Transcranial Focused Ultrasound Targeting the Amygdala May Increase Psychophysiological and Subjective Negative Emotional Reactivity in Healthy Older Adults

Background The amygdala is highly implicated in an array of psychiatric disorders but is not accessible using currently available noninvasive neuromodulatory techniques. Low-intensity transcranial focused ultrasound (TFUS) is a neuromodulatory technique that has the capability of reaching subcortical regions noninvasively. Methods We studied healthy older adult participants (N = 21, ages 48–79 years) who received TFUS targeting the right amygdala and left entorhinal cortex (active control region) using a 2-visit within-participant crossover design. Before and after TFUS, behavioral measures were collected via the State-Trait Anxiety Inventory and an emotional reactivity and regulation task utilizing neutral and negatively valenced images from the International Affective Picture System. Heart rate and self-reported emotional valence and arousal were measured during the emotional reactivity and regulation task to investigate subjective and physiological responses to the task. Results Significant increases in both self-reported arousal in response to negative images and heart rate during emotional reactivity and regulation task intertrial intervals were observed when TFUS targeted the amygdala; these changes were not evident when the entorhinal cortex was targeted. No significant changes were found for state anxiety, self-reported valence to the negative images, cardiac response to the negative images, or emotion regulation. Conclusions The results of this study provide preliminary evidence that a single session of TFUS targeting the amygdala may alter psychophysiological and subjective emotional responses, indicating some potential for future neuropsychiatric applications. However, more work on TFUS parameters and targeting optimization is necessary to determine how to elicit changes in a more clinically advantageous way.


Supplemental Methods 5.1. Criteria
Participants prescribed beta-blockers were included contingent on skipping their medication for their study visits.
These screenings also included an MRI eligibility screening to ensure that participants were safe to be MRI scanned.
Following written consent at the first visit, the MoCA was administered to ensure cognitive eligibility with a cut-off score of 19 per HCPA protocol (1,2).

TFUS 5.2.1. TFUS Administration
The brain region targeted during each study session was randomized and counterbalanced across participants such that 43% received amygdala TFUS during the first study session and 57% received entorhinal cortex TFUS during their first session.Examples of amygdala and entorhinal cortex TFUS targeting are provided in Figure 2, as well as a chart detailing sample demographics in Table 1.

TFUS Target Selection
The right amygdala was chosen as the target for this study as it is believed to be involved in the processing of negatively valenced emotions to a greater degree than the left amygdala (3).This made it the most logical target for this project, given the goal of modulating anxiety and emotional reactivity to negative stimuli.
The entorhinal cortex was chosen as an active control region as it allowed the study to investigate the regional specificity of TFUS targeting and brain region engagement due to its close proximity to the amygdala.The left entorhinal cortex was chosen in particular in order to limit the likelihood of impacting the active region when trying to target the control region.Furthermore, a separate aim of this project (not reported in this manuscript) was to gauge the ability of TFUS to the left entorhinal cortex to impact learning and memory, and the left entorhinal cortex is thought to play a disproportionate role in memory formation (4).The inclusion of the left entorhinal cortex in this manner thereby enabled the investigation of both the impact of amygdala TFUS on anxiety and emotional reactivity with the entorhinal cortex as an active control, and the investigation of the impact of entorhinal cortex TFUS on learning and memory with the amygdala as an active control region.
Although the entorhinal cortex is functionally connected to the amygdala and this connection plays a role in the emotional enhancement of memory, both in healthy and pathological (e.g.intrusive recollections in PTSD) contexts, their primary structural and functional networks are distinct.As such, these brain regions are implicated in generally distinct neurologic and psychiatric syndromes (e.g.anxiety disorders vs amnestic syndromes), and demonstrating the selective utility of tFUS targeting these two regions forms the foundation for further exploration of tFUS of these two distinct regions as a potential clinical treatment for distinct clinical phenomena.

Behavioral & Psychophysiological Measures Pre-TFUS and Post-TFUS
Before and after each TFUS session, participants completed an emotional reactivity and regulation task (ERRT) utilizing neutral and negatively valenced images from the International Affective Pictures Set (IAPS), and selfreported emotional valence and arousal were measured in response to the stimuli.Heart rate was measured during the ERRT to investigate psychophysiological responses.

State-Trait Anxiety Inventory
The State-Trait Anxiety Inventory (STAI) was collected before and after each study visit to determine overall anxiety levels from before to after TFUS (Fig S1).Form Y-1 of the STAI was the most updated form available upon the beginning of this study and was used both before and after TFUS in both study visits.The STAI was completed by participants using self-ratings for the questions assessing symptoms of anxiety (Fig S1 ) and is broken up into two components: STAI-State (STAI-S) and STAI-Trait (STAI-T), where the STAI-S assesses anxiety symptoms at the time of administration, whereas STAI-T refers to overall "general" tendencies attributed to the longer term personality trait of personality over longer periods of time.
Given that STAI-Trait scores characterizes anxiety as a stable personality trait and the pre and post-TFUS assessments measured were approximately 3 hours apart, STAI-T scores were unlikely to shift.Thus, STAI-T scores were omitted from the main report.Scores for the STAI-S and STAI-T are summed up for a general score, but the subscores for STAI-S were reported in the main text only.For the four sessions (pre/post Amygdala/Entorhinal cortex TFUS), ERRT form administration was also counterbalanced.The tasks were administered using EPrime 2.0.

ERRT Physical Set Up
Before starting the ERRT, ECG electrodes were positioned on cleaned and prepped skin of the upper right of the chest two inches below the collarbone and one inch below the bottom of the left outer ribcage on participants.ECG data was collected using the BioPac Systems Inc.BioNomadix Amplifier connected to the BioPac USB-TTL Trigger Box, AcqKnowledge v4.2 software, and EPrime 2.0.The ERRT task was administered using EPrime 2.0.Upon stimulus presentation, a trigger from the ECG via the TTL include a marker in the ECG recording in AcqKnowledge.

ERRT Administration
During stimulus onset, a mark was sent from the BioPac MP500 was recorded in the AcqKnowledge v4.2 software recording ECG data throughout the entire ERRT session.

ERRT Instructions
VIEW and WATCH trials were functionally identical in that both instructed subjects to simply attend to the image.However, we chose to use different cues for these two conditions in order to equate the four trial types in the level of anticipatory knowledge subjects had about the upcoming picture.These trials were intended to elicit unregulated forms of emotional response.
For REAPPRAISE trials, subjects were asked to feel neutral in response to the aversive image by altering their construal of the image, such as by imagining that the depicted scenario would improve over time or by adopting the perspective of a detached observer.

Stimuli
The average valence ratings for the ERRT's selected negative images per the IAPS data set by Lang et al. 2008 was 1.88 (SD ±3.69) and arousal rating average 4.16 (SD ±2.89).This task is mildly aversive due to its low valence and mid-level arousal ratings, given that these ratings are based on scales 1 (low) to 9 (high) -making these negative stimuli images relatively unpleasantly valenced (1.88 ±3.69) and mildly arousing emotionally (4.16 ±2.89).Rating averages for valence and arousal, as well as stimulus general subject matter, were maintained throughout all four task forms for both negative and neutral stimuli in ERRT forms.

Supplemental Results
For cardiac response to IAPS images baselined to preceding inter-trial interval, Figure S2 and Table S1 show the effects involving Pre/Post (pre-TFUS, post-TFUS), TFUS Target (amygdala, entorhinal cortex), and Instructions (view during neutral image, view during negative image, reappraise negative images) on the change in inter-beat interval from inter-trial intervals to IAPS image presentation.There was no significant effect of instructions on cardiac response to IAPS images.There was a significant interaction effect between Pre/Post and Target (χ²(1) = 5.42, p=0.019), indicating a significant increase in inter-beat interval pre-to post-TFUS targeting the amygdala as compared to pre-to post-TFUS targeting the entorhinal cortex.Additionally, the baselined inter-beat interval was greater post-TFUS targeting the amygdala than pre-TFUS targeting the amygdala (Z=2.05,p=0.04).Evaluation of simple effects reveals that baselined inter-beat interval pre-amygdala TFUS was significantly lower than baselined inter-beat interval pre-entorhinal cortex TFUS (Z=3.22,p <0.001), with inter-beat interval decreasing in response to stimuli preamygdala but increasing pre-entorhinal cortex.This difference between groups was no longer present post-TFUS.All other effects involving Pre/Post were not significant (ps>0.133)."Reactivity" refers to the contrast between viewing neutral images and negative images."Reappraisal"refers to the contrast between viewing negative images and reappraising negative images.Individual dots represent individual observations for each participant and each condition (e.g., Amygdala: Reactivity); circles represent "Reactivity" observations, and triangles represent "Reappraisal" observations.For each IAPS trial, the average inter-beat interval 5-seconds post-stimulus was subtracted by the average inter-beat interval of the preceding inter-trial rest period.Table S2B.Self-reported arousal in response to IAPS stimuli.Significant values bolded.There was a statistically significant Pre/Post x Target x Instructions interaction (χ²(2) = 15.99,p<0.001).Simple effects revealed an increase in negative emotional reactivity ("WATCH" negative images contrasted against "VIEW" neutral images) from pre-to post-TFUS targeting the amygdala (Z=4.21,p<0.001).Furthermore, there was a significantly greater increase in negative emotional reactivity from pre-to post-TFUS targeting the amygdala compared to pre-to post-TFUS targeting the entorhinal cortex (Z=-3.96,p<0.001).All other simple effects involving Pre/Post were not statistically significant (ps>0.062).
("WATCH", "VIEW", "REAPPRAISE") for one of three trial types: negative view (passively view a negative image; labeled "WATCH"), neutral view (passively view a neutral image; labeled "VIEW"), or negative reappraise (actively reappraise a negative image; labeled "REAPPRAISE; Figure 3).No effects including the factor Pre/Post were statistically significant (ps>0.162).Thus, there were no effects on negative emotional reactivity or negative emotional regulation.Table S2D.Cardiac activity between IAPS trials."Pre/Post" indicates pre-and post-TFUS."Target" refers to TFUS targeting the entorhinal cortex and amygdala.Significant results are in bold.Evaluation of simple effects reveals a baseline (Pre-TFUS) difference in the intertrial interval inter-beat interval between amygdala and entorhinal cortex (Z=-7.39,p<0.001), such that intertrial interval inter-beat interval was higher before amygdala sonication than before entorhinal cortex sonication.Further, there was a significant pre-TFUS targeting post-TFUS amygdala sonication decrease in inter-trial interval inter-beat interval (Z=-10.69,p<0.001), whereas entorhinal cortex sonication demonstrated no pre-to post-TFUS change (Z=0.79,p=0.429).Finally, inter-trial interval inter-beat interval was significantly lower postamygdala sonication than post-entorhinal cortex sonication (Z=3.59,p<0.001) (complete data available in Supplementary Table S2D).TFUS targeting the amygdala decreased inter-beat intervals (i.e., increased heart rate) during inter-trial intervals of the mildly aversive IAPS task.S3A, and a graph with the individual points and averaged findings are included in the main manuscript as Figure 5.

Discussion
No auditory or tactile sensations reported in either tFUS administration.Additionally, in our corresponding neuroimaging manuscript (Kuhn et al. 2023), no significant effect in the auditory cortex (all FDR-corrected p-values > 0.05).

Figure S2A .
Figure S2A.Valence Rating Scale and Instruction Screen Valence Self-Rating Screen using the Samikin Likert Rating Scale.Instructions read: "Please use the scale below to rate how happy or unhappy this image made you feel" (Lang et al. 1980, 2008).Valence rating is 1 (negative) to 9 (positive).

Figure S2B .
Figure S2B.Arousal Rating Scale and Instruction Screen Arousal Self-Rating Screen using the Samikin Likert Rating Scale.Instructions read: "Please use the scale below to rate how calm or excited this image made you feel" (Lang et al. 1980, 2008).Arousal rating is 1 (calm) to 9 (excited).

Figure S3B .
Figure S3B.Cardiac Changes During IAPS Stimuli Presentation in ERRT Task in Response TFUS Targeting Amygdala vs Entorhinal Cortex Cardiac changes were calculated by taking the difference of the post-stimulus IBI from the preceding inter-trial intervals as specified in the Results (see Section 3).Effects are collapsed across Watch, View, and Reappraise.Individual data points represent mean cardiac change values within each participant across all IAPS images within each level of the factors.Error bars represent the standard error of the mean across all 5-second intervals for each IAPS image within each level of the factors.Note: higher inter-beat interval values indicate a decrease in heart rate.

Figure S4A .
Figure S4A.Individual Participant Inter-Beat Intervals During Inter-Trial Intervals by Target Individual inter-beat intervals during inter-trial intervals involving TFUS Target (amygdala, entorhinal cortex) and Pre/Post (pre-TFUS, post-TFUS) on the average inter-beat interval between IAPS trials.TFUS targeting the amygdala decreased inter-beat intervals (i.e., increased heart rate) during inter-trial intervals of the mildly aversive IAPS task.Increased inter-trial intervals indicate a lower heart rate.A violin plot of data is included in the main manuscript in Figure 7.

Figure S4B .
Figure S4B.Inter-Beat Interval During Stimuli Presentation Individual inter-beat intervals during inter-trial intervals involving TFUS Target (amygdala, entorhinal cortex) and Pre/Post (pre-TFUS, post-TFUS) on the average inter-beat interval between IAPS trials.TFUS targeting the amygdala decreased inter-beat intervals (i.e., increased heart rate) during inter-trial intervals of the mildly aversive IAPS task.Increased inter-trial intervals indicate a lower heart rate.A violin plot of data in the main manuscript in Figure 7.

Table S2A .
Cardiac Response to IAPS Images.

Table S2A .
Cardiac Response to IAPS Images."Pre/Post"indicatespre-and post-TFUS."Target"refersto TFUS targeting the entorhinal cortex and amygdala."Instructions"refersto viewing neutral images, viewing negative images, or reappraising negative images to be less negative."StimulusTiming"refers to the five 1-second intervals of time from which cardiac inter-beat intervals were extracted during the 5-second IAPS images, subtracting the 5-second inter-beat intervals prior to trial onset.Significant results are in bold.Note that only effects involving Pre/Post were included to exclusively evaluate the effect(s) of TFUS.See Supplemental FigureS4.

Table S2C .
Self-Reported Valence in Response to IAPS Stimuli

Table S2C .
Self-reported valence in response to IAPS stimuli.

Table S3A .
Individual Participant Arousal Ratings Aggregated by Session TrialsTable rows are in a randomized order; each row across in each table represents a unique participant's visits, with the order of visits also randomized.
Figure S5A.Individual Participant Arousal Ratings Aggregated by Session Trials Individual points are available in table Supplemental Table

Table S3B . Individual Valence Ratings Aggregated by Session Trials
Supplementary Table S3B.Individual Participant Valence Ratings Aggregated by Session TrialsTable rows are in a randomized order; each row across in each table represents a unique participant's visits, with the order of visits also randomized.