How stress and social struggles fuel America’s obesity crisis

UCLA researcher says policy changes can provide long-term solutions, but doctors tailor care based on patient experiences

University of California - Los Angeles Health Sciences

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

As obesity in America continues to rise at alarming rates, researchers are finding that diet and exercise are not the only driving factors. A new scientific review from UCLA Health explains how stress, hardship and other social challenges can reshape a person’s gut bacteria and brain performance in ways that make it harder to keep weight off. 

Published in the journal Clinical Gastroenterology and Hepatology, the paper describes how social determinants of health, such as income, education, healthcare access, neighborhood disadvantages, experiences of discrimination, adverse childhood life events, and isolation and loneliness, are key drivers in the onset and worsening of obesity.

About 40% of American adults have obesity, which contributes to about $173 billion in annual healthcare costs. A recent study found obesity-related cancer deaths tripled in the U.S. between 1999 and 2020.

Led by Dr. Arpana Church, the scientific review reveals how the brain-gut microbiome acts as a bridge between a person’s environmental influences and their risk of obesity through the production of various signaling molecules including appetite-stimulating hormones, inflammatory markers and neuroactive metabolites. These chemical changes, in turn, affect what a person decides to eat, how often they eat, the quantity of food they eat, what types of food they crave, metabolic function and exercise habits.

“Our findings reveal that tackling obesity requires more than focusing on individual choices — it demands recognizing the powerful role that social and environmental forces play in shaping gut health, behavior and long-term health outcomes," said Church, who co-directs the Goodman-Luskin Microbiome Center at UCLA Health. "Reversing the escalating obesity epidemic in America demands a dual approach — personalized, equitable care for individuals and bold, systemic policy reforms that address the root causes.

“Research shows that social determinants of health — such as access to nutritious food, safe spaces for physical activity, access to resources such as education plumbing, grocery stores, and quality healthcare — are powerful drivers of obesity risk, underscoring the need to address the conditions in which people live, work and grow.”

Lower socioeconomic status often limits health literacy and drive reliance on inexpensive, energy-dense processed foods. In many disadvantaged and communities, limited access to healthy options- coupled with chronic exposure to stress, violence, and structural racism, - creates the perfect storm for obesity. Adding to this, social isolation disrupts brain networks that regulate appetite and decision-making, increasing the risk of unhealthy eating patterns and weight gain.

These social and dietary factors don’t just influence behavior but also lead to real, physical changes in the brain-gut microbiome system. Chronic exposure to unhealthy foods alters brain structure, influencing networks that regulate motivation, reward processing, and emotional regulation as well as reducing gray matter volume. At the same time, poor diets disrupt the makeup of the gut bacteria, fueling inflammation that further undermines self-control and motivation- thus reinforcing a cycle of emotional eating and cravings and dependence on ultra-processed, fast foods. 

Chronic stress, including racism-related stress and social isolation, also alter brain pathways and the gut microbes, promoting inflammation and impairing self-control, which further drives obesity risk, Church said. “Neighborhood disadvantage is also linked to reduced gut microbiome diversity and a higher presence of harmful bacteria — factors that further impair metabolism and heighten the risk of obesity and related diseases.”

Furthermore, these changes can begin prenatally and in early childhood. Environmental stressors and social adversity influence microbiome composition and brain-gut communication, setting the stage for lifelong obesity susceptibility. 

Church said that while lasting change requires bold policy action to address the root causes of obesity, individuals can still take proactive steps on a personal level to navigate these challenges and support their health in the meantime. 

For example, in the face of difficult circumstances, individuals can support their health by prioritizing nutritious foods within their budget, building social connections, and engaging in stress-reducing activities like journaling, connecting with nature and physical exercise Prioritizing empathy and gratitude can also help people improve their health despite difficult circumstances. 

"At the same time, healthcare providers have a vital role to play, not only by screening for current social determinants of health, but also by recognizing how these factors accumulate and evolve over time, which is rarely accounted for in today’s clinical practice," said Church 

“By understanding these influences and tailoring treatment plans to account for biological and psychosocial challenges, providers can offer more personalized plans that improve outcomes, empowers individuals to take control over their health, and at the same time supports long-term wellness,” Church continued.

Article: Biopsychosocial and Environmental Factors that Impact Brain-Gut-Microbiome Interactions in Obesity; Riya Sood et al.; Published online Sept. 4, 2025, Clinical Gastroenterology and Hepatology

---

Journal

Clinical Gastroenterology and Hepatology

DOI

10.1016/j.cgh.2025.07.045 

Method of Research

Literature review

Subject of Research

People

Article Title

Biopsychosocial and Environmental Factors that Impact Brain-Gut-Microbiome Interactions in Obesity

COI Statement

Arpana Church is a scientific consultant to Yamaha. All other authors have nothing to disclose.

Researchers uncover similarities between human and AI learning

Brown University

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

PROVIDENCE, R.I. [Brown University] — New research found similarities in how humans and artificial intelligence integrate two types of learning, offering new insights about how people learn as well as how to develop more intuitive AI tools.

 Led by Jake Russin, a postdoctoral research associate in computer science at Brown University, the study found by training an AI system that flexible and incremental learning modes interact similarly to working memory and long-term memory in humans.

“These results help explain why a human looks like a rule-based learner in some circumstances and an incremental learner in others,” Russin said. “They also suggest something about what the newest AI systems have in common with the human brain.”

Russin holds a joint appointment in the laboratories of Michael Frank, a professor of cogntive and psychological sciences and director of the Center for Computational Brain Science at Brown’s Carney Institute for Brain Science, and Ellie Pavlick, an associate professor of computer science who leads the AI Research Institute on Interaction for AI Assistants at Brown. The study was published in the Proceedings of the National Academy of Sciences.

 Depending on the task, humans acquire new information in one of two ways. For some tasks, such as learning the rules of tic-tac-toe, “in-context” learning allows people to figure out the rules quickly after a few examples. In other instances, incremental learning builds on information to improve understanding over time — such as the slow, sustained practice involved in learning to play a song on the piano.

 While researchers knew that humans and AI integrate both forms of learning, it wasn’t clear how the two learning types work together. Over the course of the research team’s ongoing collaboration, Russin — whose work bridges machine learning and computational neuroscience — developed a theory that the dynamic might be similar to the interplay of human working memory and long-term memory.

 To test this theory, Russin used “meta-learning”— a type of training that helps AI systems learn about the act of learning itself — to tease out key properties of the two learning types. The experiments revealed that the AI system’s ability to perform in-context learning emerged after it meta-learned through multiple examples. 

One experiment, adapted from an experiment in humans, tested for in-context learning by challenging the AI to recombine similar ideas to deal with new situations: if taught about a list of colors and a list of animals, could the AI correctly identify a combination of color and animal (e.g. a green giraffe) it had not seen together previously? After the AI meta-learned by being challenged to 12,000 similar tasks, it gained the ability to successfully identify new combinations of colors and animals.

The results suggest that for both humans and AI, quicker, flexible in-context learning arises after a certain amount of incremental learning has taken place. 

“At the first board game, it takes you a while to figure out how to play,” Pavlick said. “By the time you learn your hundredth board game, you can pick up the rules of play quickly, even if you’ve never seen that particular game before.”

 The team also found trade-offs, including between learning retention and flexibility: Similar to humans, the harder it is for AI to correctly complete a task, the more likely it will remember how to perform it in the future. According to Frank, who has studied this paradox in humans, this is because errors cue the brain to update information stored in long-term memory, whereas error-free actions learned in context increase flexibility but don’t engage long-term memory in the same way. 

For Frank, who specializes in building biologically inspired computational models to understand human learning and decision-making, the team’s work showed how analyzing strengths and weaknesses of different learning strategies in an artificial neural network can offer new insights about the human brain. 

“Our results hold reliably across multiple tasks and bring together disparate aspects of human learning that neuroscientists hadn’t grouped together until now,” Frank said. 

 The work also suggests important considerations for developing intuitive and trustworthy AI tools, particularly in sensitive domains such as mental health.  

 “To have helpful and trustworthy AI assistants, human and AI cognition need to be aware of how each works and the extent that they are different and the same,” Pavlick said. “These findings are a great first step.”

The research was supported by the Office of Naval Research and the National Institute of General Medical Sciences Centers of Biomedical Research Excellence.

---

Journal

Proceedings of the National Academy of Sciences

DOI

10.1073/pnas.2510270122 

Article Title

Parallel trade-offs in human cognition and neural networks: The dynamic interplay between in-context and in-weight learning

UCR researchers fortify AI against rogue rewiring

Preventing bots from dispensing dangerous information

University of California - Riverside

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

Open-source AI systems can return harmful results when run on lower-powered devices. 

view more 

Credit: Stan Lim/UCR

As generative AI models move from massive cloud servers to phones and cars, they’re stripped down to save power. But what gets trimmed can include the technology that stops them from spewing hate speech or offering roadmaps for criminal activity.

To counter this threat, researchers at the University of California, Riverside, have developed a method to preserve AI safeguards even when open-source AI models are stripped down to run on lower-power devices.

Unlike proprietary AI systems, open‑source models can be downloaded, modified, and run offline by anyone. Their accessibility promotes innovation and transparency but also creates challenges when it comes to oversight. Without the cloud infrastructure and constant monitoring available to closed systems, these models are vulnerable to misuse.

The UCR researchers focused on a key issue: carefully designed safety features erode when open-source AI models are reduced in size. This happens because lower‑power deployments often skip internal processing layers to conserve memory and computational power. Dropping layers improves the models’ speed and efficiency, but could also result in answers containing pornography, or detailed instructions for making weapons.

“Some of the skipped layers turn out to be essential for preventing unsafe outputs,” said Amit Roy-Chowdhury, professor of electrical and computer engineering and senior author of the study. “If you leave them out, the model may start answering questions it shouldn’t.”

The team’s solution was to retrain the model’s internal structure so that its ability to detect and block dangerous prompts is preserved, even when key layers are removed. Their approach avoids external filters or software patches. Instead, it changes how the model understands risky content at a fundamental level.

“Our goal was to make sure the model doesn’t forget how to behave safely when it’s been slimmed down,” said Saketh Bachu, UCR graduate student and co-lead author of the study.

To test their method, the researchers used LLaVA 1.5, a vision‑language model capable of processing both text and images. They found that certain combinations, such as pairing a harmless image with a malicious question, could bypass the model’s safety filters. In one instance, the altered model responded with detailed instructions for building a bomb.

After retraining, however, the model reliably refused to answer dangerous queries, even when deployed with only a fraction of its original architecture.

“This isn’t about adding filters or external guardrails,” Bachu said. “We’re changing the model’s internal understanding, so it’s on good behavior by default, even when it’s been modified.”

Bachu and co-lead author Erfan Shayegani, also a graduate student, describe the work as “benevolent hacking,” a way of fortifying models before vulnerabilities can be exploited. Their ultimate goal is to develop techniques that ensure safety across every internal layer, making AI more robust in real‑world conditions.

In addition to Roy-Chowdhury, Bachu, and Shayegani, the research team included doctoral students Arindam Dutta, Rohit Lal, and Trishna Chakraborty, and UCR faculty members Chengyu Song, Yue Dong, and Nael Abu-Ghazaleh. Their work is detailed in a paper presented this year at the International Conference on Machine Learning in Vancouver, Canada.

“There’s still more work to do,” Roy-Chowdhury said. “But this is a concrete step toward developing AI in a way that’s both open and responsible.”

Single hair strand could provide biomarker for ALS, Mount Sinai study finds

The Mount Sinai Hospital / Mount Sinai School of Medicine

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

Hair samples were collected from ALS-positive cases and ALS-negative controls at a regional ALS centre and supplemented with a nationwide collection of ALS-positive cases in the US.

view more 

Credit: Mount Sinai Health System

New York, NY (September 4, 2025) – Researchers at the Icahn School of Medicine at Mount Sinai have shown for the first time that a single strand of hair can reveal unique elemental patterns that distinguish people with amyotrophic lateral sclerosis (ALS) from healthy individuals. The findings, published in eBioMedicine, suggest that a simple, non-invasive hair-based test could one day speed ALS diagnosis and improve patient care.

ALS is a progressive and fatal neurodegenerative disease, typically taking 10 to 16 months from symptom onset to diagnosis in the United States. Earlier detection is critical since it improves the patient's quality of life and therefore potentially prolonging survival, but current fluid- and imaging-based biomarkers are often invasive, expensive, and difficult to integrate into routine clinical care. According to the ALS Association, the average survival time is three years, though about 20 percent of people with ALS live five years, 10 percent survive 10 years, and 5 percent live 20 years or longer.

Using advanced laser ablation-inductively coupled plasma-mass spectrometry, a technique that analyzes the elemental and isotopic composition of solid samples by using a laser to vaporize tiny particles, the research team analyzed single hair strands from 391 people (295 ALS-positive cases and 96 controls). Each hair strand provided up to 800 time points of data, representing elemental fluctuations at approximately two to four-hour intervals. The investigators measured 17 elements—including copper, zinc, magnesium, and lead—and used sophisticated information-theory tools to assess how elemental patterns shifted over time.

The study adds to evidence that copper plays a central role in ALS. Patients with ALS showed significantly lower synchrony (less coordinated, less similar, and less aligned in time) in copper-based elemental networks compared with controls, suggesting systemic dysregulation in copper metabolism. Male patients demonstrated weaker copper-zinc dynamics, while female patients showed disruptions in chromium-nickel patterns.

“Our study demonstrates that hair can serve as a window into the body’s elemental balance,” said Manish Arora, BDS, MPH, PhD, the Edith J. Baerwald Professor and Vice Chair of the Department of Environmental Medicine at the Icahn School of Medicine at Mount Sinai, and senior author of the study. “By analyzing the biodynamics of elements such as copper over time, we can detect disruptions associated with ALS in a simple, non-invasive way. This approach has the potential to transform how we diagnose ALS, making it faster, easier, and more accessible for patients.”

“This is the first-ever study to use hair strands to identify elemental dysregulation in ALS,” said Vishal Midya, PhD, Assistant Professor of Environmental Medicine at the Icahn School of Medicine and senior author of the study. “We found that patients with ALS had measurable differences in copper biodynamics that were not present in controls. These findings provide proof-of-concept that hair could serve as a simple and scalable diagnostic tool.”

While the study does not yet provide a diagnostic test, it shows that hair could be a promising new biomarker for ALS. A simple hair-based test might help shorten the time to diagnosis, allowing patients to start treatment and support sooner. Earlier care, including medications, nutrition plans, assistive devices, therapy, and counseling, can improve quality of life and may even extend survival.

This research was conducted in collaboration with Linus Biotechnology, Inc., Dartmouth University, and Columbia University, with support from the National Institutes of Health and the Centers for Disease Control and Prevention.

Read the full manuscript here: https://www.thelancet.com/journals/ebiom/article/PIIS2352-3964(25)00351-2/fulltext?rss=yes

###

About the Icahn School of Medicine at Mount Sinai

The Icahn School of Medicine at Mount Sinai is internationally renowned for its outstanding research, educational, and clinical care programs. It is the sole academic partner for the seven member hospitals* of the Mount Sinai Health System, one of the largest academic health systems in the United States, providing care to New York City’s large and diverse patient population. 

The Icahn School of Medicine at Mount Sinai offers highly competitive MD, PhD, MD-PhD, and master’s degree programs, with enrollment of more than 1,200 students. It has the largest graduate medical education program in the country, with more than 2,600 clinical residents and fellows training throughout the Health System. Its Graduate School of Biomedical Sciences offers 13 degree-granting programs, conducts innovative basic and translational research, and trains more than 560 postdoctoral research fellows.

Ranked 11th nationwide in National Institutes of Health (NIH) funding, the Icahn School of Medicine at Mount Sinai is among the 99th percentile in research dollars per investigator according to the Association of American Medical Colleges.  More than 4,500 scientists, educators, and clinicians work within and across dozens of academic departments and multidisciplinary institutes with an emphasis on translational research and therapeutics. Through Mount Sinai Innovation Partners (MSIP), the Health System facilitates the real-world application and commercialization of medical breakthroughs made at Mount Sinai.

-------------------------------------------------------

* Mount Sinai Health System member hospitals: The Mount Sinai Hospital; Mount Sinai Brooklyn; Mount Sinai Morningside; Mount Sinai Queens; Mount Sinai South Nassau; Mount Sinai West; and New York Eye and Ear Infirmary of Mount Sinai.  

---

Journal

EBioMedicine

Subject of Research

Human tissue samples

Article Title

Dysregulation of hair-strand-based elemental biodynamics in amyotrophic lateral sclerosis

Article Publication Date

3-Sep-2025

 

Bio-oil made with corn stalks, wood debris could plug orphaned fossil fuel wells

Iowa State University

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

Injecting bio-oil made from corn stalks and forest debris into the deep shafts of abandoned crude oil wells could be a viable form of carbon sequestration, a new Iowa State University study found.

view more 

Credit: Deb Berger/Iowa State University

AMES, Iowa – Filling abandoned oil and gas wells with bio-oil made from plant-based leftovers like corn stalks and forest debris could help remove carbon dioxide from the atmosphere, returning carbon underground in deep shafts once used to extract it.

The emerging practice, the focus of a recent study by an Iowa State University research team led by mechanical engineering professor Mark Mba-Wright, has a two-birds-one-stone appeal. Unwanted organic matter collected from forests and fields helps sequester carbon in long-term storage while also reducing the emissions and safety risks posed by the hundreds of thousands of orphaned U.S. oil wells.

“On the one hand, you have these underutilized waste products. On the other hand, you have abandoned oil wells that need to be plugged. It’s an abundant resource meeting an urgent demand,” Mba-Wright said.

Based on research by Mba-Wright’s team, a network of 200 mobile bio-oil production facilities could be an economically and technically feasible expansion of the technology, which is already in limited commercial use. The study, recently published in Energy Conversion and Management, estimated that the proposed system could sequester carbon dioxide for about $152 per ton, making it competitive with other methods of carbon dioxide removal but with far less upfront investment.

“One of the innovations here is that you can do carbon capture with units the size of a skid loader or a combine. You can start small,” Mba-Wright said.

Projecting costs

The core of the proposed system is fast pyrolysis, a process of transforming dried particles of biological material into liquid bio-oil by exposing them in an oxygen-free environment to a few seconds of high heat – temperatures that can exceed 1,000 degrees Fahrenheit. The organic matter holds carbon it pulled from the air via photosynthesis when alive.

The solid byproduct, biochar, can be sold to farmers as a soil amendment. The gas byproduct is captured to be reused as a fuel to help generate the intense heat pyrolysis requires. But the main aim of fast pyrolysis is producing bio-oil, the dense and carbon-rich fluid that forms as the vapor released in the process condenses.

Various uses for bio-oil have been identified and more are being studied. But injecting it into empty fossil fuel wells would maximize bio-oil’s carbon capture potential and take advantage of existing underground well shafts that otherwise cost about $1 million to cap. Filling a crude oil well, at an average width of about 1.6 feet and depth of nearly 2.6 miles, takes more than 216,000 gallons of liquid. The 2021 bipartisan infrastructure law allocated $4.7 billion to seal about 120,000 abandoned wells, but estimates cited by the study suggest there are 300,000 to 800,000 undocumented orphaned wells in the U.S.

Under the proposed system, each mobile fast pyrolysis unit would process about 10 tons of feedstock a day. In the Midwest, the main biomass source studied was corn stover, the portion of the plant left in the field after harvest. In the West, it would be wood debris removed from forests to lessen wildfire threats. Researchers also studied switchgrass and oriented strand board as feedstock for the mobile pyrolizers, which would deposit bio-oil in centralized terminals for transport to well sites.

Units would cost about $1.3 million to build, and the bio-oil would need to sell for at least $175 per ton, the study estimated. Carbon removal costs vary by feedstock, with the rate for wood-based materials estimated at about $100 per ton. Abatement costs fall even lower when calculated to account for the carbon in biochar and a learning rate – a factor that estimates how capital and operating costs would reduce over time.

“The more units they build, the better they would get at building them,” Mba-Wright said.

Not an ‘either/or’

The study was funded in part by Charm Industrial, a San Francisco-based startup which has already struck several carbon-removal deals with large corporations to use vacant wells to permanently store bio-oils. Companies increasingly seek carbon-removal credits to meet commitments to reduce emissions, said Peter Reinhardt, CEO and co-founder of Charm Industrial.

"We hear it time and again: after taking a close look among their options, leading carbon-removal buyers find that bio-oil sequestration represents one of the highest-quality and most cost-effective approaches,” Reinhardt said.

Charm approached Iowa State seeking an independent and detailed assessment of the system’s potential, Mba-Wright said. 

“While they were confident about the technology itself, they were looking for some validation of how much carbon could be sequestered and how economical the process could be,” he said. “There are a lot of steps involved in getting this to work at scale.”

A key takeaway is that the system stacks up well against the dominant method of removing atmospheric carbon, a technology called direct air capture that extracts carbon dioxide from air. Direct air capture systems have similar per-ton abatement costs, but they’re far more expensive to build and have few other associated benefits, Mba-Wright said.

“What we’re trying to show here is that carbon removal doesn’t need to be either/or. There are a lot of opportunities,” he said.

The techno-economic analysis by Mba-Wright and his colleagues will help companies make reliable investments in their net zero portfolios, and it highlights a path to new revenue streams in rural areas where the biomass is collected, Reinhardt said.

"Iowa State’s experts showed that bio-oil sequestration using corn stover can deliver a high-value, durable carbon removal product that outcompetes other technologies, while providing new markets for crop residues and delivering new economic value to the rural economy,” he said. “As the carbon-removal sector grows, Charm is grateful to work with farm and forest communities to grow this opportunity."

---

Journal

Energy Conversion and Management

DOI

10.1016/j.enconman.2025.119980 

Method of Research

Computational simulation/modeling

Subject of Research

Not applicable

Article Title

Enhancing carbon removal via scalable on-site pyrolysis and well-plugging systems

Article Publication Date

1-Oct-2025