Vestibular-Oriented Research Meeting, June 25 – 29, 2023

Podium Abstract 1Overview of funding opportunities and research supported by the National Institute on Aging

Coryse St. Hillaire-Clarke, Ph.D.

Program Director, Sensory and Motor Disorders of Aging Program, Division of Neuroscience, NIA

The National Institute on Aging (NIA) leads a broad scientific effort to understand the nature of aging and to extend the healthy, active years of life. NIA is also the primary federal agency supporting and conducting Alzheimer’s disease (AD) research. Over the last several years the institute has seen an unprecedented increase in federal appropriations for AD and AD-related dementias (ADRD) which has translated into generous funding pay lines. The NIA has had a longstanding interest in funding research to investigate the mechanisms underlying age-related changes in sensory and motor function both in the context of healthy aging and neurodegenerative disease. The presentation will provide an overview of NIA’s interests in vestibular research and balance, current funding pay lines, notices of funding opportunities of relevance to vestibular researchers and training opportunities for junior investigators.

Podium Abstract 2Overview of funding opportunities and research supported by the National Institute on Deafness and other Communication Disorders

Amy Poremba, Ph.D.

Program Director, NIDCD

Dr. Amy Poremba will present an overview of current research funding opportunities for students, postdoctoral fellows, early-stage investigators, and established investigators. Past and current funding levels will be presented for an overview of the priorities of each institute and current priorities for funding will be addressed.

Keynote Abstract 1Screening for Vestibular Disorders

Helen S Cohen, EdD, OTR, FAOTA

Bobby R Alford Department of Otolaryngology – Head and Neck Surgery, Baylor College of Medicine

When patients enter the health care system with complaints that might be consistent with vestibular disorders, or when they mention such complaints in the course of visits focused on other health problems, they are not immediately referred for objective tests of vestibular system (VNG). Instead, the health care provider may take a history, or give the patient a history questionnaire. The clinician might then use some screening tests that can be easily performed in the exam room. Only then, if the clinician is a medical provider who is knowledgeable about the vestibular system and if that provider has access to a facility where VNG’s are performed, the patient might be referred for a VNG. Alternatively, the patient might just be referred for a basic hearing test battery, or no further tests might be ordered, depending on the background of the clinician and the resources available. Therefore, having valid and reliable screening tests is important. Despite the large number of screening tests and history questionnaires in the literature, the value of these mechanisms for accurate screening of vestibular disorders remains unclear. Screening is notoriously unreliable, more art than science. This talk will review the literature on screening for vestibular disorders, including history questionnaires, quality-of-life self-ratings, head impulse testing, Fukuda stepping test, Romberg testing, tandem walking, blood pressure screening, and use of the basic hearing test battery to predict VNG outcome. These tests are relevant to medical providers, rehabilitation clinicians, and also epidemiologists who perform local and population-based studies of vestibular disorders.

Acknowledgement: Funded by NIH/NIDCD R01 DC009031.

Podium Abstract 3Executive functions in patients with bilateral and unilateral peripheral vestibular dysfunction

Corina G. Schöne^{1, 2, 3}, Dominique Vibert ², Fred W. Mast ¹

¹Department of Psychology, University of Bern, Bern, Switzerland.

²Department of Otorhinolaryngology, Head and Neck Surgery, Inselspital, University Hospital Bern, University of Bern, Bern, Switzerland.

³Doctoral Program for Brain and Behavioral Sciences, University of Bern, Bern, Switzerland.

Patients with peripheral vestibular dysfunction (PVD) suffer from cognitive difficulties in several visuo-spatial as well as non-visuo-spatial domains. Associations of PVD and executive problems have been made, but no study investigated a range of different executive domains in patients with PVD. We administered a broad neuropsychological test battery of basic and complex executive functions to 84 patients with different kinds of PVD (34 chronic bilateral, 29 chronic unilateral, 21 acute unilateral) and 49 healthy controls matched for age, education, and sex. Descriptive data analysis showed that patients with PVD scored lower in specific executive domains than healthy controls. Results indicate that patients with PVD should be screened for specific executive impairments and could benefit from a cognitive training targeting executive functions.

Podium Abstract 4Otolith-ocular Function and Compensatory Effect of Neck Following Vestibular Loss

Yuchen Yang, B.S.^1,2; Jing Tian, Ph.D.¹; Jorge Otero-Millan, Ph.D.^1,3; Michael C Schubert, Ph.D. ^2,4; Amir Kheradmand, M.D.^1,2,5

¹Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD

²Department of Otolaryngology-Head and Neck Surgery, Johns Hopkins University School of Medicine, Baltimore, MD

³Herbert Wertheim School of Optometry & Vision Science, University of California, Berkeley, CA

⁴Department of Physical Medicine and Rehabilitation, Johns Hopkins University School of Medicine, Baltimore, MD

⁵Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, MD

Vestibular inputs converge broadly with other sensory information across various brain functions. These multimodal contributions range from control of gaze and posture to higher-level mechanisms that mediate spatial orientation and motion perception. At the level of vestibulo-ocular reflex (VOR), there is a close interaction between vestibular signals and proprioceptive inputs that encode the position of the neck. Neck proprioceptors provide information about the position of the head relative to the trunk and they can be used by the brain to optimize gaze control with respect to the head and body positions. The neural pathways that mediate these responses are not well understood, but physiological observations suggest that the brain can adapt to perturbations in vestibular function by using inputs from the neck. Here we studied the contribution of neck inputs to otolith-ocular function at different stages of recovery from vestibular loss. We used video-oculography (VOG)-based measurement of static ocular counter-roll (vOCR) determined 30 seconds after the dynamic tilt to assess otolith function. To elucidate a compensatory effect of neck proprioception, vOCR was measured with 30° static whole-body tilt (WBT) and head-on-body tilt (HBT). The vOCR with WBT is primarily driven by the otolith inputs, whereas with HBT there is also a neck contribution. Fifty-six subjects were recruited including patients with acute (9 days [mean]), subacute (61 days), and chronic (985 days) unilateral loss of vestibular function, as well as a group of healthy controls. The gain of vOCR on the lesion side was lower with WBT than with HBT in the acute and subacute groups but not in the chronic group or controls (HBT: acute: 0.11±0.01, subacute: 0.14±0.01, chronic: 0.13±0.02, healthy control: 0.17±0.01/ WBT: acute: 0.08±0.01, subacute: 0.11±0.01, chronic: 0.13±0.02, healthy control: 0.18±0.01). The time course of vOCR response was affected as well with a reduced amplitude and slower response in the acute stage of vestibular loss. The vOCR with WBT, as opposed to HBT, could distinguish among the patient groups at various stages of vestibular loss. These findings show a compensatory effect of neck proprioception that can enhance otolith-ocular function following vestibular loss and vOCR as a potential valuable clinical marker to track recovery following vestibular loss.

Podium Abstract 5Bone Conducted Vibration for the Symptomatic Relief of Vertigo - Clinical Trial Phase 2 Results

Didier A. Depireux, Daniel Stolzberg, Alesia G. Robinson, Jonathan Akers, Jack Daggitt, Samuel Owen

Otolith Labs, Washington, DC

Vertigo—the false perception of motion or rotation—is often the result of damage and dysfunction of vestibular end organs. Despite the high incidence of vestibular vertigo in the adult population (~7.4%, Agrawal et al 2013), available treatments require pharmaceuticals, surgical intervention, and/or extensive physical therapy.

The OtoBand, a lightweight, battery powered, externally worn device developed by Otolith Labs, is a novel device that can rapidly reduce or relieve vestibular vertigo by sending a low frequency vibration to vestibular end organs through bone conduction.

We report results from an ongoing phase 2 double-blind counter-balanced sham-controlled study designed to quantify changes in vertigo severity associated with use of the OtoBand in participants during recurring episodes of vertigo. This study was carried out in a telehealth setting to accommodate COVID restrictions and to test the device in a real-world setting. Inclusion criteria included a Dizziness Handicap Inventory score > 35, chronic vertigo (condition for at least 90 days) with at least one episode of vertigo per week, and provided a self-reported diagnosis of their vestibular pathology (benign paroxysmal positional vertigo, Meniere’s disease, vertiginous migraine, or vestibular neuritis/labyrinthitis).

In this study, the OtoBand device produced a single frequency between 50 and 65 Hz at a patient-controlled force level between 0.4 N_RMS to 0.8 N_RMS. A sham device operating at a different frequency and a different range of patient-controlled force levels was also supplied with an identical form factor to the OtoBand, with vibrations from both devices easily perceived by participants. Participants were provided with an OtoBand and a sham device in a random, counterbalanced order for two weeks each. Participants were instructed to wear their device whenever experiencing an episode of vertigo and to report on the severity of vertigo episode at several time points, and an overall rating of efficacy after each use.

With “responder” defined as a participant who rated the device as helping on at least half of uses, a significant main effect of OtoBand over sham device was found using a generalized linear mixed-effects logistic model (odds ratio = 3.13, 95%, CI = [1.75–5.61], p < 0.001). This indicates that the OtoBand was found to be significantly more helpful to participants during vertigo episodes than the sham.

A continuing pilot study with a larger sample size and medically confirmed medical diagnoses is currently underway, with a large-scale pivotal study planned for mid-2023.

Keynote Abstract 2CRISPR-Cas9 Mediated Targeting of a DFNA9 Pathogenic Variant for Gene Therapy in an Auditory and Balance Disorder

CC Morton^1,2,3,4, NG Robertson¹, S Vijayakumar⁵, C Gurumurthy⁶, BP Kleinstiver^2,7, R Sherwood^2,8, MV Ivanchenko^2,9, KT Booth^2,9, DP Corey^2,9, H Romi^1,2

¹Departments of Obstetrics and Gynecology, Brigham and Women’s Hospital

²Harvard Medical School, Boston, MA

³Broad Institute of MIT and Harvard, Cambridge, MA

⁴Manchester Center for Audiology and Deafness, Manchester, UK

⁵Creighton University, Omaha, NE

⁶University of Nebraska, Omaha, NE

⁷Center for Genomic Medicine and Department of Pathology

⁸Department of Medicine, Brigham and Women’s Hospital

⁹Department of Neurobiology

Impressive advances in gene discovery in the auditory system have occurred in the last decades, making specific targeted approaches for therapeutics realistic goals of great interest. Hearing loss (HL) is an increasingly significant health problem in populations worldwide, with a substantial proportion due to genetic causes. Some auditory disorders are also accompanied by abnormalities in the vestibular system. Given the health burden and ongoing rapid discoveries, it is now essential to pursue new strategies for specific early interventions that could prevent or mitigate severity and progression of hearing and vestibular disease. One such disorder is DFNA9, an adult-onset sensorineural HL with balance dysfunction, reported with 41 missense variants and two in-frame deletions in COCH, encoding cochlin, the most abundantly detected protein in the inner ear. This disease model is representative of the majority of dominant genetic HL disorders due to a deleterious gain-of-function of the mutant cochlin. COCH mutations result in a distinct aggregative histopathology in inner ear structures, pathognomonic for DFNA9, and multimerization of mutant cochlin in vitro. Dominant negative properties of mutant cochlin are consistent with evidence that COCH haploinsufficiency in DFNA9 does not lead to the pathology in humans, and heterozygous mice in the Coch null mouse model have normal hearing function similar to wild-type littermates. Cochlin contains a 5’ motif known as LCCL (Limulus clotting factor C, Cochlin and mammalian Late gestation lung protein [Lgl1]) in which 20/41 COCH variants are located; 17 of 41 variants are clustered in a 3’ vWFA2 domain. The LCCL domain has a novel fold and is proposed to protect Limulus from bacterial infection and could serve a similar defensive function in the inner ear. Phenotypically, the onset of HL and vestibular symptoms in humans with COCH variants in the 5’ region of the gene is in the 5^th-6^th decade whereas the onset of HL is in the 2^nd-3^rd decade and is not characterized generally by a vestibular disorder. A genomic medicine approach resulting in gene silencing of the mutant COCH allele (p.A449T) in a humanized mouse model is underway, with disruption of the p.A449T dominant mutation using CRISPR-Cas9 maintaining the normal allele and rescuing hearing and balance.

We first generated a fibroblast cell line using a skin biopsy from an individual with the COCH p.A449T variant to assess whether we could rescue hearing in vivo. Using EasiCRISPR, we created a humanized mouse model in CBA/J with the COCH p.A449T mutation and flanking human sequence to match the gRNAs targeting the human mutation. We transfected cells with a plasmid carrying SaCas9-KKH and a guide RNA targeting A449T, and assessed editing with next-gen sequencing. The Cas9 successfully created indel disruption only in the variant allele, with an editing efficiency of 43% and with no disruption of the normal allele. Auditory brainstem response (ABR) as well as distortion product otoacoustic emission (DPOAE) analyses of 3-4 month-old mice showed significantly elevated thresholds for both Coch^A449T/+ (heterozygous) and Coch^A449T/A449T (homozygous) mice, as compared to their Coch^+/+ (wild-type) littermates, at all tested frequencies. There was no significant difference between the heterozygotes and homozygotes, reflecting the dominant nature of the mutation. Vestibular sensory evoked potentials (VsEP) were measured also at 3-4 months of age and thresholds were significantly elevated for heterozygous and homozygous mice compared to the wild-type littermate controls. Coch p.A449T mice showed severe swimming abnormalities: both heterozygous and homozygous mutants exhibited underwater tumbling, with homozygous mutants performing much worse than heterozygous animals. Round window micro-injection of an AAV encoding CRISPR-Cas9 and guide RNA has been performed and phenotype rescue is being assessed.

We have shown successful targeting of a COCH pathogenic variant using a CRISPR-Cas9 mediated approach in vitro and are testing rescue in a novel mouse model in vivo. If successful, this strategy is expected to inform gene therapy in other autosomal dominant HL disorders, the majority of which are due to missense pathogenic variants rather than truncation or loss of gene function. Notably, these AD disorders are often characterized by late-onset presentation, providing a window of opportunity for therapeutic intervention.

Podium Abstract 6Manipulating gene expression in adult vestibular type II hair cells promotes transdifferentiation toward the type I hair cell phenotype

Antonia González-Garrido¹, Omar López-Ramírez¹, Brandon Cox², Jennifer Stone³, Ruth Anne Eatock¹

¹University of Chicago, Department of Neurobiology, Chicago, IL

²Southern Illinois University School of Medicine, Department of Pharmacology, Springfield, IL

³University of Washington School of Medicine, Department of Otolaryngology, Virginia Merrill Bloedel Hearing Research Center, Seattle WA

The vestibular epithelia of mammals have two markedly distinct hair cell types, I and II. Type I and II hair cells have different hair bundles, calcium-binding proteins, synaptic connections with both afferent and efferent nerve fibers, and expression of ion channels. Type I cells emerged in evolution as tetrapods moved onto land, possibly in response to novel challenges of locomotion and balance out of water. Adult type II hair cells retain some features of immature hair cells, such as expression of the transcription factors Atoh1 and Sox2, whereas type I hair cells are more differentiated. This difference may explain why new type II cells, but not type I cells, can form in adult epithelia after damage (Golub et al. 2012; González-Garrido et al. 2021). The regenerated type II cells alone do not restore vestibular function, suggesting that type I hair cells are essential. To explore the differentiation of type I and II hair cells from each other, and as a stepping stone to functional regeneration of the inner ear, we have investigated whether adult type II cells can be transformed to type I cells by deleting expression of Sox2.

Our results show that deleting Sox2 causes a significant number of type II cells to acquire anatomical (Stone et al. 2021) and physiological traits that are type I markers. Of note, primary afferent neurons alter their synaptic contacts with the hair cells from bouton synapses to the extraordinary calyceal terminals typical of type I synapses. Whole-cell recordings from the newly generated calyces showed the unusual non-quantal transmission that is a key functional distinction of type I-calyceal synapses. This transmission is thought to be mediated by a large and negatively activated potassium conductance described only in type I hair cells. Indeed, our whole-cell recordings from Sox2-deleted type II hair cells show transformation of the type II potassium conductance toward type I-like size and voltage dependence. These results suggest that a relatively small number of manipulations can produce functional differentiation of newly generated hair cells in adult vestibular inner ear.

References

(1) González-Garrido A, Pujol R, López-Ramírez O, Finkbeiner C, Eatock RA, Stone JS. The differentiation status of hair cells that regenerate naturally in the vestibular inner ear of the adult mouse. J Neurosci. 2021;41(37):7779-7796. doi: 10.1523/JNEUROSCI.3127-20.2021. PMCID: PMC8445055.

(2) Stone JS, Pujol R, Nguyen TB, Cox BC. The transcription factor Sox2 is required to maintain the cell type-specific properties and innervation of type II vestibular hair cells in adult mice. J Neurosci. 2021;41(29):6217–33. doi: 10.1523/JNEUROSCI.1831-20.2021. PMCID:PMC8287988.

(3) Golub JS, Tong L, Ngyuen TB, Hume CR, Palmiter RD, Rubel EW, Stone JS. Hair cell replacement in adult mouse utricles after targeted ablation of hair cells with diphtheria toxin. J Neurosci. 2012 32(43):15093-105. doi: 10.1523/JNEUROSCI.1709-12.2012. PMCID: PMC3544304.

Funding: NIDCD R01 DC013771 (JS, BC, RAE) and R01 DC012347 (RAE)

Podium Abstract 7Sox2 is required to maintain the morphology and afferent innervation of type II hair cells and to regulate motor behaviors of adult mice

Hans Baertsch¹, Hendrik Dorssers¹, Tot Nguyen¹, James O. Phillips¹, Ruth Anne Eatock², Brandon Cox³, Jennifer Stone¹

¹University of Washington School of Medicine, Department of Otolaryngology, Virginia Merrill Bloedel Hearing Research Center, Seattle WA

²University of Chicago, Department of Neurobiology, Chicago, IL

³Southern Illinois University School of Medicine, Department of Pharmacology, Springfield, IL

Amniotes have two types of vestibular hair cell, type I and type II, that have very different morphology, synapses, molecular properties, and physiology. In adult mice, deletion of Sox2 from type II hair cells causes them to transdifferentiate toward the type I fate - i.e., to lose type II-specific anatomic and molecular features and to acquire features unique to type I hair cells (Stone et al. 2021). In addition, the bouton afferent terminals that typically contact type II hair cells are replaced by a partial or full calyx-type terminal as the hair cell transdifferentiates.

We asked whether this type II-to-I shift affects motor behavior and vestibular reflexes. We found that, instead of undergoing the typical reduction in locomotor activity as they age, experimental mice (those with type II-to-I transdifferentiation) had elevated locomotor activity, as reflected by increased velocity and distance traveled during brief (5-minute) periods in a bright cage. Experimental mice also had more exploratory behavior, as measured in the elevated plus maze or in 72-hour open field. Some experimental mice had better performance (lasted longer) on the rotarod than control mice. Mice had a small decrease in gains for the angular vestibulo-ocular reflex in the horizontal plane at 4 months but not at 8 months after the fate switch was initiated. However, mice with type II-to-I transdifferentiation exhibited no circling or head-bobbing behaviors, which often occur after substantial loss of either hair cells or hair cell function.

Because we found no evidence for off-target Sox2 deletion from vestibular pathway neurons, we surmise that the observed changes were due to type II-to-I hair cell transdifferentiation. In future studies, we hope to determine if changes in behavior reflect the gain of type I hair cells, the loss of type II hair cells, and/or destabilization of the vestibular periphery and how these changes affect central neurophysiology.

References

(1) Stone JS, Pujol R, Nguyen TB, Cox BC. The transcription factor Sox2 is required to maintain the cell type-specific properties and innervation of type II vestibular hair cells in adult mice. Journal of Neuroscience. 2021;41(29):6217-33. PMC8287988.

Funding: DC013771 (JS, RAE, BC, JOP); The Hamilton and Mildred Kellogg Charitable Trust (JS); The Whitcraft Family (JS).

Podium Abstract 8Characterizing Hair Cell and Afferent Neuron Properties in Human Vestibular Organs

Olivia Kalmanson, MD, MS; Frances Meredith, PhD; Mohammad Al-Amin, PhD; Anna Dondzillo, PhD; Tiffany Vu, BS; Samuel Gubbels, MD; Katie Rennie, PhD

Department of Otolaryngology-Head & Neck Surgery, School of Medicine, University of Colorado, Anschutz Medical Campus, Aurora, CO

The vast majority of what is known about human vestibular function has been inferred from rodent models due to the inaccessibility of human inner ear organs. There are a few reports of electrophysiological recordings from cells isolated from human inner ear (Oghalai et al. 1998, Lim et al. 2014, Quinn et al. 2021), but detailed investigations of freshly harvested adult vestibular cells are still lacking. Here, we obtained vestibular neuroepithelia from adult translabyrinthine surgical patients and performed electrophysiological and immunohistochemical studies.

Colorado Multiple Institution Review Board approval was obtained (COMIRB# 19-1340). The surgical team obtained ampullae and utricles from consenting patients undergoing translabyrinthine surgical approaches. Neuroepithelia were transferred in-ice cold sterile saline to the laboratory within one hour of harvest. For electrophysiological recordings, the tissue was transferred to L-15 solution following incubation in L-15/bovine serum albumin and cells were mechanically dissociated as previously described (Rennie & Streeter 2006). Whole cell patch clamp recordings were performed on isolated hair cells and afferent terminals to characterize voltage-dependent currents. Pharmacological agents were applied via extracellular perfusion. For immunohistochemistry, end organs were fixed and stained with a combination of antibodies targeting Myosin7a, CtBP2 and Tubulin3. Imaging was performed with a Zeiss LSM 780 confocal microscope.

Six patients (ages 29-77 years, 4M, 2F) with vestibular schwannomas and non-serviceable ipsilateral hearing loss prior to surgery were studied. None reported vertigo. Recordings were obtained from type I and type II hair cells and afferent terminals up to 5 hours following dissociation. Hair cells had a mean capacitance of 8.9±5.4pF (n = 7, SD). Type I hair cells exhibited low voltage-activated currents and large outward currents in response to standard voltage protocols. Mean resting potential was -60.4±7.2 mV (n=14) and peak outward was 3.37±1.30 nA at +20 mV (n=15). Putative type II HCs exhibited higher input resistance and outward current above -50mV. A hyperpolarization-activated current (I_h) was observed in 4 hair cells. Afferents exhibited rapid transient inward Na⁺ current, outward current (I_K) and I_h. Peak I_K was 3. 77±2.51 nA (n=4) and I_h was 331±191 pA (n=3). Single action potentials occurred with depolarizing steps in current clamp. Perfusion with the K⁺ channel blocker 4-aminopyridine (1 mM) blocked low voltage-activated currents in type I cells and inactivating I_K in calyx afferents.

Immunohistochemical labeling of cristae and utricles allowed identification of hair cells, afferent terminals, and ribbon synapses.

We performed electrophysiological recordings and immunostaining on vestibular cells extracted during human surgical inner ear procedures. Whole cell recordings from hair cells and afferent terminals revealed ionic currents that qualitatively resemble currents in cells from rodent vestibular epithelia. Rapid access to adult human vestibular epithelia allows translational studies crucial for better understanding human peripheral vestibular function.

Podium Abstract 9Quantitative Aspects of Vestibular Mitochondria

Anna Lysakowski¹, Ph.D.; Guy Perkins², Ph.D.; Team Mito undergraduates³

¹Dep of Anatomy and Cell Biology, Univ of Illinois at Chicago, IL

²National Center for Microscopy & Imaging Research, Univ California, San Diego, CA

³Dep of Biological Sciences, Univ of Illinois at Chicago

In this study, we quantified multiple aspects of mitochondrial structure in the vestibular periphery. Our main hypothesis in this research is that mitochondria are structurally heterogeneous, depending on which hair cell type is under consideration and the specific locations of mitochondria in relation to other hair cell organelles. One such structural relationship that we have reported on previously is the polarization of mitochondrial crista junctions (CJs), openings of the internal mitochondrial membrane into the intermembrane space that is likely responsible for transport of molecules from the mitochondrial matrix out to the cytoplasm.

Inner ear vestibular tissue obtained from Long-Evans rats under an approved UIC IACUC procedure was used for electron microscope tomography. Data were processed with the IMOD (v. 3.13.6) software package (Kremer et al., 1996) and serial tomograms were joined using the etomo subroutine (IMOD). Manual segmentation was done with the 3dmod subroutine (IMOD) to obtain final models. Reconstructions were visualized using 3dmod (IMOD), including its SLICER option, to track relevant features in three dimensions. Several IMOD drawing tools (Sculpt, Join, Warp, Interpolator) and a proximity analysis subroutine (Mtk) with its randomization feature, were also used.

Quantitative analyses were performed to test our hypothesis, including counts of total mitochondria within both types of hair cells (with a comparison to cochlear hair cells), efferent boutons, and afferent calyces, counts of CJs on either side of a mitochondrion in relation to other cellular organelles, analyzing polarization ratios (side toward vs side away) vs distances, bioenergetic calculations of ATP production, and an Mtk proximity analysis test was performed using CJ randomization as a control. Results support our hypothesis of structural heterogeneity of mitochondria, as CJs are polarized toward specific organelles within the cell and appear to be non-randomly distributed. Finally, vestibular hair cells have less than half the number of mitochondria compared to cochlear hair cells, although certain of these are quite a bit larger than cochlear mitochondria.

These results have functional implications for the various structural differences. CJ polarization toward organelles that use energy can transport ATP more efficiently. For example, the cuticular plate is an organelle that functions to return stereocilia rootlets back to their original position. Our results show a non-random orientation of mitochondrial crista junctions toward the cuticular plate. These structural differences may provide key information to better understand their function and to address mitochondrial deafness and dizziness disorders.

Funding: NIDCD R21-DC013181 (AL) and UIC Chancellor’s Undergraduate Research Awards (CURAs, to members of Team Mito, DR, ZA, PS, JL).

References

Kremer JR, Mastronarde DN, McIntosh JR. Computer visualization of three-dimensional image data using IMOD. J Struct Biol. 1996;116:71-76.

Keywords: Mitochondria, vestibular hair cells, vestibular afferents, vestibular efferents.

Podium Abstract 10Novel methodology and experimental ratings for real-time computational detection of pilot spatial disorientation

Jordan B. Dixon, Ph.D.¹, Tristan Endsley, Ph.D.¹, Torin K. Clark, Ph.D.²

¹The Charles Stark Draper Laboratory, Inc., Cambridge MA

²University of Colorado – Boulder, Boulder CO

Pilot spatial disorientation (SD) continues to be a leading cause of Class A mishaps and fatalities in aviation even after the broad introduction of SD training. Active countermeasures, such as adaptive displays or auditory cueing, have potential to further assist pilots with maintaining appropriate orientation of the aircraft. A necessary component of such a system, however, is a method by which to estimate possible SD being experienced and detect when intervention is warranted in real-time. Computational models of human orientation perception provide an unobtrusive means by which to estimate the impact of misperceptions, over time, on overall SD state. To overcome limitations of previously proposed SD models, we developed an experimental methodology to collect quantitative ratings of SD in response to inertial motion stimuli. This enabled identification and characterization of computational tools best suited for capturing SD as rated empirically. The state-of-the-art ‘Observer’ model of human orientation perception was leveraged in series with candidate algorithms to capture the temporal dynamics of SD. Importantly, these algorithms combine multiple, multidimensional, disparate aspects of orientation perception to compute a single, unidimensional metric of SD that can be readily used as a trigger for an active countermeasure system. To establish appropriate configurations for the computational tool, we collected experimental ratings of SD in a ground-based motion simulator. The Tilt Translation Sled was configured to provide inertial motion stimuli in the coronal plane (lateral drift and roll-tilts) in coupled relationships to represent a helicopter in hover, in a degraded visual environment, in the presence of wind. Nine subjects provided SD ratings in response to 50 experimental motion trials ranging from 15-30 seconds. At some point in each trial, the subject display switched from a mission-related visual search task, to instrument displays, representing a pilot making a cross check of instruments. At this moment they were tasked to compare their perception at the moment of the transition to the veridical information depicted on the instrument displays. Subjects rated their SD on a visual analog scale which they calibrated to during training trials. From a multidimensional trade space of algorithm types and parameter values, a subset of candidate computational tools was identified that reasonably predicted mean empirical SD estimates. An important finding from this investigation was that there was no single optimal computational tool, and model-based detection of SD likely should be tuned to specific use-cases. Our characterization methodology was designed to be generalizable such that the computational framework can be readily applied to different environments, vehicles, situational contexts, etc. Signal detection performance has been shown to impact operator usability of assistive systems, and as such, use of the system may also warrant personalization by the individual pilot. A follow-on methodology involving human-in-the-loop control of the simulated helicopter is proposed in order to characterize potential performance and safety benefits of integrating a real-time SD detection and pilot aiding system into flight decks.

Podium Abstract 11Perception-Model Analysis of the Somatogravic Illusion in an Airplane Accident

EL Groen, TK Clark, MMJ Houben, JE Bos, RJ Mumaw. Corresponding author: Eric Groen, TNO

Background: It is difficult for accident investigators to objectively determine whether spatial disorientation may have contributed to a fatal airplane accident. In this presentation, we evaluate three methods to reconstruct the possible occurrence of the somatogravic illusion based on flight data recordings from an airplane accident.

Methods: The outputs of two vestibular models were compared with the “standard” method, which uses the unprocessed gravito-inertial acceleration (GIA).

Results: All three methods predicted that the changing orientation of the GIA would lead to a somatogravic illusion when no visual references were available. However, the methods were not able to explain the first pitch-down control input by the pilot flying, which may have been triggered by the inadvertent activation of the go-around mode and a corresponding pitch-up moment. Both vestibular models predicted a few seconds delay in the illusory tilt from GIA due to central processing and sensory integration.

Conclusions: While it is difficult to determine which method best predicted the somatogravic illusion perceived during the accident without data on the pilot’s pitch perception, both vestibular models go beyond the GIA analysis in taking into account validated vestibular dynamics, and they also account for other vestibular illusions. In that respect, accident investigators would benefit from a unified and validated vestibular model to better explain pilot actions in accidents related to spatial disorientation.

Reference

Groen EL, Clark TK, Houben MMJ, Bos JE, Mumaw RJ. Objective Evaluation of the Somatogravic Illusion from Flight Data of an Airplane Accident. Safety. 2022; 8(4):85. https://doi.org/10.3390/safety8040085.

Podium Abstract 12A review of sensorimotor readaptation following spaceflight

Scott J Wood, PhD

NASA Johnson Space Center, Houston TX

The need to move and maintain awareness of spatial orientation in altered gravity environment drives sensorimotor adaptation and learning to acquire a new set of synergies optimized for the novel environment. The perceptual and motor coordination problems experienced postflight reflect the recalibration of predicted versus actual movement feedback that is required for readaptation back to the natural gravitational state. The duration of microgravity exposure has a significant effect on both the magnitude of the sensorimotor decrements and the time course of recovery to preflight performance levels. This effect of spaceflight duration is illustrated by comparing results from Shuttle and the International Space Station (ISS) astronauts for reentry motion sickness, neurological exams, computerized dynamic posturography, and functional field tasks (prone-to-stand and tandem walk). Exponential time constants of recovery also vary by task complexity and visual conditions, e.g., 21 vs 87 hr for tandem walk eyes open versus eyes closed following 6-month ISS missions. Neither reentry motion sickness nor postflight functional performance appear altered with prior flight experience. While anecdotal reports suggest prior flight experience may result in some carry over of fine motor strategies needed to control motion in the novel “microgravity” state, this postflight recalibration to “normal gravity” appears necessary with each subsequent mission. The early sensorimotor decrements have implications for the completion of critical mission tasks during and following g-transitions. Interventions are necessary to optimize crew performance for success on upcoming exploration missions.

Keynote Abstract 3Vestibular dysfunction in patients with sporadic and NF2-related vestibular schwannomas.

Richard Lewis

Vestibular schwannomas (VS), benign tumors that develop from Schwann cells of the vestibular nerve, are usually sporadic (sVS) but are also highly characteristic of neurofibromatosis type 2 (NF2), where they typically are bilateral (NF2/VS). VS tumors present with progressive vestibular symptoms, but the pathophysiology underlying vestibular dysfunction in VS patients remains uncertain, and the relationship between pre-intervention (e.g., surgery, chemotherapy, radiation therapy) vestibular metrics with post-intervention outcomes is not understood. Here I present preliminary data from an observational study that examines vestibular pathophysiology in sVS and NF2/VS patients and the vestibular implications of standard interventions by analyzing vestibular function within a conceptual framework based on understanding the changes in afferent vestibular signal and noise. Signal (afferent firing encoding vestibular parameters) is reflected behaviorally by response accuracy (response amplitude relative to the ideal amplitude), while noise (random afferent firing), quantified as the signal-to-noise ratio, is reflected by response precision (inverse of response variability). In addition to quantifying vestibular abnormalities in the two VS populations, we also examined the relationship between vestibular dysfunction, clinical vestibular disability, and non-vestibular (MRI imaging, molecular biologic) parameters. Our results suggest different patterns of dysfunction of vestibular-mediated behaviors in sVS and NF2/VS, in which response accuracy and precision were degraded in sVS but only response precision will be degraded in NF2/VS. Relating vestibular dysfunction to clinical vestibular disability (e.g., dizziness, ataxia, quality of life) suggests that in both sVS and NF2/VS, vestibular imprecision is the main correlate of clinical disability. Furthermore, our preliminary results suggest that relationship between MRI and vestibular function shows both common and disease-specific patterns in sVS and NF2/VS, and that molecular biologic measures relate to vestibular function in both VS groups in a manner that recapitulates auditory-molecular biologic relationships. Lastly, we examined the relationship between pre-intervention vestibular (and other) metrics and post-intervention outcomes, including surgical outcomes is sVS and NF2/VS, where our results suggest that poor precision (e.g., high afferent noise) and/or poor accuracy (e.g., low afferent signal) pre-operatively will associated with the best surgical outcomes; chemotherapy (bevacizumab) outcomes in NF2/VS which suggest that poor vestibular precision is associated with better vestibular outcomes after chemotherapy than better precision; and radiation therapy (RT) outcomes, where hypothesize that good pre-RT precision (low noise) and accuracy (high signal) will be associated with worse chronic post-RT vestibular outcomes.

Podium Abstract 13Diagnostic Differentiation of Meniere’s Disease and Vestibular Migraine with Vestibular Perceptual Thresholds and Inner Ear Imaging Characteristics

Cameron Fattahi^1,2,3, Amsal Madhani^1,3, Susan King^1,3, Amy Juliano, MD⁴, Faisal Karmali, PhD^1,3, Richard F. Lewis, MD^1,3,5, Divya A. Chari, MD^1,2,3,5

¹Department of Otolaryngology - Head and Neck Surgery, Massachusetts Eye and Ear, Harvard Medical School, Boston, MA, USA

²Department of Otolaryngology - Head and Neck Surgery, University of Massachusetts Medical School, Worcester, MA, USA

³Jenks Vestibular Physiology Lab, Massachusetts Eye and Ear Infirmary, Boston, MA, USA.

⁴Department of Radiology, Massachusetts Eye and Ear, Harvard Medical School, Boston, MA, USA

⁵Department of Neurology, Massachusetts Eye and Ear, Harvard Medical School, Boston, MA, USA

Background: Patients presenting with recurrent episodic vertigo and dizziness such as Meniere’s disease (MD) and vestibular migraine (VM) can present a diagnostic challenge as they can both produce similar symptoms of recurrent vertigo, tinnitus, motion intolerance, and hearing loss. Current diagnostic criteria are based on patient history with little contribution from objective measures. Vestibular perceptual thresholds, which refer to the smallest appreciable stimulus detected by the participant in various translations and rotations of body movement, have been shown to have the potential to differentiate different types of vestibular dysfunction. In addition, recent evidence suggests that delayed gadolinium (Gd)-enhanced high-resolution magnetic resonance imaging (MRI) of the inner ear may enable visualization of vestibular and cochlear endolymphatic hydrops. Herein, we compare vestibular perceptual thresholds and imaging of the inner ear in patients with definite MD, patients with definite VM, and normal control (NC) subjects.

Methods: MD patients (n=16), VM patients (n=26), and healthy normal control subjects (n=34) underwent perceptual threshold testing. Five perceptual thresholds were compared among subjects: three linear motion translation at 1.0 Hz (inter-aural (y-translation), naso-occipital (x-translation), and earth-vertical (z-translation)), one rotatory motion at 1.0 Hz (yaw rotation about an earth-vertical axis), and roll-tilt at 0.2 Hz. MD subjects underwent clinical vestibular testing (cervical vestibular evoked myogenic potential [cVEMP] and video head impulse test [vHIT]) and 4-hr delayed intravenous Gd-enhanced MRI of the inner ear. All subjects completed a dizziness handicap inventory (DHI).

Results: MD subjects had significantly elevated z-translation thresholds compared to VM and NC subjects (t=2.134, p=0.02 and t=2.134, p=0.002, respectively). MD subjects had significantly elevated x-translation thresholds compared to VM subjects (t=2.094, p=0.046). MD subjects also demonstrated significantly elevated roll-tilt thresholds compared to VM subjects (t=2.078, p=0.024). Perceptual thresholds from all other motions were not significantly different among the groups (p>0.05). Z-thresholds in MD patients positively correlated with endolymphatic volume (r2=0.65, p=0.009). Clinical vestibular tests (cVEMP, vHIT)) and subjective DHI scores did not correlate with perceptual thresholds in any motion or imaging findings (p>0.05).

Conclusions: Perceptual threshold testing may provide insight into the underlying pathophysiology of the disease process (e.g. elevated z-translation thresholds in MD patients may support saccular dysfunction while elevated roll-tilt thresholds may suggest permanent peripheral damage of the inner ear) and may improve diagnostic precision of episodic vestibular disorders. Further, threshold measurements may provide a physiologic marker for endolymphatic hydrops. Future avenues for research include whether determining threshold measurements and inner ear imaging can be used to differentiate comorbid MD and VM.

Podium Abstract 14Electrodermal and Postural Responses in Dizzy Adults: Diagnostic Indicators of Vestibular Migraine

Doaa S. Al-Sharif^1,2, Carole A. Tucker^1,3, Donna L. Coffman⁴, and Emily A. Keshner¹

¹Department of Health and Rehabilitation Sciences, Temple University, Philadelphia PA 19122

²Currently at Department of Medical Rehabilitation, King Faisal Medical Complex, Taif, Saudi Arabia

³Currently at Department of Physical Therapy, University of Texas Medical Branch, Galveston, TX 77555

⁴Department of Epidemiology and Biostatistics, Temple University, Philadelphia PA 19140

Introduction: No reliable biometric measurement of vestibular involvement with migraine is currently available.

Objective: Measures of autonomic nervous system and postural responses could serve as quantifiable indicators of vestibular involvement with migraine.

Methods: A convenience sample of 22 young healthy adults (34±9 years old) and 23 young adults (34±8 years old) diagnosed with vestibular migraine (VM) participated. A rod and frame test and clinical outcome measures of dizziness and mobility were administered. Participants stood on foam while viewing two dynamic virtual environments. Trunk acceleration in three planes and electrodermal activity (EDA) were assessed with wearable sensors. Linear mixed models were used to examine magnitude and smoothness of trunk acceleration and tonic and phasic EDA. A Welch’s t-test and associations between measures were assessed with a Pearson Correlation Coefficient. Effect sizes of group mean differences were calculated using Cohen’s d.

Results: Visual dependence was present in 83% of the VM population. Individuals with VM exhibited lower baseline EDA (t(4.17)= -7.2, p =0.001) and greater normalized trunk accelerations in the vertical (t(42.5)= 2.861, p =0.006) and medial (t(46.6)= 2.65, p = 0.01) planes than healthy participants. Tonic EDA activity increased significantly across the period of the trial (F (1,417) =23.31, p =0.001) in the VM group. Significant associations appeared between vertical trunk acceleration and EDA, Dizziness Handicap Inventory, and Activities of Balance Confidence tools.

Conclusions: Higher tonic EDA activity in healthy adults results in more accurate postural reactions. Results support the supposition that EDA activity and postural acceleration are significantly different between VM and healthy individuals when accommodating for postural instability and visual-vestibular conflict.

Podium Abstract 15Vestibular Function and Balance in People with Diabetes Mellitus

Frank E. DiLiberto PT, PhD^1,2, Heather E.R. Kamath AuD², Maxine L. Olson MS^1,2, Janet O. Helminski PT, PhD¹, Marcello Cherchi MD, PhD³, Michael C. Schubert PT, PhD⁴

¹Department of Physical Therapy, Rosalind Franklin University of Medicine and Science, North Chicago, IL

²Audiology and Speech Pathology Department, Captain James A. Lovell Federal Health Care Center, North Chicago, IL

³Department of Neurology, Northwestern University, Chicago, IL

⁴Department of Otolaryngology Head and Neck Surgery, Johns Hopkins School of Medicine, Baltimore, MD

Over 400 million people worldwide suffer from diabetes mellitus (DM) and related medical problems.¹ Sensory system dysfunction driving balance impairment, such as retinopathy and peripheral neuropathy of the feet, are common consequences of DM, elevate fall risk, and serve as a barrier to obtaining the amount of physical activity needed for disease management and health. Unfortunately, the same biochemistry propagating retinopathy and peripheral neuropathy may also adversely affect vestibular function^2,3; potentially further exacerbating balance impairments. Therefore, the purpose of this ongoing descriptive study is to characterize vestibular function and determine its relationship to balance in people with DM and peripheral neuropathy (DMPN). Twenty people with DMPN (19 male, 70.8 ± 8.4 years old, 32.4 ± 4.9 kg/m², DM duration 19.1 ± 10.5 years, HbA1c 7.1 ± 0.9 %), 14 people with DM (13 male, 69.6 ± 8.7 years, 30.5 ± 5.2 kg/m², DM duration 18.4 ± 8.0 years, HbA1c 7.6 ± 2.0 %), and 16 healthy controls (15 male, 72.2 ± 8.1 years old, 31.1 ± 4.9 kg/m²) matched for biological sex, age, and body mass index, have been enrolled thus far. Cervical and ocular vestibular evoked myogenic potential (VEMP) tests were used to measure otolith function as total response (sum of left and right inter-amplitudes) and categorized as normal versus abnormal function (absent response or > 33% asymmetry). Sinusoidal harmonic acceleration and step velocity rotational chair testing protocols were used to measure horizontal semicircular canal function as vestibulo-ocular reflex (VOR) gain, phase, and GainTC (VOR gain at .64 Hz x the average of four time constants).⁴ Balance and physical activity were quantified with the Functional Gait Assessment (FGA) and one-week of activity watch data (average daily steps), respectively. One-way ANOVA, Kruskal-Wallis, or multiple linear models were used to assess group differences in outcomes. Group differences were not detected in VEMP outcomes or GainTC (all p >.17). An effect of frequency was observed for VOR gain and phase (both p <.01), and a group by frequency interaction (p = .03), attributable to greater phase lag at .64 Hz in DMPN (mean = -14.0°) versus DM (-4.2°) and control groups (-7.6°), was also observed. The DMPN group registered the worst balance (FGA = 18.1; p <.01) and the lowest amount of physical activity (daily steps = 5175.1). Presently, study findings suggest otolith function of people with DMPN is similar to people with DM without PN of similar disease duration and glycemic control, and similar to age and body mass index matched controls. However, a unique and abnormal finding of an excessive phase lag at .64 Hz during sinusoidal harmonic acceleration, along with expected deficiencies in balance and physical activity were observed in the DMPN group. With continued study, we intend to assess within and between group associations of vestibular function, balance, and physical activity. As such, final study findings are anticipated to inform and refine balance rehabilitation efforts in the promotion of physical activity and health in people with DMPN.

References

(1) Global Report on Diabetes. World Health Organization;2016.

(2) Myers SF, Ross MD, Jokelainen P, Graham MD, McClatchey KD. Morphological evidence of vestibular pathology in long-term experimental diabetes mellitus. I. Microvascular changes. Acta Otolaryngol. 1985;100:351-364.

(3) Myers SF, Tormey MC, Akl S. Morphometric analysis of horizontal canal nerves of chronically diabetic rats. Otolaryngol Head Neck Surg. 1999;120(2):174-179.

(4) Hain TC, Cherchi M, Perez-Fernandez N. The Gain-Time Constant Product Quantifies Total Vestibular Output in Bilateral Vestibular Loss. Front Neurol. 2018;11(9):396.

Acknowledgments: Funded by VA RR&D SPiRE Rx-003584-01A1

Podium Abstract 16Impaired stationarity perception is associated with increased VR sickness

Savannah Halow^1,2; Allie Hamilton¹; Eelke Folmer²; Paul R. MacNeilage¹

¹Department of Psychology, University of Nevada, Reno

²Department of Computer Science, University of Nevada, Reno

Stationarity perception refers to the ability to perceive the surrounding visual environment as world-fixed during self-motion. Perception of stationarity depends on mechanisms that evaluate the congruence between retinal/oculomotor signals and vestibular head movement signals. In a series of psychophysical experiments, we systematically varied the congruence between retinal/oculomotor and vestibular head movement signals to find the range of visual gains that is compatible with perception of a stationary environment. On each trial, human subjects wearing a head-mounted display executed a trained yaw head movement (~15 deg over 1 sec) and reported whether the visual gain was perceived to be too slow or fast. A psychometric fit to the data across trials yields the visual gain most compatible with stationarity (point of subjective equality, PSE) and the sensitivity to visual gain manipulation (just-noticeable difference, JND). Across conditions, we varied 1) the spatial frequency of the visual stimulus, 2) the retinal stimulus location (central versus peripheral), and 3) the scene-fixed versus head-fixed nature of the oculomotor signal. At the end of each condition, participants completed the Simulator Sickness Questionnaire (SSQ) to assess symptoms and severity of induced virtual reality (VR) sickness. Results demonstrate that stationarity perception is most precise (lowest JND) and accurate (PSE»1) with scene-fixed fixation, central retinal stimulus location, and low spatial frequency visual patterns. Other circumstances lead to a reduced gain perceived as stationary (PSE<1) and impaired sensitivity (increased JND). A linear mixed-effects (LME) model was fit to investigate how psychophysical performance (PSEs, JNDs) related to sickness. Results indicate a significant association between PSEs and the Nausea subscore of the SSQ, as well as between the JNDs and the oculomotor and disorientation subscores of the SSQ. To our knowledge, these are the first results to demonstrate an association between these psychophysical measures and VR sickness.

Acknowledgements: Research was supported by NIGMS of NIH under grant number P20 GM103650 and by NSF under grant number IIS-1911041

Podium Abstract 17Vestibular Adaptation Mitigates Vertigo and Nystagmus Associated with High Strength MRI

Jacob Pogson^1,2,5,6, Ari Shemesh¹, Dale Roberts^1,2, David S Zee^1,2,3,4, Jorge Otero-Milan⁷*, Bryan Ward²*

¹Neurology, ²Otolaryngology, ³Ophthalmology, ⁴Neuroscience, The Johns Hopkins University, Baltimore, MD, United States

⁵Royal Prince Alfred Hospital, Camperdown, NSW Australia

⁶Sydney Medical School, The University of Sydney, Camperdown, NSW Australia

⁷University of California Berkely, Berkely, CA, United States

*These authors contributed equally to this work and share senior authorship

Vestibular symptoms of vertigo, dizziness, imbalance and sometimes nausea are often reported by people in high-field strength magnets (>4 Tesla) used for MRI. Magnetic vestibular stimulation (MVS) explains these symptoms as arising from utricular currents interacting with the magnetic field to create a Lorentz force that deflects the semicircular canal cupulae to a constant position, simulating a constant-acceleration of the head which produces a constant-velocity slow-phase nystagmus response. Because of adaptation there is a decrease in the nystagmus response the longer the patient stays within the magnet, and a corresponding secondary response with an oppositely directed nystagmus after the subject exits the MRI bore. We sought a simple technique to mitigate these powerful, uncomfortable sensations as experienced during entry and exit from the MRI bore.

First, we used a 3D linear control systems model of the vestibulo-ocular reflex (VOR), which incorporated adaptation, central velocity storage, and rotational feedback elements. We simulated three entry and exit durations (20, 120, 300 seconds) into and out of the magnetic field of an MRI machine and found that the horizontal VOR nystagmus peak slow-phase response (SPV) was lower the longer the duration to enter or exit the magnet. Second, we experimentally recorded the nystagmus response from 4 healthy human subjects as they underwent the same protocol with a 7-Tesla magnet. After entry all subjects were kept in the magnet for 5-minutes. Objectively, we confirmed that the horizontal nystagmus peak SPV was less with longer durations, even though the nystagmus SPV just prior to exiting was not changed by the entry duration. Subjectively, longer durations also mitigated vestibular symptoms, and no vestibular symptoms were reported during the 300 second entrance and exit protocol.

We conclude that longer duration (>120 seconds) entry and exit of the MRI bore allows vestibular adaptation to mitigate objective and subjective symptoms of MVS, but not the eventual velocity of the constant nystagmus while within the magnet. A 120 second duration could be an effective and practical trade-off suitable for clinical application.

Podium Abstract 18Moving MRI: Imaging a Moving Body with a Moving MRI magnet

Artan Kaso¹, Jingting Yao¹, Nikhil Patel¹, Andre van der Kouwe¹, Yin-Ching Iris Chen¹, Yi-Fen Yen¹, Jerome L. Ackerman¹, Daniel Merfeld², Peter Le³

¹Massachusetts General Hospital and Harvard Medical School, Boston, MA

²Ohio State University College of Medicine, Columbus, OH

³Air Force Research Laboratory, 711th Human Performance Wing, Wright-Patterson AFB, OH

Moving MRI (mMRI) is a novel method we are developing to study how the brain processes signals from the vestibular system and other sources during bodily motion such as tilts and rotations. Moving a subject’s head relative to the magnetic field of an MRI scanner creates image artifacts and confounding physiological stimuli, obscuring neuronal activation arising from motion. In mMRI, the scanner magnet and subject move in synchrony such that the subject’s head remains stationary in the magnet, suppressing motion artifacts and extraneous magnetic field-induced stimuli. We expect this to enable for the first time high quality functional MRI (fMRI) scans revealing the brain networks that process vestibular and other inputs reflecting bodily motion. Additionally, mMRI may prove useful for mapping brain tissue deformation and fluid flow when the body undergoes large scale motions, with implications for traumatic brain injury and flight physiology research.

In this report we detail our progress in creating the first mMRI scanner. The major engineering challenge in reaching this goal is developing a means to safely and controllably move an MRI magnet. All components of the motion platform near the magnet must be completely nonmagnetic and not harmed by the magnet’s field. We use a conduction cooled liquid cryogen-free 1.5T compact superconducting extremity (arm and leg) magnet. Our motion platform comprises a pair of pneumatic cylinders that can tilt the magnet up to 15 degrees from horizontal. The tilt angle is controllable by the host computer of the MRI scanner.

We expect that tilting a magnet should result in field shifts and distortion due to mechanical strain of the magnet structure. Additionally, residual relative motion between the magnet assembly (main magnet, gradient coils, RF coil) and the subject will result in image artifacts. We are making progress in reducing these artifacts to produce MR images of adequate quality. For example, we found a simple linear relationship between the static tilt angle and field shift. By recording the angle during scanning, degraded images obtained with the magnet in motion may be postprocessed to reduce the field shift and relative motion artifacts.

Funding: NIH R01EB029818 and R01AR075077

Disclaimer: The views expressed in this article reflect the results of research conducted by the authors and do not necessarily reflect the official policy or position of the Department of the Air Force, Department of Defense, nor the United States Government. P.L. is an employee of the U.S. Government. This work was prepared as part of his official duties. Title 17 U.S.C. 105 provides that copyright protection under this title is not available for any work of the U.S. Government. Title 17 U.S.C. 101 defines a U.S. Government work as work prepared by a military service member or employee of the U.S. Government as part of that person’s official duties.

Keynote Abstract 4Natural statistics and multisensory processing in spatial orientation: Behavioral modeling and clinical implications

Pieter Medendorp

Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, The Netherlands

Perception of spatial orientation involves a weighted fusion of visual, vestibular, proprioceptive and somatosensory signals, as well as cognitive signals about statistical regularities within the natural environment. When one of these signals deteriorates or breaks down, such as the vestibular signal in bilateral vestibulopathy, compensation can occur by relying more on the remaining cues. How sensory signals are weighted and reweighted in spatial orientation has remained difficult to quantify since the quality of these signals can typically not be measured in isolation. I will present our recent work combining experimental psychophysics with a reverse engineering approach based on Bayesian inference principles to characterize sensory reweighting across the life span and in individual bilateral vestibular patients. This personalized quantification approach could aid in the diagnostics and prognostics of multisensory integration deficits in vestibular disorders, and contribute to the evaluation of the effect of rehabilitation therapies, including balance training exercises.

Podium Abstract 19Computational Models of Active Human Motion Control

Moein Nazifi, Ph.D.¹; Tara Nibhanupudy¹; Elena López-Contreras Gonzalez¹; Kassia Love, B.Sc.¹; Yoonchan Rah, M.D.¹, Faisal Karmali, Ph.D.¹; Megan Reissman, Ph.D.², TimothyReissman, Ph.D.²; Torin Clark, Ph.D.³

¹Jenks Vestibular Physiology Lab, MEEI, Harvard Medical School Boston MA

²University of Dayton, Dayton, OH

³University of Colorado Boulder, Boulder, CO

Active control of motion is one of the most critical tasks for humans, with broad impacts on quality of life. It includes both control of the body (e.g., maintaining balance and gait) and of vehicles such as automobiles, airplanes, and spacecraft. While the quality of the active control is highly variable among individuals, it is not completely known why these discrepancies exist. Hence, to study the effects of vestibular processing on motion control, we developed computational models of humans controlling a machine using a joystick while onboard the machine. In our closed-loop models, machine motion results in inputs to peripheral and central vestibular processing, which provides an estimate of motion that is compared to desired motion. Finally, the brain determines a compensatory motor command, which causes the machine to move via a joystick. An innovation in modeling was the inclusion of a minimum sensory threshold before determining a motor command, under the assumption that selecting a joystick input requires a command input larger than corresponding neural noise [1]. We aimed to A) develop a closed-loop human-machine model that includes state-of-the-art spatial orientation and motor-control components [2, 3], and B) validate and refine this model using human experiments.

Modeling (Matlab/Simulink) and experimental validation were performed for an Earth-vertical yaw rotation task to limit vestibular organ stimulation to the semicircular canals. Healthy subjects (18-40 years) were seated in a standard clinical motorized rotary chair (Neurokinetics) in complete darkness while listening to white noise. Subjects were instructed to use a joystick to minimize chair velocity while experiencing a pseudo-random unpredictable sum-of-sines “disturbance” ranging between 0.004-1.7 Hz.

The closed-loop model predicted drastic frequency dependence in the quality of nullification of disturbances while the overall performance was mediocre. The most effective nullification was predicted to occur in a small band of medium-range frequencies (0.01-0.1 Hz) due to combined interactions of the high-pass filter vestibular dynamics, low-pass filter motor control dynamics, and low-pass filter machine dynamics. Human experimental data mostly followed the overall frequency response predicted by the model. Model parameters of the central nervous system delay and motor control effort were tuned to enhance the prediction accuracy.

Our novel computational model utilizes up-to-date spatial orientation and motor control blocks and accurately predicts the experimental human motion control quality. Such models are valuable as they help understand, predict, and potentially mitigate adverse performance episodes in humans.

References

(1) Love K. Bachelor’s thesis, Harvard College. 2022.

(2) Merfeld D. M., et al. J Vestib Res. 1993;3(2), 141-61

(3) McRuer D., Weir D. H. Ergonomics. 1969;12(4), 599-633.

Acknowledgements: Funded by DoD/ONR (MURI, N000142012163).

Podium Abstract 20The otolith vermis: a systems neuroscience theory of the Nodulus and Uvula

Jean Laurens

Ernst Strüngmann Institute (ESI) for Neuroscience in Cooperation with Max Planck Society, Frankfurt, Germany

The Nodulus and Uvula (NU), which are lobules X and IX of the cerebellar vermis, play a central role in vestibular information processing. Decades of research have involved the NU in a multitude of functions, including the resolution of sensory ambiguity in the inner ear’s otolith organs, the use of gravity signals to detect head rotations, the control of the velocity storage, and the differential processing of head motion that is self-generated or externally imposed.

Here I examine these findings through the lens of the internal model hypothesis, a theoretical framework for information processing. I propose that the NU implement a forward internal model to anticipate the activation of the otoliths and generate sensory prediction errors to refine internal estimates of self-motion or facilitate learning.

I show that a Kalman filter based on this framework can explain the various functions of the NU, as well as neurophysiological observations and the consequences of NU lesions. This underscores the NU’s central role in processing information from the otoliths, and supports their designation as the “otolith vermis.”

Keynote Abstract 5Are Head Direction Cell Responses Commutative in 3D – and Why it Matters

Jeffrey S. Taube, Dartmouth College

Abstract for Keynote Address to Vestibular Conference

Our research interests encompass understanding the neurobiological mechanisms that underlie our abilities to navigate and our sense of spatial orientation. In particular, my research program has focused on understanding our ‘sense of direction’ and its underlying neurobiological mechanisms. To this end and using rodents as a model, we have studied a population of so-called ‘head direction cells’, which discharge as a function of the animal’s directional heading. My talk will briefly discuss our current understanding of where and how the head direction signal is generated in the brain. I will then focus on how head direction cells respond in 3-D, particularly when the animal is moving in planes other than Earth horizontal, and why understanding this issue is important.

Podium Abstract 21Integration of Self-Motion and Visual Cues for Navigation in Younger and Healthy-Aging Older Adults

Corey S. Shayman, B.S., Hunter C. Finney, B.S., Misty M. Myers, Andoni M. Katsanevas, Phoenix Hines, Peter C. Fino, Ph.D., Jeanine K. Stefanucci, Ph.D., Sarah H. Creem-Regehr, Ph.D.

Department of Psychology, University of Utah, Salt Lake City, UT

Our brains must combine information from our self-motion cues (proprioception and vestibular) as well as our vision to help us navigate and update our cognitive spatial map. It is unclear whether individuals across the lifespan combine these senses optimally for navigation (Bates & Wolbers, 2014). To quantify sensory combination and weighting strategies, well-aging older and younger adults completed a homing task (Chen et al., 2017). The Varjo VR3 head-mounted display was used to display visual cues in immersive VR. Participants physically walked towards a target via multiple waypoints with visual landmarks to help them encode their spatial location. Participants then walked back to the target with either visual-cues only, self-motion cues only, or congruent cues (both vision and self-motion). They also completed a condition where visual cues were covertly shifted relative to self-motion cues to quantify sensory weighting. In other words, when participants were forced to make a trade-off regarding two “correct” locations, how they split the difference allowed us to determine sensory cue reliance. Both younger and older adults demonstrated improved accuracy in the combined cue condition relative to the single cue conditions. Younger adults visual-only performance was similar to their combined cue performance. Younger adults visual-only error was about 20 cm while their combined cue accuracy was about 18 cm. Older adults had higher error in all conditions (visual error was about 50 cm, self-motion was about 90 cm, and combined cue error was about 22 cm) but received greater relative multisensory benefit. Regardless of age, participants weighted vision higher than self-motion cues (observed visual weight = 60%, self-motion = 40%). This closely aligns with the sensory weights predicted from participants’ single-cue variability (predicted visual weight = 55%, self-motion = 35%). Both older adults and younger adults show evidence of optimal integration, suggesting that sensory cue-combination strategy may not explain age-related declines in navigational abilities related to spatial updating. High inter-individual variability was found across groups, suggesting that sensory weighting strategies may be inconsistent during the experiment. Our study demonstrates that older adults may have decreased navigation ability due to sensory cue combination deficits, though additional work is needed to determine why some continue to optimally integrate their senses in older age while others do not. Our study shows that simple behavioral interventions, such as training vestibular and proprioception cue use for spatial updating, may improve navigational deficits caused by impairments in sensory integration for some older individuals.

References

(1) Bates S. L., Wolbers T. How cognitive aging affects multisensory integration of navigational cues. Neurobiology of Aging. 2014;35(12), 2761–2769.

(2) Chen X., McNamara T. P., Kelly J. W., Wolber T. Cue combination in human spatial navigation. Cognitive Psychology. 2017;95, 105–144.

Acknowledgements: This work was funded by the University of Utah Center on Aging and the American Otological Society

Podium Abstract 22Navigation deficits are associated with vestibular and cognitive dysfunction in chronic mild traumatic brain injury

Linda J. D’Silva, Lucas Meitler, Jacob Sosnoff, Jeffrey M. Hausdorff, Prabhakar Chalise, Hannes Devos.

Purpose/Hypothesis: 1) To compare navigation ability in the community setting between people with persistent symptoms after mild traumatic brain injury (mTBI) and age-matched controls, 2) to examine the vestibular and cognitive contributions to navigation ability in chronic mTBI.

Fourteen people between 40-64 years of age (mean age: 53.75 ± 7.3 years) who were between 3 months to 2 years post-injury [mean: 46 weeks (range: 18-88)] and fourteen age and sex-matched controls (mean age 53.79 ± 7.4).

Navigation ability was tested by time in seconds to walk a predetermined path in the hospital setting using a map. Measures of vestibular function included dynamic balance measured by the Functional Gait Assessment (FGA), and path integration measured by end-point error on the Triangle Completion Test (TCT). Global executive function was assessed using the MoCA, simple search ability was tested by normalizing the Trail-making test (subtracting TMT-A from TMT-B), visuospatial cognition was measured using the symbol digit modalities test (SDMT), and symptoms were measured by the post-concussion symptom scale (PCSS). Based on normality of data, either independent sample t-tests or Mann Whitney U tests were used to compare between group outcomes. Linear regression was used to examine the relationship between navigation ability and the independent variables of vestibular and cognitive function.

Results: Participants with mTBI took longer to navigate to their destination (363.36 ± 91.29 versus 269.85 ± 20.28 seconds, p=0.001) compared to controls. They had lower FGA scores (20.5 (IQR: 9) versus 28.5 (IQR: 2), p<0.001), greater endpoint error on the TCT (67.63 ± 34.5 versus 44.86 ± 14.3 cm, p=0.03), longer time to complete the TMT (61.81 (IQR: 46.51) versus 40.46 (IQR: 12.8), p=0.01), lower scores on the SDMT (65.43 ± 13.2 versus 83.43 ± 15.46, p=0.003), and higher symptom severity on the PCSS (p<0.001). Independent variables that explained 77% of the variance in navigation ability were FGA (R²=0.60, p<0.001), and MoCA score (R²=0.77, p<0.001).

Conclusions: Navigation in an ecologically valid condition in the hospital took significantly longer for people with persistent symptoms after a mTBI compared to controls. People with mTBI and persistent symptoms have poorer dynamic balance, impaired path integration, poorer executive function, and visuospatial deficits compared to age-matched controls. Dynamic balance and global cognitive function were most predictive of navigation deficits. Vestibular and cognitive deficits that persist months after a mTBI and can lead to reduced community activities.

References

(1) D’Silva LJ, Chalise P, Obaidat S, Rippee M, Devos H. Oculomotor Deficits and Symptom Severity Are Associated With Poorer Dynamic Mobility in Chronic Mild Traumatic Brain Injury. Frontiers in neurology. 2021;12(1223).

(2) Leland A, Tavakol K, Scholten J, Mathis D, Maron D, Bakhshi S. The Role of Dual Tasking in the Assessment of Gait, Cognition and Community Reintegration of Veterans with Mild Traumatic Brain Injury. Mater Sociomed. 2017;29(4):251-256.

(3) Cogan AM, Haines CE, Devore MD, Lepore KM, Ryan M. Occupational Challenges in Military Service Members With Chronic Mild Traumatic Brain Injury. Am J Occup Ther. 2019;73(3):7303205040p7303205041-7303205040p7303205049.

(4) Shema-Shiratzky S, Hillel I, Mirelman A, et al. A wearable sensor identifies alterations in community ambulation in multiple sclerosis: contributors to real-world gait quality and physical activity. J Neurol. 2020;267(7):1912-1921.

(5) Kersey J, McCue M, Skidmore E. Domains and dimensions of community participation following traumatic brain injury. Brain injury. 2020;34(6):708-712.

Funding: the University of Kansas Research Institute through the Lied Clinical Research Pilot Grant program.

Keynote Abstract 6From Balance to Velocity Storage: Two Research Pearls from a Clinician’s Point of View

Chris Zalewski, Ph.D.

National Institute on Deafness and Other Communication Disorders, National Institutes of Health

Clinic research often involves investigations that attempt to find relationships between disease characteristics and fundamental physiologic or genetic characteristics associated with a particular disease. Such phenotype-genotype investigations for rare diseases and syndromes are common to the NIH. And while such investigations are extremely important to the understanding of disease and function, such descriptive studies are often limited with respect to the generalization of findings beyond the disorder and population the research is intended to serve. However, such investigations can also be extremely instrumental in facilitating research towards a better understanding of normal physiology through understanding how an individual adapt in the face of abnormal function. However, our understanding of patient outcomes is fundamentally linked to how our clinical assessments evaluate our patients and the stimuli we use to elicit patient outcomes.

Patient’s outcome measures are fundamentally influenced by the stimuli used to evaluate them. Are we always using the correct stimuli to measure what we intended to evaluate? How many assumptions are we making with respect to the relationship between the stimulus, the physiology, and function? This presentation will focus on two research studies that highlight generalizable vestibular and balance principles that may serve to better our understanding of how we both select the stimuli we use to evaluate our patient’s physiology as well as how we interpret our patient’s outcome measures.

The first study focused on the influence of an individual’s personal limits of stability (LOS) on their reported outcomes sway measure. Limits of stability were determined from 60 healthy volunteers recruited into three age groups of young, middle-aged, and elderly. Individual measures of LOS were significantly lower than theoretical estimates of the LOS in healthy adults. Implications of an individual’s personal LOS for determining postural stability and risk for falls during the sensory organization test will be discussed.

The second study investigated the implications of stimulus frequency when assessing velocity storage (VS) for patients with peripheral and central vestibular pathology. Rotational data from approximately 300 patients and healthy controls from 535 visits were analyzed to explore the optimal sinusoidal acceleration and velocity step stimuli to best recruit and reflect VS function and dysfunction. Correlations and multiple regression analyses were performed to highlight the relationships and contributions of various sinusoidal and step stimuli to provide evidence confirming a known low-frequency dependency of the VS and investigate unique clinical phenotypes sparing the common sacrificing of VS as a consequence of vestibulopathies.

Both studies will attempt to further elucidate the relationship between stimuli and clinical response to improve our clinical practice and our understanding of patient outcomes.

Podium Abstract 23Rapid gaze-shift adaptation during self-generated vestibular prosthetic stimulation

Kantapon Pum Wiboonsaksakul, Charles C Della Santina, Kathleen E Cullen

The brain must differentiate between externally-generated and self-generated sensory inputs to build stable perception and generate appropriate behavior. Specifically for vestibular prosthesis users, the brain must learn to utilize prosthetic vestibular input to maintain visual and postural stability while also suppress/cancel these reflexes when they are counterproductive to behavioral goals. Here, we leveraged a gaze-shift task—a naturalistic behavior that requires the gating/canceling of vestibular input—to directly investigate how self-generated prosthetic stimulation affects reorientation behaviors. In a monkey with bilateral vestibular deficits, we implanted a vestibular prosthesis that senses head rotation and transforms this movement into vestibular nerve stimulation, substituting for the damaged periphery. The monkey was trained to make eye-head coordinated gaze shifts between horizontal targets while the head, eye, and gaze positions were recorded. Each session comprised a three-block learning paradigm: baseline gaze-shifts, gaze-shifts with prosthetic stimulation, and washout without stimulation. We hypothesized that 1) prosthetic stimulation would first engage vestibular reflex pathways, resulting in impeded head movements and truncated gaze-shifts but also that 2) the brain would then adapt to this new sensory input and no longer engage the reflex when it is counterproductive. Consistent with our predictions, gaze position error initially increased after stimulation onset and then exponentially decayed within ~80 trials. Prosthesis-evoked vestibulo-ocular reflex (VOR) gain and change in head position impeded by the prosthesis-evoked vestibulo-collic reflex (VCR) showed a similar decay though both at a slower rate (~200 trials), suggesting that the suppression/canceling of VOR and VCR both contributed to the observed improvement in gaze accuracy. In addition, early washout trials and catch trials during learning showed oppositely-directed gaze position error, indicating a central adaptation in addition to the observed reflex suppression/cancellation. This central adaptation could be due to the updating of the gaze controller or the updating of a forward internal model to predict and cancel self-generated prosthetic sensory input. Together, these results show that the brain can quickly adapt to self-generated prosthetic stimulation to improve behavioral performance. Importantly, these findings provide new insights on how prosthetic inputs interact with different vestibular pathways in a context-specific way.

Podium Abstract 24Discharge properties of neurons in the 8^th nerve, vestibular nucleus and abducens nucleus may explain suboptimal VOR characteristics in response to unilateral and bilateral neuroprosthetic stimulation.

James Phillips, Leo Ling, Christopher Phillips, Amy Nowack, Yoshiko Kojima, Jay Rubinstein, Shawn Newlands

Introduction: Vestibular neuroprostheses aim to restore vestibular function after the loss of vestibular hair cells. These devices were designed to restore primarily the angular vestibulo-ocular reflex (aVOR) by selectively stimulating afferent fibers of each ampullar nerve. These devices have been successful in providing controlled vestibular activation in patients, and clinically relevant recovery of function. However, aVOR restoration has been suboptimal in human subjects; e.g., low or pulsatile eye velocity, incorrect timing, and fluctuating direction errors. To understand the mechanisms underlying the suboptimal aVOR response to neuroprosthetic stimulation (eVOR), we recorded eye movement behavior and neural activity during such stimulation.

Methods: Rhesus monkeys were implanted with unilateral or bilateral vestibular neuroprostheses designed for human use. Each device generated biphasic pulse stimulation (BPS) modulated either in current amplitude or pulse frequency. Eye movements were recorded in the dark with scleral coils in awake behaving animals. Unit activity was recorded from 8^th nerve afferents, identified vestibular nuclear neurons, and abducens neurons.

Results: Neuroprosthetic BPS produced electrically elicited VOR (eVOR) in the plane of the targeted (implanted) canal(s) above threshold current levels. As current was increased, eye movements were elicited out of plane. High eVOR eye velocities could only be achieved at current levels eliciting significant out of plane eye movements. eVOR directions suggested canal-canal or canal-otolith summation. Low pulse rate stimulation elicited brief velocity pulses, while higher pulse rates elicited sustained eVOR velocity. Velocity saturated with high pulse rate, and then decreased at higher rates.

Afferent fibers displayed time locked discharge in response to BPS with different thresholds. The probability of eliciting a spike increased with increasing current level. Individual afferent fibers often responded to BPS of sites in multiple canals. Differences were observed between afferents types.

Secondary vestibular neurons displayed time locked discharge in response to BPS above current thresholds corresponding to behavioral thresholds. The probability of a spike increased with increasing current amplitude and decreased with increasing pulse frequency, typically saturating below P=1.0. Changes in current level often changed the timing but not the number of elicited spikes. Most neurons responded to stimulation from multiple canal electrodes at higher current levels. Differences were observed between behaviorally identified neuron types.

Abducens neurons displayed complex response characteristics comparable to vestibular neurons, but at longer latency.

Conclusion: These results suggest that neural activation with eVOR is qualitatively different than that underlying natural aVOR. These differences limit the ability of neuroprosthetic stimulation to reproduce natural VOR responses.

Acknowledgements: R21 DC-018083 (Newlands/Phillips), R01 DC014002 (Rubinstein), HHS-N-260-2006-00005-C (Phillips), OD010425 (P51 for WaNPRC).

Podium Abstract 25Pitch and roll perturbations under conditions of controlled acceleration and vestibular prosthesis

Bassil Ramadan¹, Olivia M.E. Leavitt¹, Kathleen E. Cullen^1,2,3

¹Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD

²Neuroscience, Johns Hopkins University, Baltimore, MD

³Otolaryngology, Johns Hopkins University, Baltimore, MD

Several therapies and implantable prosthetics have emerged to treat bilateral vestibular loss (BVL). These prosthetics translate signals from a gyroscope into vestibular afferent stimulation. While prosthetics have found significant success in humans (Chow et al., 2021), there are a number of open questions regarding how to best optimize prosthetic stimulation paradigms under various conditions. Notably, the vestibular system relies on input from two different peripheral channels, regular and irregular afferents, which encode head motion with substantially different temporal dynamics (Sadeghi 2007). Thus, this difference is an important consideration in the context of optimizing biomimetic stimulation protocols. Our goal was to first assess and compare postural responses in normal Rhesus Macaque monkeys versus monkeys with bilateral vestibular labyrinthectomy. We then investigated which prosthetic stimulation protocols (static, regular, irregular, or super-irregular) best restored function following bilateral vestibular labyrinthectomy.

To assess postural responses, we applied sinusoidal and transient perturbations in the pitch and roll axes. In experiments, unrestrained macaque sat in their natural stance within a chamber secured to a hexapod motion platform. Postural response for pitch and roll were assessed using sinusoidal perturbations at five different frequencies (0.25, 0.5, 1, 2, and 5Hz) at both a 5degrees/sec and 10degrees/sec peak angular velocity. We further assessed responses to un-cued, transient tilt perturbations (with accelerations of 200, 500, and 1000 deg/s² and peak velocities of 20, 40, and 60 deg/s.) Measurements included IMU head and platform movement tracking, markerless pose estimation with DeepLabCut, and weight distribution tracking via force plate.

Overall, comparison of postural responses in normal Rhesus Macaque monkeys and those with BVL revealed disparities in both axes. Notably, for both normal and BVL animals, postural responses were symmetrical between leftward and rightward roll perturbations but were asymmetrical between anterior and posterior pitch. In the roll axis, transient tilts of the BVL animal showed an earlier, hypermetric, and reversed head motion response compared to normal animal. The BVL animal demonstrated better compensation for anteriorly-directed pitch when compared to posteriorly-directed pitch perturbations. These results pair well with the previously suggested vision- and proprioceptive-dependent BVL condition (Cullen and Brooks 2015). For sinusoidal perturbations in both axes, the BVL animal demonstrated greater postural sway (i.e. greater head-in-space movement) than normal, given the motion was predictable, the BVL animal showed better recovery for sinusoid motion than for the transients.

In addition, we found that unilateral anterior canal prosthetic stimulation protocols, which mimicked patterns of the native vestibular afferents, restored some function following bilateral vestibular labyrinthectomy. Further, we found the greatest restoration of function by matching the dynamics of irregular afferents, the primary contributors to vestibulospinal pathways.

These studies begin to elucidate the contribution of vestibular signals to postural control and how these vary with the predictability, direction, and intensity of the perturbation, while also testing future stimulation paradigms and their optimized conditions for performance. As vestibular implants become more widespread, understanding how they interact with the brain and how we can optimize them for various daily circumstances will aid in improving life for those suffering vestibular loss.

References

Chow MR, Ayiotis AI, Schoo DP, Gimmon Y, Lane KE, Morris BJ, Rahman MA, Valentin NS, Boutros PJ, Bowditch SP, Ward BK, Sun DQ, Treviño Guajardo C, Schubert MC, Carey JP, Della Santina CC. Posture, Gait, Quality of Life, and Hearing with a Vestibular Implant. N Engl J Med. 2021 Feb 11;384(6):521-532. doi: 10.1056/NEJMoa2020457. PMID: 33567192; PMCID: PMC8477665.

Sadeghi SG, Chacron MJ, Taylor MC, Cullen KE. Neural variability, detection thresholds, and information transmission in the vestibular system. J Neurosci. 2007 Jan 24;27(4):771-81. doi: 10.1523/JNEUROSCI.4690-06.2007. PMID: 17251416; PMCID: PMC5053814.

Cullen KE, Brooks JX. Neural correlates of sensory prediction errors in monkeys: evidence for internal models of voluntary self-motion in the cerebellum. Cerebellum. 2015 Feb;14(1):31-4. doi: 10.1007/s12311-014-0608-x. PMID: 25287644; PMCID: PMC4320652.

Acknowledgements: This work was supported by National Institutes of Health grant R01DC002390 (to KEC and CCDS) and grant R01DC018061 (to KEC)

Keynote Abstract 7Vertigo and motion perception: More than meets the eye?

Benjamin Crane, MD, PhD

University of Rochester

The vestibular system includes both the otoliths and semicircular canals. The semicircular canals have been extensively studied as the vestibulo-ocular reflex (VOR) that is responsible nystagmus. However, many patients have dizziness symptoms that go beyond this and may be attributed to the linear acceleration sensing organs including the utricle and saccule. Central integration with vision also plays an important part in self motion perception. After unilateral loss of vestibular function, it is well described that humans as well as animals have changes in vertical perception which can also manifest as ocular counter-rotation and head tilt. However, study of the function of the otolith organs in human heading perception has been relatively limited.

Normal human heading perception has some unusual and unexpected properties. The perception of the lateral aspect of headings is overestimated with both visual and inertial stimuli. The differences in the coordinate system used by inertial (body based) and visual headings (retina based) and the problem this presents for multisensory integration. Although it has been hypothesized that these are converted to a common coordinate system prior to multisensory integration this appears not to be the case. The underlying neurophysiology compensates for this by having people look in the direction they want to go, rather than undergo a coordinate system reconciliation.

Visual inertial heading integration will be considered in terms of the factors required. These include the relative reliability of these stimuli, timing, direction differences, visual field size, the perception of common causation, and motion profile. Relative reliability determines the weight given to each stimulus when they are integrated but doesn’t seem to influence if they are integrated or not. Visual-vestibular integration occurs over a surprisingly large difference in direction and even when individuals do not consciously perceive the stimuli to have common causation. Visual field size had surprisingly little effect. Timing differences of up to 250 ms are well tolerated. The visual motion profile does not need to match the inertial motion for multisensory integration to occur.

Individuals with unilateral vestibular hypofunction were found to reliably have inertial heading perception deviated towards the side of their vestibular lesion. This deviation was reliably present even decades after the lesion suggesting it does not recover spontaneously. It is hypothesized that this deviation may be the cause of chronic dizziness and fatigue symptoms in this population.

Adaptation of inertial heading perception could be induced by exposing people to rotating environments or shifting situations in which the visual heading was consistently offset in one direction. The potential to use this for vestibular rehabilitation is being explored.

Podium Abstract 26Assessing noisy Galvanic Vestibular Stimulation capabilities for inducing stochastic resonance in lateral translation vestibular perceptual thresholds

David R. Temple, PhD and Torin K. Clark, PhD

Smead Aerospace Engineering Sciences, University of Colorado, Boulder CO

We are studying the use of noisy Galvanic Vestibular Stimulation (nGVS) as a means of inducing a phenomenon known as stochastic resonance (SR), where a particular non-zero level of added sensory noise might enhance the response to a weak periodic input signal (A. P. Mulavara et al., 2011). Several studies have noted improved vestibular roll tilt perceptual thresholds (Galvan-Garza et al., 2018), as well as balance (Goel et al., 2015; A. P. Mulavara et al., 2011) and locomotor performance (A. Mulavara et al., 2015; Temple et al., 2018) while utilizing nGVS, that was attributed to possible SR induced enhancements of vestibular afferents. However, other studies have noted little evidence of SR occurring with nGVS in young healthy individuals (Assländer et al., 2021). We aim to systematically test various nGVS waveforms’ impact on 1 Hz lateral translation vestibular perceptual thresholds in healthy individuals, to investigate which may best enhance vestibular performance. Vestibular perceptual thresholds will be computed from 100 trials of left/right translations adjusted by a three-down-one-up (3D1U) algorithm, using three different nGVS profiles, tested at five different noise amplitudes (including zero-noise sham) each. A machine learning algorithm previously developed (Voros et al., 2022) will be utilized to objectively classify presence of SR-like phenomena for each subject and nGVS profile. Furthermore, to investigate a potential means of reducing the number of trials needed to compute vestibular thresholds, subjects will report confidence (50-100%) for each of their translation reports. If vestibular thresholds computed with confidence reports and fewer trials are found to display similar values and SR-like phenomena seen when utilizing 100 trials, then these methods could be utilized to reduce the time needed to measure vestibular thresholds in individuals, potentially avoiding fatigue and improving threshold testing methods. Such benefits could be valuable in assessing future SR-induced vestibular performance improvements, as well as measuring vestibular perceptual thresholds in general. Overall, this study could be instrumental in determining if nGVS can serve as a viable countermeasure to improve vestibular performance in those with potential deficits, such as astronauts returning to gravitational environments after long-duration exposure to microgravity.

References

(1) Assländer et al. Scientific Reports. 2021;11(1).

(2) Galvan-Garza et al. Brain Stimulation. 2018;11(4).

(3) Goel et al. PLoS One. 2015;10(8).

(4) Mulavara et al. Frontiers in Systems Neuroscience. 2015;9.

(5) Mulavara et al. Experimental Brain Research. 2011;210(2).

(6) Temple et al. Frontiers in Physiology. 2018;9.

(7) Voros et al. Journal of Neuroscience Methods. 2022;374.

Acknowledgements: We would like to thank Talie Stone and Lanna Klausing for their contributions to this research project. This work is supported by the Translational Research Institute for Space Health through NASA Cooperative Agreement NNX16AO69A.

Podium Abstract 27The Impact of Hypoxia on Vestibular Thresholds

Max Teaford¹, Kyle Pettijohn², Anne Crecelius³, Alannah Gernon², Zachary J. Mularczyk², Daniel M. Merfeld^1,2

¹The Ohio State University Wexner Medical Center, Columbus, OH

²Naval Medical Research Unit - Dayton, Dayton, OH

³University of Dayton, Dayton, OH

Pilots, mountaineers, and even airline passengers are exposed to a variety of factors that may impact their ability to sense where they are relative to the environment including hypoxia. Past studies have demonstrated that severe hypoxia adversely impacts vestibular function in animals. Despite the potential for hypoxia to impact our ability to sense where we are within space no prior studies have examined the impact of hypoxia on vestibular function in humans. Given this state of knowledge, we performed a study to begin to investigate how mild hypoxia impacts human vestibular thresholds. In order to address this question,15 participants completed multiple sessions of a z-translation threshold test on a MOOG movement platform while breathing gases with an oxygen content consistent with 0 feet (21% O₂) and 8,000 feet (15% O₂) above sea level. Thresholds were determined based upon participants’ responses on a 2 alternate forced choice recognition task using a staircase to adaptively select the next stimulus. We found that 86.67% of the participants’ median z translation thresholds were higher when breathing gases with an oxygen content comparable to 8,000 feet than when breathing gases with an oxygen content comparable to 0 feet. The median z translation threshold was 24.11% higher in the hypoxia condition. In other words, larger movements up/down were needed for participants to reliably sense z-translation motion. The results of the present study suggest that mild hypoxia (i.e., roughly equivalent to that experienced at 8,000 feet; the altitude limit to which commercial air lines must pressurize) can impact our ability to reliably sense whether we moved up or down. Specifically, when hypoxic it takes larger movements for humans to sense the motion. Additional research is needed to determine the exact mechanism(s) of this effect, if this impacts other elements of human spatial orientation, and if this generalizes to other types of motions (e.g., roll-tilts).

Acknowledgements: This research was funded by The Office of Naval Research (MURI award number N00014–20-1–2163).

Podium Abstract 28Influence of Visual Afference on Self-Motion Perceptual Training

Andrew R. Wagner, DPT, PhD and Daniel M. Merfeld, PhD

Department of Otolaryngology — Head and Neck Surgery, The Ohio State University Wexner Medical Center, Columbus, OH

Aging, as well as vestibular pathology, are each characterized by changes in vestibular self-motion perception. Specifically, the perception of a passive whole-body roll tilt motion stimulus has been found to (1) degrade with age [1] and (2) correlate strongly with balance [2,3], suggesting that changes in vestibular perception may contribute to subclinical changes in postural control. Recent evidence showing that roll tilt perception can be improved with training [4,5] supports that self-motion perception may represent a potentially modifiable risk factor for balance dysfunction. However, efforts to determine the impact of vestibular perceptual training on balance control are constrained by a lack of knowledge into the training parameters best suited to drive improvements in self-motion perception. We recently showed that a 5-day training protocol, consisting of thirteen EPOCHs of 100 trials each, led to a 17.68 to 23.77% improvement in roll tilt perceptual thresholds [5]. Our protocol used a simple feedback schema whereby participants were (a) tilted in the dark, (b) asked to judge the direction of tilt, and (c) provided distinct binaural auditory tones for correct and incorrect answers. Despite a significant effect at the group level, we observed two disparate cohorts within the training group (N = 20). Approximately half of the training group (11/20, 55%) improved at the post test assessment by more than 2-standard deviations (i.e., 33.2%), and were labeled as “responders” [4]; by extension, 45% of the cohort showed little change between the baseline and posttest assessments and were label “non-responders”. To determine if the heterogenous response to the training protocol was related to the chosen feedback schema we designed the present study to determine if learning could be enhanced with the provision of a visual feedback cue. In a new cohort of 10 healthy adult volunteers (Age 25.3±3.3), we replicated the prior study, but added visual feedback after each response during the return to an upright position. At posttest we found a nominal and insignificant (p>0.05) increase in roll tilt thresholds of 1.7±56%, and consistent with our prior study, the response to the training protocol varied widely within the group. However, unlike the previous study where only a simple auditory tone was used [4], three out of the 10 participants who received visual feedback showed a large increase in thresholds, averaging a 69.7±52.35% increase in thresholds at posttest. Interestingly, in the remaining 7 participants we found a robust decrease in thresholds of 32.36±16.29%. These data support that a visual feedback cue may enhance learning in a subset of individuals, while in others it may provide potentially conflicting feedback resulting in a worsening of self-motion perception. Additional studies are required to determine the source of the variability in perceptual training, including the potential influence of baseline visual motion perception.

References

(1) Bermudez-Rey, et al. Frontiers in Neurology. 2016;7(162).

(2) Karmali, et al. Frontiers in Neurology. 2017;8(578).

(3) Wagner, et al. Frontiers in Integrative Neuroscience. 2021;15.

(4) Klaus MP, et al. Attention Perception & Psychophysics. 2020;82(4).

(5) Wagner AR, et al. Journal of Neurophysiology. 2022;128(3).

Acknowledgements: Funded by NIA 1R01AG073113.

Podium Abstract 29Modeling the impact of galvanic vestibular stimulation on human self-orientation perception in the presence of physical stimuli

Aaron R. Allred, Caroline Austin, Nicholas Boggess, David Temple, Lanna Klausing, Skylar Edwards, Torin K. Clark

University of Colorado – Boulder, Boulder, CO

Galvanic vestibular stimulation (GVS) has a myriad of promising applications ranging from rectifying vestibular hypofunction and improving self-motion thresholds to enhancing simulator immersion and countering motion sickness. However, the utility of GVS is contingent on our ability to predict resultant human perception, particularly in the presence of physical motion. To this end, we developed a computational model of human perception due to both physical and GVS-induced virtual motion, based on the “observer model” for spatial orientation perception during passive motions (Clark et al., 2019; Merfeld et al., 1993; Merfeld & Zupan, 2002). Foundational to this effort, recent works have quantified the frequency dynamics of bipolar GVS applied to the mastoid processes of macaque monkeys (Forbes et al., 2023; Kwan et al., 2019). Incorporating these findings, we present an observer model, for passive motions in the dark, augmented to account for the influence of GVS on vestibular sensory afferents in humans. To account for anatomical differences between humans and primate models, and to model the individual afferent contributions of both physical and electrical stimuli, we collected perception of roll tilt data in human participants (N=13) during static roll tilts across a range of GVS intensities. Perception of tilt was measured via a subjective haptic orientation task, and direct current GVS was supplied to participants in the bipolar configuration, both concordant and discordant with physical tilt. Perceptions of tilt were found to be affected by GVS (with the sign of the current corresponding to the right mastoid’s polarity) averaging about -0.49 deg/mA, with consistent effects observed across individuals. To further inform and validate this model, we have designed a secondary experiment to dynamically assess perceptions of roll tilt in subjects during dynamic roll tilt in the presence of a dynamic GVS waveform, both with low (<0.5Hz) frequency content across a range of peak stimulus intensities. The realization of this model offers the potential to computationally assess and design GVS waveforms to achieve specific perceptual responses in humans.

References

Clark et al (2019); Progress in Brain Research (Vol. 248, pp. 65–90).

Merfeld et al. (1993); Journal of Vestibular Research, 3(2), 123–139.

Merfeld and Zupan (2002); Journal of Neurophysiology, 87(2), 819–833.

Forbes et al (2023); The Journal of Neuroscience, JN-RM-0987-22.

Kwan et al (2019); Nature Communications, 10(1), 1904.

Acknowledgements: This work was supported by a NASA Space Technology Graduate Research Opportunities Award.

Podium Abstract 30Model of Motion Perception Following Sudden Transitions of Visual Cue Availability

J. L. Voros¹, T. K. Clark¹

¹Ann and H. J. Smead Aerospace Engineering Sciences, The University of Colorado Boulder, 80309

Existing models of motion perception do not capture motion perception during all common flight scenarios. For example, we are not aware of any existing model that aptly predicts the dynamics of motion perception following sudden transitions in the availability of visual cues (e.g. as an aircraft pilot experiences flying into or out of a cloud). It is important that we understand spatial orientation and motion perception in commonly occurring flight scenarios because spatial disorientation is associated with adverse failure modes (aircraft crashes). By developing computational models of orientation perception that are robust to sudden changes in visual cues, we can better understand the potential spatial disorientation that occurs around such transitions.

For the present study, we first collected empirical human subject data on motion perception during extended periods of Earth vertical yaw motion. The specific Earth vertical yaw rotations that our human subjects experienced were designed to elicit different perceptions of orientation with and without visual cues. Therefore, when a sudden change in the availability of visual cues occurred during a period of misperception of angular velocity in the no visual cues case, we were able to characterize the ensuing pattern of orientation perception. Data collection was performed on 28 unique subjects. The subjects were tasked with pressing a button every 90 degrees in order to record their perception of motion. Visual scenarios included two control conditions: no visual cues (i.e., in the dark) and visual cues (i.e., rotating dots providing angular vection) the entire time. The two test conditions were: suddenly losing visual cues and suddenly gaining visual cues.

We hypothesized that the existing pathway for sensory conflict associated with visual angular vection is effectively low pass filtered before being integrated with vestibular pathways. Therefore, we modified an existing model of orientation perception [1] and were able to have the modified form [with low pass filtering] match out results. While speculative, this low pass filter mechanism may be associated with the ambiguity in visual motion cues being caused by self-motion or scene-motion.

References

[1] T. K. Clark, M. C. Newman, F. Karmali, C. M. Oman, and D. M. Merfeld, Mathematical models for dynamic, multisensory spatial orientation perception, in Progress in Brain Research, vol. 248, Elsevier, 2019, pp. 65–90.

[2] M. C. Newman. A multisensory observer model for human spatial orientation perception, MS Thesis, Massachusetts Institute of Technology. Cambridge, MA, 2009.

Podium Abstract 31Why velocity storage changes in response to peripheral vestibular damage

Amsal Madhani, Richard F. Lewis, Faisal Karmali

Harvard Medical School, Jenks Vestibular Physiology Lab, Mass Eye and Ear

Peripheral vestibular cues are processed by a central mechanism known as velocity storage. This mechanism plays an important role in directing the dynamic response to yaw rotation. For example, during constant angular velocity rotation, horizontal eye movements and perceived angular velocity exponentially decay with a velocity storage time constant of ~15-30 s in normal individuals. Notably, the vestibular periphery input decays much faster, around 6 s. Peripheral vestibular damage leads to a substantial reduction in the velocity storage time constant. For example, the diagnostic range for complete unilateral loss is 6-12 s. However, the underlying mechanism linking vestibular damage and changes in the time constant is not fully understood. In this study, we aimed to address this gap. We built on a theory that the brain uses Bayesian optimal processing to compute the ideal velocity storage time constant and does so based on the statistics of vestibular noise and experienced motion. More specifically, although a longer time constant would improve the accuracy of the central estimate of angular head velocity, it may also amplify neural noise, which degrades precision. Thus, in determining the optimal time constant. the brain may be striving to strike a balance between accuracy and precision. In this study, we applied a Bayesian Kalman filter model to investigate the optimal velocity storage time constant for unilateral damage. To account for the unknown effect of unilateral damage on vestibular noise, we developed five scenarios based on literature. As detailed in our paper (Madhani et al. 2023), for all scenarios, predictions were significantly reduced compared to those of normal individuals and consistent with clinical observations. Furthermore, interactions between age-related hair cell loss and peripheral vestibular damage were also modeled. Together, results suggest that the brain optimizes the velocity storage time constant based on the vestibular signal-to-noise ratio and emphasize the highly consequential role of noise in neurocomputations performed by the brain.

Reference

Madhani, Amsal, Richard F. Lewis, and Faisal Karmali. Journal of the Association for Research in Otolaryngology. 2022;23.4:551-566.

Acknowledgements: NIH/NIDCD R01-DC018287

Podium Abstract 32The role of sensory noise in closed-loop sensorimotor control

Faisal Karmali¹, Kassia Love¹, Moein Nazifi¹, Tara J. Nibhanupudy¹, Torin Clark², Marissa Rosenberg³, Raquel Galvan-Garza⁴, Megan. E. Reissman⁵, Timothy Reissman⁵, Y.C. Rah¹, E. López-Contreras Gonzalez¹

¹Harvard Medical School, Jenks Vestibular Physiology Lab, Mass Eye and Ear, Boston, MA

²Univ. of Colorado - Boulder, Boulder, CO

³Space X, Los Angeles, CA

⁴Lockheed Martin, Arlington, VA

⁵Department of Mechanical Engineering, University of Dayton, Dayton, OH

Sensory, perceptual, and motor processes are affected by noise, which is defined as random disturbances in neural processes [1]. In closed-loop tasks like postural control, it is thought that performance is degraded due to neural noise in sensory and motor systems. However, it is complicated to study the role of noise in closed-loop tasks because noise result in variability in the motor output, which is fed back into the sensory inputs, potentially causing more variability in the motor output. Thus, the problem has been studied using both computational models and experiments. Examples of computational approaches include posture models that posit that postural sway arises from neural noise [2]. Conversely, postural sway and performance in manual control are correlated across subjects with perceptual thresholds, which assay sensory noise according to signal detection theory; notably, these correlations do not consider closed-loop behaviors. We brought together these approaches to test the hypothesis that sensory noise degrades performance in closed-loop orientation control, using: 1) closed-loop computational models; 2) experimental closed-loop task performance; and 3)vestibular perceptual threshold as an assay of sensory noise. We modeled tasks in which subjects sat on a motorized chair that could be controlled using a joystick. In one study we performed experiments and modeling for yaw-axis control, in which subjects experienced a random motion and were instructed to use the joystick to keep themselves at rest. We also modeled results from a published study showing a correlation between performance in a roll-tilt self-orientation control task and vestibular thresholds [4]. Performance in different gravity environments was also studied using a centrifuge. In the closed-loop model, we assumed based on decision-making theory that subjects would make joystick inputs only when they had reliable vestibular cues - i.e., the sensory input exceeded sensory noise. Model predictions closely matched experimental [4] results, including frequency responses. Subjects with lower vestibular thresholds (i.e., less neural noise) had less variable manual control performance, with a similar slope between manual control performance and thresholds for model and experimental results. Thus, the results strengthen support for the hypothesis that vestibular noise worsens closed-loop control performance. Model variations had little impact on results, showing robustness. Manual control performance was better (i.e., less variable) in higher G levels in both model and experimental [4] results, supporting the hypothesis that better performance results from increased vestibular signal-to-noise ratio in higher G level.

References

(1) Diaz-Artiles & Karmali. Neuroscience, 2021. 468: p. 282.

(2) Maurer C, Peterka RJ. J Neurophysiol. 2005 Jan;93(1):189-200.

(3) van der Kooij & Peterka. J Comput Neurosci, 2011. 30(3): p. 759.

(4) Rosenberg. J Neurophysiol, 2018. 120(6): p. 3187.

Acknowledgements: NIH/NIDCD R01-DC018287 and DC018287S1, ONR MURI N000142012163 and NASA NCC 9-58

Podium Abstract 33Links between vestibular perception and cognition in healthy aging

Megan J. Kobel, AuD, PhD¹, Andrew R. Wagner, DPT, PhD¹, Daniel M. Merfeld, PhD¹

¹The Ohio State University Medical Center—Department of Otolaryngology—Head & Neck Surgery

A growing body of evidence suggests relationships between age-related changes in vestibular function and cognition. However, literature to date has focused on vestibular reflexive pathways and potential links between vestibular perception and cognition have not been thoroughly examined. Thus, we assessed a comprehensive vestibular threshold test battery and standard assessments of cognitive function previously seen to be linked to vestibular dysfunction across a broad age-range (21-86 years of age; n=52). Overall, both vestibular perceptual thresholds and cognitive function demonstrated significant age-related changes. Regression analyses controlling for age and demographic variables identified a relationship between vestibular perceptual thresholds and cognitive assessments for only low frequency roll tilt (i.e., medio-lateral earth horizontal rotation) thresholds, reflecting canal-otolith integration, and earth-vertical superior-inferior translation thresholds, probing saccular function. No other significant relationships to cognitive measures were seen for other measures probing perception predominantly mediated by the semi-circular canals or utricle. Thus, our evidence suggests a link between cognition and modality specific vestibular perception in healthy aging.

Funding: Department of Defense Congressionally Directed Medical Research Programs (CDMRP) Award Number W81XWH192000.

Podium Abstract 34Sensory contributions to frontal plane balance control during stance and stepping-in-place gait in younger and older healthy adults

Robert J. Peterka

National Center for Rehabilitative Auditory Research (NCRAR), VA Portland Health Care System, Portland OR

Control of body orientation during stance and metronome paced stepping-in-place (SiP) gait was measured in 20 younger (25-43 years) and 20 older (65-82 years) healthy adults. Balance was perturbed by continuously applied pseudorandom rotational stimuli (six 36.6-s cycles) at 3 amplitudes (1°, 3°, 6° peak-to-peak) that evoked frontal plane sway in 3 test conditions: surface-tilt with eyes open (SS/EO) and closed (SS/EC), and visual scene tilt (VS/EO). Lateral body displacement measures at hip and shoulder levels were used to estimate center-of-mass (CoM) tilt angle with respect to an axis at ankle height centered between the feet. For each test, the frequency-dependent relationship between the stimulus tilt and the evoked CoM sway angle was represented by a frequency response function (FRF) that characterized response sensitivity (gain) and timing (phase). Parameters of a relatively simple feedback model of balance control were estimated from fits to average FRFs. Parameters included 1) sensory weights that represent the relative contribution of different sensory systems to balance control with a proprioceptive weight (W_prop) estimated from SS/EO and SS/EC tests, and a visual weight (W_vis) estimated from VS/EO tests, 2) neural controller parameters that translate integrated sensory information to corrective torque, 3) torque feedback parameters, and 4) time delay. This report focuses on the sensory contributions. For each of the 3 test conditions, for both stance and SiP, and for both younger and older subjects, W_prop and W_vis decreased with increasing stimulus amplitude demonstrating amplitude-dependent sensory reweighting. For each of the 3 test conditions, for both stance and SiP, and at each stimulus amplitude W_prop and W_vis were larger for older than for younger subjects indicating age-related differences in sensory utilization for balance. For both younger and older subjects on surface-tilt tests, W_prop on SiP tests was about 40% smaller than on stance tests indicating a reduced ability to use proprioception for balance control during SiP. On SS/EC tests smaller W_prop implies increased vestibular reliance (larger W_vest) during SiP compared to stance. Additionally, comparing younger and older subjects, the higher reliance on proprioception on the SS/EC test in older subjects indicates their reduced ability to use vestibular cues for balance. For both younger and older subjects on VS/EO tests, W_vis was remarkably larger (~10 times) on SiP than on stance tests indicating a greatly enhanced reliance on vision for balance during SiP gait. Additionally, older subjects were less able to reduce W_vis with increasing stimulus amplitude on SiP than younger subjects. At the 1° stimulus amplitude, W_vis in older subjects was only 5% larger than in younger subjects, but at the 6° amplitude, W_vis was 2 times larger in older than in younger subjects.

Acknowledgements: Supported by VA Rehabilitation Research and Development (RR&D) Merit Award I01RX001951 to Robert Peterka. Additional resource and facilities support provided by VA RR&D NCRAR Center Award #C2361C/I50 RX002361 at the VA Portland Health Care System. The content is solely the responsibility of the author and does not necessarily represent the official views of the Department of Veterans Affairs.

Poster Abstracts