Evidence of elevated situational awareness for active duty soldiers during navigation of a virtual environment.

U.S. service members maintain constant situational awareness (SA) due to training and experience operating in dynamic and complex environments. Work examining how military experience impacts SA during visual search of a complex naturalistic environment, is limited. Here, we compare Active Duty service members and Civilians' physiological behavior during a navigational visual search task in an open-world virtual environment (VE) while cognitive load was manipulated. We measured eye-tracking and electroencephalogram (EEG) outcomes from Active Duty (N = 21) and Civilians (N = 15) while they navigated a desktop VE at a self-regulated pace. Participants searched and counted targets (N = 15) presented among distractors, while cognitive load was manipulated with an auditory Math Task. Results showed Active Duty participants reported significantly greater/closer to the correct number of targets compared to Civilians. Overall, Active Duty participants scanned the VE with faster peak saccade velocities and greater average saccade magnitudes compared to Civilians. Convolutional Neural Network (CNN) response (EEG P-300) was significantly weighted more to initial fixations for the Active Duty group, showing reduced attentional resources on object refixations compared to Civilians. There were no group differences in fixation outcomes or overall CNN response when comparing targets versus distractor objects. When cognitive load was manipulated, only Civilians significantly decreased their average dwell time on each object and the Active Duty group had significantly fewer numbers of correct answers on the Math Task. Overall, the Active Duty group explored the VE with increased scanning speed and distance and reduced cognitive re-processing on objects, employing a different, perhaps expert, visual search strategy indicative of increased SA. The Active Duty group maintained SA in the main visual search task and did not appear to shift focus to the secondary Math Task. Future work could compare how a stress inducing environment impacts these groups' physiological or cognitive markers and performance for these groups.

Decoding neural activity to assess individual latent state in ecologically valid contexts.

Objective.Currently, there exists very few ways to isolate cognitive processes, historically defined via highly controlled laboratory studies, in more ecologically valid contexts. Specifically, it remains unclear as to what extent patterns of neural activity observed under such constraints actually manifest outside the laboratory in a manner that can be used to make accurate inferences about latent states, associated cognitive processes, or proximal behavior. Improving our understanding of when and how specific patterns of neural activity manifest in ecologically valid scenarios would provide validation for laboratory-based approaches that study similar neural phenomena in isolation and meaningful insight into the latent states that occur during complex tasks.Approach.Domain generalization methods, borrowed from the work of the brain-computer interface community, have the potential to capture high-dimensional patterns of neural activity in a way that can be reliably applied across experimental datasets in order to address this specific challenge. We previously used such an approach to decode phasic neural responses associated with visual target discrimination. Here, we extend that work to more tonic phenomena such as internal latent states. We use data from two highly controlled laboratory paradigms to train two separate domain-generalized models. We apply the trained models to an ecologically valid paradigm in which participants performed multiple, concurrent driving-related tasks while perched atop a six-degrees-of-freedom ride-motion simulator.Main Results.Using the pretrained models, we estimate latent state and the associated patterns of neural activity. As the patterns of neural activity become more similar to those patterns observed in the training data, we find changes in behavior and task performance that are consistent with the observations from the original, laboratory-based paradigms.Significance.These results lend ecological validity to the original, highly controlled, experimental designs and provide a methodology for understanding the relationship between neural activity and behavior during complex tasks.

Gaze Behavior During Navigation and Visual Search of an Open-World Virtual Environment.

Eye tracking has been an essential tool within the vision science community for many years. However, the majority of studies involving eye-tracking technology employ a relatively passive approach through the use of static imagery, prescribed motion, or video stimuli. This is in contrast to our everyday interaction with the natural world where we navigate our environment while actively seeking and using task-relevant visual information. For this reason, an increasing number of vision researchers are employing virtual environment platforms, which offer interactive, realistic visual environments while maintaining a substantial level of experimental control. Here, we recorded eye movement behavior while subjects freely navigated through a rich, open-world virtual environment. Within this environment, subjects completed a visual search task where they were asked to find and count occurrence of specific targets among numerous distractor items. We assigned each participant into one of four target conditions: Humvees, motorcycles, aircraft, or furniture. Our results show a statistically significant relationship between gaze behavior and target objects across Target Conditions with increased visual attention toward assigned targets. Specifically, we see an increase in the number of fixations and an increase in dwell time on target relative to distractor objects. In addition, we included a divided attention task to investigate how search changed with the addition of a secondary task. With increased cognitive load, subjects slowed their speed, decreased gaze on objects, and increased the number of objects scanned in the environment. Overall, our results confirm previous findings and support that complex virtual environments can be used for active visual search experimentation, maintaining a high level of precision in the quantification of gaze information and visual attention. This study contributes to our understanding of how individuals search for information in a naturalistic (open-world) virtual environment. Likewise, our paradigm provides an intriguing look into the heterogeneity of individual behaviors when completing an un-timed visual search task while actively navigating.

"Blue Sky Effect": Contextual Influences on Pupil Size During Naturalistic Visual Search.

Pupil size is influenced by cognitive and non-cognitive factors. One of the strongest modulators of pupil size is scene luminance, which complicates studies of cognitive pupillometry in environments with complex patterns of visual stimulation. To help understand how dynamic visual scene statistics influence pupil size during an active visual search task in a visually rich 3D virtual environment (VE), we analyzed the correlation between pupil size and intensity changes of image pixels in the red, green, and blue (RGB) channels within a large window (~14 degrees) surrounding the gaze position over time. Overall, blue and green channels had a stronger influence on pupil size than the red channel. The correlation maps were not consistent with the hypothesis of a foveal bias for luminance, instead revealing a significant contextual effect, whereby pixels above the gaze point in the green/blue channels had a disproportionate impact on pupil size. We hypothesized this differential sensitivity of pupil responsiveness to blue light from above as a "blue sky effect," and confirmed this finding with a follow-on experiment with a controlled laboratory task. Pupillary constrictions were significantly stronger when blue was presented above fixation (paired with luminance-matched gray on bottom) compared to below fixation. This effect was specific for the blue color channel and this stimulus orientation. These results highlight the differential sensitivity of pupillary responses to scene statistics in studies or applications that involve complex visual environments and suggest blue light as a predominant factor influencing pupil size.

Decoding P300 Variability Using Convolutional Neural Networks.

Deep convolutional neural networks (CNN) have previously been shown to be useful tools for signal decoding and analysis in a variety of complex domains, such as image processing and speech recognition. By learning from large amounts of data, the representations encoded by these deep networks are often invariant to moderate changes in the underlying feature spaces. Recently, we proposed a CNN architecture that could be applied to electroencephalogram (EEG) decoding and analysis. In this article, we train our CNN model using data from prior experiments in order to later decode the P300 evoked response from an unseen, hold-out experiment. We analyze the CNN output as a function of the underlying variability in the P300 response and demonstrate that the CNN output is sensitive to the experiment-induced changes in the neural response. We then assess the utility of our approach as a means of improving the overall signal-to-noise ratio in the EEG record. Finally, we show an example of how CNN-based decoding can be applied to the analysis of complex data.

Analyzing P300 Distractors for Target Reconstruction.

P300-based brain-computer interfaces (BCIs) are often trained per-user and per-application space. Training such models requires ground truth knowledge of target and nontarget stimulus categories during model training, which imparts bias into the model. Additionally, not all non-targets are created equal; some may contain visual features that resemble targets or may otherwise be visually salient. Current research has indicated that non-target distractors may elicit attenuated P300 responses based on the perceptual similarity of these distractors to the target category. To minimize this bias, and enable a more nuanced analysis, we use a generalized BCI approach that is fit to neither user nor task. We do not seek to improve the overall accuracy of the BCI with our generalized approach; we instead demonstrate the utility of our approach for identifying targetrelated image features. When combined with other intelligent agents, such as computer vision systems, the performance of the generalized model equals that of the user-specific models, without any user specific data.

EEGNet: a compact convolutional neural network for EEG-based brain-computer interfaces.

OBJECTIVE: Brain-computer interfaces (BCI) enable direct communication with a computer, using neural activity as the control signal. This neural signal is generally chosen from a variety of well-studied electroencephalogram (EEG) signals. For a given BCI paradigm, feature extractors and classifiers are tailored to the distinct characteristics of its expected EEG control signal, limiting its application to that specific signal. Convolutional neural networks (CNNs), which have been used in computer vision and speech recognition to perform automatic feature extraction and classification, have successfully been applied to EEG-based BCIs; however, they have mainly been applied to single BCI paradigms and thus it remains unclear how these architectures generalize to other paradigms. Here, we ask if we can design a single CNN architecture to accurately classify EEG signals from different BCI paradigms, while simultaneously being as compact as possible. APPROACH: In this work we introduce EEGNet, a compact convolutional neural network for EEG-based BCIs. We introduce the use of depthwise and separable convolutions to construct an EEG-specific model which encapsulates well-known EEG feature extraction concepts for BCI. We compare EEGNet, both for within-subject and cross-subject classification, to current state-of-the-art approaches across four BCI paradigms: P300 visual-evoked potentials, error-related negativity responses (ERN), movement-related cortical potentials (MRCP), and sensory motor rhythms (SMR). MAIN RESULTS: We show that EEGNet generalizes across paradigms better than, and achieves comparably high performance to, the reference algorithms when only limited training data is available across all tested paradigms. In addition, we demonstrate three different approaches to visualize the contents of a trained EEGNet model to enable interpretation of the learned features. SIGNIFICANCE: Our results suggest that EEGNet is robust enough to learn a wide variety of interpretable features over a range of BCI tasks. Our models can be found at: https://github.com/vlawhern/arl-eegmodels.

A novel method linking neural connectivity to behavioral fluctuations: Behavior-regressed connectivity.

BACKGROUND: During an experimental session, behavioral performance fluctuates, yet most neuroimaging analyses of functional connectivity derive a single connectivity pattern. These conventional connectivity approaches assume that since the underlying behavior of the task remains constant, the connectivity pattern is also constant. NEW METHOD: We introduce a novel method, behavior-regressed connectivity (BRC), to directly examine behavioral fluctuations within an experimental session and capture their relationship to changes in functional connectivity. This method employs the weighted phase lag index (WPLI) applied to a window of trials with a weighting function. Using two datasets, the BRC results are compared to conventional connectivity results during two time windows: the one second before stimulus onset to identify predictive relationships, and the one second after onset to capture task-dependent relationships. RESULTS: In both tasks, we replicate the expected results for the conventional connectivity analysis, and extend our understanding of the brain-behavior relationship using the BRC analysis, demonstrating subject-specific BRC maps that correspond to both positive and negative relationships with behavior. Comparison with Existing Method(s): Conventional connectivity analyses assume a consistent relationship between behaviors and functional connectivity, but the BRC method examines performance variability within an experimental session to understand dynamic connectivity and transient behavior. CONCLUSION: The BRC approach examines connectivity as it covaries with behavior to complement the knowledge of underlying neural activity derived from conventional connectivity analyses. Within this framework, BRC may be implemented for the purpose of understanding performance variability both within and between participants.