Research Day - 2025 Program

Research Day 2025

December 11th | The Ohio Union, Cartoon Room

Schedule of Events

Keynote Presentation

Keynote Presentation: Dr. Lucina Q. Uddin

Bio: After receiving a Ph.D. in cognitive neuroscience from the Psychology Department at the University of California Los Angeles, Dr. Uddin completed a postdoctoral fellowship in the Child Study Center at New York University. For several years she worked as a faculty member in Psychiatry & Behavioral Science at Stanford University. She recently returned to UCLA where she directs the Brain Connectivity and Cognition Laboratory and the Center for Cognitive Neuroscience Analysis Core in the Semel Institute for Neuroscience and Human Behavior. Within a cognitive neuroscience framework, Dr. Uddin’s research combines functional and structural neuroimaging to examine the organization of large-scale brain networks supporting the development of social cognition and executive function. Her current projects focus on understanding dynamic brain network interactions underlying cognitive inflexibility in neurodevelopmental conditions such as autism spectrum disorder. Dr. Uddin’s work has been published in the Journal of Neuroscience, JAMA Psychiatry, and Nature Neuroscience.

Title: Towards a universal taxonomy of functional brain networks

Abstract: While the idea that the human brain is composed of multiple large functional networks has been gaining traction in cognitive and network neuroscience, the field has yet to reach consensus on several key issues regarding terminology. What constitutes a functional brain network? Are there “core” functional networks, and if so, what are their spatial topographies? What naming conventions, if universally adopted, will provide the most utility and facilitate communication amongst researchers? Can a taxonomy of functional brain networks be delineated? The Workgroup for HArmonized Taxonomy of NETworks (WHATNET) was formed in 2020 as an Organization for Human Brain Mapping–endorsed best practices committee to provide recommendations on points of consensus, identify open questions, and highlight areas of ongoing debate in the service of moving the field toward standardized reporting of network neuroscience results. The group has developed a Network Correspondence Toolbox (NCT) to permit researchers to examine and report spatial correspondence between their novel neuroimaging results and multiple widely used functional brain atlases. The adoption of the NCT will make it easier for researchers to report their findings in a standardized manner, thus aiding reproducibility and facilitating comparisons between studies to produce interdisciplinary insights.

Watch Dr. Uddin's talk here:

Please note that audio is not available for the first 2 minutes of the video.

Featured Faculty Presentations

Comparing resting state with task-based functional connectivity for connectivity fingerprinting

Abstract: For the last 20 years, resting-state fMRI has been used to reconstruct the network of circuitry throughout the brain, generating a set of relational metrics known as functional connectivity. Task-based fMRI can also be used to estimate functional connectivity, after the effects of the task are removed from the signal. This has recently led numerous researchers to argue that the acquisition of resting-state fMRI may be redundant and a waste of resources if researchers are already collecting task-based fMRI.
Our prior work has demonstrated that an individual’s pattern of functional connectivity can be used to accurately predict their own, unique pattern of neural activity, an approach that we have called Connectivity Fingerprinting. In these studies, we used resting-state fMRI to train and test Connectivity Fingerprinting models. But are these models robust to different sources of functional connectivity?
Here, we compare and contrast Connectivity Fingerprinting models that were trained and tested with resting-state or task-based functional connectivity, across regions involved in high-level vision, language, and spatial working memory. We find that connectivity derived from task data predicts responses just as well as connectivity derived from resting-state data. We observed these findings even when the training and test data are mixed (e.g. models trained with resting state data and applied to task data). In other words, task-based functional connectivity captures fine-grained individual differences in brain responses to a variety of other tasks. This supports the growing argument that the acquisition of resting-state fMRI may be unnecessary when other task-based fMRI is available.

Student Oral Presentations

Uncovering the neural representations of color appearance

Ramith Ambati^1,2, William Narhi-Martinez^3,4, Angela M. Brown¹, Julie D. Golomb³, Delwin T. Lindsey^2,3

¹ College of Optometry, The Ohio State University; ²Graduate Program in Vision Science, The Ohio State University; ³Department of Psychology, The Ohio State University; ⁴Department of Psychology, Yale University.

Introduction: Humans can perceive a rich set of millions of colors, but the neural mechanisms underlying color appearance are not fully understood. One influential theory is Hering’s “color-opponent” theory of color appearance. The theory asserts that color appearance is encoded as unary or binary combinations of 4 elemental chromatic sensations: redness, yellowness, greenness, and blueness, where redness is opposed to greenness, blueness is opposed to yellowness, and combinations of opposed sensations cannot occur. For example, the appearance of “orange” is understood to be a combination of redness and yellowness. Hering proposed that his color-opponent processes originated in the human retina. There is abundant behavioral evidence for Hering’s color-opponent theory, but despite clear evidence of spectral opponency in visual processing as early as the retina, no one has yet found neural responses anywhere in the ascending visual pathways that are consistent with the color-opponent processes predicted by Hering. To further examine this issue, we designed an fMRI study using multi-voxel pattern analysis and Representational Similarity Analysis to investigate different hypothetical formats of visual color representation.

Methods: Participants performed a series of Hering-style hue-scaling tasks in the fMRI scanner. On each trial, participants viewed a 10°, 1.75-sec duration colored rotating spiral on a gray background. The spiral was colored with one of the following 8 hues: each participant’s personal unique red, yellow, green, or blue hues, or one of their personal four binary-composite hues: orange (yellow-red), lime (yellow-green), cyan (blue-green), or purple (blue-red) as determined by a separate psychophysical testing session while in the scanner. During a block of trials, participants performed one of four hue-scaling tasks, during which they rated the amount of redness, yellowness, greenness, or blueness present in the colored spirals by 4-alternative button press (“none, “some”, “mostly”, “all”). Each participant performed each of the four hue-scaling tasks twice (8 blocks per participant). We performed a whole-brain searchlight analysis using Representational Similarity Analysis to compare similarity patterns based on neural activations in cerebral cortex to hypothetical similarity patterns derived from models of representational similarity that were based on stimulus colors (visual hue only) and/or the hue-scaling task (which is expected to produce different representations of a given hue depending on which task participants performed).

Results: Our searchlight analysis revealed cortical areas in and around hV4 where color information was represented according to pure hue similarity, regardless of task. However, representational patterns of hue-scaling were found in dorsal regions of parietal and frontal cortex that were shaped by both hue and task in ways fully compatible with Hering’s theory.

Discussion: These results demonstrate the increasing complexity of color representation when moving from visual areas of lower- to higher-level processing. Moreover, our findings reveal, for the first time, neurophysiological correlates of Hering’s theory of color appearance.

Keywords: color processing, Hering color-opponency, hV4, parietal cortex, frontal cortex, fMRI, searchlight, representational similarity analysis

Altered Cerebellar-Cerebrum Dynamics in Social Gaze Processing: Implications for Schizophrenia and Social Cognitive Deficits

Aravind Kalathil, Aubrey Moe, Scott Blain, Ivy Tso

Ohio State University Department of Psychiatry

Background: People with schizophrenia (SZ) exhibit deficits in social cognition, including abnormal eye gaze processing (GP) associated with top-down inhibition from cortical social-cognition regions. While these cerebral regions have been implicated in GP, recent research has also linked cerebellar-cerebral connections to cognitive functioning. How these cerebellar-cerebral dynamics are altered in SZ is unknown. Here, we utilized dynamic causal modeling (DCM) to investigate how posterior cerebellar activity influences social processing nodes during GP. We hypothesized that there would be bidirectional cerebellar-cerebral connections during GP with cerebellar outputs to cerebral nodes being more upregulated in SZ during explicit GP.

Methods: Two datasets were used to see if bi-directional cerebellar-cerebral dynamics were present and altered in SZ. The primary dataset contained 39 SZ and 33 healthy controls (HC). The replication dataset contained 27 SZ and 22 HC participants. All participants underwent a GP task during BOLD fMRI with explicit and implicit blocks. A GLM contrast of explicit-implicit GP was used to identify four social processing nodes for DCM analysis: Crus II (cerebellum), inferior parietal lobule (IPL), insula, and fusiform. Connectivity of these nodes during explicit & implicit GP and modulation of each connection during explicit gaze discrimination was used to identify the cerebellar parameters that contributed most to explaining GP activity. Three models were also created for structural model comparison: a full model with bidirectional connections, self-connections, and modulated parameters during explicit GP, a "no modulation" model without cerebellar connection modulation during explicit gaze, and a "no connection" model lacking cerebellar-cerebral connections.

Results: Both datasets showed excitatory cerebellar inputs from the insula (95% posterior probability/Pp) and inhibitory cerebellar outputs (95% Pp) to the fusiform in both HC and SZ groups. Cerebellar outputs to the insula and fusiform were consistently upregulated during explicit GP (95% Pp). Inputs from the fusiform were also upregulated during explicit GP (95% Pp). Cerebellar output to IPL was lower in SZ compared to HC in both the primary dataset (95% Pp) and secondary dataset (75% Pp). In both datasets, the full model received very strong evidence (>99.99% Pp) compared to the no modulation in HC. In SZ, the primary dataset’s full model also had strong evidence (97% Pp) and the secondary dataset had positive evidence (74% Pp) relative to the no modulation model.

Discussion: Our model comparison results provide strong evidence that cerebellar connections to cerebral social processing nodes explain GP brain activity well. This is more prevalent in HC compared to SZ showing that cerebellar connectivity alterations may be related to social cognition deficits. Furthermore, the parameter estimates also show that cerebellar connectivity to social processing regions is reduced in SZ and not upregulated to the same extent as HC during explicit gaze. Due to the cerebellum’s physical isolation from other brain structures and its accessibility for noninvasive interventions such as transcranial magnetic stimulation, this may provide a potential therapeutic target for social functioning deficits. Future directions will focus on neuromodulation of the cerebellum to determine causally if limiting cerebellar activity influences social functioning."

Keywords: gaze processing, schizophrenia, cerebellum, dynamic causal modeling

Interoception and the Neural Representation of Self and Other in Adolescence

Jingyi Luo¹, Ruofan Ma¹, Mallory Feldman², Jessica Flannery³, Eva Telzer², & Kristen Lindquist¹

¹The Ohio State University; ²University of North Carolina at Chapel Hill; ³The University of Georgia

Introduction: Having a sense of self relies in part on the ability to represent one’s own internal states as different from others. Self-other representation may thus rely on interoception, which is one’s ability to perceive visceral bodily signals. Adolescence is a time when self-other representation and interoception are both developing. The present study aims to (1) investigate how self-other neural representations are associated with interoception and (2) examine whether perspective taking motivation can further moderate these relationships.

Methods: Adolescents (11 to 14 years old, N = 63) first completed a heart rate discrimination task assessing their interoceptive accuracy and completed the Perspective Taking subscale from the Interpersonal Reactivity Index (IRI) scale. They then performed an in-scanner self-other appraisal task consisting of separate runs for representing one’s own attitudes (self1), their close peers’ attitudes (other), and their own attitudes again after mentalizing their close peers (self2) towards a set of prosocial and risky behaviors. We utilized a multivariate similarity analysis to examine neural similarity between conditions in hubs (i.e., vmPFC, dmPFC, and bilateral anterior insula) within the allostatic-interoceptive network, a brain network linked to generating and representing internal states. Multi-level models were separately constructed to examine Aim 1 and 2.

Results: We found significant effects of interoception on self-other neural similarity across different conditions where less interoceptive accuracy predicted less self1-other neural similarity within the dmPFC (b = -0.008, 95% CI [-0.015, -0.0003], p = .041) and more self1-self2 neural similarity within the vmPFC (b = 0.009, 95% CI [0.001, 0.017], p = .019). This effect of interoception can be further qualified by considering one’s motivation to take others’ perspectives into the model. Specifically, we found significant interaction between interoception and motivation to perspective take on self-other neural similarity within the vmPFC (other-self2: b = -0.003, 95% CI [-0.005, -0.0002], p = .039; self1-self2: b = 0.003, 95% CI [0.001, 0.005], p = .0075). Individuals with poor interoceptive accuracy and low motivation to take the perspective of others have less self-other similarity after mentalizing peers’ attitudes and relatively less stability of self-representation. Having more motivation to take others’ perspectives counteracts the self-instability effect for people with poor interoception.

Discussion: Interoception contributes to individual differences in self and other representations in different contexts. Individuals with poorer interoception and lower motivation to engage in perspective taking may be more likely to adjust their self-representation to be more different from others after thinking of their close peers and thus lower self stability across time. However, those with more motivation to take others’ perspectives can compensate for the unstable self-representation even for people who are bad at interoception. Interestingly, we didn’t find the similar moderation effect of perspective taking motivation on self-other or self-to-self neural similarity among people with relatively high interoceptive accuracy. This finding suggests that interoceptive accuracy may play as a stronger determinant in forming self and other neural representation over other external factors such as perspective taking motivation.

Keywords: Interoception, Self-other Representation, Multivariate Similarity Analysis, Perspective Taking, Adolescence

Age-dependent associations between measures of gait, balance, and white matter integrity in community-dwelling adults

Grace Amadon^1,2, Deb A. Kegelmeyer³, Anne D. Kloos³, Jasmeet P. Hayes^1,2, & Scott M. Hayes^1,2

¹Department of Psychology, The Ohio State University; ²Chronic Brain Injury Program, The Ohio State University; ³Division of Physical Therapy, The Ohio State University

Introduction: Mobility and white matter integrity each show measurable decline with age and are critical for maintaining functional independence. Many studies examine single mobility measures, such as gait speed or balance, in isolation, and often focus on clinical populations. Few studies have examined the relationship between multiple domains of mobility and white matter integrity across the adult lifespan in community-dwelling adults. Understanding these relationships could identify key targets for interventions aimed at preserving both mobility and brain health with age.

Methods: Data from 127 community-dwelling adults (mean age (sd)=54.3 years (17.2), age range: 18-85 years, 58% female) with complete mobility and magnetic resonance imaging (MRI) data from the Fitness, Aging, Stress, and Traumatic Brain Injury Exposure Repository (FASTER; S. Hayes & J. Hayes) were examined. Mobility was assessed using accelerometers and gyroscopes worn during comfortable pace walking and static balance tasks. Spatiotemporal gait characteristics and deviations from center of mass during progressively difficult static balance conditions were measured. White matter integrity was examined using fractional anisotropy (FA) derived from Diffusion Tensor Imaging. Principal component analysis (PCA) was used to reduce multiple mobility measures into data-driven domains. Tract-Based Spatial Statistics (TBSS) was applied to conduct a whole-brain voxel-wise analysis examining associations between mobility domains and FA, as well as Age x Mobility interactions. Post-hoc analyses explored the relationship between mobility and FA in older adults (65+ years).

Results: PCA revealed four distinct domains of mobility: static balance, gait, gait variability, double support variability. Age x Double Support Variability interactions emerged in the inferior longitudinal fasciculus (t=4.25, p<.005), right splenium (t=3.89, p<.005), and left splenium (t=4.46, p<.005), such that among young adults, greater double support variability was associated with lower FA, whereas greater double support variability was associated with higher FA among older adults. There were no associations among middle-aged adults in these regions. Post-hoc analyses among older adults revealed worse static balance was associated with lower FA in the body of the corpus callosum (t=4.51, p<.005), and greater double support variability was associated with lower FA in the left superior corona radiata (t=5.10, p<.005).

Discussion: Greater double support variability was associated with lower FA in young adults but higher FA in older adults in regions that may be associated with the integration of visuospatial, somatosensory, and motor information necessary for maintaining stability. This pattern may be explained by normal age-related trajectories in structural brain health, with posterior tracts like the splenium maturing later and showing differential FA patterns across the lifespan. Results specific to older adults in the present study highlight static balance and double support variability as mobility domains associated with structural integrity in later life, particularly related to interhemispheric communication and motor control. Combined, these results highlight a multidimensional approach to understanding mobility-white matter integrity relationships and point to specific domains that could inform interventions aimed at maintaining functional independence and brain health across the adult lifespan.

Keywords: Aging, Balance, Gait, Magnetic Resonance Imaging, Mobility, White Matter Integrity

Poster Presentations

An examination of the association between neighborhood deprivation, physical activity, and white matter integrity in aging

Olivia Horn, Ann J. Lee, Jasmeet P. Hayes, Scott M. Hayes

Department of Psychology, The Ohio State University; Chronic Brain Injury Program, The Ohio State University

Introduction: Neighborhood deprivation, or socioeconomic disadvantage of a geographic area, has been previously associated with lower brain white matter integrity in middle-aged and older adults. Previous studies have also identified physical activity as a modifiable lifestyle variable that is positively associated with white matter microstructure. However, prior research has often examined independent, rather than cumulative, effects of neighborhood deprivation and physical activity on white matter using self-reported measures of physical activity, which may not accurately represent activity levels. Understanding these associations is essential for public health policy promoting healthy brain aging. The current study addresses gaps in the literature by examining the independent and interactive effects of neighborhood deprivation and objectively measured physical activity on white matter integrity in middle-aged and older adults.

Methods: 101 community-dwelling middle-aged and older adults (ages 38-86 years, mean age=62.1 years, mean education=16.6 years, 91.1% White) were recruited from the Fitness, Aging, Stress, and TBI Exposure Repository (FASTER). Neighborhood deprivation was quantified with the Area Deprivation Index derived from current household ZIP codes. Physical activity was objectively measured with a wrist-worn ActiGraph accelerometer over seven days. White matter integrity was examined using fractional anisotropy metrics derived from diffusion tensor imaging tract-based spatial statistics obtained from magnetic resonance imaging (MRI). Whole-brain voxelwise analysis was conducted to evaluate the independent and interactive effects of neighborhood deprivation and physical activity on white matter integrity using general linear modeling. Post-hoc analyses examined neighborhood deprivation and physical activity in separate models rather than together to explore whether associations with white matter integrity varied by age.

Results: Adjusting for sex and age, higher neighborhood deprivation was associated with decreased fractional anisotropy in two clusters within the inferior longitudinal fasciculus. In contrast, we did not observe a main effect of physical activity or Neighborhood Deprivation × Physical Activity interaction effect. In separate post-hoc models, neighborhood deprivation was negatively associated with fractional anisotropy in the inferior longitudinal fasciculus, uncinate fasciculus, and superior longitudinal fasciculus, and physical activity was positively associated with fractional anisotropy in the medial lemniscus. Neither relationship depended on age.

Discussion: Greater neighborhood deprivation related to lower white matter integrity in several tracts, independent of age or physical activity levels, highlighting the importance of considering the role of contextual disadvantage in brain health. Contrasting prior studies, we did not observe associations between physical activity and white matter integrity, which may reflect the elevated physical activity levels of our sample. Overall, our findings indicate that community-level neighborhood deprivation may have stronger relationships with white matter microstructure than individual-level health behaviors such as physical activity. These results emphasize the relevance of social determinants of health in brain aging during mid-to-late adulthood and provide insight into potential targets for public health policies to support brain health across aging.

Keywords: aging, white matter, neighborhood deprivation, physical activity

Neuropsychiatric Symptoms of Depression and Corpus Callosum Integrity Across Cognitive Diagnostic Stages

Sinae Park, Erica Howard, Alyssa A. Miller, Dr. Jasmeet P. Hayes

Department of Psychology, The Ohio State University; Chronic Brain Injury Initiative, The Ohio State University 

Introduction: Depressive symptoms frequently accompany the onset of Alzheimer’s disease (AD) and are associated with poorer quality of life and greater care burden. Despite their clinical relevance, the neural substrates of depressive symptoms in AD remain poorly understood, particularly which brain regions are involved. Prior work has shown that depressive symptoms across adulthood are associated with reduced white matter integrity in the corpus callosum, suggesting that disruption of interhemispheric communication may play a role. However, it remains unknown whether corpus callosum microstructure also helps explain depressive symptoms that emerge in later life, particularly in the context of concomitant cognitive decline. This study examined whether regional corpus callosum integrity accounts for the association between diagnostic group (cognitively normal [CN], mild cognitive impairment [MCI], or AD) and late-life depressive symptoms, and whether these relationships differ across the AD spectrum neurodegenerative disease.

Methods: This cross-sectional study included 66 AD, 101 MCI patients, and 85 CN participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Participants were included based on availability of baseline Neuropsychiatric Inventory Questionnaire (NPIQ) depression scores, completed by participants’ study partners, and baseline diffusion tensor imaging (DTI) scans. Fractional anisotropy (FA) values were obtained from DTI scans for three sections of the corpus callosum (CC): genu, body, and splenium. Indirect effect models examined whether CC regions accounted for the association between diagnosis (AD, MCI, CN) and depression NPIQ scores, adjusting for age and sex. PROCESS macro for R (Hayes, 2018) was used for these analyses, employing a 95% bootstrap confidence interval based on 10,000 bootstrap replicates to estimate statistical significance. To examine the neural correlates of depression in earlier stages, a separate analysis excluded AD participants to examine these effects among MCI and CN groups only.

Results: Indirect effects models revealed that diagnosis (AD, MCI, CN) had a significant indirect effect on depressive symptoms through the genu of the corpus callosum. Both the indirect effect (ab = 0.0803, 95% CI [0.0039, 0.2200]) and the direct effect (c′ = 0.7373, p = .0009) were significant. After excluding the AD group, analyses indicated that diagnosis (MCI vs. CN) had a significant indirect effect on depressive symptoms though the splenium of the corpus callosum (ab= 0.1769, 95% CI [0.0278, 0.3997]), while the direct effect was nonsignificant (c′ = 0.7373, p = .0899). Other indirect effects models examining the effect of three corpus callosum sections on the relationship between diagnosis and depressive symptoms were not significant.

Discussion: White matter integrity in the genu and splenium of the corpus callosum influenced the relationship between cognitive diagnosis and depressive symptoms. The splenium, which connects posterior sensory and association regions, showed early degeneration in mild cognitive impairment that disrupted posterior networks and indirectly affected mood. The genu, linking prefrontal regions involved in executive control and emotion regulation, played a greater role in depressive symptoms during later disease stages. Longitudinal studies are needed to determine whether progressive callosal changes predict the onset or worsening of depression.

Keywords: Corpus callosum, genu, splenium, Alzheimer's disease, NPI-Q

Structural Limbic Volume as a Mediator of Behavior Problems in Pediatric TBI

Sophia Jeng¹, Elisabeth A. Wilde², William A. Cunningham³, Kathryn Vannatta^1,4, Keith Owen Yeates⁵, Kristen R. Hoskinson^1,4

¹Center for Biobehavioral Health, Abigail Wexner Research Institute at Nationwide Children’s Hospital, Columbus Ohio; ²Department of Neurology, University of Utah School of Medicine, Salt Lake City Utah; ³Department of Psychology, University of Toronto, Ontario Canada; ⁴Department of Pediatrics, The Ohio State University College of Medicine, Columbus Ohio; ⁵Department of Psychology, University of Calgary, Alberta Canada

Introduction: Youth with externalizing behavior problems are both more likely to sustain a TBI compared to youth without externalizing behavior problems, & they are more likely to experience behavior problems post-injury. In other populations (e.g., youth with early traumatic experiences, healthy youth), structural volume of limbic regions have been associated with the scope of behavior problems; in contrast, in youth with TBI, existing literature largely focuses on fronto-limbic connection & functional network communication strength rather than structural volume, & volumetric studies typically focus on more global regions (e.g., total cortical volume) rather than focusing on the limbic regions explicitly. Therefore, we examined the extent to which bilateral limbic volume alone contributes to variability in externalizing symptoms in youth with TBI.

Methods: 19 youth with moderate to severe TBI (msTBI, Mage=11.45, SD=2.73, 13 male), 13 youth with complicated-mild TBI (cmTBI, Mage=11.45, SD=2.73, 9 male), & 26 youth with orthopedic injury (OI, Mage=11.45, SD=2.73, 17 male) underwent MRI in a 3Tesla Siemens scanner including a high-resolution structural sequence. Freesurfer 7.0 quantified volumes of limbic regions (hippocampi, amygdale, cingulate gyrus, parahippocampal gyrus, entorhinal cortex, thalamus, nucleus acumbens) that were then summed to form a bilateral limbic volume composite. Behavior problems were rated by parents on the Child Behavior Checklist, including Anxious/Depressed, Withdrawn/Depressed, Attention, Aggressive, Rule Breaking, Oppositional/Defiant, & Conduct Problems subscales. One-way ANOVAs quantified group differences in behavior problems & bilateral limbic volume, using CBCL raw scores & controlling for age & sex given maturational changes in brain volume during adolescence. Partial correlations, controlling for age & sex, examined links among bilateral limbic structure & behavior problems. Finally, ordinary least squares path analysis (PROCESSv5 Model 4, 5,000 bootstrap re-samples) examined whether limbic volume accounted for significant variance in behavior problems, above & beyond injury group, age, & sex.

Results: We found significant group differences in parent-reported aggressive (Eta2=.13, p=.022, msTBI>OI) & oppositional/defiant behaviors (Eta2=.11, p=.043, msTBI>OI) & differences in attention & conduct problems that approached significance with larger than medium effect sizes (Eta2=.10, p=.068, msTBI>cmTBI=OI; & Eta2=.10, p=.059, msTBI>OI, respectively). Youth with msTBI also had smaller bilateral limbic volume (Eta2=.14, p=.033, msTBI<cmTBI=OI). Limbic volume was significantly associated with attention (r=-.342, p=.015) & oppositional/defiant (r=-.287, p=.044) behavior. Finally, the indirect effect of msTBI on attention problems via reduced bilateral limbic volume was significant (LLCI=0.041, ULCI=2.113) whereas the direct effect of msTBI on attention problems, with limbic volume in the model, was no longer significant (p=.370, LLCI=-0.998, ULCI=3.334). Limbic volume did not mediate the link between msTBI & oppositional/defiant problems, nor did either model show mediation for links with cmTBI.

Discussion: These findings suggest that limbic volume, in addition to limbic structural and functional connectivity found in the literature, may meaningfully explain behavioral morbidities following childhood TBI. Despite modest subgroup sample sizes, these results contribute to a growing literature regarding TBI.

Keywords: Pediatric Neuropsychology. Traumatic Brain Injury

Sex-Based Differences in Network-Level Dedifferentiation in Aging

Quinn K. Meyerson, Xiangrui Li, Madhura Phansikar, Ruchika S. Prakash

Department of Psychology, The Ohio State University; Center for Cognitive and Behavioral Brain Imaging, The Ohio State University

Introduction: Neural dedifferentiation is an age-related process where brain regions become less specialized, and has been associated with declines in cognitive performance. Entropy, a novel measure of neural dedifferentiation, has been linked with advancing age and declines in fluid cognition. However sex-based differences in dedifferentiation remain poorly understood. Recent work found that females showed lower whole-brain entropy at all ages, indicating higher neural specialization, while males showed greater entropy increases with aging, demonstrating poorer aging outcomes. The objective of this study is to characterize network-level sex-based entropy differences in a large, diverse cohort, providing a representative and fine-scale view on dedifferentiation.

Participants and Methods: We examined resting state functional magnetic resonance imaging (fMRI) data from 2,803 adults (Ages 50–92, mean age = 65.01, SD = 8.68, female = 1,745) in the Health & Aging Brain Study–Health Disparities (HABS-HD), a large, community-based study of cognitive aging in a diverse population of older adults. We derived entropy in the 13 networks of the Glasser parcellation, and the whole-brain level. Sex-based differences were examined using a linear regression of the form Entropyᵢⱼ = β₀ + β₁(Sexᵢ) + β₂(AgeCenteredᵢ) + β₃[(Sexᵢ) × (AgeCenteredᵢ)] + εᵢⱼ.False discovery rate (FDR) correction adjusted for multiple comparisons.

Results: Age was positively associated with whole-brain entropy (p = .004, partial η² = 0.008) and entropy in seven networks: frontoparietal, visual1, visual2, ventral-multimodal, posterior-multimodal, orbito-affective, and subcortical (all < .05), with effect sizes ranging from partial η² = 0.003 (visual2) to 0.023 (subcortical). No sex-based differences were found in whole-brain entropy, but six networks showed differences, with female sex associated with higher specialization in the language, auditory, subcortical, orbito-affective, ventral-multimodal, and posterior-multimodal networks (all < .05). Effect sizes ranged from a partial η² = 0.002 (auditory) to 0.011(language). No significant age-by-sex interaction was found.

Discussion: Females demonstrated lower network entropy controlling for age, suggesting greater specialization in networks focused on language, auditory, and reward processing, multimodal integration and subcortical regions. These effects were small, showing that sex-based differences in entropy are subtle and must be interpreted in context. Age was associated with entropy across seven networks and the whole-brain, suggesting the relationship is not uniform. No sex-by-age interactions were observed, indicating age trajectories are not sex-dependent. These network-level analyses provide a nuanced view of sex differences in dedifferentiation, supporting prior findings of lower female entropy while showing these differences are not uniform across networks or racial groups. Further, our findings demonstrate the value of entropy as a flexible measure of specialization across topological scales. Future research should examine the underlying mechanisms behind these sex-based differences, including hormonal influences (e.g., menopause), modifiable risk factors such as lifestyle and health behaviors, and social and structural determinants.

Keywords: aging, sex differences, functional connectivity

Disrupted Emotion Regulation and Reward Network Effective Connectivity Associated with Suicidal and Non-Suicidal Self-Injury in First-Episode Psychosis

Cindy An^1,2, Seema Dhaher¹, Melissa Kilicoglu¹, Aubrey Moe^1,3, Mindy Westlund Schreiner^1,2, Jessica Turner¹

¹Department of Psychiatry and Behavioral Health, The Ohio State University; ²Nationwide Children’s Hospital; ³Department of Psychology, The Ohio State University

Background: Individuals with first-episode psychosis (FEP) have elevated risk for suicide, the leading cause of death in the first five years following diagnosis. Non-suicidal self-injury (NSSI) significantly predicts suicidal behavior yet suicide and NSSI studies often exclude psychosis. We investigated effective connectivity in emotion regulation and reward network regions among participants with lifetime history of NSSI or suicide attempt (SA) with and without FEP.

Methods: Resting-state fMRI data was acquired for 18 individuals with FEP and 18 non-clinical controls (NCC). Group Iterative Multiple Model Estimation (GIMME) developed effective connectivity models using signal from bilateral emotion regulation and reward network regions: middle cingulate cortex (MCC), posterior cingulate cortex (PCC), caudate, putamen, posterior superior temporal gyrus (STG), orbitofrontal cortex (OFC), and insula. Two models grouped participants by diagnosis and NSSI (present[+] vs. absent[-]).

Results: FEP was associated with increased STG and putamen connectivity. NSSI+ (n=5 NCC, 13 FEP) had reduced insula to STG and MCC, reduced PCC and STG to OFC, and increased putamen to STG and MCC connectivity against NSSI- (n=13 NCC, 5 FEP). NSSI was positively correlated with insula activity (p=0.001). SA frequency was positively correlated with superior temporal (p=0.022) and STG to insula activity (p=0.001) and negatively correlated with OFC (p=0.009) and putamen to STG activity (p=0.012).

Conclusion: NSSI in FEP is associated with increased connectivity within emotion regulation regions and disrupted between-emotion regulation and reward network interactions modulated by the STG and striatum. Findings are consistent with existing NSSI and SA literature, potentially supporting using similar strategies to address self-injury within FEP.

Keywords: Non-Suicidal Self-Injury, First-Episode Psychosis, Functional Connectivity

Effective Connectivity in an Impulsivity Network

Grace Hodges and Jessica Turner

Ohio State University Wexner Medical Center

Introduction: Previous meta-analyses identified a network of cortical nodes underlying impulsivity in the Balloon Analog Risk Taking (BART) task. We examined the effective connectivity within this network to assess the flow of information underlying impulsivity.

Methods: Our 3T rs-fMRI scans come from Welsh Advanced Neuroimaging Database (n=170). Impulsivity was measured using the BART number of balloon pumps, which was performed outside the scanner. Seeds for this analysis are the bilateral insula, dorsal ACC, right DLPFC, right putamen, and left caudate. The data were processed using ENIGMA HALFpipe to acquire time series from each seed. We used the lowest and highest quartile of BART mean pumps to select high and low impulsivity groups. After quality control and group selection, the sample included 65 healthy participants (48f), ages 18-63. We modeled effective connectivity using the group iterative multiple model estimation (GIMME). We used both supervised and unsupervised clustering algorithms.

Results: The best-fitting models showed directed simultaneous connectivity from the left caudate to the right putamen, and a separate pathway from the dACC to the right insula, then to the left insula. The high impulsivity group also had a connection from the rDLPFC to the dACC in supervised and unsupervised clustering models. However, the strength of that connection in each participant was not related to impulsivity in a regression analysis.

Conclusion: The impulsivity network identified in meta-analyses of the BART task has been shown in resting state fMRI to relate to impulsivity; however, we find that effective connectivity within this network is not strongly related to impulsivity as measured in the BART. The directed connection from rDLPFC to dACC may be found in the high impulsivity group, but it is not sufficient to identify impulsivity. This analysis used a non-clinical population, and the effects within a clinical sample remain to be examined.

Keywords: resting-state fMRI, impulsivity, effective connectivity

Reorganization of Cortico-Cerebellar Networks Reflects Progression from Familial Risk to Psychotic Illness

Kami Pearson1^,2, Katrina Aberizk², Caroline Demro³, Bryon A. Mueller³, Scott R. Sponheim^3,4, Jessica Turner²

¹Neuroscience Graduate Program, OSU; ²Wexner Medical Center, OSU; ³Department of Psychiatry and Behavioral Sciences, University of Minnesota; ⁴Minneapolis Veterans Affairs Health Care System

Background: Altered effective connectivity (EC) among cortical, subcortical, and cerebellar regions is implicated in psychosis. Familial liability may influence these patterns, but distinctions between risk- and illness-related cortico-cerebellar interactions remain unclear. This study examined EC differences across healthy controls (HC, n=45), individuals with schizophrenia (SZ, n=74), affective psychosis (AP, n=40), and their first-degree relatives (SZrel, n=46; APrel, n=25). We hypothesized that patients would show reorganized cortico-cerebellar connectivity compared to relatives, reflecting a shift from familial vulnerability to disease expression.

Methods: Resting-state fMRI data obtained from the Psychosis Human Connectome Project (488 whole-brain volumes, 3T Siemens Prisma) was preprocessed using ENIGMA HALFpipe. Blood-oxygen level dependent (BOLD) time series from prefrontal, insular, striatal, and cerebellar triple network representations were modeled using group iterative multiple model estimation (GIMME). The following models were analyzed independently: HC, SZrel, APrel, SZ and AP.

Results: Compared to HCs, relatives showed reduced prefrontal connectivity, reversed cerebellar control network (C13) influences on the right ventrolateral prefrontal cortex (vlPFC), and enhanced striatal-insular coupling. SZ uniquely demonstrated right caudate to C13 connectivity, which is absent in HC, AP, and relatives. Similarly, AP exclusively showed cerebellar salience(C8) and default mode (C17) projections to the putamen, insula, and vlPFC. A lagged path from the vlPFC to C13 appeared in both relatives and patients, but not HCs, suggesting transition from familial risk to clinical manifestation.

Conclusions: Cortico-cerebellar connectivity distinguishes familial risk from disease expression in psychosis. Fronto-cerebellar network reorganization may reflect vulnerability. These findings emphasize the cerebellum’s central role in psychosis development.

Keywords: BOLD fMRI, Affective Psychosis, Schizophrenia, Effective Connectivity

Dynamic read-out of spatial remapping across saccades with time-resolved EEG alpha band decoding

¹Tzu-Yao Chiu, ²Zitong Lu, ¹Julie D. Golomb

¹The Ohio State University; ²Massachusetts Institute of Technology

Introduction: Our perceptual experience of the world remains remarkably stable even though saccades induce drastic shifts in retinal input. In order to achieve stability across saccades, a remapping process is required, such that spatial information at the previous retinotopic (eye-centered) location has to be removed and updated to reflect the spatiotopic (world-centered) location. While previous neurophysiological studies found mostly pre-saccadic predictive remapping, behavioral studies on attentional remapping showed mixed evidence for both pre-saccadic remapping of attention and retinotopic attentional trace lingering after saccade completion. Here, we used multivariate decoding of EEG alpha band oscillations to efficiently and more precisely track spatial remapping across the peri-saccadic time window.

Methods: We recorded eye-tracking and EEG data from human subjects performing a spatial memory task, in which they memorized a cued peripheral location during sustained fixation periods (no-saccade trials) or across a saccade. Subjects participated in two sessions: one in which they memorized the cue’s spatiotopic location and another in which they memorized its retinotopic location. We trained linear classifiers to decode the location of the memory cue using alpha band oscillations on no-saccade trials, and then tested the classifiers’ preference timepoint-by-timepoint on saccade trials to track the dynamics of spatial attention in the peri-saccadic time window.

Results: In both tasks, results in the pre-saccadic time window showed robust decoding of the currently attended location with no reliable decoding of the predictive remapped location. Post-saccadically, in the spatiotopic memory task, decoding of the memory cue’s retinotopic location lingered early after the saccade and decoding of the task-relevant spatiotopic location emerged later after saccade completion, whereas results in the retinotopic task showed significant and prolonged decoding of the task-relevant retinotopic location and no spatiotopic decoding. Further analysis showed a significant task difference starting around 120ms after the saccade. Finally, brain-behavior correlation analysis showed that relative decoder preference in the post-saccadic time window correlated with behavioral memory precision, such that subjects with stronger spatiotopic neural decoding had more precise spatiotopic memory reports.

Discussion: Overall, the results provide converging evidence for the gradual remapping of spatial attention after saccades and offer an exciting new approach to study novel aspects of the fine-grain time course of remapping with human non-invasive neural recordings.

Keywords: Spatial attention, Remapping, Visual stability

Cortical Myelin Differences Across the Lifespan: Associations with Age, Sex, and Traumatic Brain Injury

Taweh Hunter, Erica Howard, Nicole Saltiel, Scott M. Hayes, Jasmeet P. Hayes

Department of Psychology, The Ohio State University

Objective: Demyelination has been implicated in cognitive aging, neurodegenerative disease, and pathological protein aggregation (Bartzokis, 2004;Depp et al., 2023). Prior research on brain aging has primarily focused on myelinated fibers within subcortical and deep white matter tracts using diffusion tensor imaging (Sexton et al. 2011). However, this method is less sensitive to shorter myelinated axons that extend to the cortical surface. Cortical myelin is critical for synchronized cognitive functioning (Timmler, 2019), underscoring the importance of examining both deep and cortical myelin. Age, sex, and environmental exposures such as traumatic brain injury (TBI) influence dementia risk and cognitive decline (Levine et al., 2021; Murman etal., 2015; Livingston et al., 2024), yet their associations with intracortical myelin remain poorly understood. Here, we used T1-weighted/T2-weighted(T1w/T2w) ratio mapping, a magnetic resonance imaging (MRI) method proposed as a proxy measure of cortical myelin, to examine demographic and clinical variables known to be associated with advanced cognitive aging and risk for dementia.

Participants and Methods: Eighty-two adults (48.7%female; age range: 18 – 86 years; mean age = 50.6, years) from the Fitness, Aging, Stress, and TBI Exposure Repository (FASTER) study at The Ohio State University completed a structural MRI scan on a 3T Siemens Prisma scanner. Participant-level cortical myelin was quantified via T1w/T2wprocessing using the Human Connectome Project pipeline (Glasser et al., 2011), and groupwise whole-brain T1w/T2w ratios were obtained via summation of sulcal-gyral parcellation values (Destrieux et al., 2011). TBI history was assessed using the Boston Assessment of TBI-Lifetime interview, and 42.6% of participants endorsed previous TBI exposure. A linear regression model examined the main and interaction effects of sex and age on cortical myelin content, covarying for an imaging quality assurance metric. Exploratory Wilcoxon rank-sum tests compared cortical myelin content between TBI and control participants, disaggregated by sex.

Results: The regression model revealed significant main effects of age (p < .0001) and sex (p < .0001) on global cortical myelin content. Specifically, older age was associated with lower cortical myelination, and male sex was associated with lower cortical myelin content. We did not observe a significant age-by-sex interaction, suggesting that females have higher myelin content through adulthood. Exploratory TBI analyses revealed that males with a history of TBI had lower myelin levels compared to females with a TBI history.

Conclusions: The present study provides evidence of significant age and sex differences in cortical myelin using T1w/T2w mapping. In the presence of TBI, older males may be particularly vulnerable to cortical demyelination, while females may exhibit greater resilience. To our knowledge, this is the first report of sex differences in cortical myelination as well as a possible sex-specific vulnerability to TBI-related demyelination. Our findings support the growing consensus that sex plays a significant role in brain aging and injury response. This work emphasizes the importance of incorporating cortical myelin measures into models of neurodegenerative risk and cognitive aging.

Keywords: Neuroimaging, T1w/T2w, Brain Aging, Sex Differences

Affective Variability and Dynamic Brain Network Functional Connectivity Patterns During Induced Affect

Yuritza Y. Escalante¹, Ruofan Ma¹, Gretchen E. Wulfekuhle², Taylor West^2,3, Barbara Fredrickson², Jessica R. Cohen², Kristen A. Lindquist¹

¹Ohio State University; ²University of North Carolina at Chapel Hill; ³University of Sussex

Introduction: People experience a range of pleasant and unpleasant affect states across days and weeks defined as affective variability, or the average deviation from mean levels of affect (i.e, standard deviation) across a week (Dejonckheere et al., 2019). We increasingly understand how brain network connectivity is linked to affective experiences, but we understand less about how dynamics in brain connectivity across time might be linked to dynamics in affective experiences. The purpose of this study is to examine links between daily affective variability and dynamic functional connectivity of the brain’s intrinsic networks during an affect induction.

Methods: Eighty participants (aged 18-69) completed 7 days of daily diaries, and on the 8th day they completed a positive and negative affect induction during fMRI. The fMRI data was parcellated into the canonical default mode network (DMN), dorsal attention network (DAN), fronto-parietal network (FPN), salience network (SN), and ventral attention network (VAN; using the Seitzman et al., 2020 functional atlas). We computed both static functional connectivity (sFC; average functional connectivity across the task) and dynamic functional connectivity (dFC; variability in functional connectivity across the task) within each network as within-module degree (WD). sFC and dFC of each network’s connections with the rest of the brain were computed as participation coefficients (PC). Linear regression was used to investigate the extent to which static and dynamic WD and PC are associated with affective dynamics in daily life.

Results: We found that individuals who experienced greater affective variability in the week before an fMRI scan had more dynamic functional connectivity (dFC; variability in functional connectivity across time) between the DMN, SN, FPN, and DAN and the rest of the brain (i.e., higher PC). However, these patterns were not observed when examining sFC. These findings suggest that variability in affect is particularly associated with dFC within the observed networks, and is not a product of sFC.

Discussion: One hypothesis for these results is that the psychological mechanisms associated with the activation of the specified brain networks are more involved during negative versus positive affect. Moreover, negative affect is thought to signal a need to change one’s circumstances (Clore & Huntsinger, 2009), and thus may require greater integration of networks with the rest of the brain. Collectively, these findings underscore the importance of understanding dynamics across time when examining affect, examining variation along the bipolar axis of affect, and the inclusion of an affect induction task to capture the dFC patterns associated with affect.

Keywords: Affective Variability, Experience Sampling Methods, Dynamic Functional Connectivity, Brain Networks

Vermal cerebellum connectivity differences between 3 Tesla and 7 Telsa fMRI within subclinical anxiety and healthy adults

Zachary Brodnick^1,2, Jessica Turner²

¹Neuroscience Graduate Program, The Ohio State University; ²Wexner Medical Center, The Ohio State University

Background: The cerebellar vermis is linked to emotional processing. We examined whether its connectivity reflects anxiety levels, and if 7T fMRI is more sensitive than 3T in the same individuals.

Methods: The present study included healthy adults aged 18–63 years from the Welsh Advanced Neuroimaging Database (n = 104). All participants completed 3T, 7T fMRI scans, and the Generalized Anxiety Disorder-7 (GAD-7) survey. Participants were considered to be sub-clinically anxious if they reported a GAD-7 total score ≥5. We tested differences in healthy and subclinical anxious participants via seed-to-voxel functional connectivity maps of the posterior vermis (VI, VII, VIII, and IX) using a general linear model that included age and sex, in both 3T and 7T datasets. Paired t-tests on vermal seed connectivity maps assessed differences between 7T and 3T scans across healthy and sub-clinically anxious groups.

Results: Distinct vermal seeds showed reliable differences in their connectivity maps, and significant differences were observed for 7T>3T scanner-related differences in the healthy group (pFWE<0.05, cluster extent ≥ 20 voxels). Regions of heightened connectivity in 7T across vermal seeds included the thalamus, caudate, cingulum, insula, and supplemental motor area. Within the anxiety subgroup analyses, only localized cerebellar regions showed heightened 7T connectivity, and no significant between group differences were shown.

Conclusions: Resting-state static connectivity was not related to GAD7 scores in this analysis. However, the 7T fMRI may be more sensitive to cerebello-cortical and subcortical connectivity, particularly in vermal connectivity to the emotional processing and salience network regions, which is promising for studies of anxiety disorders.

Keywords: rs-fMRI, Cerebellar Vermis, 7T fMRI, Anxiety

Task-derived Functional Connectivity is Sufficient to Capture Individual Differences in Functional Brain Region Activations

Isaac Liao, Zeynep M. Saygin, David E. Osher

The Ohio State University, Department of Psychology

Introduction: Most studies of brain networks use resting-state functional connectivity (FC) to characterize the brain’s intrinsic functional organization. However, recent work has questioned whether we should rely exclusively on resting-state data to define this organization (Finn, 2021). Here, we tested whether FC estimated from residualized task fMRI data can predict task responses in visual, language, and multiple-demand regions at the individual subject level. While previous studies have demonstrated that responses of visual regions can be predicted from resting-state FC at the level of individual subjects (Saygin et al., 2012; Osher et al., 2016; Molloy et al., 2024, Tavor et al., 2016, Osher et al., 2019), it is unclear whether residual connectivity from task fMRI of a completely different domain (e.g. vision) can a) predict responses to higher-level cognitive tasks like working memory and linguistic processing; and b) if it can do so in a selective manner (i.e. for fROIs within each network).

Methods: We collected structural MRI, resting-state fMRI, and task fMRI from 35 participants (plus an additional 15 for out-of-sample comparisons). Task scans included a dynamic visual localizer, an auditory language localizer, and a spatial working memory task. Responses of functional ROIs (fROIs) were estimated using standard contrasts from the three tasks (visual category > others, sentences > nonsense, hard > easy). Resting-state FC was computed between each fROI vertex and Glasser-atlas regions. Linear models were trained to predict fROI responses from connectivity, and task-derived FC were then applied to these models.

Results: Models generalized well across FC types, with resting-state FC performing comparably to task FC in nearly all cases. When tested on 13 out-of-sample participants, model accuracy remained high and was consistently higher than within-sample accuracies, indicating minimal overfitting. Task-derived connectivity is as effective as resting-state connectivity for predicting task-evoked fROIs across multiple brain regions and cognitive domains. This was observed even for disparate domains (e.g., for visual task-derived FC predicting auditory language fROIs and the laterality therein). Task data, regardless of the specific task performed, are sufficient to capture fine-grained individual differences in brain network architecture.

Discussion: Previous studies have shown that resting-state FC can be used to predict fine-grained individual differences in brain responses at the individual level across a wide range of tasks (Hiersche et al., 2025; Cole et al., 2016; Tavor et al., 2016; Molloy et al., 2024; Ngo et al., 2022; Zheng et al. 2022). Our study extends these findings to task-derived FC and an fROI-based approach that offers higher precision and fine-grained information. Our findings have several implications. First, task data alone can be used to predict networks of activity without the need of resting-state acquisition. Second, by residualizing task effects, we can use the same dataset both to independently define fROIs and to estimate fROIs for other mental domains, thereby obviating the need for separate fMRI localizer experiments. Finally, the approach allows researchers to extract information about task responses of interest even in datasets that lack the corresponding task. Together, these findings broaden the utility of task fMRI for investigating the functional organization of brain networks.

Keywords: Functional Connectivity, Task fMRI, resting state

Neural representations of human and AI agents during mentalizing

Dan Zhu, Jingping Yang, Dylan Wagner

Department of Psychology, The Ohio State University

Introduction: With the rapid development of artificial intelligence (AI), people are beginning to form social bonds with AI chatbots that mimic human-to-human relationships. Previous research has demonstrated heightened neural activation in the medial prefrontal cortex (mPFC) when participants interacted with humans compared to computer-based agents. However, this prior work occurred before the advent of Large Language Models (LLMs). Does interacting with modern LLMs that are able to present a simulation of human-like conversation and even engage in emotional prosody (i.e., ChatGPT advanced voice mode) increase the likelihood that people will attribute a mind to AI agents? In the present fMRI study, we investigated whether people are more likely to mentalize when interacting with AI partners introduced through voice or text-based conversations in comparison to a simple computer program and whether these patterns resemble interactions with human partners.

Methods: Twenty-eight right-handed adults between the ages of 18 to 35 participated in this study. First, participants engaged in a task designed to increase intimacy between people (i.e., the Fast Friend Task). Participants became familiar with three partners respectively: a human partner, a virtual partner using ChatGPT advanced voice mode, and another virtual partner using ChatGPT text mode. During the subsequent fMRI session, participants thought they were playing a rock-paper-scissors game with each partner and with a random computer program. In reality, all four conditions involved random computer opponents. Following the scan, participants completed measures assessing attitudes and experience with AI chatbots.

Results: Analysis of the rock-paper-scissors game revealed that participants showed greater neural activity in the mPFC, a key region involved in mentalizing, when they believed they were playing against human compared to all other categories (AI voice, AI text, random computer program). There was no difference between these other three categories. Multivariate pattern analyses further supported the univariate results by showing that the neural representation of voice and basic modes resembled computer rather than human. Finally, we tested if any individual differences were associated with mPFC neural activity and found no evidence that attitudes towards AI or prior experience modulated the response during the rock-paper-scissors game.

Discussion: Our study shows that despite the impressive capabilities of modern LLM-based chatbots, individuals continue to perceive them as mindless tools. Even after engaging in a ten-minute conversation with AI partners, participants nevertheless only showed mentalizing-related brain activity in the mPFC for human partners, even though in our task all conditions were random. Analysis of several individual difference measures related to mentalizing or to people’s attitudes towards AI showed no association with brain activity during the rock-paper-scissors task, suggesting that positive attitudes or a general tendency to anthropomorphize do not necessarily translate into a higher likelihood of imbuing AI agents with a mind. Together, these findings suggest that AI chatbots do not evoke mentalizing process in early interactions.

Keywords: AI chatbots, mentalizing, mPFC, fMRI

Connectivity-based predictions of putative language regions in the fetal and infant brain

Kelly J. Hiersche^1,2,3, Xinnan Wang^1,2,3, Anna Quatrale^1,2,3, Journie Dickerson^3,4, Rachael F. Holt^3,4, Zeynep M. Saygin^1,2,3

¹Department of Psychology, The Ohio State University; ² Center for Cognitive and Behavioral Brain Imaging, The Ohio State University; ³Center for Cognitive and Brain Sciences, The Ohio State University ⁴Department of Speech and Hearing Science, The Ohio State University

Introduction: Both adults and young children (by age 3) have domain-specific, left lateralized language networks. However, given that exposure to speech begins in utero and language skills rapidly emerge during infancy, the brain likely shows specialization much earlier or is at least set up to support language processing. One mechanism supporting brain specialization is functional connectivity. Previously, this relationship has been leveraged to predict idiosyncratic task-based activation (‘connectivity fingerprint’, CF modeling) in both children and adults. Given an individual’s functional connectome is stable across infancy, and that resting-state networks are present in-utero, it may be possible to leverage the relationship between connectivity and function to predict putative language activation very early in development. Therefore, in this study, we ask: 1. Does the fetal or infant brain show speech selectivity? 2. Can we predict speech activation in the fetal and infant brain using mature CF fingerprints that support language processing?

Methods: We collected resting state data and a language localizer in 32 fetuses (30-38 weeks gestation), as well as 24 infants (ages 5-7 months old), as part of two on-going longitudinal studies. Approximately 20 minutes of resting state data were collected for each fetal subject, and a single run of speech localizer (mother read vs hum conditions). Data was preprocessed according to the following: masked from the maternal compartment, motion corrected, combined across echoes, registered to a template space, and denoised. Voxel-to-parcel functional connectivity matrices were calculated, and general linear models were used to identify speech activation within subjects. In infants, we collected resting state data (5-15 min), and a language localizer (meaningful language vs texturized sound), and preprocessed following traditional steps. Probabilistic atlases were created to determine if the fetal or infant brain showed consistent (i.e. overlapping) speech activation across participants. CF models were ℓ2 regularized linear ridge regression models trained in adults (N=34) and children (N=58) who were scanned on a similar language fMRI task and who had resting state data. We applied the final connectivity fingerprint model from either the child or adult group to each of the fetal and infant participants’ connectivity to predict their own voxelwise language activation in inferior frontal and superior temporal regions (canonical language network).

Results: Task-based fMRI results showed bilateral activation to speech in temporal regions in the infants; remarkably, we also found fetal activation to speech in the left superior temporal cortex. We found that child CF models were able to predict fetal speech activation within temporal but not frontal regions. When applied to infant connectivity, the child model was able to predict both frontal and temporal activation. Adult CF models did not accurately predict fetal or infant activation, suggesting that fetal and early postnatal connectivity patterns for language may be like those in young children but not in adults.

Discussion: We observe language activation in infants as young as 5 months and remarkably also in the fetal temporal cortex, demonstrating a scaffold for high-level cognition and the prenatal emergence of language processing. Connectivity fingerprints may drive the emergence of this functional organization, but these fingerprints are not static and seem to change through development.

Keywords: language, development, fMRI, fetal, infant, connectivity fingerprint

A Multivariate Analysis of Visual and Auditory Language Responses in the Proto-VWFA of Pre-Reading Children

Lauren Rydel^1,2,3, Kelly J. Hiersche^1,2,3, Zeynep M. Saygin^1,2,3

¹Department of Psychology, The Ohio State University; ²Center for Cognitive and Behavioral Brain Imaging, The Ohio State University; ³Center for Cognitive and Brain Sciences, The Ohio State University

Introduction: The visual word form area (VWFA) is unique among ventral visual areas in that it shows protracted development, emerging only after a child can read. Prior work demonstrated connectivity to high-level language regions informs the eventual location of the VWFA, but the functional profile of the proto-VWFA is still unclear. Univariate word selectivity may emerge from initial multivariate representations that differentiate between visual words and other visual categories (e.g., faces, objects) or between linguistic vs. non-linguistic auditory stimuli. In this study, we investigate whether pre-reading children exhibit multivariate responses to words or auditory language in the VWFA. Additionally, we examine the similarity between the visual and auditory response profiles of readers’ and prereaders’ VWFA and whether a multivariate response precedes the univariate region.

Methods: We scanned 111 children ages 3-12 years (24 longitudinal, including prereaders who transitioned into readers) on two runs of a high-level visual fMRI task and a subset on two runs of an auditory language task. We used functional regions of interest, multivariate representation analyses, and representational similarity analyses to understand the functional profile of emerging VWFA.

Results: We find that while pre-readers lack univariate word responses, they do show multivariate representations for written words. Representational similarity analyses showed that activation to words is dissimilar from other stimulus categories in the pre-VWFA and that this dissimilarity is almost identical to that of the VWFA in readers. The proto-VWFA in prereaders did not show multivariate (or univariate) representations for auditory language, much like results in readers. Finally, our longitudinal data supports these results, showing the proto-VWFA holds a multivariate representation for words, but not for auditory language, before the univariate VWFA emerges.

Discussion: Overall, our results suggest that multivariate representations may scaffold univariate selectivity of the VWFA as a child learns to read; interestingly, these do not seem to emerge from auditory representations to language, suggesting that any responses of this region to auditory language in older children or adults are likely a result of experience and challenge the theory that the proto-VWFA is part of the distributed language network.

Keywords: ventral temporal cortex, visual word form area, multi-voxel pattern analysis, multivariate response, representational similarity analysis, pre-readers, auditory language

Voxel-Based Morphometry of 3T vs 7T: Divergent Grey Matter Density Estimates and Their Correlation with Extraversion

Jack Wegner, Zachary Brodnick, Jessica Turner

Wexner Medical Center at The Ohio State University

Introduction: Dimensions of personality such as neuroticism have been correlated with decreased grey matter density (GMD) while traits like extraversion were correlated with increased GMD in the medial orbitofrontal cortex and decreased GMD in the amygdala. However, there have been no studies that have analyzed personality traits with differing levels of magnetic resonance strength. Most studies use 3 Tesla Magnetic Resonance Imaging (3T MRI) due to their cost and prevalence as opposed to 7T MRI. The aim of this study is to investigate the difference between 3T and 7T in how they detect GMD as well as their sensitivity to relationships with personality traits, using extraversion as an example.

Methods: This study used 107 healthy subjects (aged 18-63) from the Welsh Advanced Neuroimaging Database (WAND). Extraversion was assessed using the mean subscores of each subject from The Big Five Inventory-2 Short Form (BFI-2-S). Preprocessing and 2nd-level analyses through Voxel-Based Morphometry (VBM) were performed using Statistical Parametric Mapping (SPM) and Computational Anatomy Toolbox (CAT12).

Results: When contrasting 3T over 7T, increased GMD was observed in the right inferior temporal gyrus, the right palladium, both sides of cerebellar VIIB and the left side of VIII. When contrasting 7T over 3T, there were significant differences in GMD across the entire brain. With an uncorrected significance level (p<0.001), the most prevalent area of positively correlated grey matter density and extraversion between both 3T and 7T was in the opercular part of the right inferior frontal gyrus.

Discussion: Since it’s impossible for the GMD of subjects to be changing between scans, there must be an overlying difference in the way MRI scans are processed into usable files between 3T and 7T. The positive correlation of GMD and extraversion in the right pars opercularis aligns with the region’s moderate role in attention and social cognition.

Keywords: 3T, 7T, VBM, Extraversion, Grey Matter Density

The VWFA as a Neurochemical and Functional Bridge Between Language and Visual Cortex

Laura Bradley & Zeynep Saygin

Department of Psychology

Introduction: The visual word form area (VWFA) sits at a unique intersection of high-level visual and language systems, yet the biological factors that anchor it to these systems remain poorly understood. A central question is whether the VWFA is more similar in its underlying organization to its visual neighborhood or by its connectivity to language network. Functional and structural brain asymmetries support uniquely human abilities like language, and connectivity patterns in children can predict later VWFA location, suggesting that both local structure and long-range connections may constrain its emergence.

Methods: Here, we combine neurotransmitter density profiles and functional connectivity to characterize the architecture of the VWFA. To address this, we examined category-selectivity regions in the ventral temporal cortex (VTC), including VWFA, face-selectivity regions (FFA, OFA), and object-selective regions (LO, PFS), alongside frontal and temporal language regions. Neurochemical architecture was quantified from mean PET-derived receptor density maps (1,238 adults, 20 neurotransmitters/transporters). We applied hierarchical clustering and representational similarity analysis to characterize relationships among neurotransmitter profiles across ROIs.

Results: Hierarchical clustering and RSA revealed that the frontal language regions are most distinct from temporal language and visual regions; face rois cluster with each other, object rois with each other. The VWFA is neurochemically most similar to temporal language regions.

Discussion: The distinct neurochemical profile of the VWFA from other visual regions and its similarity to temporal language regions suggest that neurochemical affinity may be the physical mechanism by which privileged connectivity between VWFA and temporal language cortex is instantiated in development.

Keywords: Neurotransmitter, Language, Vision

Theory of Mind Task Performance in Critical Congenital Heart Disease

Declan E. Alford, Aaron McAllister, May Ling Mah, Kathryn Vanatta, Kristen R. Hoskinson

Abigail Wexner Research Institute at Nationwide Children's Hopsital, Nationwide Children's Hospital, The Ohio State University College of Medicine

Introduction: Moderate to complex congenital heart disease (mcCHD) can impact the development of the central nervous system, reflected in deficits in neurocognitive abilities. This could be due to changes in brain development, and such is linked with poor physical health and underperformance. One skill vulnerable to neurocognitive deficit is theory of mind: an individual’s ability to understand other individuals by ascribing mental states to them. Theory of mind deficits threaten children’s success in establishing and maintaining social relationships, which can have adverse effects on their personal lives, especially as social nuance becomes more complicated during adolescence. Although extensive research has been conducted on attention and executive functioning deficits in this population, little has been done specifically examining theory of mind. Therefore, this study aims to find the link between mcCHD and theory of mind in children and adolescents.

Methods: The present study includes 20 youth with mcCHD (Mage = 11.78, Male = 12) and 20 healthy controls (HC: Mage = 12.27, Male = 15). All participants completed the Coding and Symbol Search subtests of the Weschler Intelligence Scale for Children-Fifth Edition (WISC-V) to measure processing speed, and all youth with mcCHD and 12 HC also completed the Vocabulary and Symbol Search subtests of the Wechsler Abbreviated Scale of Intelligence-Second Edition (WASI-II) as a screener for overall cognitive function - Theory of mind was assessed using the Jack and Jill paradigm. Group differences were quantified via independent samples T-tests or chi square analyses, as appropriate, and correlations quantified using bivariate Pearson statistics. Given the pilot sample size, both statistical significance and effect size were examined.

Results: We found significant differences in theory of mind performance between the CHD and the HC groups (t = .002, d = 1.022), as well as poorer performance on the Symbol Search subtest (t = .027, d = 0.631) and overall Processing Speed Index (t = .045, d = .657) on the WISC-V. Theory of mind performance, but not performance on control conditions of the Jack and Jill paradigm, were strongly associated with processing speed (rs = RANGE).

Discussion: Our results suggest that core neurocognitive deficits found in youth with mcCHD extend to aspects of social cognition as well, specifically theory of mind. There is existing literature that discusses the link between multifocal white matter pathology (a typical neuroanatomical finding in mcCHD) and processing speed, and it is possible that impacted theory of mind is a concerning downstream functional implication. If a child with mcCHD cannot accurately assess theory of mind under a time constraint and adequately interpret the feelings and perspectives of their peer group, psychosocial consequence may ensue, thus increasing the social strain and difficulties this population faces.

Keywords: Congenital Heart Disease, Theory of Mind

An examination of age-related changes in the relationships between cardiorespiratory fitness, working memory, and frontoparietal network efficiency: Preliminary results from the FASTER study

Jillian H. Graham, Jessica A. Cloud, Jasmeet P. Hayes, Scott M. Hayes

Department of Psychology, The Ohio State University; Chronic Brain Injury Program, The Ohio State University

Introduction: Brain function and working memory performance decline with age, yet individual trajectories vary. This includes neural network dynamics at rest, such as efficiency, which measures how effective systems are at communicating and how resistant that communication is to failure. Modifiable lifestyle variables, such as cardiorespiratory fitness (CRF), may attenuate age-related decline. Although prior studies have investigated these relationships, they focus on older adult samples, leading to a lack of understanding of how these relationships may vary across the adult lifespan. Therefore, to address the gap in the literature, the current study investigates the relationships between CRF, working memory, and frontoparietal network efficiency across the adult lifespan.

Methods: Adults (n=123, mean age=51.0 years; ages 18-81) were selected from the Fitness, Aging, Stress, and TBI Exposure Repository (FASTER; S. Hayes & J. Hayes). CRF (VO2peak) was assessed via graded maximal exercise testing on a cycle ergometer. T1-weighted and resting-state functional magnetic resonance scans were collected using a Siemens 3T Prisma scanner with a 32-channel head coil. Resting-state network efficiency was calculated as global and local efficiency in the frontoparietal network (FPN). Global efficiency measures the capacity for parallel information processing over an entire network, whereas local efficiency measures how effectively information is processed within localized subnetworks. The Wechsler Adult Intelligence Scale-Fourth Edition (WAIS-IV) Digit Span sequencing subtest was used to assess working memory. Hierarchical linear regression models were used to assess the moderating effect of age on the relationships between cardiorespiratory fitness, working memory, and resting-state network efficiency. Model 1 assessed the effect of age and CRF on FPN efficiency, and Model 2 assessed the effect of age and FPN efficiency on working memory.

Results: In Model 1, main effects of age, VO2peak, and an Age X VO2peak interaction were observed on FPN global and local efficiency. The main effect of age was negative, whereas the main effect of VO2peak was positive for local efficiency and negative for global efficiency. The Age X VO2peak interaction revealed that the relationship between VO2peak and both local and global efficiency became more negative with increasing age. In Model 2, FPN global and local efficiency were negatively associated with working memory. An Age X Efficiency (both global and local) interaction was observed, such that the relationship between FPN efficiency and working memory became more negative with increasing age.

Discussion: These findings indicate that within the FPN, the relationships between cardiorespiratory fitness, working memory, and network efficiency depend on age. Future research should use a larger sample size, which may further increase power and sensitivity to detect additional significant associations. Understanding how CRF influences age-related changes in resting-state network efficiency will provide insight into optimal neural network integrity that may benefit cognitive maintenance in aging.

Keywords: CRF; working memory; resting-state network efficiency

Examining the relationship between hippocampal perfusion and pattern separation performance: Preliminary findings from the FASTER study 

Justin M. Palmer^1,2, Jessica Cloud^1,2, Jasmeet P. Hayes^1,2, and Scott M. Hayes^1,2

¹Department of Psychology, The Ohio State University; ²Chronic Brain Injury Program, The Ohio State University

Introduction: Aging is associated with declines in cerebral perfusion, which can reduce the necessary nutrients delivered to brain tissue that support cognition. Some studies have suggested that alterations to the neurovasculature can be an early marker of Alzheimer’s disease progression. However, previous literature among cognitively normal older adults is mixed, with some studies indicating that higher perfusion is predictive of better cognitive performance, while others suggest the opposite. Many of these studies focus on whole-brain perfusion or flow within large arterial territories, which may lack the specificity to assess areas most vulnerable to cognitive changes in aging. The hippocampus is highly susceptible to alterations in perfusion, yet few studies have directly examined the relationship between hippocampal perfusion and memory performance using mnemonic discrimination tasks, which are particularly sensitive to hippocampal integrity.

Methods: Cross-sectional data from the Fitness, Aging, Stress, and TBI Exposure Repository (FASTER) were used for the current analysis. Cognitively normal adults (n=209; mean age=51 (SD=18); 124 females) completed the continuous recognition version of the Mnemonic Similarity Task. Participants identified whether objects presented sequentially were “old,” “new,” or “similar” compared to previously presented objects. Lure discrimination index was calculated as the rate of “similar” responses to lure objects of various levels of difficulty (easy, medium, or hard), corrected for false alarms. Participants also underwent brain magnetic resonance imaging (MRI), which included pseudo-continuous arterial spin labeling to quantify cerebral blood flow. Perfusion images were processed with Bayesian-Inference for Arterial Spin Labeling toolbox (BASIL) and FreeSurfer was used for volumetric segmentation. Multiple linear regression was used to analyze the relationship between hippocampal perfusion and age, covarying for sex, education, history of moderate/severe TBI, and presence of a current major depressive episode. Lure discrimination index was analyzed using repeated measures ANOVA that included difficulty level as the repeated measure and hippocampal perfusion and age as the between-subjects variables, covarying as described above, in addition to total cerebral blood flow. Two separate models assessing left and right hippocampal perfusion were conducted.

Results: The regression model assessing age and hippocampal perfusion explained 24% of the variance (F(5,203)=13.03, p<0.001). Hippocampal perfusion decreased with older ages (b=-0.20, SE=0.04, t=-4.5, p<0.001) and was lower among males (b=-10.21, SE=1.54, t=-6.64, p<0.001). Repeated measures ANOVA indicated that the lure discrimination index decreased with older ages (F(2,201)=90.40, p<0.001) and with increasing levels of difficulty (F(2,414)=10.01, p<0.001). Additionally, better lure discrimination index was marginally predicted by greater levels of perfusion within the left hippocampus (F(2,201)=2.83, p=0.09). Perfusion within the right hippocampus did not predict lure discrimination index.

Discussion: Our results suggested that hippocampal perfusion was lowest among older adults and among males. Preliminary results indicated a potential laterality effect: greater perfusion in the left hippocampus predicted better memory performance. Future work will incorporate genetic risk for Alzheimer's disease, which may moderate these relationships.

Keywords: perfusion; hippocampus; pattern separation; aging

Indicating pediatric contact sport participation using ABCD functional connectivity data

Nii-Ayi Aryeetey¹, Ginger Yang², Jaclyn Caccese², Zeynep M. Saygin²

¹The Ohio State University, Columbus, Ohio; ²Nationwide Children’s Hospital, Columbus, Ohio

Introduction: What are the changes in functional brain organization, if any, that occur after a child incurs repetitive head impacts due to participation in contact sports? Our previous work has shown that after only one season of tackle football, children show higher levels of effortful working memory activation as compared to their pre-season scan, and as compared to non-contact and non-athlete controls (who were also scanned longitudinally), but do not show improvements in working memory performance, indicating increased effort on the task in the absence of a behavioral benefit.

Methods: Here we leverage the ABCD database to identify any differences in the multiple demand network (involved in effortful tasks) and other networks during rest in a large sample of children who have or have not participated in contact sports, and identify any confounding factors that may bias these group differences. We will use these results to constrain analyses in our own data where we have a baseline scan (before the initial season of contact sports) and can infer that changes in neural organization are more likely caused by head impacts during the season. In the ABCD data, we selected children who had functional connectivity data and data on whether they played a contact sport. This resulted in a cohort of 5,820 boys aged 8 to 11 years.

Results: A general linear model predicting contact sport participation showed modest yet significant accuracy (deviance p < 0.001). Many of the significant predictors included connectivity between dorsal attention, ventral attention, and frontoparietal networks.

Discussion: These significant predictors will be used as networks of interest in ongoing analyses of our in-lab youth tackle football rsFC data to pinpoint longitudinal functional connectivity differences between youth tackle football athletes and non-contact controls.

Keywords: ABCD, resting state, functional connectivity, sports, children

Representation of 3D visual space: Population receptive field modeling of positions in depth

Stephanie Kristensen, Julie Golomb

Ohio State University

Introduction: Accurate perception of positions in depth is essential for human behavior such as grasping. During natural vision, binocular disparity (the slight image difference between left and right eyes) is an important cue to perceive depth. Neural selectivity for binocular disparity is distributed across the visual cortex in regions as early as V1, though some prior studies using fMRI suggest a transition from 2D to 3D processing along the visual hierarchy with more specialized processing of 3D depth perception in later visual areas. Our goal is to describe how neural populations across human cortex respond to changes in perceived depth and whether there exists a systematic organization of depth-related neuronal tuning, similar to 2D retinotopic maps.

Methods: We performed an fMRI experiment during which participants (N=8) passively viewed a random dot motion stereogram while wearing red/green anaglyph glasses. The stimulus depicts a full-field plane of dots located at varying positions-in-depth relative to fixation. Within a session, the plane moved forward or backward through 13 depth positions ranging from –18 arcmin (perceptually near, in front of fixation) to 18 arcmin (perceptually far, behind fixation). We adapted standard population receptive field (pRF) modeling and used a 1D Gaussian to identify the preferred depth and tuning width (range of preferred disparities) for each voxel. We also incorporated a parameter estimating each voxel’s sensitivity to depth changes vs depth-invariant visual stimulation.

Results: The results revealed that voxels in early visual cortex tend to exhibit primarily depth-invariant visual preferences, while some later visual areas exhibit voxel-specific tuning to position-in-depth. In particular, a cluster along the transverse occipital sulcus (overlapping with V3a/b and IPS0) shows a preference for higher binocular disparities, with spatially separated voxels preferring negative or positive disparities. Another cluster near visual region LO shows consistent preferences for near-zero disparities.

Discussion: Overall, this study indicates we can decode positions in depth from later visual areas using population receptive field modeling.

Keywords: 3D space, depth, binocular disparity, fMRI, pRF

Will I Remember You? Anxiety, Memory and Frontal Theta

Hollen Knoell, Blythe Karras, Dr. Andrew Engell

Center for Biobehavioral Health, Abigail Wexner Research Institute, Nationwide Children’s Hospital; Kenyon College Psychology Department

Introduction: Anxiety disorders are associated with disruptions to cognitive function. For example, source memory (i.e., recollection) is negatively impacted in episodic memory tasks. Given the previously established relationship between frontal theta and episodic memory encoding in the neurotypical brain, we hypothesized that the impact on source memory performance was due to anxiety induced differences in frontal theta power. We therefore predicted that higher trait anxiety would show a decrease in source memory performance associated with a decrease in frontal theta during encoding. We investigated this in both the high (6-8 Hz) and low (4-6 Hz) theta bands.

Method: Participants (N = 97) completed an episodic memory task for novel emotional faces while we recorded continuous EEG from a 72-channel cap. During encoding, participants viewed 96 faces and used a 7-point Likert scale to rate each face’s dominance or trustworthiness. During retrieval, participants viewed a randomized series of 144 faces: the 96 faces displayed during encoding and 48 novel faces. Participants were asked to indicate whether each face was “old” or “new” (i.e., recognition). If they reported that a face was old, they were then asked to indicate which characteristic–dominance or trustworthiness–they evaluated for that face (i.e., recollection). Finally, participants completed the State Trait Anxiety Inventory.

Results: Contrary to our prediction, trait anxiety was not negatively correlated with source memory performance and we did not observe a subsequent source memory effect in frontal theta power. We further explored the data with a more liberal statistical approach that did not correct for multiple comparisons (ps ≦ .05, uncorrected). This approach revealed that anxious individuals showed a significant increase in medial frontal high-theta from 0-400ms during source memory encoding, regardless of subsequent recollection performance. There were also several effects observed in the 800–1200ms window. Across all participants, subsequent misses induced significantly more medial frontal high-theta than hits. Medial frontal low-theta induced by subsequent misses was correlated with trait anxiety. Finally, trait anxiety was negatively associated with the difference between hits and misses, with more anxiety inducing more low-theta for misses than hits. Despite the liberal statistical approach, we still did not observe a correlation between trait anxiety and source memory performance.

Discussion: The results of our planned analyses do not support our hypothesis. However, using more liberal statistical thresholds, we found that medial frontal high-theta was more prominent during encoding of subsequent misses than hits, which contradicts prior findings. The more liberal approach also found that high anxiety was correlated with an overall increase in medial frontal high-theta during source memory encoding. This is inconsistent with our hypothesis that decreased frontal theta during encoding mediates poorer memory performance in individuals with high anxiety. This unexpected pattern of theta and memory performance results are consistent with attentional control theory (ACT). ACT posits that anxiety causes disruptions in processing efficiency rather than overall performance. From this perspective, the greater medial frontal high-theta during encoding for higher trait anxiety might reflect greater use of cognitive resources in order to achieve similar levels of performance.

Keywords: Frontal Theta, Anxiety, Episodic Memory

Research Day - 2025 Program

Research Day - 2025 Program