Medha
Tirumalareddy

Papers

Sponsor: George Mangun, Ph.D. Psychology Stress isn't just felt; it can be heard. Subtle patterns in the human voice may reveal stress even before individuals consciously recognize it. While prior research has focused primarily on biochemical and neural indicators, the present study investigates speech as a non-invasive modality for stress detection. By leveraging acoustic features embedded in vocal signals, stress- related physiological responses may be inferred without complex measurements. One hundred neurotypical participants completed a modified Trier Social Stress Test, which involved delivering a prepared speech before evaluative judges followed by an unexpected mental arithmetic task. Salivary samples were collected during the task to assess cortisol levels. Participants were categorized into high- and low-stress groups based on cortisol responses. Speech recordings from the prepared speech were analyzed to extract acoustic features such as Mel-frequency cepstral coefficients, spectral characteristics, and jitter. Principal component analysis was used to reduce feature dimensionality, and machine learning classifiers were trained to distinguish between high- and low-stress conditions. The resulting speech- based model achieved classification accuracy exceeding 80%, outperforming chance-level prediction (50%) and demonstrating that stress-related physiological responses, as indexed by cortisol, can be decoded from vocal features. In the Face of Marital Conflict: Parental Warmth as a Predictor of Adolescent Respiratory Sinus Arrhythmia (RSA) Authors: Medha Tirumalareddy, Javen Sevilla

Sponsor: George Mangun, Ph.D. Psychology Stress isn't just felt; it can be heard. Subtle patterns in the human voice may reveal stress even before individuals consciously recognize it. While prior research has focused primarily on biochemical and neural indicators, the present study investigates speech as a non-invasive modality for stress detection. By leveraging acoustic features embedded in vocal signals, stress- related physiological responses may be inferred without complex measurements. One hundred neurotypical participants completed a modified Trier Social Stress Test, which involved delivering a prepared speech before evaluative judges followed by an unexpected mental arithmetic task. Salivary samples were collected during the task to assess cortisol levels. Participants were categorized into high- and low-stress groups based on cortisol responses. Speech recordings from the prepared speech were analyzed to extract acoustic features such as Mel-frequency cepstral coefficients, spectral characteristics, and jitter. Principal component analysis was used to reduce feature dimensionality, and machine learning classifiers were trained to distinguish between high- and low-stress conditions. The resulting speech- based model achieved classification accuracy exceeding 80%, outperforming chance-level prediction (50%) and demonstrating that stress-related physiological responses, as indexed by cortisol, can be decoded from vocal features. In the Face of Marital Conflict: Parental Warmth as a Predictor of Adolescent Respiratory Sinus Arrhythmia (RSA) Authors: Medha Tirumalareddy, Javen Sevilla

Abstract profile. Full document pending author claim.

Authors:

Medha Tirumalareddy

Date Created:

Not specified

Course Title:
Professor:

Not specified

About Paper:

Respiratory sinus arrhythmia (RSA) is a cardiac index of parasympathetic (PNS) activity and a biomarker of emotion regulation during adolescence. Parental warmth has been associated with children's regulatory capacities, but this may depend on the background quality of family relationships. This study will examine whether more parental warmth predicts adolescents' greater RSA and if high levels of marital conflict disrupt this association. Participants included 229 adolescents and their parents from a longitudinal study of Mexican-origin youths. Parental warmth was assessed from videotaped parent- adolescent interactions at 10 and 12 years and coded using the Iowa Family Interaction Rating Scales. Marital conflict was assessed using parent reports on the Dyadic Adjustment Scale. Lastly, adolescent baseline RSA was collected at 17 years with an electrocardiogram during a laboratory visit. We will perform multiple regression analyses testing whether parental warmth predicts adolescent baseline RSA and whether marital conflict moderates this relation. Findings will clarify the role of parental warmth and marital conflict in adolescent PNS regulation. Understanding how family relational dynamics become biologically embedded during adolescence may inform intervention and policy efforts targeting improved family functioning. Momentum in College Basketball - A Study on Game Patterns and Indicators Preceding Scoring Surges in Men's NCAA Basketball Chance Tokubo

Source:

UC Davis / Psychology / 2026

Topics:

No topics listed

Co-authors:

Medha Tirumalareddy