Culture

Humanity's best friend

For some 15,000 years, dogs have been our hunting partners, workmates, helpers and companions. Could they also be our next allies in the fight against COVID-19?

According to UC Santa Barbara professor emeritus Tommy Dickey(link is external) and his collaborator, BioScent researcher Heather Junqueira, they can. And with a review paper(link is external) published in the Journal of Osteopathic Medicine they have added to a small but growing consensus that trained medical scent dogs can effectively be used for screening individuals who may be infected with the COVID-19 virus.

This follows a comprehensive survey of research devoted to the use of trained scent dogs for detecting COVID. "The most striking result is that studies have already demonstrated that dogs can identify people who are COVID-19 positive," Dickey said of their findings. "Not only that," he added, "they can do it non-intrusively, more rapidly and with comparable or possibly better accuracy than our conventional detection tests."

Not surprisingly, the magic lies in canine sense of smell, which gives dogs the ability to detect molecules in tiny concentrations -- "one part in a quadrillion compared with one part in one billion for humans," according to the paper. Add to that other optimizations for smell, such as a large nasal area and the structure of their noses, which allows inflow through the nostrils and outflow through nasal folds. Further, with 125-300 million olfactory cells and a third of their brains devoted to interpreting odors, dogs are well equipped with the ability to sniff out the volatile organic compounds that indicate the presence of COVID.

"The dogs are basically smelling the sweat of the person," Dickey said of a series of experiments by French and Lebanese researchers(link is external) testing canines' capacity to sense COVID infection. Although the virus itself has no odor, metabolic products excreted by COVID-positive individuals through their sweat glands were detected by the 18 dogs selected for the study (16 Belgian malinois, one German shepherd and one Jack Russell terrier) with an accuracy rate of 83-100% after only four days of training. True failures, according to the study, could be attributed to "distractive external smells or movements by a TV filming crew."

"One dog twice indicated positive results that could not be confirmed," Dickey said. "Two weeks later they found that both people who gave those samples had to be hospitalized with COVID."

Meanwhile, a German research group employed eight scent detection dogs(link is external) in a randomized, double-blind controlled pilot study. The group trained the dogs for a week and then set them to sniffing 1,012 samples of saliva or tracheobronchial secretions. They returned an average detection rate of 94% with a sensitivity (ability to detect a true positive) of 67.9% to 95.2% and a specificity (ability to detect a true negative) of 92.4% to 98.9%. This pilot study used positive samples from severely affected individuals and negative samples from people with no symptoms. Future studies, according to that paper, could focus more on identifying different phases of infection or perhaps the detection of different disease phenotypes.

Using dogs to detect disease is not new. In fact, co-author Junqueira has previously published results showing that her scent dogs (beagles, bassett hounds and mixes of the two) can effectively detect non-small cell lung cancer.

"Canines are capable of detecting other types of cancer as well as malaria, Parkinson's disease and diabetes," Junqueira said, adding that "medical scent dog research has really only gained traction in recent years and that it will take many more peer-reviewed papers before the idea of using dogs for disease detection hits the mainstream."

Dickey's own interest in the subject was sparked over the course of his work as a therapy dog handler of three Great Pyrenees (over 3,000 therapy dog visits), a longtime avocation he leaned on after cancer forced him to retire from UC Santa Barbara's Department of Geography in 2013. "I loved UCSB," he said. "I loved teaching and I brought my therapy dogs to class all the time. I just had the life, what can I say?"

In fact, he couldn't stay away -- he and his dogs have been on hand to see the UCSB community through hard times, offering their shaggy coats, wet noses and calm demeanors through tragedy and stress. In retirement, Dickey has published three therapy dog books for children, some of which recount stories of their UCSB therapy dog adventures. In addition, he and his canines have presented educational demonstrations at the California Science Center and the Los Angeles Public Library, work that sparked his interest in the power of a dog's sense of smell for medical detection.

So when the new disease called COVID-19 hit, Dickey was primed to ask: Can dogs detect the novel coronavirus? Naturally, there was little in the way of refereed research on the topic, so he teamed up with Junqueira, who was already conducting her own COVID detection research(link is external) with her scent dogs in Florida. "One of our big motivations was to write a peer-reviewed paper that basically gave a progress report," he said. "Where are we? Is this stuff really possible?"

Dickey and Junqueira found that researchers used a variety of dogs. "There were a lot of Belgian malinois that were used, and dogs who have been trained on explosives and on colon cancer. So they were pro sniffers," Junqueira said. "Other groups, such as the one behind a Colombian study(link is external), were motivated by the need to find a quick, accurate and cost-effective form of COVID early detection." The Colombian group utilized a variety of dogs -- four Belgian malinois, one Alaskan malamute-Siberian mix and an American pit bull terrier.

"The pit bull had been previously mistreated," Dickey said, "but they rehabilitated him, and he was perfectly capable and doing a great job at sniffing." After almost two months of training and thousands of samples later, this Colombian canine cohort performed with a remarkable 95.5% sensitivity and 99.6% specificity.

During the various blind, controlled experiments, the total time for detection was a matter of minutes or less. Such speed is a huge asset in real world scenarios. In particular, a U.K.-based research group(link is external) has outlined their plans to train and ultimately deploy dogs at United Kingdom airports and ports of entry as part of the COVID-19 screening process.

With all the sniffing going on in the presence of an airborne disease, it's natural to be concerned over whether dogs can catch and transmit COVID-19. It's still the topic of ongoing research, but evidence points to a low likelihood of transmission, according to the paper, though precautionary measures should be taken to protect everyone involved.

"Current research supports the use of scent detection dogs for pilot COVID-19 screening studies involving humans in venues such as airports and sporting events," Dickey said. "In addition, the JOM paper points out that another line of research can utilize medical scent detection dogs involving the development of medical electronic noses."

In principle, you wouldn't even need a dog to sniff out COVID if you could mimic the way it smells and processes scents, according to the researchers. Through sensors and artificial intelligence, they said, it might be possible to someday match a dog's performance using wearable electronic noses, similar to wristband sensors for reporting heart beat rate and patterns, blood pressure and oxygen, that could monitor a person's sweat for metabolites and biomarkers that could indicate diseases such as COVID-19.

Credit: 
University of California - Santa Barbara

Spinal fluid of people with Alzheimer's risk gene signals inflammation

DURHAM, N.C. - People who have a gene variant associated with an increased risk of developing Alzheimer's disease also tend to have changes in the fluid around their brain and spinal cord that are detectable years before symptoms arise, according to new research from Duke Health.

The work found that in people who carry the APOE4 gene variant, which is found in roughly 25 percent of the population, the cerebrospinal fluid contains lower levels of certain inflammatory molecules. This raises the possibility that these inflammatory molecules may be collecting in the brain where they may be damaging synapses, rather than floating freely in the cerebrospinal fluid.

The findings, which were published online last month in the Journal of Alzheimer's Disease, provide a potential means to identify the earliest mechanisms occurring among APOE4 carriers that might contribute to Alzheimer's disease before people develop memory problems or other symptoms of dementia.

"Our work suggests a potential role for a long-studied molecule called C-reactive protein (CRP), which is typically elevated when there's inflammation, as a factor in the increased Alzheimer's disease risk seen in APOE4 carriers," said lead author Miles Berger, M.D., Ph.D., associate professor in Duke's Department of Anesthesiology.

Berger and colleagues analyzed data from targeted cerebrospinal fluid of Alzheimer's Disease Neuroimaging Institute research participants. Controlling for Alzheimer's disease clinical status, they identified protein level variations in the cerebrospinal fluid from people with an increasing number of APOE4 gene variant copies. They found that people with more APOE4 copies had lower CRP levels circulating in their cerebrospinal fluid.

Berger said this is consistent with the current risk profile associated with APOE4 carriers. People with a single APOE4 variant have about a three- to four-fold increased risk of developing Alzheimer's disease, while those who carry two APOE4 variants have a greater than 10-fold risk.

"We found that spinal fluid CRP levels are lower in people with the APOE4 allele, before they ever develop dementia or even mild cognitive impairment," Berger said. "This suggests that CRP might be actively involved in damaging synapses. We think CRP is doing this together with a cascade of inflammatory proteins called complement, which sequentially activate each other like a row of dominos falling.

"Altered complement activity has also been found at autopsy in the brains of patients who suffered from Alzheimer's disease, and the complement pathway can both damage synapses and target them for destruction," Berger said. "Our results raise the possibility that processes like these operating over many years and even decades in APOE4 carriers could eventually result in Alzheimer's disease pathology and cognitive decline."

Berger said there are other inflammatory diseases for which CRP can be decreased rather than elevated, notably lupus. He said therapies used to control CRP for conditions such as rheumatoid arthritis and vasculitis might warrant investigation for Alzheimer's disease, but additional studies are needed to further illuminate the Duke team's findings.

"Maybe those with more APOE4 variants have excessive deletion of synapses throughout life, until it gets to point where brain can't process information anymore," Berger said. "It would be like taking a book and randomly deleting every 1000th letter. For a while, you could do that and the book would still make sense. But after time, too many letters would be gone and you would lose the information in the book. We think that might be what's happening in the brain."

Credit: 
Duke University Medical Center

Learn what you live? Study finds watching others can reduce decision bias

image: New research finds first evidence that watching and learning from others can help reduce bias and improve decision-making. In business, the results could help improve hiring practices or increase cost savings.

Image: 
Indiana University

New research from the Indiana University Kelley School of Business shows first evidence that watching and learning from others can help reduce bias and improve decision-making.

The research, published in the journal Organizational Behavior and Human Decision Processes, used a computer game designed to decrease bias to see if people who watched others play the game could in turn reduce their own bias. Through three experiments, researchers found that watching others solve bias-related problems helped the observers learn about decision biases and improve on their own. Their study showed this observational learning reduced decision biases such as anchoring - or, relying too much on an original bit of information -- and also improved how the observers take advice.

"Everyone has biases when it comes to making decisions," said study co-author Haewon Yoon, assistant professor of marketing at the Indiana University Kelley School of Business on the IUPUI campus. "We've found these biases can be mitigated by watching others and observing how they make decisions. For example, as people observe others playing these interactive games, they're able to see inside the game player's decision process and learn from their mistakes. Likewise, in a business setting, a training module or video showing employees how managers or co-workers demonstrate decision biases -- or avoid such biases -- could reduce that employee's own decision biases. This is a more cost-effective way to teach employees than through extensive training with individualized feedback."

"From childhood on, we learn by watching others like our parents, siblings, and friends: 'What is safe to eat? How do I do that? How should we behave in social contexts?'" said Carey Morewedge, professor of marketing at Boston University Questrom School of Business and co-author of the study. "This kind of observational learning teaches us important lessons about the world around us. Unfortunately, it also leads us to absorb many of the biases we observe and exhibit those biased behaviors ourselves. This new research shows that we can learn to improve our decision-making and unlearn some of our biases by watching others. We glean unique insights from seeing others solve problems that help us make better and less biased decisions."

The researchers explain there is value in reducing decision bias in both personal and professional domains - Take, for example, buying a house. Research has shown that whatever price the seller puts it up for sale at - No matter how much - buyers tend to adjust based on that price. These researchers argue that if you watch someone else committing or correcting the bias, you could take away insights that help you better make those decisions. For hiring managers, this could be used while onboarding employees - or interviewing employees. Asking others to watch and learn helps reduce bias.

"Social learning interventions like observational learning are not only promising in their effectiveness; they are relatively inexpensive to implement and scalable," said Irene Scopelliti, professor at the Business School (formerly Cass) and co-author of the study. "The findings could benefit all kinds of cases where people have to make decisions under uncertainty (i.e., without all the facts), from which gift to buy a friend to major business, law and policy decisions. We hope this strategy for debiasing decision making is added to the many training interventions used by teachers, government officials and industry leaders to help people make better decisions."

Credit: 
Indiana University

Tuning the circadian clock, boosting rhythms may be key to future treatments and medicines

image: A network of communicating clocks generates circadian rhythms. This artwork is a play on its striking parallel to the states of the U.S.A. - entities that are governed semi-autonomously but operate as part of a larger federated system. In the image, the clock network is illustrated as globe-style map with its semi-sovereign regions (for example the brain, liver, and epidermis [skin]). The land has many features that represent circadian clock components; rivers and roads connect the different regions just as blood and nerves relay signals between tissues in the body.

Image: 
Imogen Reekie

Irvine, CA - February 11, 2021 - Subconsciously, our bodies keep time for us through an ancient means - the circadian clock. A new University of California, Irvine-led article reviews how the clock controls various aspects of homeostasis, and how organs coordinate their function over the course of a day.

"What is fascinating is that nearly every cell that makes up our organs has its own clock, and thus timing is a crucial aspect of biology," said Kevin B. Koronowski, PhD, lead author and a postdoctoral fellow in Biological Chemistry at the UCI School of Medicine. "Understanding how daily timing is integrated with function across organs has implications for human health, as disruption of the clock and circadian rhythms can be both a cause and effect of diseases from diabetes to cancer."

The circadian clock generates a ~24 hour rhythm that controls behavior, hormones, the immune system and metabolism. Using human cells and mice, researchers from the Paolo Sassone-Corsi Laboratory at UCI's Center for Epigenetics and Metabolism aim to uncover the physiological circuits, for example between the brain and liver, whereby biological clocks achieve coherence. Their work, titled, "Communicating clocks shape circadian homeostasis," was published today in Science.

Circadian clocks align internal processes with external time, which enables diverse lifeforms to anticipate daily environmental changes such as the light-dark cycle. In complex organisms, clock function starts with the genetically encoded molecular clock or oscillator within each cell and builds upward anatomically into an organism-wide system. Circadian misalignment, often imposed in modern society, can disrupt this system and induce adverse effects on health if prolonged.

"Strategies to tune our clocks and boost rhythms have been promising in pre-clinical studies, which illustrates the importance of unraveling this aspect of our biology and unlocking the potential it holds for treatments and medicines of the future," said Koronowski.

Without electrical light, high-speed travel, constant food availability and around the clock work-life schedules, our ancestors' clocks were in constant harmony with the environment. However, due to these pressures of modern society, aligning our internal time with geophysical time has become a challenge in today's world. Chronic misalignment - when eating and sleeping patterns conflict with the natural light-dark cycle - is associated with an increased risk of metabolic syndrome, cardiovascular disease, neurological conditions, and cancer. A large portion of the global workforce has atypical hours and may be particularly vulnerable.

"It has become urgent that we uncover the molecular underpinnings of the relationship between the circadian clock and disease," explained Koronowski. "Deciphering the means by which clocks communicate across metabolic organs has the potential to transform our understanding of metabolism, and it may hold therapeutic promise for innovative, noninvasive strategies to promote health."

Credit: 
University of California - Irvine

'Left behind' adolescent women must be prioritised within sustainable development agenda

The needs of millions of overlooked, 'left behind' adolescent women must become a more significant priority within international efforts to end poverty by 2030, a UK Government-commissioned report is urging.

The University of Cambridge report, which was commissioned by the Foreign, Commonwealth and Development Office, argues that there is an urgent need to do more to support marginalised, adolescent women in low and middle-income countries; many of whom leave education early and then face an ongoing struggle to build secure livelihoods.

Amid extensive evidence which highlights the difficulties these women face, it estimates that almost a third of adolescent women in many such countries are not in education, training, or work.

'Adolescents' (technically people aged 10 to 19) comprise about one sixth of the world's population. Women in this age group are some of the most vulnerable people in the world. The report argues that unless more is done to support them, it is unlikely that the UN's 2030 Sustainable Development Goals - which include ending poverty, ensuring inclusive education, and empowering women and girls - will be met.

In particular, the document highlights the need for more concerted efforts to be made to prevent gender discrimination in labour markets, strengthen social safety nets for women, and provide both formal education and continued training for the huge numbers of adolescent women who, it says, 'have missed out on acquiring relevant skills to enhance their livelihood opportunities.'

Professor Pauline Rose, Director of the Research for Equitable Access and Learning (REAL) Centre at the Faculty of Education, University of Cambridge, said: "Marginalised adolescent girls are those who experience extreme poverty, live in rural areas, have disabilities, are affected by conflict, or belong to disadvantaged groups. We need to prioritise these young women both in education and as they transition into work. Millions are being left behind by a range of interlocking problems, and strong, sustained political leadership is needed to turn that around."

The Government has identified girls' education as a key focus of the UK's presidency of the G7 group of industrial countries this year, and gender equality will be mainstreamed across the different ministerial tracks. The new report raises gender inequality - both in education and employment - as major areas of concern for the international community.

The report further stresses that adolescence is a make-or-break time for many girls in low- and lower-middle-income countries and should therefore be a focal point of international efforts. During this period, many young women leave education early, either to work, or because they are expected to marry and start a family. Often, they do so without having acquired basic literacy or numeracy. In addition, very few have the transferable skills or training that they need to succeed in the world of work.

The document draws on more than 150 sources to evidence both the scale of the problem and the nature of the barriers that marginalised adolescent girls face. For many, a quality education remains a far-off dream. In sub-Saharan Africa, for example, fewer than one in 10 girls from poor households in rural areas complete lower secondary education.

Many also struggle to find secure employment. Data from 30 low- and middle-income countries suggests that 31% of young women are not in education, employment or training, compared with 16% of boys. Those who do find jobs frequently work for low wages, in unsafe settings and without any sort of social safety net.

One of the main reasons for this, the report says, is a lack of access to appropriate skills development and training. For example, one in three unemployed adolescent girls in the Asia-Pacific region, and one in five in sub-Saharan Africa, report that the entry requirements for their preferred career path exceed their education and training.

Compounding these problems, gender discrimination in both labour markets and wider society is an accepted norm in many countries. Among many other examples, this manifests itself in inheritance laws which transfer land and property to sons but not daughters; the tendency to force girls who struggle to find work into early marriage and childbearing; and widespread gender-related violence. One study in Nigeria, cited in the report, found that two-thirds of young female apprentices had experienced physical violence - and 39% said that their employer was the most recent perpetrator.

While the research also identifies many successful individual programmes around the world that address some of these issues, it stresses the need for policy-makers internationally to prioritise adolescent girls in larger-scale, systemic reforms.

It makes numerous recommendations about how that can be done, including:

Implementing measures and laws that challenge gender discrimination in education, labour markets and wider society.

Curriculum reforms to develop women's transferrable skills in school, supported by skills development programmes outside the education system.

Catch-up programmes for those who have missed out on a basic education.

Strengthening social safety nets, which have been shown to benefit women in particular.

Providing sexual and reproductive health services and information for all adolescent girls.

Providing counselling and rehabilitation services that offer practical support to adolescent girls who have been forced into unsafe work settings.

The report highlights the particular role that female political leaders and parliamentarians can play in driving forward a more integrated agenda for marginalised young women, and in challenging patriarchal norms that hold back gender equality.

It also warns that many of the trends documented are currently at risk of becoming worse as a result of COVID-19. "The best way that Governments can signal their commitment to this problem is by putting women and girls at the forefront of COVID-19 recovery efforts and ambitions to build back better," Rose said. "It is vital that this includes a strong focus on adolescent girls."

Credit: 
University of Cambridge

Combination of copanlisib with cetuximab improves tumor response

image: Gene set enrichment plot of the EGFR and PI3K-AKT-mTOR signaling pathways for the copanlisib and cetuximab cohorts. (A) Cetuximab-responsive cohort (labeled as "REST' on the plot) presented with a differentially stronger activity of EGFR signaling (MSigDB id M563) compared to the cetuximab-resistant HNSCC samples (p = 0.040; false discovery rate, q = 0.022). (B) Copanlisib-responsive cohort (labeled as "Intermediate' on the plot) was found to harbor a differentially stronger activity of PI3K-AKT-mTOR pathway (MSigDB id M5923) compared to the copanlisib-resistant HNSCC samples (p = 0.03; q = 0.067).

Image: 
Correspondence to - Konrad Klinghammer - konrad.klinghammer@charite.de

Oncotarget published "Combination of copanlisib with cetuximab improves tumor response in cetuximab-resistant patient-derived xenografts of head and neck cancer" which reported that HNSCC is frequently associated with either amplification or mutational changes in the PI3K pathway, making PI3K an attractive target, particularly in cetuximab-resistant tumors.

Here, the authors explored the antitumor activity of the selective, pan-class I PI3K inhibitor copanlisib with predominant activity towards PI3Kα and δ in monotherapy and in combination with cetuximab using a mouse clinical trial set-up with 33 patient-derived xenograft models with known HPV and PI3K mutational status and available data on cetuximab sensitivity.

Combination treatment with copanlisib and cetuximab was superior to either of the monotherapies alone in the majority of the models, and the effect was particularly pronounced in cetuximab-resistant tumors.

While no correlation was observed between PI3K mutation status and response to either cetuximab or copanlisib, increased PI3K signaling activity evaluated through gene expression profiling showed a positive correlation with response to copanlisib.

Together, these data support further investigation of PI3K inhibition in HNSCC and suggests gene expression patterns associated with PI3K signaling as a potential biomarker for predicting treatment responses.

Dr. Konrad Klinghammer from The Charité - Universitätsmedizin Berlin said, "Head and neck squamous cell carcinoma (HNSCC) represents the sixth most common type of cancer with 0.65 million new cases and 0.33 million deaths annually worldwide."

Upregulated PI3K/AKT/mTOR signaling has also been shown to increase resistance to radiotherapy and cytostatic drugs, and PI3K has been defined as an alternate signaling pathway in cetuximab-resistant HNSCC.

Accordingly, based on the cancer genome atlas data, up to 56% of HNSCC display either amplification or mutational changes in the PI3K pathway, and according to the head and neck cancer tissue array data, the PI3K/AKT/mTOR pathway is upregulated in over 90% in both HPV-positive and negative HNSCCs.

Upregulated PI3K/AKT/mTOR signaling, the occurrence of mutational changes in the PI3K pathway, and emerging clinical data from trials with PI3K inhibitors make PI3K an attractive target for cancer therapy, particularly in patients with recurrent/metastatic HNSCC.

Copanlisib is a highly selective, pan-class I PI3K inhibitor with preferential activity against the p110α and p110Δ isoforms that lead to the downregulation of PI3K signaling.

To this end, a panel of altogether 33 HPV-positive or negative HNSCC PDX models were selected based on PI3KCA mutation status and cetuximab sensitivity to evaluate the efficacy of copanlisib given as monotherapy and in combination with cetuximab, and furthermore, the capacity of copanlisib to overcome resistance to cetuximab.

The Klinghammer Research Team concluded in their Oncotarget Research Paper "we demonstrated an improved tumor control by combining copanlisib with cetuximab using 33 patient-derived xenograft models and thereby establishing a preclinical rationale for the evaluation of this combination therapy in a clinical setting. A phase Ib/II study of copanlisib in combination with cetuximab in HNSCC patients harboring a PI3KCA mutation/amplification and/or a PTEN loss is currently ongoing."

Credit: 
Impact Journals LLC

Researchers unravel what makes someone a COVID-19 super-spreader

Scientists and public health experts have long known that certain individuals, termed "super-spreaders," can transmit COVID-19 with incredible efficiency and devastating consequences.

Now, researchers at Tulane University, Harvard University, MIT and Massachusetts General Hospital have learned that obesity, age and COVID-19 infection correlate with a propensity to breathe out more respiratory droplets -- key spreaders of SARS-CoV-2, the virus that causes COVID-19. Their findings were published in Proceedings of the National Academy of Sciences.

Using data from an observational study of 194 healthy people and an experimental study of nonhuman primates with COVID-19, researchers found that exhaled aerosol particles vary greatly between subjects. Those who were older with higher body mass indexes (BMI) and an increasing degree of COVID-19 infection had three times the number of exhaled respiratory droplets as others in the study groups.

Researchers found that 18% of the human subjects accounted for 80% of the exhaled particles of the group, reflecting a distribution of exhaled aerosol particles that follows the 20/80 rule seen in other infectious disease epidemics - meaning 20% of infected individuals are responsible for 80% of transmissions.

Aerosol droplets in nonhuman primates increased as infection with COVID-19 progressed, reaching peak levels a week after infection before falling to normal after two weeks. Notably, as infection with COVID-19 progressed, viral particles got smaller, reaching the size of a single micron at the peak of infection. Tiny particles are more likely to be expelled as people breathe, talk or cough. They can also stay afloat much longer, travel farther in the air and penetrate deeper into the lungs when inhaled.

The increase in exhaled aerosols occurred even among those with asymptomatic cases of COVID-19, said Chad Roy, PhD, corresponding author and director of infectious disease aerobiology at the Tulane National Primate Research Center.

"We've seen a similar increase in droplets during the acute infection stage with other infectious diseases like tuberculosis," Roy said. "It seems likely that viral and bacterial infections of the airway can weaken airway mucus, which promotes the movement of infectious particles into this environment."

The generation of respiratory drops in the airways varies between people depending on their body composition, said lead author David Edwards, PhD, professor of the practice of biomedical engineering at Harvard University.

"While our results show that the young and healthy tend to generate far fewer droplets than the older and less healthy, they also show that any of us, when infected by COVID-19, may be at risk of producing a large number of respiratory droplets," Edwards said.

Credit: 
Tulane University

A language learning system that pays attention -- more efficiently than ever before

Human language can be inefficient. Some words are vital. Others, expendable.

Reread the first sentence of this story. Just two words, "language" and "inefficient," convey almost the entire meaning of the sentence. The importance of key words underlies a popular new tool for natural language processing (NLP) by computers: the attention mechanism. When coded into a broader NLP algorithm, the attention mechanism homes in on key words rather than treating every word with equal importance. That yields better results in NLP tasks like detecting positive or negative sentiment or predicting which words should come next in a sentence.

The attention mechanism's accuracy often comes at the expense of speed and computing power, however. It runs slowly on general-purpose processors like you might find in consumer-grade computers. So, MIT researchers have designed a combined software-hardware system, dubbed SpAtten, specialized to run the attention mechanism. SpAtten enables more streamlined NLP with less computing power.

"Our system is similar to how the human brain processes language," says Hanrui Wang. "We read very fast and just focus on key words. That's the idea with SpAtten."

The research will be presented this month at the IEEE International Symposium on High-Performance Computer Architecture. Wang is the paper's lead author and a PhD student in the Department of Electrical Engineering and Computer Science. Co-authors include Zhekai Zhang and their advisor, Assistant Professor Song Han.

Since its introduction in 2015, the attention mechanism has been a boon for NLP. It's built into state-of-the-art NLP models like Google's BERT and OpenAI's GPT-3. The attention mechanism's key innovation is selectivity -- it can infer which words or phrases in a sentence are most important, based on comparisons with word patterns the algorithm has previously encountered in a training phase. Despite the attention mechanism's rapid adoption into NLP models, it's not without cost.

NLP models require a hefty load of computer power, thanks in part to the high memory demands of the attention mechanism. "This part is actually the bottleneck for NLP models," says Wang. One challenge he points to is the lack of specialized hardware to run NLP models with the attention mechanism. General-purpose processors, like CPUs and GPUs, have trouble with the attention mechanism's complicated sequence of data movement and arithmetic. And the problem will get worse as NLP models grow more complex, especially for long sentences. "We need algorithmic optimizations and dedicated hardware to process the ever-increasing computational demand," says Wang.

The researchers developed a system called SpAtten to run the attention mechanism more efficiently. Their design encompasses both specialized software and hardware. One key software advance is SpAtten's use of "cascade pruning," or eliminating unnecessary data from the calculations. Once the attention mechanism helps pick a sentence's key words (called tokens), SpAtten prunes away unimportant tokens and eliminates the corresponding computations and data movements. The attention mechanism also includes multiple computation branches (called heads). Similar to tokens, the unimportant heads are identified and pruned away. Once dispatched, the extraneous tokens and heads don't factor into the algorithm's downstream calculations, reducing both computational load and memory access.

To further trim memory use, the researchers also developed a technique called "progressive quantization." The method allows the algorithm to wield data in smaller bitwidth chunks and fetch as few as possible from memory. Lower data precision, corresponding to smaller bitwidth, is used for simple sentences, and higher precision is used for complicated ones. Intuitively it's like fetching the phrase "cmptr progm" as the low-precision version of "computer program."

Alongside these software advances, the researchers also developed a hardware architecture specialized to run SpAtten and the attention mechanism while minimizing memory access. Their architecture design employs a high degree of "parallelism," meaning multiple operations are processed simultaneously on multiple processing elements, which is useful because the attention mechanism analyzes every word of a sentence at once. The design enables SpAtten to rank the importance of tokens and heads (for potential pruning) in a small number of computer clock cycles. Overall, the software and hardware components of SpAtten combine to eliminate unnecessary or inefficient data manipulation, focusing only on the tasks needed to complete the user's goal.

The philosophy behind the system is captured in its name. SpAtten is a portmanteau of "sparse attention," and the researchers note in the paper that SpAtten is "homophonic with 'spartan,' meaning simple and frugal." Wang says, "that's just like our technique here: making the sentence more concise." That concision was borne out in testing.

The researchers coded a simulation of SpAtten's hardware design -- they haven't fabricated a physical chip yet -- and tested it against competing general-purposes processors. SpAtten ran more than 100 times faster than the next best competitor (a TITAN Xp GPU). Further, SpAtten was more than 1,000 times more energy efficient than competitors, indicating that SpAtten could help trim NLP's substantial electricity demands.

The researchers also integrated SpAtten into their previous work, to help validate their philosophy that hardware and software are best designed in tandem. They built a specialized NLP model architecture for SpAtten, using their Hardware-Aware Transformer (HAT) framework, and achieved a roughly two times speedup over a more general model.

The researchers think SpAtten could be useful to companies that employ NLP models for the majority of their artificial intelligence workloads. "Our vision for the future is that new algorithms and hardware that remove the redundancy in languages will reduce cost and save on the power budget for data center NLP workloads" says Wang.

On the opposite end of the spectrum, SpAtten could bring NLP to smaller, personal devices. "We can improve the battery life for mobile phone or IoT devices," says Wang, referring to internet-connected "things" -- televisions, smart speakers, and the like. "That's especially important because in the future, numerous IoT devices will interact with humans by voice and natural language, so NLP will be the first application we want to employ."

Han says SpAtten's focus on efficiency and redundancy removal is the way forward in NLP research. "Human brains are sparsely activated [by key words]. NLP models that are sparsely activated will be promising in the future," he says. "Not all words are equal -- pay attention only to the important ones."

Credit: 
Massachusetts Institute of Technology

COVID containment measures in Philadelphia associated with rise in gun violence

image: Jessica H. Beard, MD, MPH, FACS, Assistant Professor of Surgery and Director of Trauma Research at the Lewis Katz School of Medicine at Temple University.

Image: 
Temple University Health System

(Philadelphia, PA) - While the COVID-19 pandemic brought most of the country to a standstill in March 2020, Philadelphia trauma surgeons noticed an alarming trend in the incidence of firearm violence. Instead of decreasing with containment measures, firearm-injured patients were presenting at even higher rates to Temple University Hospital and other trauma centers around the city.

A team led by Jessica H. Beard, MD, MPH, FACS, Assistant Professor of Surgery and Director of Trauma Research at the Lewis Katz School of Medicine at Temple University (LKSOM), sought to determine the magnitude of Philadelphia's increase in firearm violence during the COVID-19 pandemic. They also aimed to understand potential causes of the increase by trying to pinpoint when the increase occurred. They hypothesized that two major events of 2020 could explain the increase in firearm violence in Philadelphia: the enactment of public health policies designed to contain COVID-19 and a national reckoning with systemic racism, including widespread protests sparked by the police killing of George Floyd.

In a new research letter published in JAMA, Dr. Beard and her team used the Philadelphia Police registry data of shooting victims from January 1, 2016 through November 26, 2020 to determine when the spike in firearm violence began. They examined changes in the number of individuals shot per week in the city following the enactment of Philadelphia's first COVID-19 containment policy (closure of non-essential businesses) on March 16, 2020, and following the killing of George Floyd on May 25, 2020.

The team found that there were 7,159 individuals shot in the city overall during the 256 weeks included in the study. Prior to Philadelphia's first COVID-19 containment policy, there was an average of 25 individuals shot per week. In the weeks after that policy was put in place on March 16, 2020, the average number of individuals shot per week climbed to 46. A time-series analysis found that the increase in firearm violence was strongly associated with the enactment of COVID-19 containment policies. The killing of George Floyd was not associated with any significant increase in the rate of gun violence in the study model.

These findings indicate a significant and sustained increase in firearm violence in Philadelphia following enactment of COVID-19 containment policies.

"In the city of Philadelphia, shootings are often geographically concentrated in lower-income communities," Dr. Beard said. "These communities have not only been disproportionately affected by the coronavirus disease itself, but the pandemic and its associated policies have also exacerbated issues that were already present, including unemployment, poverty, structural racism and place-based economic disinvestment, which are empirically tied to firearm violence in Philadelphia. Our research shows that the measures put in place to contain the pandemic for health and safety reasons had a significant and sustained association with increased firearm violence in the city. In addition to mitigating the impact of COVID-19 in Philadelphia, we must also come together to address what is now an epidemic of gun violence in the city."

Credit: 
Temple University Health System

Biomarkers that could help determine who's at risk for severe COVID-19 symptoms

One of the many mysteries still surrounding COVID-19 is why some people experience only mild, flu-like symptoms, whereas others suffer life-threatening respiratory problems, vascular dysfunction and tissue damage. Now, researchers reporting in ACS' Analytical Chemistry have used a combination of metabolomics and machine learning to identify possible biomarkers that could both help diagnose COVID-19 and assess the risk of developing severe illness.

Although some pre-existing conditions, such as diabetes or obesity, can increase the risk of hospitalization and death from COVID-19, some otherwise healthy people have also experienced severe symptoms. As most of the world's population awaits vaccination, the ability to simultaneously diagnose a patient and estimate their risk level could allow better medical decision-making, such as how closely to monitor a particular patient or where to allocate resources. Therefore, Anderson Rocha, Rodrigo Ramos Catharino and colleagues wanted to use mass spectrometry combined with an artificial intelligence technique called machine learning to identify a panel of metabolites that could do just that.

The cross-sectional study included 442 patients who had different severities of COVID-19 symptoms and tested positive by a reverse transcriptase-polymerase chain reaction (RT-PCR) test, 350 controls who tested negative for COVID-19 and 23 people who were suspected of having the virus despite a negative RT-PCR test. The researchers analyzed blood plasma samples from the participants with mass spectrometry and machine learning algorithms, identifying 19 potential biomarkers for COVID-19 diagnosis and 26 biomarkers that differed between mild and severe illnesses. Of the COVID-19-suspected patients, 78.3% tested positive with the new approach, possibly indicating these were RT-PCR false negatives. Although the identified biomarkers, which included metabolites involved in viral recognition, inflammation, lipid remodeling and cholesterol homeostasis, need to be further verified, they could reveal new clues to how SARS-CoV-2 affects the body and causes severe illness, the researchers say.

Credit: 
American Chemical Society

Substance in the blood of pregnant women fights pathological immune reaction

A team of scientists from Immanuel Kant Baltic Federal University with their colleagues from the Institute of Ecology and Genetics of Microorganisms of the Ural Branch of the Russian Academy of Sciences (Perm) studied the effect of trophoblastic β1-glycoprotein in the blood of pregnant women on pro-inflammatory immune cells. Thanks to trophoblastic β1-glycoprotein, a woman's body does not adversely react to the fetus and supports its normal development until birth. It turned out that trophoblastic β1-glycoproteins also suppressed the development of pro-inflammatory lymphocytes and reduced their activity. The results of the work could be used to develop medicinal drugs for pregnancy maintenance and treatment of various autoimmune diseases.

For a pregnant woman's body, a fetus is a source of antigens. However, there is a natural protection mechanism that prevents a mother's immune system from fighting the fetus and helps it adapt to it. This happens thanks to a fine balance between the pro-inflammatory and anti-inflammatory components, namely the Th17 and Treg populations of lymphocytes, respectively. Th17 lymphocytes belong to the group of helpers that support other immune cells and are identified by the presence of the CD4 receptor. On the contrary, Treg lymphocytes suppress the activity of the immune system. Treg cells prevail in the normal course of pregnancy and in case of any issues, the count of Th17 begins to grow.

"An increased share of Th17 is associated with preeclampsia, a condition of pregnant women that causes high blood pressure, edema, and protein in the urine and in severe cases may lead to multi-organ failure. Also, high Th17 can be the reason for preterm delivery, miscarriage, or repeated pregnancy loss of unknown etiology. Even when the course of pregnancy is relatively normal, increased Th17 levels can interfere with the development of a baby's nervous system and lead to higher risks of neuropsychic disorders," explained Larisa Litvinova, the Head of the Center for Immunology and Cellular Biotechnology, BFU.

Th17 immune regulation disorders are the cause of autoimmune diseases such as asthma, psoriasis, rheumatoid arthritis, Crohn's disease (chronic intestine inflammation), multiple sclerosis, and many others. However, these diseases can go into the remission stage in pregnant women. The team concluded that the bodies of pregnant women must produce some active substances to fight the inflammation caused not only by the presence of a fetus but also by an actual disease.

These substances are known as β1-glycoproteins, special protein-carbohydrate molecules that regulate congenital and adaptive (or acquired) immunity. The authors of the work tested the effect of these compounds on CD4+ lymphocytes with the specific receptor. To do so, the team took samples of venous blood from healthy 21-39 years old pregnant women. After that, immune cells were separated from the blood and cultivated together with trophoblastic β1-glycoproteins. The method of its preparation was patented at the Institute of Ecology and Genetics of Microorganisms, Ural Branch of the Russian Academy of Sciences.

It turned out that trophoblastic β1-glycoproteins suppressed the division of CD4+ lymphocytes and the production of cytokines by them. Cytokines are small peptide molecules that act as signal agents in cases of inflammation. Based on these results, the authors suggested using medicinal drugs with trophoblastic β1-glycoproteins against pregnancy complications to save the lives and health of both mothers and children.

Credit: 
Immanuel Kant Baltic Federal University

'Handy pen' lights up when exposed to nerve gas or spoiled food vapors

image: The handy pen's tip changes color when exposed to harmful gases.

Image: 
Adapted from <i>ACS Materials Letters</i> <b>2021</b>, DOI: 10.1021/acsmaterialslett.0c00516

Exposure to some odorless, colorless and tasteless gases, such as nerve agents, can be toxic or even lethal. And having the ability to detect other types of vapors could save people from eating spoiled or rotten food. Easy-to-use portable devices could, therefore, go a long way toward protecting the public. Now researchers reporting in ACS Materials Letters have created a pen-like sensor that changes color when exposed to harmful gases.

Humans can't detect many toxic vapors, such as poisonous nerve agents or volatile amines released from spoiled foods, so a sensor that can notice these gases' very minute concentrations would be useful. Fluorescence-based sensors are a potential solution because they are inexpensive and can reveal trace amounts of compounds. However, some fluorescing compounds clump together once they react with gases, reducing their intensity, and they can require complex fabrication processes. Yet other fluorophores produce more intense light when they are clumped together -- aggregation-induced emission fluorogens (AIEgens). Most of the current detection methods using AIEgens are liquid-based, requiring gases to be dissolved in solution before analysis, and are not easily portable. So, Zhe Jiao, Pengfei Zhang, Haitao Feng, Ben Zhong Tang and colleagues wanted to adapt AIEgens to be integrated into a needle-thin fiber, creating a handheld device whose tip "turns on" in the presence of a particular gas.

The researchers developed two AIEgen-based "handy pens," one for identifying the nerve agent diethyl chlorophosphite (DCP) and the other for amines produced by rotting food. First, they coated silicon dioxide polymer fibers with a thin sol-gel layer to immobilize AIEgens. Next, they added an AIEgen that changes color when it reacts with DCP on one set of fibers, and an AIEgen that reacts with amines on another set. The coated fibers were then placed at the end of a pen-like device with a UV light source inside. The DCP sensor's tip changed from a yellow fluorescence to blue within 30 minutes of exposure to DCP. The amine sensor's tip was initially a mild blue-gray color, but it generated a vibrant yellow-colored light within five minutes when it was exposed to volatile amine vapors. Both sensors reverted to the original hue when exposed to neutralizing vapors, demonstrating that they were reversible. Finally, the team used the amine-responsive handy pen to distinguish between a salmon sample that had been refrigerated and one that had been left at room temperature for 48 hours. The researchers say that other handy pens could be easily developed by using different vapor-sensitive AIEgens, which could be applied to food safety, environmental monitoring or public safety applications.

Credit: 
American Chemical Society

Reductions in CFC-11 emissions put ozone recovery back on track

A potent ozone-depleting chemical whose emissions unexpectedly spiked in recent years has quickly dropped back to much lower levels, putting the recovery of the stratospheric ozone layer back on track, according to a new study by scientists at MIT, the University of Bristol, and other institutions in South Korea, the U.S., Japan, Australia, and Switzerland.

The chemical in question is CFC-11, a chlorofluorocarbon that was once commonly used for refrigeration, insulation, and other purposes. When emitted to the atmosphere, CFC-11 can loft into the stratosphere, where the sun's ultraviolet radiation breaks the chemical down to release chlorine -- a noxious chemical that then eats away at ozone, stripping away the Earth's natural shield against UV rays.

CFC-11 and other chlorofluorocarbons are now banned under the Montreal Protocol, an international treaty under which every country agreed to phase out the chemicals' production and use by 2010. But in 2018, a team of scientists reported a concerning spike in global emissions of the chemical beginning in 2013. In 2019, a second team reported that a significant portion of the emissions could be traced to eastern China, predominately the Shandong and Hebie provinces.

Now, in two papers published today in Nature, the same teams report that global annual emissions of CFC-11 into the atmosphere have declined sharply, by about 20,000 U.S. tons, from 2018 to 2019. The researchers traced a substantial fraction of the global emission reductions to the very same regions of eastern China where they had previously reported the original spike. The results are consistent with evidence that the country has taken successful actions to stamp out illegal production of this ozone-depleting chemical.

"This is tremendously encouraging," says Ronald Prinn, the director of the Center for Global Change Science at MIT and a co-author on both papers. "If emissions of CFC-11 had continued to rise or even just leveled off, there would have been a much bigger problem building up. The global monitoring networks really caught this spike in time, and subsequent actions have lowered emissions before they became a real threat to recovery of the ozone layer."

A brief history of the spike

Both the original spike and subsequent drop in CFC-11 emissions were detected by the researchers using two independent networks.

One is a global monitoring network operated by the National Oceanographic and Atmospheric Administration (NOAA), comprising about 30 stations. Researchers at each site collect air samples and send them to a central laboratory, where the air is analyzed for CFC-11 and many other trace gases. The weekly measurements, from sites around the world, give an accurate average picture of the chemical species circulating in the global atmosphere.

The other network is the Advanced Global Atmospheric Gases Experiment, or AGAGE, an MIT-led effort funded in part by NASA, with more than a dozen monitoring stations situated on coastal and mountain sites around the world. The AGAGE stations take automated on-site measurements of passing air parcels about every hour, monitoring for more than 50 trace gases, including CFC-11, to provide detailed records of the regional and global atmosphere.

In a 2018 Nature report, the researchers analyzed measurements from NOAA and observed that, from 2014 to 2016, global emissions of CFC-11 grew by more than 14,000 U.S. tons a year -- a 25 percent increase from emissions between 2002 and 2012. In a subsequent 2019 Nature report, regional measurements taken by AGAGE stations in Hateruma, Japan, and Gosan, South Korea, along with three-dimensional modeling, showed that about half or more of these emissions came from eastern China, primarily from the factory-heavy Shandong and Hebei provinces.

Following these 2018 and 2019 reports, the scientists continued to track the chemical through the atmosphere, at both global and regional levels.

In the first of the two new Nature papers, they analyze both NOAA and AGAGE global data and report a dramatic turnaround: From 2018 to 2019, CFC-11 annual emissions dropped throughout the global atmosphere by about U.S. 20,000 tons, returning to levels prior to 2012, following the chemical's 2010 global phaseout.

In the second paper, based on AGAGE measurements, the scientists observed that CFC-11 emissions specifically from eastern China hit a peak around 2017. At some point soon afterward, levels began to drop, although the researchers cannot say exactly when the regional turnaround occurred, as the South Korean station sustained typhoon-related damage that resulted in some data gaps. Despite these gaps, the group observed a decline in CFC-11 annual emissions, by about 11,000 U.S. tons from eastern China, through 2019.

As the researchers write in the paper, "it seems that any substantial delay in ozone-layer recovery has been avoided, perhaps owing to timely reporting, and subsequent action by industry and government in China."

"Continuous vigilance"

However, there is still work to be done. While it appears that CFC-11 emissions from eastern China have declined, indicating that significant illegal production of the chemical there has ceased, these emissions only account for roughly half of the global emissions. Where the remainder could have come from is still unknown.

In general, CFC-11 is currently emitted in large amounts through leakages during new production and during subsequent use in refrigeration and manufacture of foams. The chemicals can also leak out from "banks" of old, discarded refrigerators and foams, though at a much slower and more diffuse rate than the rapid regional increase observed in 2013.

"The challenge now is to ask, where's the rest of it coming from?" says Prinn, the TEPCO Professor of Atmospheric Science in MIT's Department of Earth, Atmospheric and Planetary Sciences. "We will need to expand measurements and modeling to identify new sources, and continue to keep watch. Hopefully, emission levels will continue to drop."

Going forward, the scientists hope to add more stations to the AGAGE network, so that they might identify and quantify other regional sources of CFC-11, particularly in rapidly industrializing parts of the world.

"Clearly this story shows that, in order to ensure that countries are adhering to international agreements like the Montreal Protocol, continuous vigilance is required," Prinn says. "You can't stop measuring these chemical species and assume the problem is solved."

Credit: 
Massachusetts Institute of Technology

Creating more sustainable fragrances with biotech

In the face of a changing climate and crop diseases, manufacturers of products containing natural flavors and fragrances are pivoting to a new way to source ingredients. Companies have been partnering with biotechnology firms to manufacture scents and flavors using fermented microbes, which experts say are more sustainable. A new story in Chemical & Engineering News (C&EN), the weekly newsmagazine of the American Chemical Society, details how the industry is brewing up new fragrances.

Although the availability of natural fragrances and flavors like citrus and vanilla is dwindling, the demand for them has increased, writes Senior Business Editor Melody Bomgardner. In recent years, flavor and fragrance companies have been working with the biotech industry to shore up supply chains and avoid issues like seasonality and poor harvests, without having to use synthetic compounds. Fermentation-derived ingredients can be listed as natural in the U.S. and Europe, which appeals greatly to consumers. Biotech firms and major chemical companies are stepping up their production of fermented products while also making the process more efficient and less expensive.

To be commercially viable, engineered microbes need to make a flavor or scent molecule at a much higher concentration than what's found in plants. Once microbial engineers figure out which plant genes they need to get microbes to produce a new flavor or scent molecule, it's relatively easy to make variants of that structure, they say. For example, γ-decalactone, which smells like peaches, is just one of 20 fruity and buttery molecules that can be made from a lactone process. These new fragrance molecules can be used in products such as hard seltzers, laundry detergents and fine perfumes. Another advantage of fermented molecules is consistency of flavor and smell, which is hard to achieve with naturally derived ingredients. With sustainability being a must-have for consumers and manufacturers alike, experts are hopeful that this is a versatile, cost-effective solution.

Credit: 
American Chemical Society

Mobile game that uses implicit learning improved children's short-term food choices

Rates of overweight and obesity in children are rising around the world, with serious long-term consequences for health and health care costs. In prior research, video and mobile games have helped children eat healthier and exercise more. A new study examined how Indian 10- and 11-year-olds' food choices were affected by playing a pediatric dietary mobile game that uses implicit learning--educating players without making them aware of the lessons through innovations in neurocognitive training and immersive technology. The study found that the game significantly improved children's food choices immediately after play.

The study was conducted by researchers at Carnegie Mellon University (CMU), Hofstra University, Johns Hopkins University Center for Communication Programs (CCP), FriendsLearn, The Mithra Trust, Mind in Motion, the Center for Communication and Change - India (CCC-I), and Seethapathy Clinic and Hospital. The randomized controlled trial was designed and conducted by researchers from CCC-I and CCP and the data modeling and analytics were led by CMU. The study appears in JMIR mHealth and uHealth.

"While many factors contribute to overweight and obesity, dietary decisions are a leading cause," explains Rema Padman, trustees professor of management science and healthcare informatics at CMU's Heinz College, who led the study. "Video games that are perceived by children as a fun activity rather than a learning tool present a great opportunity to change children's health behaviors by delivering relevant knowledge implicitly. We are studying such gamified interventions as 'digital vaccine' candidates that have the potential to influence lifestyle behavior changes and lead to better health outcomes." (Digital vaccines are a subcategory of digital therapeutics, which are evidence-based, prevention approaches that use digital technologies, such as gamified applications delivered via mobile devices, to encourage positive behavior.)

Most video games for children use explicit education strategies, such as providing answers, feedback, instructions, or suggestions to players. This study examined how an action video game called fooya!, which uses implicit learning to promote healthy eating and physical activity in children, affected actual food choices. In the game, an avatar fights robots that represent unhealthy foods, and the avatar's speed and body shape vary in response to the type of food it eats.

Evidence of the effects of games that use implicit education strategies on pediatric healthy eating is limited. Using data about the clicks made by players as they played fooya!, researchers analyzed the relationship between patterns of game play and behavioral outcomes related to dietary health.

The study involved 104 children ages 10 and 11 years from three schools in Chennai, India. The children were randomly assigned to a treatment group that played fooya! or a control group that played a board game that did not feature dietary education. Children played the games for 20 minutes each in two sessions. After playing, they were shown three pairs of healthy and unhealthy food items from three categories--drinks (water and a carbonated soft drink), savory snacks (cashews and potato chips), and sweet snacks (raisins and a chocolate bar)--and asked to choose two items to eat.

Children who played fooya! were more likely to choose healthy foods immediately after playing the game, the study found. Children's food choices were not influenced by how many levels of the game they played, as previous research on this topic has found, but by food facts children read while playing the game: Reading more facts about healthy foods was associated with healthier food choices, while reading more facts about unhealthy foods was associated with more unhealthy food choices, a finding the authors called counterintuitive. Nonetheless, children searched for more food facts about healthy food than about unhealthy food, which drove the overall positive effect of playing the game.

"This finding will also influence how we communicate to children about healthy food choices for behavior change," says Uttara Bharath Kumar, technical advisor for social and behavior change at CCP. "It is consistent with what we know from behavioral science that fear and negative communication do not work as well as positive messaging and promoting self-efficacy--the notion that 'you can do it!'"

"Nutrition and lifestyle are at the root of lifelong risk of noncommunicable and infectious diseases," explains Bhargav Sri Prakash, founder & CEO of FriendsLearn, the life science and health technology company that serves as the research translation and innovation partner of the Digital Vaccine Project at CMU. "These findings indicate the potential for societal impact by developing rigorous evidence-based science for 'digital vaccines' based on neurocognitive computing and analytics. As we build a platform of scalable, rich, game-like, engaging experiences, we aim to protect the health of children and families through science."

Among the limitations of the study, the authors note, are its small size, homogeneous groups of children, and short-term food choices, limiting the generalizability of the findings. Longitudinal studies on this subject can help determine longer-term effects, the authors suggest.

"By examining the complex interactions between game-playing patterns and health behaviors, our findings can inform the design and use of more effective mobile games for improving children's dietary health," notes Yi-Chin Kato-Lin, assistant professor of information systems and business analytics at Hofstra University, who collaborated on the study. "For example, video game designers may want to limit the display of unhealthy foods in their games."

Credit: 
Carnegie Mellon University