Travel Awards

One of the most rewarding and unique features of the symposium is that we provide travel awards to domestic and international students to come to UW-Madison and learn with us. These scholarships go to students at all levels of their education including: undergraduates, grad students, PhD students, post-doctoral fellows, MD students and residents training in psychiatry.

This year, the HealthEmotions Research Institute supported the expenses of 12 trainees from the US and around the world to come to Madison and participate in the symposium. This is a wonderful opportunity for students at all levels to interact with world class scientists, meet UW-Madison faculty, and forge connections with others conducting research in the broad field of affective neuroscience.

Congratulations to our Travel Award Winners!

2026 Travel Award Recipients

This is an accordion element with a series of buttons that open and close related content panels.

Jacob Faibishenko, BS | "Bridging Space and Time in Relational Memory: Behavioral and ERP Evidence of Enhanced Spatiotemporal Integration by Emotion"

Name: Jacob Faibishenko, BS
University: University of Illinois at Urbana-Champaign
Department: Psychology and Beckman Institute for Advanced Science & Technology
Program or Lab:
Dolcos SCoPE Neuroscience Lab
Mentor or PI: Dr. Florin Dolcos and Dr. Sandra Dolcos

Poster Title: Bridging Space and Time in Relational Memory: Behavioral and ERP Evidence of Enhanced Spatiotemporal Integration by Emotion”

Prior research has primarily investigated the effects of emotion on spatial and temporal associations separately. Here, we jointly examined the influence of negative arousal on both types of associations to provide a more comprehensive understanding of emotion’s impact on episodic relational memory (RM). During encoding, participants (N=133) viewed a series of images involving dissociable foreground and background components and were instructed to make temporal associations between consecutive images. Subsequently, participants’ RM was tested using three retrieval tasks assessing: (1) subjective RM (recollection-based), (2) spatial objective RM (foreground-background matches), along with the link between subjective and objective RM accuracy (Double-Hit), and (3) temporal RM (sequential associations) and accuracy in the spatiotemporal associations (Triple-Hit). First, replicating and extending earlier evidence (Bogdan et al., 2024), emotion enhanced subjective recollection confirmed by objective spatial RM (Double-Hit). Second, we also present novel evidence that emotion enhances spatiotemporal integration – i.e., emotion enhanced subjective and spatial objective RM, while also upregulating associations with subsequent neutral information (Triple-Hit). Finally, preliminary analyses of ERP data (N=39) revealed patterns of centroparietal positivity linked to emotional memory encoding, suggesting that the identified behavioral patterns may be linked to attentional mechanisms. Overall, these findings promote a broader understanding of emotion’s impact on episodic memory and lay the foundation for developing interventions aimed at improving memory for contextual associations in healthy individuals and reducing memory decontextualization in populations affected by PTSD and age-related cognitive decline.

Eugenia Giampetruzzi, BA | "Resting State Amygdala-vmPFC Connectivity Moderates Negative Affective Reactivity to Daily Stress in Adolescence"

Name: Eugenia Giampetruzzi, BA
University: Standford University
Department: Department of Psychology
Program or Lab:
Stanford Neurodevelopment Affect and Psychopathology Lab
Mentor or PI: Ian H. Gotlib

Poster Title: “Resting State Amygdala-vmPFC Connectivity Moderates Negative Affective Reactivity to Daily Stress in Adolescence”

Background: While adolescence is characterized by normative increases in affective reactivity, anomalous increases in negative affect (NA) experienced in response to stressful events can increase adolescents’ risk for psychopathology. Neural models have implicated a functional imbalance between subcortical affect-generating and prefrontal cortical brain regions in contributing to maladaptive affective functioning. We do not know, however, whether these laboratory-based models translate to real-world behavior and, more specifically, whether adolescents’ momentary affective reactivity to daily stress is moderated by their fronto-amygdala resting-state functional connectivity (rsFC).
Methods: Adolescents (N=124; Mage=16.15 years) completed a 14-day ecological momentary assessment protocol assessing levels of stress severity, NA, and positive affect (PA). Given preliminary findings linking stress severity with NA but not PA, preregistered mixed-effects models tested in a subsample of participants with imaging data (N=79) whether rsFC between the amygdala and nodes of the pre-frontal cortex moderated within-person associations between stress and NA. Likelihood ratio tests compared interaction models across four seed-by-system sets.
Results: Left amygdala connectivity to a regulatory set of regions including the dorsolateral, dorsomedial, ventrolateral, and ventromedial (vmPFC) prefrontal cortex moderated the within-person association between daily stress severity and NA. Edge-level analyses localized this effect to the left amygdala-vmPFC pathway, such that stronger connectivity was associated with a flatter stress-NA slope.
Conclusions: Individual differences in fronto-amygdala connectivity, particularly left amygdala-vmPFC coupling, may help to explain the heterogeneity in adolescents’ affective responses to everyday stress. Findings support the ecological validity of fronto-limbic imbalance models by linking resting-state circuitry to real-world stress reactivity.

Byeol Kim Lux, MS | "The Neural Cost of Witnessing Suffering: Network Disruption and Sustained Arousal After Vicarious Trauma"

Name: Byeol Kim Lux, MS
University: Dartmouth College
Department: Psychological and Brain Sciences
Program or Lab:
Cognitive Affective Neuroscience Lab
Mentor or PI: Tor Wager

Poster Title: “The Neural Cost of Witnessing Suffering: Network Disruption and Sustained Arousal After Vicarious Trauma”

Vicarious trauma (VT), witnessing others’ suffering, has been linked to moral injury, PTSD, and chronic disease, yet its neurobiological mechanism in healthy humans remains undefined. Existing laboratory stressors lack the moral salience and ecological validity of trauma exposure. In this study, participants (N = 87) completed two fMRI sessions in which they viewed either a 30-minute documentary depicting industrial animal slaughter (VT) or a matched film showing positive human-animal interactions (Vicarious Community, VC). Following movie-viewing, they underwent resting-state fMRI. Whole-brain network maps were derived using template-constrained independent component analysis, and functional connectivity was compared across conditions. Principal component analysis of in-scanner emotion ratings identified two dominant dimensions: valence and empathic engagement. Empathic engagement during VT was positively associated with Personal Distress (Interpersonal Reactivity Index). We found significant functional connectivity changes between conditions (FDR q < 0.05). In VT, the sensorimotor network (SMN) showed reduced positive connectivity with the precuneus (t = -4.60), and the central executive network (CEN) showed weakened negative connectivity with posterior insula and SMN in VT (t > 4.43). Importantly, the magnitude of these connectivity changes correlated with individual differences in personal distress (|rs| > 0.25; ps < 0.05). VT also produced sustained increases in post-movie heart rate, and greater heart rate differences were associated with larger reductions in SMN-CEN connectivity in VT (r = 0.28, p < 0.05). Together, these findings suggest that VT disrupts integration among interoceptive, sensorimotor, and executive control systems, linking network reconfiguration to autonomic arousal and individual differences in empathic distress.

Katherine Kim, ScB | "Neural Responses to Error Processing Predict Treatment Outcomes in Youth with Anxiety"

Name: Katherine Kim, ScB
University: National Institutes of Health (NIH)
Department: National Institute of Mental Health (NIMH)
Program or Lab:
Neuroscience and Novel Therapeutics Unit (NNT)
Mentor or PI: Melissa A. Brotman, PhD

Poster Title: “Neural Responses to Error Processing Predict Treatment Outcomes in Youth with Anxiety”

Background: Inhibitory control, the ability to regulate attention and behavior in a goal-directed manner, is often impaired in youth with anxiety disorders. Cognitive behavioral therapy (CBT) is effective; however, many youths fail to fully respond, highlighting the need to identify neural markers that predict treatment outcomes. This study examined whether baseline neural activation during error processing predicts anxiety severity post-treatment.
Methods: Fifty-one youth with anxiety disorders (Mage = 14.01, SD = 2.55) completed a modified flanker task during fMRI prior to 12 sessions of CBT. Incongruent trials required participants to respond to the target arrow surrounded by conflicting arrows. Treatment response was assessed using the clinician-administered Pediatric Anxiety Rating Scale (PARS). Whole brain analyses examined the association between baseline error-related neural activation during error vs correct incongruent trials and post-treatment PARS, controlling for pre-treatment PARS, age, and number of errors made. Results were thresholded at p<.005, and cluster size=56 at 𝛼=.05.
Results: Higher post-treatment PARS scores were associated with increased neural activation during incongruent error trials in the right cuneus (b= 0.34, t(93)= 2.47, p=.015) and right superior frontal gyrus (b=0.44, t(93)=3.38, p=.001). Higher post-treatment PARS scores were associated with decreased neural activation during incongruent correct trials in the left middle cingulate cortex (b=0.32, t(93)=2.56, p=.012).
Conclusions: Findings suggest that baseline neural activation in error processing regions seems to be sensitive to anxiety severity but in task-specific conditions, indicating that neural responses to errors made compared to correct responses differentially modulate the treatment-driven changes in anxiety.

Fangze Li, BA | "Hippocampal Stem and Granule Cells in Affective Behavior and Cognition"

Name: Fangze Li, BA
University: Northwestern University
Department: Neurology
Program or Lab:
Interdepartmental Neuroscience PhD Program
Mentor or PI: Dr. John Kessler

Poster Title: “Hippocampal Stem and Granule Cells in Affective Behavior and Cognition”

The hippocampal dentate gyrus (DG) plays a critical role in emotion regulation, learning, and memory, and DG granule cells (GCs) dysfunction contributes to many neurological and psychiatric disorders. DG GCs form a heterogeneous population that differs in maturation stage and functional properties. A common feature across these populations is the expression of bone morphogenetic protein (BMP) receptors. While BMP signaling is well known for its essential roles in early development and in regulating adult hippocampal neurogenesis, its effects on different existing GC populations and their behavioral consequences remain poorly understood.
Previously, we demonstrated that BMP signaling levels throughout the DG correlate with hippocampus-dependent behavior: elevated BMP signaling impairs, whereas reduced signaling improves hippocampal function. Building on this work, we found that activation of newborn GCs suppresses depression- and anxiety-like behaviors, and many classes of antidepressants act through this mechanism. One such example is ketamine. We found that ketamine’s rapid and sustained antidepressant effects arise through distinct mechanisms. A single subanesthetic dose of ketamine rapidly improves affective behavior by increasing activity of existing immature neurons in the DG without altering neurogenesis. In contrast, repeated low-dose ketamine administration produces longer-lasting behavioral effects that require decreased BMP signaling, leading to increased numbers of immature GCs.
Current work investigates how cell-autonomous BMP signaling in different GC populations regulates the structural and functional properties, influencing both affective and cognitive behaviors. Together, these studies will elucidate cell-type-specific mechanisms underlying hippocampus-dependent behavior and highlight BMP signaling as a potential therapeutic target for neurological and psychiatric disorders.

Matthew Mattoni, MA | "Positive Affect and the Neural Reward Response: An Intensive Longitudinal fMRI Study"

Name: Matthew Mattoni, MA
University: Temple University
Department: Psychology and Neuroscience
Program or Lab:
Clinical Psychology
Mentor or PI: Thomas Olino, PhD

Poster Title: “Positive Affect and the Neural Reward Response: An Intensive Longitudinal fMRI Study”

Background: At a between-person level, blunted reward responses implicated in lower positive mood and several psychological disorders. However, between-person comparisons are limited by low test-retest reliability of task-based responses and an inability to explain within-person processes (i.e., non-ergodicity). Within-person studies are needed to address distinct clinical questions, such as how changes in reward-related brain functioning relate to the change in mood (e.g., during treatment). To address this need, we collected intensively sampled neuroimaging data to examine state and trait properties of the neural reward response and within-person associations with mood.
Methods: We openly released a precision imaging dataset with 12 sessions each for four participants, acquired twice weekly. Participants completed multiple reward-related tasks, mood and alertness ratings, and a behavioral mood induction. We examined the internal consistency and retest reliability of the neural reward response, as well as associations with positive mood using multilevel models.
Results: Test-retest-reliability of the reward response was very low (0-.36), suggesting limited statistical power for between-person comparisons. At an intraindividual level, mood and alertness explained 15%-37% of the within-person variance of the anticipatory reward response across individuals.
Discussion: Results suggest that the neural reward response has limited statistical power for between-person comparisons. In contrast, our intensively sampled design showed that intraindividual variation in the reward response was moderately explained by state-like mood and alertness. Overall, results indicate with reward-related neuroimaging should substantially increase attention to within-individual study. We also highlight the need for intensively sampled neuroimaging designs to understand the neural mechanisms of positive mood and related psychopathology.

Defne Mull, MS | "Modulation of Emotional Relational Memory by Neurostimulation: Evidence from a Multimodal Investigation using HD-tDCS and 7T fMRI"

Name: Defne Mull, MS
University: University of Illinois, Urbana-Champaign
Department: Psychology Department, Beckman Institute for Advanced Science & Technology
Program or Lab:
Dolcos SCOPE Neuroscience Lab, Cognitive Neuroscience PhD Program
Mentor or PI: Dr. Florin Dolcos

Poster Title: “Modulation of Emotional Relational Memory by Neurostimulation: Evidence from a Multimodal Investigation using HD-tDCS and 7T fMRI”

Emotion reliably enhances item memory, but its impact on memory for the associated context (or relational memory, RM) remains less clear. Neural evidence suggests that interactions between the amygdala and hippocampus support emotional RM enhancement, while top-down control signals from the ventrolateral prefrontal cortex (vlPFC) modulate this interaction. The present study investigated whether neurostimulation of the left vlPFC during encoding influences emotional RM, using anodal (excitatory) or cathodal (inhibitory) high-definition transcranial direct current stimulation (HD-tDCS). Participants completed an encoding task in which attention was directed toward emotional or neutral foreground items or neutral background contexts, followed by a retrieval task assessing subjective (recollection-based) and objective (item-context matching) RM. Preliminary behavioral results (N = 19) showed that attention modulated the experienced emotion: directing attention to the background reduced the emotional ratings of negative images. Memory results revealed that recollection-based hit rates were overall higher for emotional stimuli, particularly during the foreground attentional focus. Stimulation was associated with a pattern consistent with a directional modulation of this effect: cathodal stimulation attenuated the emotional enhancement, consistent with an inhibitory modulation, but there were no effects of anodal stimulation compared to sham. Objective RM results showed a similar pattern, consistent with an inhibitory modulation by cathodal stimulation. These results will be discussed in conjunction with data obtained from combining this neurostimulation protocol with concurrent ultra-high-field (7T) fMRI, to examine contributions of the amygdala, hippocampus, and vlPFC, at nuclear-, subfield-, and laminar-level resolution, respectively.

Bernard Mulvey, PhD | "Spatial Transcriptomic Profiling of Human Hypothalamic Sex Differences: Ventromedial and Arcuate Nuclei"

Name: Bernard Mulvey, PhD
University: Johns Hopkins University/Lieber Institute for Brain Development
Department: Psychiatry
Program or Lab:
Postdoctoral Fellow
Mentor or PI: Keri Martinowich, PhD

Poster Title: “Spatial Transcriptomic Profiling of Human Hypothalamic Sex Differences: Ventromedial and Arcuate Nuclei”

The tuberal hypothalamus, which includes the ventromedial hypothalamic (VMH) and arcuate (ARC) nuclei, has been characterized extensively in rodents for their roles in controlling social and appetitive behavior as well as metabolism. Several functions the ARC and VMH have been shown to be sex-differential at molecular, physiologic, and behavioral levels in rodents, largely mediated through neuron populations expressing gonadal hormone receptors including estrogen receptor Esr1. To address the lack of modern transcriptomic data from metabolically healthy and reproductive-age adult hypothalamus of both sexes, we performed spatial transcriptomics (10x Visium platform) on postmortem tuberal hypothalamus in 8 healthy human donors ages 27-48 years old. We used our transcriptome-wide results to select 100 genes with specific, sex-differential, or spatially variable gene expression in ARC/VMH for single-molecule profiling with in-situ sequencing using the 10x Xenium platform, verifying distinctive marker genes in these human brain areas. The two datasets revealed retinoid pathway gene expression specific to ARC, including of retinoid metabolism gene CYP26A1—which is not expressed in the adult mouse ARC. Using cell-resolution data from Xenium, we localized sex-differential expression identified with Visium to specific neuron populations of VMH and ARC. Further, we found that genes more highly expressed in male VMH are enriched in autism spectrum disorder (ASD)-associated genes. We thus identified major spatial divisions of VMH and ARC, transcriptomic sex differences in these regions and their resident cell types, and novel human-specific features marking these cells and areas, creating a useful resource for studying hypothalamus in the context of neuropsychiatric disease.

Niki Sabetfakhri, MS | "Static and Dynamic Salience Network Alterations in Veterans with Co-Occurring Mild Traumatic Brain Injury and Alcohol Use Disorder"

Name: Niki Sabetfakhri, MS
University: University of Illinois Chicago College of Medicine
Department: Psychiatry
Program or Lab:
Medical Scientist Training Program, Graduate Program for Neuroscience
Mentor or PI: Dr. Olusola Ajilore

Poster Title: “Static and Dynamic Salience Network Alterations in Veterans with Co-Occurring Mild Traumatic Brain Injury and Alcohol Use Disorder”

Mild TBI (mTBI) and alcohol-use disorder (AUD) commonly co-occur in Veterans, a comorbidity marked by treatment-resistance. Both conditions disrupt the salience network (SN), which coordinates transitions between default-mode (DMN) and control network (CN). Alterations in this mechanism may lock individuals into maladaptive states that impair cognitive flexibility and sustain craving, but how mTBI+AUD jointly alters SN dynamics and relates to craving remains poorly understood.

Forty-four Veterans (22 mTBI+AUD; 22 matched controls) completed resting-state fMRI (20min, TR=555ms) and post-scan craving assessments. Static connectivity for six a priori SN edges per hemisphere (FDR-corrected) revealed reduced left CN-SN coupling (t=2.90, p=.025) and elevated right DMN-SN coupling (t=−2.92, p=.025) in the mTBI+AUD group. Higher craving across all participants correlated with elevated right DMN-SN coupling (r=.33, p=.037). These edges were recast as state-space axes, with θ (inverse tangent of CN-SN to DMN-SN coupling) summarizing salience balance. mTBI+AUD was shifted toward DMN dominance (t=2.98, p=.005), and decreased θ (increased DMN-SN coupling) correlated with craving (r=−.35, p=.021).

We next tracked CN-SN versus DMN-SN balance moment-to-moment using sliding-window connectivity (45s overlapping, ~1,000/subject). mTBI+AUD spent less time in CN-dominant configurations (p=.004; replication in second acquisition: p=.001) and transitioned between CN- and DMN-SN less frequently (p=.030; replication: p=.034). Phase-randomization null modeling confirmed reduced transitioning exceeded what static differences alone predict. Dynamic metrics did not associate with craving.

These findings suggest mTBI+AUD alters SN coupling at two levels— a static shift toward DMN dominance that tracks with craving and reduced dynamic transitioning between network configurations— offering potential targets for circuit-based intervention.

Grace Schamber, BS | "Ovarian Hormone Modulation of Prefrontal Fear Encoding Across the Lifespan"

Name: Grace Schamber, BS
University: Marquette University
Department: Biomedical Sciences
Program or Lab:
Neuroscience-Gilmartin Lab
Mentor or PI: Dr. Marieke Gilmartin

Poster Title: “Ovarian Hormone Modulation of Prefrontal Fear Encoding Across the Lifespan”

PTSD has a strong sex bias, yet the neurobiological basis of this difference remains unclear. We have identified the medial prefrontal cortex as a potential site where sex hormones modify learning circuits. In both rodents and humans, the prefrontal cortex is required for associating threat-related cues and shocks when the two events are separated in time as in trace fear conditioning (Gilmartin et al., 2010, 2013; Knight et al., 2004). Our lab has shown that neural activity in the prelimbic (PL) is necessary for trace conditioning (Gilmartin et al, 2013). Recently, we found that ovarian hormone state during learning influences fear memory strength and memory-related neural activity in the PL (Kirry et al, 2019; LaViola, in prep). As mPFC dysfunction contributes to disorders involving impaired cognitive control and fear regulation, including PTSD, understanding how hormonal state and age-related hormone loss shape encoding in this region is critical. We investigated how hormonal state and aging influence PL neural encoding during trace fear conditioning. Using viral expression of pan-neuronal GCaMP8f in PL neurons, we performed calcium imaging during fear learning. PL activity revealed distinct neuronal activation patterns in females trained in high versus low estrous states. To examine interactions between hormonal state and aging, additional experiments were conducted across multiple age groups and analyses are ongoing characterizing activity profiles across aging timepoints. Together, these studies will define how hormonal state and aging shape prefrontal neural representations of threat learning and may identify neural activity patterns that predict divergent fear outcomes.

Patlapa Sompolpong, BA | "Advancing TMS for PTSD - A Neuroimaging-guided Approach to Modulate the Threat Circuitry"

Name: Patlapa Sompolpong, BA
University: Emory University
Department: Neuroscience
Program or Lab:
Stress and Neuromodulation Lab
Mentor or PI: Dr. Sanne van Rooij

Poster Title: “Advancing TMS for PTSD – A Neuroimaging-guided Approach to Modulate the Threat Circuitry”

Background: Resting-state functional magnetic imaging (fMRI) is now being used to guide targets for transcranial magnetic stimulation (TMS), but many methodological, neurobiological, and clinical questions remain unanswered. Given that amygdala hyperreactivity is a main driver of posttraumatic stress disorder (PTSD), we developed a neuroimaging-guided approach to attenuate the amygdala in a TMS clinical trial for PTSD. We investigated target topography variability and stability to explore the potential clinical relevance of individualized TMS in PTSD.

Methods: 50 adults with PTSD symptoms were randomized to a 10-day treatment of 1Hz (36,000 pulses/day, n=26) or sham (n=24) TMS. Neuroimaging data were collected a week before and after TMS. The pre-TMS resting state scan was used to define targets within the right dorsolateral prefrontal cortex (rDLPFC) with strongest positive FC to right amygdala. Between-subjects target variability was assessed using a repeated measures general linear model (GLM). Effect of TMS on target topography was assessed in a 3*2*2 GLM with post-hoc analyses. Results: We found significant variability in target topography (F(1,45)=4,626.43 p<0.001, ηp2=0.990) and an interaction effect of treatment on target change over time such that change was only significant with active (F(1,21)=4.60, p=0.044, ηp2=0.180) but not sham TMS.

Conclusions: Variability in target topography supports the use of patient-specific TMS. Furthermore, TMS-induced changes in topography suggest circuit-guided TMS may improve PTSD through altering FC in the threat circuitry.

Haley West, MS | "Functional Connectivity Varies with Menstrual Hormones, Anxiety and Stress"

Name: Haley West, MS
University: University of Illinois Urbana-Champaign
Department: Psychology
Program or Lab:
CANOPY Lab
Mentor or PI: Wendy Heller

Poster Title: “Functional Connectivity Varies with Menstrual Hormones, Anxiety, and Stress”

The effects of menstrual hormones on internalizing symptoms are complex and understudied. These hormones vary substantially across a typical 28-day cycle. In addition, there are distinct neural and chemical relationships that indicate that estrogen and progesterone can have an effect on internalizing disorders, such as anxiety, and their supporting neural mechanisms. One example of these relationships is the pathway of cortisol production from available progesterone. The Human Connectome Project is one of the few large neuroimaging datasets that include menstrual cycle data. Using HCP data, the present project examined functional connectivity during a working memory task. Comparing hormonal birth control users vs. those with an active menstrual cycle, connectivity between right amygdala and right hippocampus varied by reported anxious arousal (main effect) and group (interaction): higher anxious arousal was related to higher connectivity, especially in the group without hormonal birth control. Perceived stress and group produced a similar interaction. In both cases, higher distress was related to higher connectivity. We further examined the active cycling group by comparing participants in luteal vs. follicular phases. Higher stress was related to higher connectivity in the luteal but not the follicular group. These findings suggest a role for menstrual hormones in an active cycle on psychological distress and identify potential neural mechanisms such as functional connectivity. Results also suggest a role for the more variable hormone levels of the active cycle in contrast to the more homogeneous levels of hormones produced by hormonal contraceptives.