0442 Predicting Sleepiness from Verbal Reaction Time During Verbal Cognitive Testing

BackgroundTo determine whether objective markers of sleepiness can be collected passively using voice data to detect sleepiness in individuals undergoing testing in situations where sleepiness is not the focal point of assessment. We assessed verbal reaction time (VRT) as a vocalic marker of subjective sleepiness in middle aged and older adults with history of insomnia and benzodiazepine-receptor-agonist (BZRA) use.MethodsAdults aged ≥55 without a diagnosis of dementia were recruited from a BZRA deprescribing clinical trial and enrolled in the present study that tested the feasibility of cognitive testing using out-of-office, self-directed mobile apps. Participants’ working/episodic memory were assessed through recorded verbal responses to Verbal Paired Associates (VPA) tests, and ecological momentary assessments (EMA) of self-reported sleepiness (1[not at all] to 4[more prominent]). Using a generalized additive model, we examined the association between VRT during VPA testing and self-reported sleepiness, adjusting for demographic, test parameters, caffeine intake, cognition, mood, and BZRA-use (p≤0.05 was considered significant). A stratified k-fold cross-validation/random forest (SKCV/RF) was performed to classify sleepiness levels, adjusting for other variables.ResultsWe analyzed 1,513 observations from 16 patients. VRT was operationalized as the time duration between recording start time and first speech epoch. Longer VRTs were positively associated with greater EMA sleepiness (p≤0.05). The SKCV/RF model yielded a mean F1-score of 0.80 ± 0.08 across folds.ConclusionsLonger VRTs correlated with greater self-reported sleepiness, indicating that voice data can be used as a marker of sleepiness in patients undergoing cognitive testing in out-of-office settings.

After experiencing the global COVID-19 pandemic, whether there have been new changes in the epidemiological characteristics of influenza has become a topic of great concern. This study aims to investigate the impact of implementation and lifting of COVID-19 control measures on influenza positivity among patients with acute respiratory infections (ARI) from 2018 to 2023. The data were collected from January 2018 to December 2023 in two designated sentinel hospitals in Jinhua. We performed an interrupted time series analysis (ITSA) using a beta regression model and a generalized additive model (GAM), adopting a two-model cross-validation strategy to assess the effect of two major interventions on influenza positivity: the COVID-19 control measures implemented in early 2020 and lifted at the end of 2022. We also analyzed influenza epidemiological characteristics and seasonality before, during, and after the pandemic. A total of 98,244 cases were included in this study, and the overall influenza positivity rate was 39.34%. Females and the 6-17-year age group had higher positivity rates. Before the pandemic, influenza primarily showed a winter peak pattern, whereas during the pandemic, the positivity rate declined significantly with no distinct peak. After the pandemic ended, an unusual dual-peak pattern emerged. The interrupted time series analysis revealed that, following the implementation of non-pharmaceutical interventions (NPIs) in early 2020, influenza positivity immediately decreased significantly in the beta regression model (β = -1.75, p = 0.003). After the lifting of measures in late 2022, a marginally lagged increasing trend was observed in the beta regression model (β = 0.14, p = 0.096) and a significant increasing trend was found in the GAM model (edf = 7.00, p < 0.001). Seasonal effects differed between the models: the beta regression model exhibited significant annual seasonal fluctuations (sin12 = 0.67, p < 0.001), while the GAM model did not exhibit a significant association independent of the time trend. COVID-19 and its control measures substantially reduced influenza positivity rates; however, once these measures were lifted, influenza activity resurged, and its seasonal epidemic pattern changed. The intensity of influenza appeared to exceed pre-pandemic levels, underscoring the importance of NPIs in controlling respiratory infectious diseases. Strengthened surveillance and optimized strategies remain necessary to mitigate the threat of influenza in the post-pandemic era.

Performance in cognitive tests is influenced by age, education and culture. Many batteries for screening and diagnosing dementia were developed in countries where low education and illiteracy are uncommon and therefore they may not be suitable for evaluation of populations with heterogeneous educational backgrounds. In this chapter, we discuss the influences of education in the diagnosis of dementia and present the results of studies with the Brief Cognitive Screening Battery (BCSB) or Brief Cognitive Battery unbiased by education (BCB-Edu) developed by Nitrini et al. (1994). The BCB-Edu has been employed for the diagnosis of dementia and Alzheimer’s disease (AD) in subjects with heterogeneous educational background (illiterate, low/medium/high levels of education). The delayed recall (DR) test of BCB-Edu proved to be more accurate than DR of Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) battery for the diagnosis of dementia in illiterate subjects [sensitivity = 93.3%, specificity = 95.7%, area under receiver operator characteristic curves (AUC-ROC) = 0.975]. DR of BCB-Edu also showed high sensitivity (82.2%) and specificity (90.4%) for the diagnosis of AD in patients with medium/high levels of education. A mathematical model including the results of core tests is also suggested to be employed in clinical practice. The BCB-Edu is a simple and useful tool for diagnosis and screening of dementia, particularly AD, in populations with heterogeneously educated individuals.

Walleye pollock Theragra chalcogramma (pollock hereafter) is a key ecological and economic species in the eastern Bering Sea, yet detailed synthesis of the spatial and temporal patterns of pollock ichthyoplankton in this important region is lacking. This knowledge gap is particularly severe considering that egg and larval distribution are essential to reconstructing spawning locations and early life stages drift pathways. We used 19 yr of ichthyoplankton collections to determine the spatial and temporal patterns of egg and larval distribution. Generalized additive models (GAMs) identified two primary temporal pulses of pollock eggs, the first occurring from 20 February to 31 March and the second from 20 April to 20 May; larvae showed similar, but slightly lagged, pulses. Based on generalized cross‐validation and information theory, a GAM model that allowed for different seasonal patterns in egg density within three unique areas outperformed a GAM that assumed a single fixed seasonal pattern across the entire eastern Bering Sea. This ‘area‐dependent’ GAM predicted the highest densities of eggs (i.e., potential spawning locations) in three major areas of the eastern Bering Sea: near Bogoslof Island (February–April), north of Unimak Island and the Alaska Peninsula (March–April), and around the Pribilof Islands (April–August). Unique temporal patterns of egg density were observed for each area, suggesting that pollock spawning may be more spatially and temporally complex than previously assumed. Moreover, this work provides a valuable baseline of pollock spawning to which future changes, such as those resulting from climate variability, may be compared.

This study evaluates the influence of routine soil index properties on the prediction of optimum moisture content (wopt) and maximum dry unit weight (γdmax), which are the primary outcomes of the Proctor compaction test, using machine learning (ML) methods. A curated database of fine-grained soils (n = 465, drawn from 15 sources) included gravel content (GC), sand content (SC), fines content (FC), liquid limit (LL), plastic limit (PL), plasticity index (PI), specific gravity (Gs), wopt, and γdmax. After correlation-based feature filtering, three models were developed: Generalized Additive Model (GAM), Random Forest (RF), and Extreme Gradient Boosting (XGBoost). The training used nested cross-validation with Bayesian optimization, corresponding to an overall 80–20 train-test split. The model performance was evaluated using R2, RMSE, MAE, MAPE, r, and the overfitting ratio calculated for the test set. For wopt, the best GAM model achieved R2 = 0.84 and RMSE = 2.16%, outperforming RF and XGBoost. For γdmax, the best GAM and XGBoost models reached R2 = 0.79 and RMSE = 0.76 kN/m3, respectively. SHapley Additive exPlanations (SHAP), model-based importance scores, and single ablation analyses consistently identified LL and PL as the most influential predictors, and FC provided secondary contributions, while GC and Gs added little once LL and PL had been included. Moreover, paired-feature ablation confirmed the joint influence of LL and PL on the prediction. Overall, all three models predicted compaction parameters with good accuracy; however, GAM models achieved comparable or better predictive metric values than the ensembles (RF and XGBoost) while offering interpretability through plots linking soil indices with the predicted outcomes. This balance of accuracy and interpretability supports GAM as the preferred model for prediction modeling.

Research studies connecting air pollution and its effect on health are much needed by the policymakers in India. Generalized additive model (GAM) was applied to quantify the effect of air pollutants on hospital visits related to respiratory diseases for an Indian city, Guwahati. A total of 6335 patients visited the only medical college and hospital in the city for respiratory diseases during 2016-2018. The multi-pollutant GAM models indicate the existence of synergistic effects of particulate matter less than 2.5 µm in diameter (PM2.5), nitrogen dioxide (NO2), and carbon monoxide (CO) on hospital visits of the patients due to respiratory diseases. The ambient PM2.5 concentrations have a strong link with hospital visits in all 6 single lag days. However, the effect of CO on hospital visits due to respiratory diseases was more robust. The percent change of hospital visits associated with each 1 unit increase in CO concentrations at lag 1 day was 15.7% (RR = 1.157; 95% CI 0.929, 1.442), which was enhanced to as high as 40.8% (RR = 1.408; 1.130,1.755) at lag3. The rise of each 10 units of concentrations of PM2.5 and 1 unit of CO at cumulative lag0-6 day was associated with the percentage change in hospital visits of 6.67% and 62.36% respectively. Single-pollutant GAM models indicate that the risk percentage of PM2.5 causing respiratory disease increased in combination with other pollutants.

The Mini-Mental State Exam (MMSE) is a brief cognitive test that assesses several cognitive domains, such as orientation, attention, concentration, memory, language, and constructional abilities. While the MMSE was found to be valid and reliable in the diagnosis of moderate dementia, its sensitivity and specificity for the diagnosis of mild dementia has been rarely examined. We assessed the specificity and sensitivity of the MMSE in a consecutive series of 44 patients with mild dementia, and a group of age-comparable normal controls. While the specificity of the MMSE for the diagnosis of mild dementia was very high (100%), the sensitivity was only 55% [20 of the 44 patients with mild dementia had an MMSE score in the normal range (≥26 points)]. On the other hand, the assessment with both the Buschke Selective Reminding and the Boston Naming tests discriminated mild Alzheimer's disease patients with normal MMSE scores from controls with a sensitivity and specificity of 64%. In conclusion, the assessment with verbal memory and naming tasks provided a significantly more sensitive measure of early dementia than the MMSE.

Theory-Guided Machine Learning Finds Geometric Structure-Property Relationships for Chemisorption on Subsurface Alloys

Event Abstract Back to Event Standardization of catch-per-unit-effort in a multi-gear multi-species fishery Lorena Couce1*, Angelo Santana2, Alberto Bilbao-Sieyro1, Yeray Perez1 and Jose J. Castro3 1 Management of the Rural Environment of Canarias SAU, Spain 2 University of Las Palmas de Gran Canaria, Departamento de Matemáticas, Spain 3 Instituto Universitario de Acuicultura Sostenible y Ecosistemas Marinos (IU-ECOAQUA), Spain Catch-per-unit-effort (CPUE) data are commonly standardized taking into account factors that influence catch rates that are not related to resource abundance. In this work we analyze catch rates and effort data of a small-scale fishery targeting multiple species in the Canary Islands. We developed a method that combines principal component scores derived from a Principal Component Analysis (PCA) of the effort data within a Generalized Additive Model (GAM) framework. From the PCA analysis we also estimated the relative fishing power of each vessel with reference to a standard vessel for which the principal components are equal to zero, and based on the results we define two categories; the first category includes the benthic and demersal species and the second category includes the oceanic and pelagic species. The GAM model described 66.1% of the total deviance and includes seasonal variables, fishing grounds, climate index, oceanographic and biological parameters and dummies for tuna season and medium-sized coastal pelagics. The variables with the highest contribution in order of importance were the tuna dummy, the first principal component, the medium-sized coastal pelagics dummy and the fishing ground. The standardized CPUE time series showed a slight positive trend throughout the analyzed period as well as a marked seasonal component due to the tuna harvest. However, when the CPUE is analyzed according to the target groups, a negative trend is observed in the herviborous and invertebrate feeders group, reef-associated fishes, shallow-water demersals and molluscs from 2010, and the benthic cephalopods show an appreciable downward trend from 2012. We consider the approach presented here has potentially useful for stock assessment in similar multi-species fisheries. Acknowledgements This work was developed within the framework of the PLASMAR Project (MAC/1.1a/030), with the support of the European Union (EU) and co-financed by the European Regional Development Fund (EDRF ) and the INTERREG V-A Spain-Portugal MAC 2014-2020 (Madeira-Azores-Canarias). Keywords: Small-scale fishery, GAM (generalized additive model), PCA principal component analysis, CPUE, standardization, Multi-gear fisheries, Multi-species fisheries Conference: XX Iberian Symposium on Marine Biology Studies (SIEBM XX) , Braga, Portugal, 9 Sep - 12 Sep, 2019. Presentation Type: Poster Presentation Topic: Fisheries, Aquaculture and Biotechnology Citation: Couce L, Santana A, Bilbao-Sieyro A, Perez Y and Castro J (2019). Standardization of catch-per-unit-effort in a multi-gear multi-species fishery. Front. Mar. Sci. Conference Abstract: XX Iberian Symposium on Marine Biology Studies (SIEBM XX) . doi: 10.3389/conf.fmars.2019.08.00161 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 02 Jun 2019; Published Online: 27 Sep 2019. * Correspondence: Dr. Lorena Couce, Management of the Rural Environment of Canarias SAU, Agüimes, Spain, lorenacm@gmrcanarias.com Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Lorena Couce Angelo Santana Alberto Bilbao-Sieyro Yeray Perez Jose J. Castro Google Lorena Couce Angelo Santana Alberto Bilbao-Sieyro Yeray Perez Jose J. Castro Google Scholar Lorena Couce Angelo Santana Alberto Bilbao-Sieyro Yeray Perez Jose J. Castro PubMed Lorena Couce Angelo Santana Alberto Bilbao-Sieyro Yeray Perez Jose J. Castro Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.

The swimming layer is one of the important factors to get maximum catches, especially on tuna longline effort. The vertical abundance of the albacore tuna was investigated based on catch data and 3-DINDESO Ocean Model data, such as sub-surface conditions of sea water potential temperature (Temp), salinity (Sal) and mass concentration of diatoms and flagellates expressed as chlorophyll (Chl) in the Eastern Indian Ocean period 2014-2015. Combining he statistical method of generalized additive model (GAM) was performed to analysis in this study. There were seven GAM models that generated with the number of ALB vertical abundance as a response variable, and Temp, Sal, and Chl as predictor variables. Sal has highly significant (P &lt; 0.001) while Chl and Temp significant (P &lt; 0.01) to ALB vertical abundance. Deduced from GAMs, indicated that a negative effect of Sal on the number of ALB was observed at salinity &gt;34.52 psu. There was a positive effect of salinity on the number of ALB, which was from 34.30 to 34.47 psu and Chl showed a positive effect of this variable on the number of ALB caught occurred between 0.01 mg/m3 and 0.12 mg/m3 in the region of high confidence level where negative effect on &gt; 0.13 mg/m3. While ALB catches abundance varied in the temperature range with the highest frequency at 24.0-24.9 °C. Sal was the most important environmental variable to the number of ALB vertically caught, followed by Chl and Temp.

Acifluorfen is a nonsystemic PPO-inhibiting herbicide commonly used for POST Palmer amaranth control in soybean, peanut, and rice across the southern United States. Concerns have been raised regarding herbicide selection pressure and particle drift, increasing the need for application practices that optimize herbicide efficacy while mitigating spray drift. Field research was conducted in 2016, 2017, and 2018 in Mississippi and Nebraska to evaluate the influence of a range of spray droplet sizes [150 μm (Fine) to 900 μm (Ultra Coarse)], using acifluorfen to create a novel Palmer amaranth management recommendation using pulse width modulation (PWM) technology. A pooled site-year generalized additive model (GAM) analysis suggested that 150-μm (Fine) droplets should be used to obtain the greatest Palmer amaranth control and dry biomass reduction. Nevertheless, GAM models indicated that only 7.2% of the variability observed in Palmer amaranth control was due to differences in spray droplet size. Therefore, location-specific GAM analyses were performed to account for geographical differences to increase the accuracy of prediction models. GAM models suggested that 250-μm (Medium) droplets optimize acifluorfen efficacy on Palmer amaranth in Dundee, MS, and 310-μm (Medium) droplets could sustain 90% of maximum weed control. Specific models for Beaver City, NE, indicated that 150-μm (Fine) droplets provide maximum Palmer amaranth control, and 340-μm (Medium) droplets could maintain 90% of greatest weed control. For Robinsonville, MS, optimal Palmer amaranth control could be obtained with 370-μm (Coarse) droplets, and 90% maximum control could be sustained with 680 μm (Ultra Coarse) droplets. Differences in optimal droplet size across location could be a result of convoluted interactions between droplet size, weather conditions, population density, plant morphology, and soil fertility levels. Future research should adopt a holistic approach to identify and investigate the influence of environmental and application parameters to optimize droplet size recommendations.

Performing high-quality and reliable cognitive testing requires significant resources and training. As a result, large-scale studies involving cognitive testing are difficult to perform in low- and middle-income settings, limiting access to critical knowledge to improve academic achievement and economic production in these populations. The NIH Toolbox® is a collection of cognitive, motor, sensory, and emotional tests that can be administered and scored using an iPad® tablet, reducing the need for training and quality monitoring; and thus, it is a potential solution to this problem. We describe our process for translation and cultural adaptation of the existing NIH Toolbox tests of fluid cognition into the Swahili and Dholuo languages for use in children aged 3-14 years in western Kenya. Through serial forward and back translations, cognitive interviews, group consensus, outside feedback, and support from the NIH Toolbox team, we produced translated tests that have both face validity and linguistic validation. During our cognitive interviews, we found that the five chosen tests (one each of attention, cognitive flexibility, working memory, episodic memory, and processing speed) were generally well understood by children aged 7-14 years in our chosen populations. The cognitive interviews informed alterations in translation as well as slight changes in some images to culturally adapt the tests. This study describes the process by which we translated five fluid cognition tests from the NIH Toolbox into the Swahili and Dholuo languages. The finished testing application will be available for future studies, including a pilot study for assessment of psychometric properties.

The aim of this research was to evaluate the neurocognitive functioning and symptom reporting of high school athletes with the Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT) battery after sustaining a single sports-related concussion. The ImPACT battery was administered to 26 athletes at an average of 6.8 days after their head injury. ImPACT composite scores, including neurocognitive measures of Verbal Memory, Visual Memory, Processing Speed, and Reaction Time, as well as a Total Symptom Score, were also obtained from an equivalent group of 25 nonconcussed football players. The composite scores of the concussed athletes were lower but not statistically different than the nonconcussed athletes. The findings were consistent with previous ImPACT research that reported no differences between concussed and nonconcussed athletes 7 days after a concussion. The symptom scores of the concussed athletes, on the other hand, were significantly higher than those who had no concussion. The similarities and differences in ImPACT test performances of the present sample of concussed high school athletes as compared with previous studies of concussed high school athletes are discussed. This study raises awareness that with high school athletes, symptom complaints may persist, even after cognitive functioning has returned to preinjury levels.

Various populations with chronic conditions are at risk for decreased cognitive performance, making assessment of their cognition important. Formal mobile cognitive assessments measure cognitive performance with greater ecological validity than traditional laboratory-based testing but add to participant task demands. Given that responding to a survey is considered a cognitively demanding task itself, information that is passively collected as a by-product of ecological momentary assessment (EMA) may be a means through which people's cognitive performance in their natural environment can be estimated when formal ambulatory cognitive assessment is not feasible. We specifically examined whether the item response times (RTs) to EMA questions (eg, mood) can serve as approximations of cognitive processing speed. This study aims to investigate whether the RTs from noncognitive EMA surveys can serve as approximate indicators of between-person (BP) differences and momentary within-person (WP) variability in cognitive processing speed. Data from a 2-week EMA study investigating the relationships among glucose, emotion, and functioning in adults with type 1 diabetes were analyzed. Validated mobile cognitive tests assessing processing speed (Symbol Search task) and sustained attention (Go-No Go task) were administered together with noncognitive EMA surveys 5 to 6 times per day via smartphones. Multilevel modeling was used to examine the reliability of EMA RTs, their convergent validity with the Symbol Search task, and their divergent validity with the Go-No Go task. Other tests of the validity of EMA RTs included the examination of their associations with age, depression, fatigue, and the time of day. Overall, in BP analyses, evidence was found supporting the reliability and convergent validity of EMA question RTs from even a single repeatedly administered EMA item as a measure of average processing speed. BP correlations between the Symbol Search task and EMA RTs ranged from 0.43 to 0.58 (P<.001). EMA RTs had significant BP associations with age (P<.001), as expected, but not with depression (P=.20) or average fatigue (P=.18). In WP analyses, the RTs to 16 slider items and all 22 EMA items (including the 16 slider items) had acceptable (>0.70) WP reliability. After correcting for unreliability in multilevel models, EMA RTs from most combinations of items showed moderate WP correlations with the Symbol Search task (ranged from 0.29 to 0.58; P<.001) and demonstrated theoretically expected relationships with momentary fatigue and the time of day. The associations between EMA RTs and the Symbol Search task were greater than those between EMA RTs and the Go-No Go task at both the BP and WP levels, providing evidence of divergent validity. Assessing the RTs to EMA items (eg, mood) may be a method of approximating people's average levels of and momentary fluctuations in processing speed without adding tasks beyond the survey questions.

Objective:Memory is a critical piece of the human experience and impairments in neural memory networks can have devastating consequences for the affected person. A subtype of memory, episodic memory generates context for the present based on past experience and allows us to make predictions about the future. Episodic memories become stable fixtures through long-term memory consolidation. It is believed that consolidation of episodic memory requires a dynamic interplay between connected hippocampal-cortical networks, mainly during sleep. Sleep oscillations, slow oscillations and thalamocortical spindles, coupled with hippocampal sharp wave ripples (SWR) is proposed to be mechanistically involved in establishing the crucial cortical-subcortical dialog. The current study aimed to determine alterations in typical sleep oscillations and oscillation coupling in patients with and without structural hippocampal damage and correlate them with neuropsychological measures believed to be sensitive to hippocampal dysfunction, i.e., Rey Auditory Verbal Learning Task (RAVLT) and Verbal Paired Associates (VPA-II).Participants and Methods:We used intracranial electroencephalography (iEEG) in 14 patients with epilepsy to directly record hippocampal and neocortical oscillations and neuropsychological measures obtained prior to implantation. Half of the participants were diagnosed with mesial temporal sclerosis (MTS) in the left hippocampus and healthy tissue in the right hippocampus. The other half did not have MTS and had either mesial temporal epilepsy without MTS or extra-temporal seizures. We analyzed hippocampal SWR output from both hippocampi and characterized neocortical slow oscillations and spindles and their coupling for each participant. We correlated electrophysiological data with behavioral results of neuropsychological testing in order to characterize the clinical relevance.Results:SWR analysis revealed significant differences in the frequency, t(7639) = 15.52, p&gt;.001, p &gt; .001), amplitude, t(7664) = -23.93, p &gt; .001, and waveforms (p &gt; .001) of SWR in the sclerotic versus healthy hippocampi. Patients with a sclerotic hippocampus but relatively preserved verbal memory scores (RAVLT, VPA-II) showed increased SWR amplitudes in the contralateral hippocampus compared to patients with low verbal memory scores. Additionally, we found differences between hemispheres in phase amplitude coupling of SWRs to spindles and SOs (p &gt; 0.001). Results of our correlational analysis were variable and dependent upon additional factors, such as age of onset and diagnosis duration.Conclusions:Results from this work will aid in establishing a criterion for characterizing a relationship between subcortical and cortical oscillations as they relate to memory performance. Besides aiding our understanding of the neural mechanisms underpinning memory consolidation this will ideally help with developing neurophysiological biomarkers that may predict possible memory decline in resective or ablative neurosurgery absent of structural lesion. In addition, this work may potentially provide first evidence of a neurophysiological biomarker directly recorded from the human hippocampus to support possible reorganization of memory functioning in the non-sclerotic hippocampus.