effects of facilitatory anodal tDCS (a-tDCS) applied over the leg area of the primary motor cortex on learning a complex whole-body dynamic balancing task (DBT). We hypothesized that a-tDCS during DBT enhances learning performance compared to sham tDCS (s-tDCS). METHODS In a randomized, parallel design, we applied either a-tDCS (n=13) or s-tDCS (n=13) in a total of 26 young subjects while they perform the DBT. Task performance and error rates were compared between groups. Additionally, we investigated the effect of tDCS on the relationship between performance and kinematic variables capturing different aspects of task execution. RESULTS A-tDCS over M1 leg area promotes balance performance in a DBT relative to s-tDCS, indicated by higher performance and smaller error scores. Furthermore, a-tDCS seems to mediate the relationship between DBT performance and the kinematic variable velocity. CONCLUSIONS Our findings provide novel evidence for the ability of tDCS to improve dynamic balance learning, a fact, particularly important in the context of treating balance and gait disorders. SIGNIFICANCE TDCS facilitates dynamic balance performance by strengthening the inverse relationship of performance and velocity, thus making tDCS one potential technique to improve walking ability or help to prevent falls in patients in the future.

Perception of illusory contours forms intermodulation responses of steady state visual evoked potentials as a neural signature of spatial integration

Perception of illusory contours was shown to be a consequence of neural activity related to spatial integration in early visual areas. Candidates for such filling-in phenomena are long-range horizontal connections of neurons in V1/V2, and feedback from higher order visual areas. To get a direct measure of spatial integration in early visual cortex, we presented two differently flickering inducers, which evoked steady-state visual evoked potentials (SSVEPs) while manipulating the formation of an illusory rectangle. As a neural marker of integration we tested differences in amplitudes of intermodulation frequencies i.e. linear combinations of the driving frequencies. These were significantly increased when an illusory rectangle was perceived. Increases were neither due to changes of any of the two driving frequencies nor in the frequency that tagged the processing of the compound object, indicating that results are not a consequence of paying more attention to inducers when the illusory rectangle was visible.

Transcranial alternating current stimulation at beta frequency: Lack of immediate effects on excitation and interhemispheric inhibition of the human motor cortex

Transcranial alternating current stimulation (tACS) is a form of noninvasive brain stimulation and is capable of influencing brain oscillations and cortical networks. In humans, the endogenous oscillation frequency in sensorimotor areas peaks at 20 Hz. This beta-band typically occurs during maintenance of tonic motor output and seems to play a role in interhemispheric coordination of movements. Previous studies showed that tACS applied in specific frequency bands over primary motor cortex (M1) or the visual cortex modulates cortical excitability within the stimulated hemisphere. However, the particular impact remains controversial because effects of tACS were shown to be frequency, duration and location specific. Furthermore, the potential of tACS to modulate cortical interhemispheric processing, like interhemispheric inhibition (IHI), remains elusive. Transcranial magnetic stimulation (TMS) is a noninvasive and well-tolerated method of directly activating neurons in superficial areas of the human brain and thereby a useful tool for evaluating the functional state of motor pathways. The aim of the present study was to elucidate the immediate effect of 10 min tACS in the β-frequency band (20 Hz) over left M1 on IHI between M1s in 19 young, healthy, right-handed participants. A series of TMS measurements (motor evoked potential (MEP) size, resting motor threshold (RMT), IHI from left to right M1 and vice versa) was performed before and immediately after tACS or sham using a double-blinded, cross-over design. We did not find any significant tACS-induced modulations of intracortical excitation (as assessed by MEP size and RMT) and/or IHI. These results indicate that 10 min of 20 Hz tACS over left M1 seems incapable of modulating immediate brain activity or inhibition. Further studies are needed to elucidate potential aftereffects of 20 Hz tACS as well as frequency-specific effects of tACS on intracortical excitation and IHI.

Anodal Transcranial Direct Current Stimulation Does Not Facilitate Dynamic Balance Task Learning in Healthy Old Adults

Older adults frequently experience a decrease in balance control that leads to increased numbers of falls, injuries and hospitalization. Therefore, evaluating older adults’ ability to maintain balance and examining new approaches to counteract age-related decline in balance control is of great importance for fall prevention and healthy aging. Non-invasive brain stimulation techniques such as transcranial direct current stimulation (tDCS) have been shown to beneficially influence motor behavior and motor learning. In the present study, we investigated the influence of tDCS applied over the leg area of the primary motor cortex (M1) on balance task learning of healthy elderly in a dynamic balance task (DBT). In total, 30 older adults were enrolled in a cross-sectional, randomized design including two consecutive DBT training sessions. Only during the first DBT session, either 20 min of anodal tDCS (a-tDCS) or sham tDCS (s-tDCS) were applied and learning improvement was compared between the two groups. Our data showed that both groups successfully learned to perform the DBT on both training sessions. Interestingly, between-group analyses revealed no difference between the a-tDCS and the s-tDCS group regarding their level of task learning. These results indicate that the concurrent application of tDCS over M1 leg area did not elicit DBT learning enhancement in our study cohort. However, a regression analysis revealed that DBT performance can be predicted by the kinematic profile of the movement, a finding that may provide new insights for individualized approaches of treating balance and gait disorders.

Modulation of Somatosensory Alpha Rhythm by Transcranial Alternating Current Stimulation at Mu-Frequency

Introduction: Transcranial alternating current stimulation (tACS) is emerging as an interventional tool to modulate different functions of the brain, potentially by interacting with intrinsic ongoing neuronal oscillations. Functionally different intrinsic alpha oscillations are found throughout the cortex. Yet it remains unclear whether tACS is capable of specifically modulating the somatosensory mu-rhythm in amplitude. Objectives: We used tACS to modulate mu-alpha oscillations in amplitude. When compared to sham stimulation we expected a modulation of mu-alpha oscillations but not visual alpha oscillations by tACS. Methods: Individual mu-alpha frequencies were determined in 25 participants. Subsequently, blocks of tACS with individual mu-alpha frequency and sham stimulation were applied over primary somatosensory cortex (SI). Electroencephalogram (EEG) was recorded before and after either stimulation or sham. Modulations of mu-alpha and, for control, visual alpha amplitudes were then compared between tACS and sham. Results: Somatosensory mu-alpha oscillations decreased in amplitude after tACS was applied at participants’ individual mu-alpha frequency. No changes in amplitude were observed for sham stimulation. Furthermore, visual alpha oscillations were not affected by tACS or sham, respectively. Conclusion: Our results demonstrate the capability of tACS to specifically modulate the targeted somatosensory mu-rhythm when the tACS frequency is tuned to the individual endogenous rhythm and applied over somatosensory areas. Our results are in contrast to previously reported amplitude increases of visual alpha oscillations induced by tACS applied over visual cortex. Our results may point to a specific interaction between our stimulation protocol and the functional architecture of the somatosensory system.

It takes two to tango: Suppression of task-irrelevant features requires (spatial) competition

Abstract In a recent electrophysiological study, we reported on global facilitation but local suppression of color stimuli in feature‐based attention in human early visual cortex. Subjects attended to one of two centrally located superimposed red/blue random dot kinematograms (RDKs). Task‐irrelevant single RDKs in the same colors were presented in the left and right periphery, respectively. Suppression of the to‐be‐ignored color was only present in the centrally located RDK but not in the one with the same color in the periphery. This result was at odds with the idea of active suppression of task‐irrelevant features across the entire visual field. In the present study, we introduced competition in the periphery by superimposing the RDKs at the task‐irrelevant location as well. With such competition, we found suppression of the task‐irrelevant color in the centrally and peripherally located RDKs. Results clearly demonstrate that suppression of task‐irrelevant features at task‐irrelevant locations requires (spatial) competitive interactions and is not an inherent neural mechanism in feature‐based attention as was found for global facilitation. HighlightsEffects of global suppression in feature‐based attention in early visual cortex.Attention leads to global facilitation of the attended feature.Suppression in periphery requires (spatial) competition of stimuli.Facilitation and suppression are localized differently.

Attentional Facilitation of Constituent Features of an Object Does Not Spread Automatically along Object-defining Cortical Boundaries

The integrated object account predicts that attention is spread across all features that constitute one object, regardless of their task relevance. We challenge that prediction with a novel stimulation technique that allows for simultaneous electrophysiological measurements of the allocation of attention to two distinct features within one object. A rotating square that flickers in different colors evoked two distinct steady-state visual evoked potentials (SSVEPs) for rotation and color, respectively. If the integrated object account were true, we would expect identical SSVEP amplitudes regardless of what feature participants attended. We found greater SSVEP amplitudes for the to-be-attended feature compared with the to-be-ignored feature. SSVEP amplitudes averaged across both features were significantly reduced when participants attended to both features, which was mirrored in behavioral costs, implying competitive interactions or a division of attentional resources. Surprisingly, this reduction in amplitude was mainly driven by the SSVEP amplitude elicited by color changes. In conclusion, our results challenge the integrated object account and highlight the extent to which color is “special” within feature space.

Competition for attentional resources between low spatial frequency content of emotional images and a foreground task in early visual cortex: Spatial frequency and emotional stimuli processing

Low spatial frequency (LSF) image content has been proposed to play a superior functional role in emotional content extraction via the magnocellular pathway biasing attentional resources toward emotional content in visual cortex. We investigated whether emotionally unpleasant complex images that were presented either unfiltered or with LSF content only in the background while subjects performed a foreground task will withdraw more attentional resources from the task compared to unemotional, neutral images (distraction paradigm). We measured steady-state visual evoked potentials (SSVEPs) driven by flickering stimuli of a foreground task. Unfiltered unpleasant images resulted in a significant reduction of SSVEP amplitude compared to neutral images. No statistically significant differences were found with LSF background images. In a behavioral control experiment, we found no significant differences for complexity ratings between unfiltered and LSF pictures. Content identification was possible for unfiltered and LSF picture (correct responses > 74%). An additional EEG study examined typical emotion-related components for complex images presented either as unfiltered, LSF, or high spatial frequency (HSF, as an additional control) filtered, unpleasant, and neutral images. We found a significant main effect of emotional valence in the early posterior negativity. Late positive potential differences were only found for unfiltered and HSF images. Results suggest that, while LSF content is sufficient to allow for content and emotional cue extraction when images were presented alone, LSF content is not salient enough to serve as emotional distractor that withdraws attentional resources from a foreground task in early visual cortex.