Emmanuelle Tognoli - US grants

On an ordinary day, people perform actions that affect other people's behavior resulting in outcomes that affect both. Such social interactions are dynamic, far ranging, and usually require an exchange of information between the parties involved. A simple example is two persons modifying their approach to an elevator to accommodate each other's passage through the narrow space. More complex situations involve a teacher providing guidance to a pupil based on the pupil's own actions or training foreign military personnel to accomplish some function that depends on the latter's previous experiences. How are such social interactions to be quantified and understood? What goes on in a person's brain when they interact with another and what principles and mechanisms govern the coordination between brains? The investigators of this project constitute an interdisciplinary team whose expertise spans physics, cognitive science and neuroscience. They use the concepts, methods and tools of coordination dynamics, the science of coordination, to investigate both the behavioral and neural underpinnings of social behavioral interactions. The basic experimental paradigm involves coordination of movements between two people or between an individual and a computer avatar endowed with human-like capabilities. Pairs of people perform simple actions in front of each other and the investigators monitor key behavioral and neural variables that reveal how strongly each affects the other. By testing specific predictions of a mathematical model of coupled dynamical systems, the investigators aim to understand how social coordination evolves in time and to determine the respective strength of one person's influence on another.

Modern technology is always seeking to enhance human experience and productivity. The notion of 'others' has been expanded to include not only real human beings but also cyber-individuals, as evident in the increasing roles played by robotics and virtual environments in everyday human transactions. In crucial experiments, a human subject interacts with a virtual partner, an avatar driven by the investigators' mathematical model of coupled dynamical systems, thereby allowing the investigators to manipulate parameters (the 'personality' or 'attitude' of the virtual partner) that are not normally accessible in studies of live interactions. Thus the present project may not only uncover rules of ordinary social coordination but also offer a principled approach to human-machine interaction. In addition, the project will disseminate knowledge about complex systems and dynamical approaches to human social behavior. The investigators will train undergraduate and graduate students in advanced methods and analysis techniques that cut across the behavioral and social sciences, the physics and mathematics of coupled dynamical systems, and neuroscience and brain imaging. The development of new measuring instruments that may apply to any arbitrary social interaction is a target. Such training aims to bring forth a new generation of interdisciplinary scientist who will be equipped to integrate studies of brain, behavior and social function within a much needed dynamical framework.

DESCRIPTION (provided by applicant): The goal of this project is to discover the dynamical principles and mechanisms at play both within and between human brains during real-time social interaction. The research plan employs a three-pronged approach that combines (1) experimental manipulations to test specific hypotheses regarding key issues in the neurophysiology of social neuroscience (2) sophisticated measurement and analysis tools from the theory of dynamical systems, including virtual partner interaction (behavioral dynamic clamp of reciprocally coupled humans and model-partners) and (3) multiscale neurocomputational modeling of both structure and function in order to advance our understanding of how individual behavior and the interaction of individuals drives basic forms of social behavior. In our previous research, we established a comprehensive framework to tackle real- time interactions between people in simple, well-defined experimental paradigms in which pairs of participants simultaneously performed and perceived each other's movements. The research program led to the discovery of the phi complex, a neuromarker of social coordination. Also clarified were the contributions of other neuromarkers, especially alpha and mu, to different phases and facets of social behavior. What is most needed now -and what we seek support for in the present Competing Renewal- is to understand the dynamical orchestration of identified neuromarkers over the course of social behavior. The experimental thread works hand-in-hand with neurocomputational modeling of social behavior, theoretical models informing experiments and vice-versa. The aims of this research program-still very much in its infancy-- are (1) to elucidate the neuromarker choreography, that is, to determine when each neuromarker is recruited and disengaged, which neuromarkers originate from which brain areas and how neuromarkers interact with each other in transient networks during the course of social behavior. All of the proposed work is geared to the prediction of efficient or deficient outcomes as assessed by detailed single trial analysis of real-time social behavior; (2) to construct a human dynamic clamp that allows for direct manipulation of the interaction between human participants and virtual partners endowed with human appearance and coordinative capacities. This new paradigm opens up the detailed parametric exploration of social behavior; and (3) to integrate the findings in a multiscale neuro- computational model of social behavior, a platform that will enable understanding of basic mechanisms of interpersonal interactions at combined neural, behavioral and social levels. Successful achievement of this program will specify the neurobehavioral routes leading to improved social function. Given the vast number of pathologies with etiological or symptomatic ties to social behavior, such information will afford many translational opportunities for the compensation or remediation of deficits in diseases such as autism, schizophrenia, depression and dementia to name just a few.

By reconnecting the previously severed sense of touch, the field of neuroprosthetics has tremendous potential to substantially improve the lives of millions of amputees and disabled people worldwide. However, the rate of progress to develop neuroprosthetic limbs has been comparatively slow relative to other areas of robotics for two primary reasons: research involving neural implants with human subjects is very expensive and a lengthy process is required to obtain FDA approval to implant electrodes in human subjects. Thus, the overall goal of this project is to develop a virtual neuroprosthesis in which a facsimile of a neural implant is externalized and housed in a well-controlled microfluidic chamber, thereby abating the intrinsic limitations of highly invasive studies with neural implants. Upper limb amputee subjects will be recruited to control a dexterous artificial hand and arm with electromyogram signals while electroencephalogram (EEG) signals are simultaneously measured. Robotic grip force measurements will be biomimetically converted into electrical pulses similar to those found in the peripheral nervous system to catalyze in. vitro nerve regeneration after neurotrauma. The synergistic contributions of this multidisciplinary project will lead to a transformative understanding of the symbiotic interaction of neural plasticity within human-robotic systems. Currently, there is no systematic understanding of how tactile feedback signals can contribute to the neural regeneration of afferent neural pathways to restore somatosensation and improve motor function in amputees fitted with neuroprosthetic limbs. Tackling this problem will be a significant breakthrough for the important field of neuroprosthetics. The proposed virtual neuroprosthesis will be much less expensive and vastly simpler to obtain IRB approval to conduct research with human subjects. Through this, the research team can conduct meaningful neuroprosthetic experiments with human subjects at a fraction of the cost while accumulating significant data much quicker.

Project summary The key benefit of our virtual neuroprosthetic framework proposed in the parent R01 grant (1R01EB025819) is that multiple hypotheses can be tested simultaneously using dorsal root ganglia (DRG) cultures in a cost- effective manner. The overall goal of this supplement grant is to augment the haptic signaling pathways that are restored in the virtual neuroprosthetic platform from one to four, consistent with human anatomy. Compared to the parent R01 in which we proposed to model only one pathway for a type of slowly adapting mechanoreceptor in the fingertip (continuously signaling the grasp force applied by the robotic hand), we propose herein to explore multiple parallel channels to mimic the complex functional anatomy of the hand which contains at least four different sensory receptor types carrying information on different aspects of the haptic experience (pressure, vibrational information used to detect slippage, texture, hand conformation, etc). These signals provide important complementary haptic information for the fine control of movements. Therefore, neurostimulation of DRG cultures mimicking the signals from different mechanoreceptors will produce distinctive spatial, temporal and functional forms of DRG regeneration post-axotomy, and the resulting functional diversity of regenerated pathways will ultimately play a key role in the overall quality of somatosensory restoration after amputation, supporting better usage of haptic feedback-enabled hand protheses. Specifically, we propose to (1) design highly realistic electrical stimulation patterns from the four principal types of slowly adapting (SA) and rapidly adapting (RA) mechanoreceptors to electrically stimulate DRGs cultured in microfluidic chambers, (2) to study the contribution of those multiple pathways, individually and synergistically, to anatomical and functional restoration, and (3) to establish if this restoration scheme surpasses one with the lower attentional load but lesser informational richness obtained when the haptic feedback is derived from only one functional pathway, as is currently under investigation within the parent R01.

Project Summary/Abstract This supplementary research proposal applies the expertise, techniques and hardware developed under the funded parent grant (R01EB025819) to study the contribution of neuronal activity to the pathogenesis of Alzheimer?s disease (AD). AD is a progressive multifactorial neurodegenerative disorder and the major type of dementia. Neuronal activity shows a complex relationship with AD, with (1) neurons and areas susceptible to more intense neuronal activity, clinically hyper-excitable or stimulated intensely exhibiting evidence of AD pathogeny in humans, animal models and cell cultures; and (2) neurons or synapses whose activity is reduced or putatively under-stimulated by lack of cognitive engagement also demonstrating altered prospects as the disease progresses. Since neuroinflammation is one of the central mechanisms in AD pathogenesis and an area currently under intensive research, elucidation of its systemic drivers at the neuronal level and pathogenic impact will provide mechanistic insights on disease progression and uncover intervention principles. We hypothesize that one of key mediators of this nonlinear relationship between neuronal activity and AD is neuroinflammation, because studies independently linked neuroinflammation to both sides of the hypothetical equation AD=f(neuronal activity). Specifically in this application, using an Alzheimer?s-in-a-dish model with neurons- microglia cocultures derived from induced pluripotent stem cells (iPSCs) of AD patients, we aim to determine the electrical parameters that modulate neuroinflammatory response and how this relates to AD progression. Our operational hypothesis is that different patterns of electrical stimulation will nonlinearly affect neuroinflammatory responses in AD neuron-microglia co-cultures, which in turn contributes to the pathogenesis of AD at the neurons? structural and functional levels. To test this hypothesis, we propose to deliver different patterns of electrical stimulation to Alzheimer?s-in-a-dish models and measure cytokine release using cytokine array (Exp. 1), analyze microglia migration behavior with impedimetric monitoring (Exp. 2) and investigate neuronal function electrophysiologically with multielectrode array (MEA) (Exp. 3). This plan is executed by an interdisciplinary team of 4 principle investigators whose skills cover broadly the needs of the research plan, including an expert in the molecular biology of neurodegeneration/regeneration, a roboticist expert in control systems; a biosensor and microfluidic nanoengineer and a complexity neuroscientist expert in electrophysiological spatiotemporal dynamics. This supplement grant is directly focused on investigating AD pathogenesis in the presence of different neuronal stimulation patterns, which resemble physiological or pathological neuroelectric events responding to environmental sources. Our proposed studies, thus well aligned with Milestone 2H outlined in the research implementation plans by National Institute of Aging (NIA), provide a novel in vitro platform with integrated electrical stimulation and cellular components and a broadly integrated analysis toolkit to gain more detailed mechanistic insights on neuronal activity-mediated AD progression.

Social interactions are beneficial for aging individuals, be they healthy or affected by degenerative disease. Social engagement not only brings support but also acts on brain structure, behavior and cognition to slow aging. The aging process, however, produces multiple changes that compromise people?s ability to interact with others (person too slow or frail to keep to regular outings with family; thought processes and verbal fluency too sluggish, and hearing too feeble, to secure sufficiently frequent turns-at-talk in group conversation with younger individuals; etc.), leading to attrition of numerous and varied social links. The goal of this research program, to be conducted by an interdisciplinary team of socio-cognitive scientists, mathematical physicists and geriatric nursing experts, is to gain understanding on first principles underlying preservation or loss of social interaction in aging. This is accomplished by mathematically modeling the underpinnings of social integration and segregation within heterogeneous groups of older and younger individuals, and by conducting observational studies of group activities involving elderly and younger adults in a memory and wellness center (storytelling, gentle yoga, music making). Our mathematical model of social coordination (empirically validated in young adults) allows to vary each agent?s ?coupling? capabilities, (behavioral or cognitive) slowing pace, the memory process of social adaptation and behavioral noise level. Preliminary evidence suggests that pace discrepancy and weak coupling lead to briefer, less frequent periods of coordination, which are fundamentally scale- and context-dependent. Theory also suggests that noise enhances the stability of heterogenous groups, while tending to disrupt that of more homogenous ones. All of those preliminary findings point to systemic effects: interactional opportunities not only depend on individuals, but also nontrivially on the match or mismatch between individuals and their social environment. Therefore, the contribution of this research on methodology and measurement in the behavioral and social sciences lies in a much-needed emphasis on this systemic of social behavior, that is, the effect that the whole exerts on the parts, a key property of complex systems. It leads to an exemplary framework for quantifying individual and collective behavior; provides analysis strategies to characterize their entanglement; and identifies cues to recognize when systemic effects are likely at play. The specific aims of this projects are, first and via models, to quantify the effect that the social environment exerts on elderly social interactions and second, via empirical observations, to develop translational work that connects the model?s first principle with verified outcomes so as to engineer social interactions that maximize connectedness. All of those advances will help to sustain behavioral and cognitive reserve and extend the span of healthy and functional aging. Because of its foundation in a general mathematical model of coordination, the findings and their methodology can also apply in a broad range of translational contexts, including the many facets of communicable health and communicable disease.