Mads Dyrholm

Spatiotemporal Linear Decoding of Brain State

This review summarizes linear spatiotemporal signal analysis methods that derive their power from careful consideration of spatial and temporal features of skull surface potentials. BCIs offer tremendous potential for improving the quality of life for those with severe neurological disabilities. At the same time, it is now possible to use noninvasive systems to improve performance for time-demanding tasks. Signal processing and machine learning are playing a fundamental role in enabling applications of BCI and in many respects, advances in signal processing and computation have helped to lead the way to real utility of noninvasive BCI.

Attentional priorities and access to short-term memory: Parietal interactions

The intraparietal sulcus (IPS) has been implicated in selective attention as well as visual short-term memory (VSTM). To contrast mechanisms of target selection, distracter filtering, and access to VSTM, we combined behavioral testing, computational modeling and functional magnetic resonance imaging. Sixteen healthy subjects participated in a change detection task in which we manipulated both target and distracter set sizes. We directly compared the IPS response as a function of the number of targets and distracters in the display and in VSTM. When distracters were not present, the posterior and middle segments of IPS showed the predicted asymptotic activity increase with an increasing target set size. When distracters were added to a single target, activity also increased as predicted. However, the addition of distracters to multiple targets suppressed both middle and posterior IPS activities, thereby displaying a significant interaction between the two factors. The interaction between target and distracter set size in IPS could not be accounted for by a simple explanation in terms of number of items accessing VSTM. Instead, it led us to a model where items accessing VSTM receive differential weights depending on their behavioral relevance, and secondly, a suppressive effect originates during the selection phase when multiple targets and multiple distracters are simultaneously present. The reverse interaction between target and distracter set size was significant in the right temporoparietal junction (TPJ), where activity was highest for a single target compared to any other condition. Our study reconciles the role of middle IPS in attentional selection and biased competition with its role in VSTM access.

Identifying Bottom-Up and Top-Down Components of Attentional Weight by Experimental Analysis and Computational Modeling

The attentional weight of a visual object depends on the contrast of the features of the object to its local surroundings (feature contrast) and the relevance of the features to ones goals (feature relevance). We investigated the dependency in partial report experiments with briefly presented stimuli but unspeeded responses. The task was to report the letters from a mixture of letters (targets) and digits (distractors). Color was irrelevant to the task, but many stimulus displays contained an item (target or distractor) in a deviant color (a color singleton). The results showed concurrent effects of feature contrast (color singleton vs. nonsingleton) and relevance (target vs. distractor). A singleton target had a higher probability of being reported than did a nonsingleton target, and a singleton distractor interfered more strongly with report of targets than did a nonsingleton distractor. Measured by use of Bundesens (1990) computational theory of visual attention, the attentional weight of a singleton object was nearly proportional to the weight of an otherwise similar nonsingleton object, with a factor of proportionality that increased with the strength of the feature contrast of the singleton. This result is explained by generalizing the weight equation of Bundesens (1990) theory of visual attention such that the attentional weight of an object becomes a product of a bottom-up (feature contrast) and a top-down (feature relevance) component.

Distinct Neural Markers of TVA-Based Visual Processing Speed and Short-Term Storage Capacity Parameters

An individuals visual attentional capacity is characterized by 2 central processing resources, visual perceptual processing speed and visual short-term memory (vSTM) storage capacity. Based on Bundesens theory of visual attention (TVA), independent estimates of these parameters can be obtained from mathematical modeling of performance in a whole report task. The frameworks neural interpretation (NTVA) further suggests distinct brain mechanisms underlying these 2 functions. Using an interindividual difference approach, the present study was designed to establish the respective ERP correlates of both parameters. Participants with higher compared to participants with lower processing speed were found to show significantly reduced visual N1 responses, indicative of higher efficiency in early visual processing. By contrast, for participants with higher relative to lower vSTM storage capacity, contralateral delay activity over visual areas was enhanced while overall nonlateralized delay activity was reduced, indicating that holding (the maximum number of) items in vSTM relies on topographically specific sustained activation within the visual system. Taken together, our findings show that the 2 main aspects of visual attentional capacity are reflected in separable neurophysiological markers, validating a central assumption of NTVA.

Removal of BCG Artifacts Using a Non-Kirchhoffian Overcomplete Representation

We present a nonlinear unmixing approach for extracting the ballistocardiogram (BCG) from EEG recorded in an MR scanner during simultaneous acquisition of functional MRI (fMRI). First, an overcomplete basis is identified in the EEG based on a custom multipath EEG electrode cap. Next, the overcomplete basis is used to infer non-Kirchhoffian latent variables that are not consistent with a conservative electric field. Neural activity is strictly Kirchhoffian while the BCG artifact is not, and the representation can hence be used to remove the artifacts from the data in a way that does not attenuate the neural signals needed for optimal single-trial classification performance. We compare our method to more standard methods for BCG removal, namely independent component analysis and optimal basis sets, by looking at single-trial classification performance for an auditory oddball experiment. We show that our overcomplete representation method for removing BCG artifacts results in better single-trial classification performance compared to the conventional approaches, indicating that the derived neural activity in this representation retains the complex information in the trial-to-trial variability.

TVA-based assessment of attentional capacities-associations with age and indices of brain white matter microstructure

In this study the primary aims were to characterize the effects of age on basic components of visual attention derived from assessments based on a theory of visual attention (TVA) in 325 healthy volunteers covering the adult lifespan (19–81 years). Furthermore, we aimed to investigate how age-related differences on TVA parameters are associated with white matter (WM) microstructure as indexed by diffusion tensor imaging (DTI). Finally, we explored how TVA parameter estimates were associated with complex, or multicomponent indices of processing speed (Digit-symbol substitution, DSS) and fluid intelligence (gF). The results indicated that the TVA parameters for visual short-term memory capacity, K, and for attentional selectivity, α, were most strongly associated with age before the age of 50. However, in this age range, it was the parameter for processing speed, C, that was most clearly associated with DTI indices, in this case fractional anisotropy (FA), particularly in the genu and body of the corpus callosum. Furthermore, differences in the C parameter partially mediated differences in DSS within this age range. After the age of 50, the TVA parameter for the perceptual threshold, t0, as well as K, were most strongly related to participant age. Both parameters, but t0 more strongly so than K, were associated WM diffusivity, particularly in projection fibers such as the internal capsule, the sagittal stratum, and the corona radiata. Within this age range, t0 partially mediated age-related differences in gF. The results are consistent with, and provide novel empirical support for the neuroanatomical localization of TVA computations as outlined in the neuronal interpretation of TVA (NTVA). Furthermore, the results indicate that to understand the biological sources of age-related changes in processing speed and fluid cognition, it may be useful to employ methods that allow for computational fractionation of these multicomponent measures.

Smooth bilinear classification of EEG

The goal of this paper is to improve on single-trial classification of electro-encephalography (EEG) using linear methods. The paper proposes to combine the classification of the spatial distribution of activity with the classification of its temporal profile. The work is based on the idea that a current source in the brain has a reproducible temporal profile with a static spatial projection to the electrodes. This assumption reduces the parameter space of a linear classifier to a rank-one factorial space. The new model limits over-fitting due to the fewer number of parameters, and furthermore, it allows us to declare a prior belief of smoothness on the spatial and temporal profiles of the source. Our experiments show that the method is useful as a classifier with an area under the ROC curve of 0.93 having only 40 target trials available for training. Investigation of the trained classifier encourages us to belief that the method can also be useful as a tool to interpret the activity in the data at hand with respect to experimental events

Visual attention capacity parameters covary with hemifield alignment.

The theory of visual attention (TVA; Bundesen, 1990. Psychological Review, 97(4), 523-547), allows one to measure distinct visual attention parameters, such as the temporal threshold for visual perception, visual processing capacity, and visual short-term memory (VSTM) capacity. It has long been assumed that visual processing capacity and VSTM capacity parameters are nearly constant from trial to trial. However, Dyrholm, Kyllingsbæk, Espeseth, and Bundesen (2011). Journal of Mathematical Psychology, 55(6), 416-429, found evidence of considerable trial-by-trial variability of VSTM capacity. Here we show that one cause of trial-by-trial variation is that some parameters depend on whether processing of relevant information occurs in only one hemifield or in both hemifields. Our results show that VSTM and visual processing capacities are higher when stimuli are distributed across the hemifields rather than located in the same hemifield. This corresponds to previous suggestions that parallel processing is more efficient across hemifields than within a single hemifield because both hemispheres are involved (e.g., Alvarez & Cavanagh, 2005. Psychological Science, 16(8), 637-643; Kraft et al., 2005. Cognitive Brain Research, 24(1), 453-463). We argue that the established view of a fixed visual attentional capacity must be relativized by taking hemifield distribution into account.

Signal Processing and Machine Learning for Single-trial Analysis of Simultaneously Acquired EEG and fMRI

Publisher Summary The simultaneous acquisition of electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) is a potentially powerful multimodal imaging technique for measuring the functional activity of the human brain. Given that EEG measures the electrical activity of neural populations while fMRI measures hemodynamics via a blood oxygenation-level-dependent (BOLD) signal related to neuronal activity, simultaneous EEG/fMRI (hereafter referred to as EEG/fMRI) offers a modality to investigate the relationship between these two phenomena within the context of noninvasive neuroimaging. Though fMRI is widely used to study cognitive and perceptual function, there is still substantial debate regarding the relation- ship between local neuronal activity and hemodynamic changes. Another rationale for EEG/fMRI is that, despite the fact that the individual modalities measure markedly different physiological phenomena, in terms of spatial and temporal resolution they are quite complementary. EEG offers millisecond temporal resolution; however, the spatial sampling density and ill-posed nature of the inverse model problem limit its spatial resolution. On the other hand, fMRI provides millimeter spatial resolution, but because of scanning rates and the low-pass nature of the BOLD response, the temporal resolution is limited. One approach that has been adopted to take advantage of this complementarity is to use fMRI activations to seed EEG source localization.

Single‐Trial Inference on Visual Attention

In this paper we take a step towards single‐trial behavioral modeling within a Theory of Visual Attention (TVA). In selective attention tasks, such as the Partial Report paradigm, the subject is asked to ignore distractors and only report stimuli that belong to the target class. Nothing about a distractor is observed directly in the subject’s overt behavior, hence behavioral modeling of such trials involves out‐marginalizing the variables that represent the distractors’ influence on behavior. In this paper we derive equations for inferring a latent representation of the distractors on a Partial Report trial. This result retrodicts a latent attentional state of the subject using the observed response from that particular trial and thus differs from other predictions made with TVA which are based on expected values of observed variables. We show an example of the result in single‐trial analysis of an occipital EEG component.