Baseline correction of overlapping event-related responses using a linear deconvolution technique

Multi-response deconvolution of auditory evoked potentials in a reduced representation space

The estimation of auditory evoked potentials requires deconvolution when the duration of the responses to be recovered exceeds the inter-stimulus interval. Based on least squares deconvolution, in this article we extend the procedure to the case of a multi-response convolutional model, that is, a model in which different categories of stimulus are expected to evoke different responses. The computational cost of the multi-response deconvolution significantly increases with the number of responses to be deconvolved, which restricts its applicability in practical situations. In order to alleviate this restriction, we propose to perform the multi-response deconvolution in a reduced representation space associated with a latency-dependent filtering of auditory responses, which provides a significant dimensionality reduction. We demonstrate the practical viability of the multi-response deconvolution with auditory responses evoked by clicks presented at different levels and categorized according to their stimulation level. The multi-response deconvolution applied in a reduced representation space provides the least squares estimation of the responses with a reasonable computational load. matlab/Octave code implementing the proposed procedure is included as supplementary material.

P3-like signatures of temporal predictions: a computational EEG study

Many cognitive processes, ranging from perception to action, depend on the ability to predict the timing of forthcoming events. Yet, how the brain uses predictive models in the temporal domain is still an unsolved question. In previous work, we began to explore the neural correlates of temporal predictions by using a computational approach in which an ideal Bayesian observer learned the temporal probabilities of target onsets in a simple reaction time task. Because the task was specifically designed to disambiguate updating of predictive models and surprise, changes in temporal probabilities were explicitly cued. However, in the real world, we are usually incidentally exposed to changes in the statistics of the environment. Here, we thus aimed to further investigate the electroencephalographic (EEG) correlates of Bayesian belief updating and surprise associated with incidental learning of temporal probabilities. In line with our previous EEG study, results showed distinct P3-like modulations for updating and surprise. While surprise was indexed by an early fronto-central P3-like modulation, updating was associated with a later and more posterior P3 modulation. Moreover, updating was associated with a P2-like potential at centro-parietal electrodes, likely capturing integration processes between prior beliefs and likelihood of the observed event. These findings support previous evidence of trial-by-trial variability of P3 amplitudes as an index of dissociable inferential processes. Coupled with our previous findings, the present study strongly bolsters the view of the P3 as a key brain signature of temporal Bayesian inference. Data and scripts are shared on OSF: osf.io/sdy8j/.

Optimizing cognitive neuroscience experiments for separating event- related fMRI BOLD responses in non-randomized alternating designs

Functional magnetic resonance imaging (fMRI) has revolutionized human brain research. But there exists a fundamental mismatch between the rapid time course of neural events and the sluggish nature of the fMRI blood oxygen level-dependent (BOLD) signal, which presents special challenges for cognitive neuroscience research. This limitation in the temporal resolution of fMRI puts constraints on the information about brain function that can be obtained with fMRI and also presents methodological challenges. Most notably, when using fMRI to measure neural events occurring closely in time, the BOLD signals may temporally overlap one another. This overlap problem may be exacerbated in complex experimental paradigms (stimuli and tasks) that are designed to manipulate and isolate specific cognitive-neural processes involved in perception, cognition, and action. Optimization strategies to deconvolve overlapping BOLD signals have proven effective in providing separate estimates of BOLD signals from temporally overlapping brain activity, but there remains reduced efficacy of such approaches in many cases. For example, when stimulus events necessarily follow a non-random order, like in trial-by-trial cued attention or working memory paradigms. Our goal is to provide guidance to improve the efficiency with which the underlying responses evoked by one event type can be detected, estimated, and distinguished from other events in designs common in cognitive neuroscience research. We pursue this goal using simulations that model the nonlinear and transient properties of fMRI signals, and which use more realistic models of noise. Our simulations manipulated: (i) Inter-Stimulus-Interval (ISI), (ii) proportion of so-called null events, and (iii) nonlinearities in the BOLD signal due to both cognitive and design parameters. We offer a theoretical framework along with a python toolbox called deconvolve to provide guidance on the optimal design parameters that will be of particular utility when using non-random, alternating event sequences in experimental designs. In addition, though, we also highlight the challenges and limitations in simultaneously optimizing both detection and estimation efficiency of BOLD signals in these common, but complex, cognitive neuroscience designs.

Subspace-constrained deconvolution of auditory evoked potentials

Auditory evoked potentials can be estimated by synchronous averaging when the responses to the individual stimuli are not overlapped. However, when the response duration exceeds the inter-stimulus interval, a deconvolution procedure is necessary to obtain the transient response. The iterative randomized stimulation and averaging and the equivalent randomized stimulation with least squares deconvolution have been proven to be flexible and efficient methods for deconvolving the evoked potentials, with minimum restrictions in the design of stimulation sequences. Recently, a latency-dependent filtering and down-sampling (LDFDS) methodology was proposed for optimal filtering and dimensionality reduction, which is particularly useful when the evoked potentials involve the complete auditory pathway response (i.e., from the cochlea to the auditory cortex). In this case, the number of samples required to accurately represent the evoked potentials can be reduced from several thousand (with conventional sampling) to around 120. In this article, we propose to perform the deconvolution in the reduced representation space defined by LDFDS and present the mathematical foundation of the subspace-constrained deconvolution. Under the assumption that the evoked response is appropriately represented in the reduced representation space, the proposed deconvolution provides an optimal least squares estimation of the evoked response. Additionally, the dimensionality reduction provides a substantial reduction of the computational cost associated with the deconvolution. matlab/Octave code implementing the proposed procedures is included as supplementary material.

Regression-based analysis of combined EEG and eye-tracking data: Theory and applications

Fixation-related potentials (FRPs), neural responses aligned to the end of saccades, are a promising tool for studying the dynamics of attention and cognition under natural viewing conditions. In the past, four methodological problems have complicated the analysis of such combined eye-tracking/electroencephalogram experiments: (1) the synchronization of data streams, (2) the removal of ocular artifacts, (3) the condition-specific temporal overlap between the brain responses evoked by consecutive fixations, and (4) the fact that numerous low-level stimulus and saccade properties also influence the postsaccadic neural responses. Although effective solutions exist for the first two problems, the latter two are only beginning to be addressed. In the current paper, we present and review a unified regression-based framework for FRP analysis that allows us to deconvolve overlapping potentials while also controlling for both linear and nonlinear confounds on the FRP waveform. An open software implementation is provided for all procedures. We then demonstrate the advantages of this proposed (non)linear deconvolution modeling approach for data from three commonly studied paradigms: face perception, scene viewing, and reading. First, for a traditional event-related potential (ERP) face recognition experiment, we show how this technique can separate stimulus ERPs from overlapping muscle and brain potentials produced by small (micro)saccades on the face. Second, in natural scene viewing, we model and isolate multiple nonlinear effects of saccade parameters on the FRP. Finally, for a natural sentence reading experiment using the boundary paradigm, we show how it is possible to study the neural correlates of parafoveal preview after removing spurious overlap effects caused by the associated difference in average fixation time. Our results suggest a principal way of measuring reliable eye movement-related brain activity during natural vision.

No Evidence for a Role of Spatially Modulated α-Band Activity in Tactile Remapping and Short-Latency, Overt Orienting Behavior

Oscillatory α-band activity is commonly associated with spatial attention and multisensory prioritization. It has also been suggested to reflect the automatic transformation of tactile stimuli from a skin-based, somatotopic reference frame into an external one. Previous research has not convincingly separated these two possible roles of α-band activity. Previous experimental paradigms have used artificially long delays between tactile stimuli and behavioral responses to aid relating oscillatory activity to these different events. However, this strategy potentially blurs the temporal relationship of α-band activity relative to behavioral indicators of tactile-spatial transformations. Here, we assessed α-band modulation with massive univariate deconvolution, an analysis approach that disentangles brain signals overlapping in time and space. Thirty-one male and female human participants performed a delay-free, visual search task in which saccade behavior was unrestricted. A tactile cue to uncrossed or crossed hands was either informative or uninformative about visual target location. α-Band suppression following tactile stimulation was lateralized relative to the stimulated hand over central-parietal electrodes but relative to its external location over parieto-occipital electrodes. α-Band suppression reflected external touch location only after informative cues, suggesting that posterior α-band lateralization does not index automatic tactile transformation. Moreover, α-band suppression occurred at the time of, or after, the production of the saccades guided by tactile stimulation. These findings challenge the idea that α-band activity is directly involved in tactile-spatial transformation and suggest instead that it reflects delayed, supramodal processes related to attentional reorienting.SIGNIFICANCE STATEMENT Localizing a touch in space requires integrating somatosensory information about skin location and proprioceptive or visual information about posture. The automatic remapping between skin-based tactile information to a location in external space has been proposed to rely on the modulation of oscillatory brain activity in the α-band range, across the multiple cortical areas that are involved in tactile, multisensory, and spatial processing. We report two findings that are inconsistent with this view. First, α-band activity reflected the remapped stimulus location only when touch was task relevant. Second, α-band modulation occurred too late to account for spatially directed behavioral responses and, thus, only after remapping must have taken place. These characteristics contradict the idea that α-band directly reflects automatic tactile remapping processes.

Latency-dependent filtering and compact representation of the complete auditory pathway response

Auditory evoked potentials (AEPs) include the auditory brainstem response (ABR), middle latency response (MLR), and cortical auditory evoked potentials (CAEPs), each one covering a specific latency range and frequency band. For this reason, ABR, MLR, and CAEP are usually recorded separately using different protocols. This article proposes a procedure providing a latency-dependent filtering and down-sampling of the AEP responses. This way, each AEP component is appropriately filtered, according to its latency, and the complete auditory pathway response is conveniently represented (with the minimum number of samples, i.e., without unnecessary redundancies). The compact representation of the complete response facilitates a comprehensive analysis of the evoked potentials (keeping the natural continuity related to the neural activity transmission along the auditory pathway), which provides a new perspective in the design and analysis of AEP experiments. Additionally, the proposed compact representation reduces the storage or transmission requirements when large databases are manipulated for clinical or research purposes. The analysis of the AEP responses shows that a compact representation with 40 samples/decade (around 120 samples) is enough for accurately representing the response of the complete auditory pathway and provides appropriate latency-dependent filtering. MatLab/Octave code implementing the proposed procedure is included in the supplementary materials.

Unfold: an integrated toolbox for overlap correction, non-linear modeling, and regression-based EEG analysis

Electrophysiological research with event-related brain potentials (ERPs) is increasingly moving from simple, strictly orthogonal stimulation paradigms towards more complex, quasi-experimental designs and naturalistic situations that involve fast, multisensory stimulation and complex motor behavior. As a result, electrophysiological responses from subsequent events often overlap with each other. In addition, the recorded neural activity is typically modulated by numerous covariates, which influence the measured responses in a linear or non-linear fashion. Examples of paradigms where systematic temporal overlap variations and low-level confounds between conditions cannot be avoided include combined electroencephalogram (EEG)/eye-tracking experiments during natural vision, fast multisensory stimulation experiments, and mobile brain/body imaging studies. However, even "traditional," highly controlled ERP datasets often contain a hidden mix of overlapping activity (e.g., from stimulus onsets, involuntary microsaccades, or button presses) and it is helpful or even necessary to disentangle these components for a correct interpretation of the results. In this paper, we introduce unfold, a powerful, yet easy-to-use MATLAB toolbox for regression-based EEG analyses that combines existing concepts of massive univariate modeling ("regression-ERPs"), linear deconvolution modeling, and non-linear modeling with the generalized additive model into one coherent and flexible analysis framework. The toolbox is modular, compatible with EEGLAB and can handle even large datasets efficiently. It also includes advanced options for regularization and the use of temporal basis functions (e.g., Fourier sets). We illustrate the advantages of this approach for simulated data as well as data from a standard face recognition experiment. In addition to traditional and non-conventional EEG/ERP designs, unfold can also be applied to other overlapping physiological signals, such as pupillary or electrodermal responses. It is available as open-source software at http://www.unfoldtoolbox.org.

Auditory Attention Causes Gain Enhancement and Frequency Sharpening at Successive Stages of Cortical Processing-Evidence from Human Electroencephalography

Previous findings have suggested that auditory attention causes not only enhancement in neural processing gain, but also sharpening in neural frequency tuning in human auditory cortex. The current study was aimed to reexamine these findings. Specifically, we aimed to investigate whether attentional gain enhancement and frequency sharpening emerge at the same or different processing levels and whether they represent independent or cooperative effects. For that, we examined the pattern of attentional modulation effects on early, sensory-driven cortical auditory-evoked potentials occurring at different latencies. Attention was manipulated using a dichotic listening task and was thus not selectively directed to specific frequency values. Possible attention-related changes in frequency tuning selectivity were measured with an adaptation paradigm. Our results show marked disparities in attention effects between the earlier N1 deflection and the subsequent P2 deflection, with the N1 showing a strong gain enhancement effect, but no sharpening, and the P2 showing clear evidence of sharpening, but no independent gain effect. They suggest that gain enhancement and frequency sharpening represent successive stages of a cooperative attentional modulation mechanism that increases the representational bandwidth of attended versus unattended sounds.

Estimation of a transient response from steady-state responses by deconvolution with built-in constraints

Evidence suggests that the steady-state response (SSR) elicited by a periodic train of auditory stimuli can largely be understood as a superposition of transient responses. This study is devoted to the problem of how to estimate that transient response from measured SSRs. The proposed method differs from previous approaches in that the solution can be constrained to be consistent with physiology-based prior knowledge or educated guesses. To achieve this goal, the transient response is not represented by a time series, but by a linear combination of auxiliary functions, called components. Constraints are introduced by assigning certain properties to the components. Only few parameters are required for that purpose, because the individual components are derived from a suitably designed mother component. After adjusting the components to the problem at hand, the component amplitudes are determined by optimizing the match between predicted and measured SSRs. This requires solving a linear inverse problem. A model simulation as well as an analysis of exemplary experimental data (auditory SSRs elicited by periodically presented clicks) prove the workability of the method. Since part of the theory is quite general, it would be relatively easy to refine and extend the method. Not only could responses other than SSRs be dealt with, it could also be realized that certain key parameters of the transient response, such as amplitude and delay, depend on stimulus repetition rate.

Auditory signal detection appears to depend on temporal integration of subthreshold activity in auditory cortex

The threshold of hearing decreases with increasing sound duration up to a limit of a few hundred milliseconds, whereas other auditory time constants are orders of magnitude shorter. A possible solution to this resolution-integration paradox is that temporal integration occurs more centrally than computations depending on high temporal resolution. But this would require information about subthreshold events in the periphery to reach higher centers. Here we show that this prerequisite is fulfilled. The auditory evoked response to a just perceptible pulse series does basically not depend on whether single pulses are below or above behavioral threshold. The failure to find evidence of temporal integration up to response latencies of 30 ms suggests that the integrator is located more centrally than primary auditory cortex. By using noise to its advantage, the auditory system apparently has established a central integration mechanism that is about as efficient as the peripheral one in the visual system.

Auditory cortex tracks the temporal regularity of sustained noisy sounds

Neuroimaging studies have revealed dramatic asymmetries between the responses to temporally regular and irregular sounds in the antero-lateral part of Heschl's gyrus. For example, the magnetoencephalography (MEG) study of Krumbholz et al. [Cereb. Cortex 13, 765-772 (2003)] showed that the transition from a noise to a similar noise with sufficient temporal regularity to provoke a pitch evoked a pronounced temporal-regularity onset response (TRon response), whereas a comparable transition in the reverse direction revealed essentially no temporal-regularity offset response (TRoff response). The current paper presents a follow-up study in which the asymmetry is examined with much greater power, and the results suggest an intriguing reinterpretation of the onset/offset asymmetry. The TR-related activity in auditory cortex appears to be composed of a transient (TRon) and a TR-related sustained response (TRsus), with a highly variable TRon/TRsus amplitude ratio. The TRoff response is generally dominated by the break-down of the TRsus activity, which occurs so rapidly as to preclude the involvement of higher-level cortical processing. The time course of the TR-related activity suggests that TR processing might be involved in monitoring the environment and alerting the brain to the onset and offset of behaviourally relevant, animate sources.