Search

The integration of artificial intelligence (AI) into human teams is widely expected to enhance performance and collaboration. However, our study reveals a striking and counterintuitive result: human-AI teams performed worse than human-only teams, especially when task difficulty increased. Using a virtual reality-based sensorimotor task, we observed that the inclusion of an active human-like AI teammate disrupted team dynamics, leading to elevated arousal, reduced engagement, and diminished communication intensity among human participants. These effects persisted even as the human teammates' perception of the AI teammate improved over time. These findings challenge prevailing assumptions about the benefits of AI in team settings and highlight the critical need for human-centered AI design to mitigate adverse physiological and behavioral impacts, ensuring more effective human-AI collaboration.

In collaborative environments, a deep understanding of multi-human teaming dynamics is essential for optimizing performance. However, the relationship between individuals' behavioral and physiological markers and their combined influence on overall team performance remains poorly understood. To explore this, we designed a triadic human collaborative sensorimotor task in virtual reality (VR) and introduced a novel predictability metric to examine team dynamics and performance. Our findings reveal a strong connection between team performance and the predictability of a team member's future actions based on other team members' behavioral and physiological data. Contrary to conventional wisdom that high-performing teams are highly synchronized, our results suggest that physiological and behavioral synchronizations among team members have a limited correlation with team performance. These insights provide a new quantitative framework for understanding multi-human teaming, paving the way for deeper insights into team dynamics and performance.

Neural biomarkers that can classify or predict disease are of broad interest to the neurological and psychiatric communities. Such biomarkers can be informative of disease state or treatment efficacy, even before there are changes in symptoms and/or behavior. This work investigates EEG-estimated functional connectivity (FC) as a Parkinson's Disease (PD) biomarker. Specifically, we investigate FC mediated via neural oscillations and consider such activity during the Simons conflict task. This task yields sensory-motor conflict, and one might expect differences in behavior between PD patients and healthy controls (HCs). In addition to considering spatially focused approaches, such as FC, as a biomarker, we also consider temporal biomarkers, which are more sensitive to ongoing changes in neural activity. We find that FC, estimated from delta (1-4Hz) and theta (4-7Hz) oscillations, yields spatial FC patterns significantly better at distinguishing PD from HC than temporal features or behavior. This study reinforces that FC in spectral bands is informative of differences in brain-wide processes and can serve as a biomarker distinguishing normal brain function from that seen in disease., Comment: This work is accepted at IEEE EMBC 2024. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications standards/publications/rights/index.html for more information

Learning collaborative behaviors is essential for multi-agent systems. Traditionally, multi-agent reinforcement learning solves this implicitly through a joint reward and centralized observations, assuming collaborative behavior will emerge. Other studies propose to learn from demonstrations of a group of collaborative experts. Instead, we propose an efficient and explicit way of learning collaborative behaviors in multi-agent systems by leveraging expertise from only a single human. Our insight is that humans can naturally take on various roles in a team. We show that agents can effectively learn to collaborate by allowing a human operator to dynamically switch between controlling agents for a short period and incorporating a human-like theory-of-mind model of teammates. Our experiments showed that our method improves the success rate of a challenging collaborative hide-and-seek task by up to 58$% with only 40 minutes of human guidance. We further demonstrate our findings transfer to the real world by conducting multi-robot experiments.

There is a growing interest in characterizing circular data found in biological systems. Such data are wide ranging and varied, from signal phase in neural recordings to nucleotide sequences in round genomes. Traditional clustering algorithms are often inadequate due to their limited ability to distinguish differences in the periodic component. Current clustering schemes that work in a polar coordinate system have limitations, such as being only angle-focused or lacking generality. To overcome these limitations, we propose a new analysis framework that utilizes projections onto a cylindrical coordinate system to better represent objects in a polar coordinate system. Using the mathematical properties of circular data, we show our approach always finds the correct clustering result within the reconstructed dataset, given sufficient periodic repetitions of the data. Our approach is generally applicable and adaptable and can be incorporated into most state-of-the-art clustering algorithms. We demonstrate on synthetic and real data that our method generates more appropriate and consistent clustering results compared to standard methods. In summary, our proposed analysis framework overcomes the limitations of existing polar coordinate-based clustering methods and provides a more accurate and efficient way to cluster circular data., Comment: Manuscript is under review in IEEE Transactions on Computational Biology and Bioinformatics. Copyright holder is credited to IEEE

Vision Transformers (ViT) have advanced computer vision, yet their efficacy in complex tasks like driving remains less explored. This study enhances ViT by integrating human eye gaze, captured via eye-tracking, to increase prediction accuracy in driving scenarios under uncertainty in both real-world and virtual reality scenarios. First, we establish the significance of human eye gaze in left-right driving decisions, as observed in both human subjects and a ViT model. By comparing the similarity between human fixation maps and ViT attention weights, we reveal the dynamics of overlap across individual heads and layers. This overlap demonstrates that fixation data can guide the model in distributing its attention weights more effectively. We introduce the fixation-attention intersection (FAX) loss, a novel loss function that significantly improves ViT performance under high uncertainty conditions. Our results show that ViT, when trained with FAX loss, aligns its attention with human gaze patterns. This gaze-informed approach has significant potential for driver behavior analysis, as well as broader applications in human-centered AI systems, extending ViT's use to complex visual environments., Comment: 25 pages, 9 figures, 3 tables

Several approximate inference methods have been proposed for deep discrete latent variable models. However, non-parametric methods which have previously been successfully employed for classical sparse coding models have largely been unexplored in the context of deep models. We propose a non-parametric iterative algorithm for learning discrete latent representations in such deep models. Additionally, to learn scale invariant discrete features, we propose local data scaling variables. Lastly, to encourage sparsity in our representations, we propose a Beta-Bernoulli process prior on the latent factors. We evaluate our spare coding model coupled with different likelihood models. We evaluate our method across datasets with varying characteristics and compare our results to current amortized approximate inference methods.

Simultaneous EEG-fMRI is a multi-modal neuroimaging technique that provides complementary spatial and temporal resolution. Challenging has been developing principled and interpretable approaches for fusing the modalities, specifically approaches enabling inference of latent source spaces representative of neural activity. In this paper, we address this inference problem within the framework of transcoding -- mapping from a specific encoding (modality) to a decoding (the latent source space) and then encoding the latent source space to the other modality. Specifically, we develop a symmetric method consisting of a cyclic convolutional transcoder that transcodes EEG to fMRI and vice versa. Without any prior knowledge of either the hemodynamic response function or lead field matrix, the complete data-driven method exploits the temporal and spatial relationships between the modalities and latent source spaces to learn these mappings. We quantify, for both the simulated and real EEG-fMRI data, how well the modalities can be transcoded from one to another as well as the source spaces that are recovered, all evaluated on unseen data. In addition to enabling a new way to symmetrically infer a latent source space, the method can also be seen as low-cost computational neuroimaging -- i.e. generating an 'expensive' fMRI BOLD image from 'low cost' EEG data.

Cognition in midlife is an important predictor of age-related mental decline and statistical models that predict cognitive performance can be useful for predicting decline. However, existing models struggle to capture complex relationships between physical, sociodemographic, psychological and mental health factors that effect cognition. Using data from an observational, cohort study, Midlife in the United States (MIDUS), we modeled a large number of variables to predict executive function and episodic memory measures. We used cross-sectional and longitudinal outcomes with varying sparsity, or amount of missing data. Deep neural network (DNN) models consistently ranked highest in all of the cognitive performance prediction tasks, as assessed with root mean squared error (RMSE) on out-of-sample data. RMSE differences between DNN and other model types were statistically significant (T(8) = -3.70; p < 0.05). The interaction effect between model type and sparsity was significant (F(9)=59.20; p < 0.01), indicating the success of DNNs can partly be attributed to their robustness and ability to model hierarchical relationships between health-related factors. Our findings underscore the potential of neural networks to model clinical datasets and allow better understanding of factors that lead to cognitive decline.

We propose a hierarchical Bayesian recurrent state space model for modeling switching network connectivity in resting state fMRI data. Our model allows us to uncover shared network patterns across disease conditions. We evaluate our method on the ADNI2 dataset by inferring latent state patterns corresponding to altered neural circuits in individuals with Mild Cognitive Impairment (MCI). In addition to states shared across healthy and individuals with MCI, we discover latent states that are predominantly observed in individuals with MCI. Our model outperforms current state of the art deep learning method on ADNI2 dataset., Comment: Machine Learning for Health (ML4H) at NeurIPS 2020 - Extended Abstract

We propose a deep generative factor analysis model with beta process prior that can approximate complex non-factorial distributions over the latent codes. We outline a stochastic EM algorithm for scalable inference in a specific instantiation of this model and present some preliminary results.

Simultaneous EEG-fMRI is a multi-modal neuroimaging technique that provides complementary spatial and temporal resolution for inferring a latent source space of neural activity. In this paper we address this inference problem within the framework of transcoding -- mapping from a specific encoding (modality) to a decoding (the latent source space) and then encoding the latent source space to the other modality. Specifically, we develop a symmetric method consisting of a cyclic convolutional transcoder that transcodes EEG to fMRI and vice versa. Without any prior knowledge of either the hemodynamic response function or lead field matrix, the method exploits the temporal and spatial relationships between the modalities and latent source spaces to learn these mappings. We show, for real EEG-fMRI data, how well the modalities can be transcoded from one to another as well as the source spaces that are recovered, all on unseen data. In addition to enabling a new way to symmetrically infer a latent source space, the method can also be seen as low-cost computational neuroimaging -- i.e. generating an 'expensive' fMRI BOLD image from 'low cost' EEG data.

Many imaging technologies rely on tomographic reconstruction, which requires solving a multidimensional inverse problem given a finite number of projections. Backprojection is a popular class of algorithm for tomographic reconstruction, however it typically results in poor image reconstructions when the projection angles are sparse and/or if the sensors characteristics are not uniform. Several deep learning based algorithms have been developed to solve this inverse problem and reconstruct the image using a limited number of projections. However these algorithms typically require examples of the ground-truth (i.e. examples of reconstructed images) to yield good performance. In this paper, we introduce an unsupervised sparse-view backprojection algorithm, which does not require ground-truth. The algorithm consists of two modules in a generator-projector framework; a convolutional neural network and a spatial transformer network. We evaluated our algorithm using computed tomography (CT) images of the human chest. We show that our algorithm significantly out-performs filtered backprojection when the projection angles are very sparse, as well as when the sensor characteristics vary for different angles. Our approach has practical applications for medical imaging and other imaging modalities (e.g. radar) where sparse and/or non-uniform projections may be acquired due to time or sampling constraints.

Cascading large-amplitude bursts in neural activity, termed avalanches, are thought to provide insight into the complex spatially distributed interactions in neural systems. In human neuroimaging, for example, avalanches occurring during resting-state show scale-invariant dynamics, supporting the hypothesis that the brain operates near a critical point that enables long range spatial communication. In fact, it has been suggested that such scale-invariant dynamics, characterized by a power-law distribution in these avalanches, are universal in neural systems and emerge through a common mechanism. While the analysis of avalanches and subsequent criticality is increasingly seen as a framework for using complex systems theory to understand brain function, it is unclear how the framework would account for the omnipresent cognitive variability, whether across individuals and/or tasks. To address this, we analyzed avalanches in the EEG activity of healthy humans during rest as well as two distinct task conditions that varied in cognitive demands and produced behavioral measures unique to each individual. In both rest and task conditions we observed that avalanche dynamics demonstrate scale-invariant characteristics, but differ in their specific features, demonstrating individual variability. Using a new metric we call normalized engagement, which estimates the likelihood for a brain region to produce high-amplitude bursts, we also investigated regional features of avalanche dynamics. Normalized engagement showed not only the expected individual and task dependent variability, but also scale-specificity that correlated with individual behavior. Our findings expand our understanding of avalanche features and are supportive of the emerging theoretical idea that the dynamics of an active human brain operate close to a critical-like region and not a singular critical-state.

In reinforcement learning (RL), sparse rewards are a natural way to specify the task to be learned. However, most RL algorithms struggle to learn in this setting since the learning signal is mostly zeros. In contrast, humans are good at assessing and predicting the future consequences of actions and can serve as good reward/policy shapers to accelerate the robot learning process. Previous works have shown that the human brain generates an error-related signal, measurable using electroencephelography (EEG), when the human perceives the task being done erroneously. In this work, we propose a method that uses evaluative feedback obtained from human brain signals measured via scalp EEG to accelerate RL for robotic agents in sparse reward settings. As the robot learns the task, the EEG of a human observer watching the robot attempts is recorded and decoded into noisy error feedback signal. From this feedback, we use supervised learning to obtain a policy that subsequently augments the behavior policy and guides exploration in the early stages of RL. This bootstraps the RL learning process to enable learning from sparse reward. Using a robotic navigation task as a test bed, we show that our method achieves a stable obstacle-avoidance policy with high success rate, outperforming learning from sparse rewards only that struggles to achieve obstacle avoidance behavior or fails to advance to the goal., Comment: 2020 IEEE International Conference on Robotics and Automation - ICRA 2020

Steady-State Visual Evoked Potentials (SSVEPs) are neural oscillations from the parietal and occipital regions of the brain that are evoked from flickering visual stimuli. SSVEPs are robust signals measurable in the electroencephalogram (EEG) and are commonly used in brain-computer interfaces (BCIs). However, methods for high-accuracy decoding of SSVEPs usually require hand-crafted approaches that leverage domain-specific knowledge of the stimulus signals, such as specific temporal frequencies in the visual stimuli and their relative spatial arrangement. When this knowledge is unavailable, such as when SSVEP signals are acquired asynchronously, such approaches tend to fail. In this paper, we show how a compact convolutional neural network (Compact-CNN), which only requires raw EEG signals for automatic feature extraction, can be used to decode signals from a 12-class SSVEP dataset without the need for any domain-specific knowledge or calibration data. We report across subject mean accuracy of approximately 80% (chance being 8.3%) and show this is substantially better than current state-of-the-art hand-crafted approaches using canonical correlation analysis (CCA) and Combined-CCA. Furthermore, we analyze our Compact-CNN to examine the underlying feature representation, discovering that the deep learner extracts additional phase and amplitude related features associated with the structure of the dataset. We discuss how our Compact-CNN shows promise for BCI applications that allow users to freely gaze/attend to any stimulus at any time (e.g., asynchronous BCI) as well as provides a method for analyzing SSVEP signals in a way that might augment our understanding about the basic processing in the visual cortex., Comment: Accepted for publication at the Journal of Neural Engineering

Reinforcement Learning AI commonly uses reward/penalty signals that are objective and explicit in an environment -- e.g. game score, completion time, etc. -- in order to learn the optimal strategy for task performance. However, Human-AI interaction for such AI agents should include additional reinforcement that is implicit and subjective -- e.g. human preferences for certain AI behavior -- in order to adapt the AI behavior to idiosyncratic human preferences. Such adaptations would mirror naturally occurring processes that increase trust and comfort during social interactions. Here, we show how a hybrid brain-computer-interface (hBCI), which detects an individual's level of interest in objects/events in a virtual environment, can be used to adapt the behavior of a Deep Reinforcement Learning AI agent that is controlling a virtual autonomous vehicle. Specifically, we show that the AI learns a driving strategy that maintains a safe distance from a lead vehicle, and most novelly, preferentially slows the vehicle when the human passengers of the vehicle encounter objects of interest. This adaptation affords an additional 20\% viewing time for subjectively interesting objects. This is the first demonstration of how an hBCI can be used to provide implicit reinforcement to an AI agent in a way that incorporates user preferences into the control system., Comment: 11 pages, 9 figures, 1 table, Submitted to IEEE Trans. on Neural Networks and Learning Systems

Multimodal biosignal acquisition is facilitated by recently introduced software solutions such as LabStreaming Layer (LSL) and its associated data format XDF (Extensible Data Format). However, there are no stand-alone applications that can visualize multimodal time series stored in XDF files. We extended SigViewer, an open source cross-platform Qt C++ application with the capability of loading, resampling, annotating, and visualizing signals stored in XDF files and successfully applied the tool for post-hoc visual verification of the accuracy of a system that aims to predict the phase of alpha oscillations within the electroencephalogram in real-time., Comment: 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society

Simultaneously recorded electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) can be used to non-invasively measure the spatiotemporal dynamics of the human brain. One challenge is dealing with the artifacts that each modality introduces into the other when the two are recorded concurrently, for example the ballistocardiogram (BCG). We conducted a preliminary comparison of three different MR compatible EEG recording systems and assessed their performance in terms of single-trial classification of the EEG when simultaneously collecting fMRI. We found tradeoffs across all three systems, for example varied ease of setup and improved classification accuracy with reference electrodes (REF) but not for pulse artifact subtraction (PAS) or reference layer adaptive filtering (RLAF)., Comment: 1 Page, IEEE EMBS Conference 2017, Korea

Objective. We investigated the neural correlates of workload buildup in a fine visuomotor task called the boundary avoidance task (BAT). The BAT has been known to induce naturally occurring failures of human-machine coupling in high performance aircraft that can potentially lead to a crash; these failures are termed pilot induced oscillations (PIOs). Approach. We recorded EEG and pupillometry data from human subjects engaged in a flight BAT simulated within a virtual 3D environment. Main results. We find that workload buildup in a BAT can be successfully decoded from oscillatory features in the electroencephalogram (EEG). Information in delta, theta, alpha, beta, and gamma spectral bands of the EEG all contribute to successful decoding, however gamma band activity with a lateralized somatosensory topography has the highest contribution, while theta band activity with a frontocentral topography has the most robust contribution in terms of real world usability. We show that the output of the spectral decoder can be used to predict PIO susceptibility. We also find that workload buildup in the task induces pupil dilation, the magnitude of which is significantly correlated with the magnitude of the decoded EEG signals. These results suggest that PIOs may result from the dysregulation of cortical networks such as the locus coeruleus (LC) anterior cingulate cortex (ACC) circuit. Significance. Our findings may generalize to similar control failures in other cases of tight man machine coupling where gains and latencies in the control system must be inferred and compensated for by the human operators. A closed-loop intervention using neurophysiological decoding of workload buildup that targets the LC ACC circuit may positively impact operator performance in such situations., Comment: Manuscript as initially submitted to Journal of Neural Engineering in March, 2016

We present an efficient algorithm for simultaneously training sparse generalized linear models across many related problems, which may arise from bootstrapping, cross-validation and nonparametric permutation testing. Our approach leverages the redundancies across problems to obtain significant computational improvements relative to solving the problems sequentially by a conventional algorithm. We demonstrate our fast simultaneous training of generalized linear models (FaSTGLZ) algorithm on a number of real-world datasets, and we run otherwise computationally intensive bootstrapping and permutation test analyses that are typically necessary for obtaining statistically rigorous classification results and meaningful interpretation. Code is freely available at http://liinc.bme.columbia.edu/fastglz.

Neural mechanisms of extraclassical receptive field phenomena in V1 are commonly assumed to result from long-range lateral connections and/or extrastriate feedback. We address two such phenomena: surround suppression and contrast dependent receptive field size. We present rigorous computational support for the hypothesis that the phenomena largely result from local short-range (< 0.5 mm) cortical connections and LGN input. Surround suppression in our simulations results from (A) direct cortical inhibition or (B) suppression of recurrent cortical excitation, or (C) action of both these mechanisms simultaneously. Mechanisms B and C are substantially more prevalent than A. We observe an average growth in the range of spatial summation of excitatory and inhibitory synaptic inputs for low contrast. However, we find this is neither sufficient nor necessary to explain contrast dependent receptive field size, which usually involves additional changes in the relative gain of these inputs.