Your Daily digest for PsyPost – Psychology News Daily Digest (Unofficial)

[Article Digests for Psychology & Social Work]

PsyPost – Psychology News Daily Digest (Unofficial)

 

(https://www.psypost.org/symbiosexuality-new-study-validates-attraction-to-established-couples-as-a-real-phenomenon/) Symbiosexuality: New study validates attraction to established couples as a real phenomenon
Aug 28th 2024, 10:00

We often think of attraction as something that draws us to another individual, but what if that pull not just toward an individual but toward the energy that flows between people in a relationship? Recent research published in the (https://link.springer.com/article/10.1007/s10508-024-02857-x) Archives of Sexual Behavior has identified a phenomenon called symbiosexuality, where individuals find themselves attracted to the dynamics within existing relationships. These findings offer a new perspective on human desire, suggesting that attraction can be more complex than we previously understood.
The concept of symbiosexuality emerged from observations in cultural and academic discourses that some people are drawn to the relationships between others rather than to individuals themselves. Despite this, the phenomenon had remained largely unexplored, with most discussions on human attraction focusing on one-to-one dynamics.
“As part of the polyamorous community, I have heard people talk openly about experiencing attraction to established couples,” explained study author Sally W. Johnston, an adjunct professor of anthropology and sociology at Seattle University. “In a review of the literature, I was surprised to find that this phenomenon has been largely ignored in social and academic discourse. Paradoxically, (https://doi.org/10.1177/13634607221107821) in a previous study (Johnston, 2024), I also found that sexual and romantic relationships between single people and couples (potetnially motivated by symbiosexual attraction) receive a lot of negative attention in polyamorous communities. I wanted to learn more about this understudied attraction.”
The primary source of data for her new study was a larger project known as The Pleasure Study, which was designed to investigate various aspects of gender identity and sexual pleasure. The survey administered in The Pleasure Study consisted of 65 questions that covered a wide range of topics, including gender, sexual orientation, relationship practices, cultural background, education, and specific experiences with couples.
Johnston zeroed in on responses from participants who indicated they had experienced attraction to a couple — defined as an attraction to two people and their relationship together, rather than to each individual separately. Out of the 373 participants in The Pleasure Study, 145 reported this type of attraction.
In addition to analyzing survey data, Johnston conducted in-depth interviews with a subset of these participants. A total of 34 interviewees, who had previously indicated their attraction to couples in the survey, provided richer, more detailed accounts of their experiences. These interviews were conducted via Zoom, recorded, transcribed, and then analyzed for thematic content.
One of the key discoveries was that symbiosexual attraction is experienced by a diverse group of people. While the sample was skewed towards queer and nonmonogamous individuals due to the recruitment methods, the variety within this group was notable. Participants varied widely in age, race, ethnicity, education, and social class, indicating that symbiosexual attraction is not confined to any particular demographic.
“I was most surprised by the large and diverse population of people that reported experiencing this attraction,” Johnston told PsyPost.
There was a particularly high representation of individuals who identified as queer or nonmonogamous, suggesting a possible correlation between these identities and the experience of symbiosexual attraction.
Through the interviews, Johnston uncovered rich descriptions of what participants found attractive about couples. Many described a fascination with the energy and dynamics between two people in a relationship. This included the cohesion, charisma, multidimensionality, and power that couples projected. Participants often spoke of being drawn to the “synergy” or “third force” created by the couple, which was perceived as greater than the sum of its parts. This attraction was distinct from being attracted to the individuals within the couple; it was specifically about the relational dynamic they created together.
Johnston also identified several common themes in the types of relationships that were particularly attractive to those with symbiosexual desires. Participants frequently mentioned being drawn to couples who displayed strong intimacy, high-quality communication, and a playful or sexually open dynamic. Some were specifically attracted to the physical appearance of the couple as a unit, while others were drawn to the inherent queerness or gender diversity within the relationship.
Another significant finding was that, for many participants, symbiosexual attraction was an unexpected and unfamiliar experience. Several described it as something they had not initially recognized or had difficulty articulating. This lack of language and recognition likely contributed to a sense of uncertainty or even hesitation about the validity or “normality” of their attraction. Despite this, for some, symbiosexual attraction was a profound and central aspect of their sexual identity, suggesting that this phenomenon could be an important component of a broader spectrum of human desire.
Johnston’s study also highlighted the potential social stigma and challenges faced by those who experience symbiosexual attraction, particularly within nonmonogamous communities. Individuals who experience this attraction are sometimes referred to as “unicorns”—a term that describes someone who is willing to engage in romantic or sexual relationships with established couples. However, this label, while popular, often oversimplifies the complexity of symbiosexual attraction, which involves a deeper pull toward the energy and dynamics within a relationship.
“There is a diverse population of people who experience symbiosexual attraction, an attraction to the energy, multidimensionality, and power shared between people in relationships,” Johnston said. “Therefore, we need to rethink the nature of human attraction and desire as only one-to-one experiences. We also need to challenge the stigma and discrimination against this attraction within the polyamorous community.”
The findings from this study offer evidence that symbiosexuality is a genuine and significant experience for many people. However, Johnston noted that “while I provide a label for this attraction, ‘symboisexual attraction,’ it is important to note that participants were not offered this label during the study when asked about their experiences of attraction to couples. It is unknown if the label will resonate for every participant.”
Further research is also needed to explore the prevalence of symbiosexuality in the general population and to understand how it intersects with other aspects of identity, such as gender and sexual orientation. It would also be beneficial to investigate the psychological and emotional impacts of symbiosexual attraction. For instance, how does this attraction influence relationship satisfaction or mental health? Are there particular life experiences or personal characteristics that make someone more likely to experience symbiosexual attraction?
“I have a follow-up study under review examining sexual and romantic experiences with couples of people who experience symbiosexual attraction,” Johnston said. “As we learn more about people’s lived experiences with this attraction, it is my hope that those who experience symbiosexual attraction will be able to find both information and validation. In addition, I hope that this work will reduce stigma in both monogamous and non-monogamous communities and expand conceptualizations of desire in sexuality studies.”
Researchers could also examine the relationship between symbiosexual attraction and compersion, which refers to the positive emotions, such as joy, pleasure, or contentment, that some people experience when witnessing or imagining their partner engaging in intimate relations with someone else.
“I believe there is an interesting connection between the experience of compersion (often discussed in polyamorous communities) and symbiosexual attraction,” Johnston said. “More research is needed on these potentially intersecting phenomena.”
The study, “(https://doi.org/10.1007/s10508-024-02857-x) Symbiosexual Attraction: An Integrated Mixed‑Methods Study,” was published April 8, 2024.

(https://www.psypost.org/mothers-adhering-to-healthy-dietary-patterns-at-lower-risk-of-having-children-with-autism/) Mothers adhering to healthy dietary patterns at lower risk of having children with autism
Aug 28th 2024, 08:00

A massive longitudinal study conducted in Norway and England has found that mothers who adhere to a healthy dietary pattern during pregnancy have 22% lower odds of having children diagnosed with autism compared to mothers with less healthy diets. The study also revealed that children of these mothers are less likely to experience social communication difficulties at ages 3 and 8. These findings were recently published in (https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2821339open.2024.22815) JAMA Network Open.
Autism, or autism spectrum disorder, is a developmental condition that manifests in challenges related to social communication, behavior, and sensory processing. The term “spectrum” is used because the symptoms and their severity can vary widely from one individual to another. Some people with autism may experience significant difficulties in communication, while others may have symptoms that are barely noticeable. Common characteristics of autism include repetitive behaviors, a strong preference for routines, and difficulty adapting to changes in the environment. It is estimated that 1 to 2% of the global population is affected by some form of autism.
The causes of autism are not fully understood, but it is believed that both genetic and environmental factors play a role. While genetics are a significant contributor, research has also identified several environmental factors that may be associated with an increased risk of autism in children. Recently, there has been growing interest in the potential link between prenatal dietary patterns and the risk of autism. Previous studies have suggested that women who take multivitamins, folic acid supplements, and consume adequate amounts of vitamin D and fish during pregnancy may have a lower risk of having children with autism.
Catherine Friel and her colleagues conducted this study to explore whether a more comprehensive approach to dietary patterns during pregnancy could be linked to autism in children. To do this, they analyzed data from two large, well-known cohort studies: the Norwegian Mother, Father, and Child Cohort Study (MoBa) and the Avon Longitudinal Study of Parents and Children (ALSPAC).
“I was drawn to this topic because while some studies have explored the relationship between prenatal diet and the risk of autism in offspring, their findings have been inconsistent and often based on small sample sizes. Smaller studies can be less reliable and so there was a clear need to investigate this link within a larger, more robust cohort, such as MoBa,” explained Friel, a dietitian and researcher affiliated with the MRC/CSO Social and Public Health Sciences Unit
at the University of Glasgow.
The MoBa study, conducted by the Norwegian Institute of Public Health, involved 95,200 pregnant women between 1999 and 2008. The ALSPAC study included 14,541 pregnant women from southwest England who were recruited between April 1991 and December 1992. Both studies followed the health and development of the mothers and their children over several years.
The participating women, all of whom had singleton pregnancies, completed detailed food frequency questionnaires. These questionnaires asked them to report how often they consumed various types of foods. The researchers used the responses to classify the women into three groups based on their adherence to a healthy dietary pattern. This pattern was characterized by high consumption of fruits, vegetables, fish, nuts, and whole grains, and low consumption of red and processed meats, sugary drinks, and foods high in fats and refined carbohydrates. The groups were categorized as having high, medium, or low adherence to this healthy dietary pattern.
To assess the potential impact of these dietary patterns, the researchers linked the MoBa data with autism diagnoses recorded in the Norwegian Patient Registry. They also analyzed assessments of social communication difficulties in the children at ages 3 and 8, completed by the mothers in the MoBa study. Similarly, in the ALSPAC study, the mothers completed a 12-item Social and Communication Disorders Checklist when their children were around 8 years old.
The results showed a significant association between high adherence to a healthy dietary pattern during pregnancy and a lower likelihood of autism in children. Specifically, mothers in the highest adherence group had 22% lower odds of having a child diagnosed with autism compared to those in the lowest adherence group.
“Given the mixed results from previous studies, we didn’t have a clear expectation of what we might find,” Friel told PsyPost. “The most surprising aspect was simply that our research uncovered a notable link between a ‘healthy’ prenatal diet and a reduced likelihood of autism in children.”
Additionally, these mothers had a 24% lower chance of having a child with social communication difficulties at age 3, according to the MoBa data. In the ALSPAC cohort, mothers with high adherence to a healthy diet also had 24% lower odds that their children would exhibit social communication difficulties at age 8.
“Our study suggests a potential link between following a ‘healthy’ prenatal dietary pattern and a reduced likelihood of having a child with autism,” Friel explained. “However, it is crucial to understand that this does not establish a cause-and-effect relationship. More research is needed to determine whether maternal diet directly influences the development of autism or if other factors might explain our findings.”
While the findings suggest that a healthy diet during pregnancy may reduce the risk of autism, it is also possible that other factors—such as genetics, socioeconomic status, or overall maternal health—could influence both the likelihood of autism and dietary choices during pregnancy. Another limitation is that the study relied on self-reported dietary data, which can be subject to errors or biases. Participants might not accurately remember or report their food intake, and the food frequency questionnaires used might not capture all aspects of their diet.
Given these limitations, more research is needed to fully understand the relationship between prenatal diet and autism.
“My long-term goal is to determine whether the associations we observed can be replicated in other large cohorts,” Friel said. “If replicated, I aim to further investigate the underlying mechanisms, whether dietary or otherwise, that could explain these findings.”
“To better understand why we observed a link between a healthy prenatal diet and a reduced risk of autism, it will be important to explore alternative explanations. These could include genetic factors, the role of paternal diet, or the impact of diet during childhood, among others. Such exploration could provide a more comprehensive understanding of the factors contributing to autism risk.”
The paper, “(https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamanetworkopen.2024.22815&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2024.22815) Healthy Prenatal Dietary Pattern and Offspring Autism,” was authored by Catherine Friel, Alastair H. Leyland, Jana J. Anderson, Alexandra Havdahl, Anne Lise Brantsæter, and Ruth Dundas.

(https://www.psypost.org/gut-health-tied-to-psychological-resilience-new-research-reveals-gut-brain-stress-connection/) Gut health tied to psychological resilience: New research reveals gut-brain stress connection
Aug 28th 2024, 06:00

Stress has a profound impact on our lives, contributing to a staggering $300 billion in healthcare costs and missed workdays in the United States alone each year. The way we handle stress, known as resilience, can make a significant difference in how we navigate life’s challenges. Now, new research from the University of California, Los Angeles (UCLA) has provided groundbreaking insights into the biology of resilience, revealing that it involves not only the brain but also the gut microbiome, the vast community of microorganisms that live in our digestive tract.
The study, published in (https://www.nature.com/articles/s44220-024-00266-6) Nature Mental Health, found that people who exhibit greater resilience tend to have more activity in brain regions associated with better cognitive functioning and emotional regulation. These resilient individuals were also more mindful, better at describing their feelings, and had different gut microbiome activity compared to those who were less resilient. This connection between the brain and the gut microbiome could pave the way for new interventions to help people manage stress more effectively.
The research team at UCLA was motivated by the increasing evidence linking stress with mental health issues like depression and anxiety. While previous studies have shown that resilience can protect against these conditions, most of the research has focused on psychological traits and social factors. The UCLA researchers wanted to take a different approach by exploring the biological underpinnings of resilience.
By identifying what a “resilient” brain and microbiome look like, the researchers hoped to open up new avenues for treatments that could enhance resilience and reduce the risk of stress-related mental health disorders. According to (https://www.uclahealth.org/departments/medicine/gastro/church-lab) Arpana Church, the senior author of the study and co-director of the (https://www.uclahealth.org/departments/medicine/gastro/microbiome) UCLA Goodman-Luskin Microbiome Center, the ultimate goal is to develop targeted interventions that can help people better cope with stress by focusing on both the brain and the gut.
“We often focus our energy on investigating disease and treating disease. Or even how stress leads to disease,” Church told PsyPost. “But I wanted to know what if we flipped the switch and looked at the other side of the coin, by investigating those individuals who despite experiencing stress do well and do not develop disease? I strongly believe that by pursuing this line of investigation we can think about preventing disease before it even begins.”
The study involved 116 healthy adults from the Los Angeles area. To assess resilience, the researchers used the Connor-Davidson Resilience Scale, a well-validated self-report questionnaire. This scale asks participants to rate how true certain statements are for them, such as “I can handle unpleasant feelings” or “I tend to bounce back quickly after hard times.” Based on their scores, participants were divided into two groups: those with high resilience and those with low resilience.
The participants underwent several types of brain imaging to capture different aspects of brain structure and function. These included high-resolution structural MRI scans to measure the volume and surface area of different brain regions, diffusion MRI to assess the connectivity between brain regions by examining the white matter tracts, and resting-state functional MRI to observe how different regions of the brain communicate with each other when the brain is at rest.
In addition to the brain imaging, participants provided stool samples a few days before their MRI scans. These samples were used to analyze the gut microbiome, focusing on both its composition (which microorganisms are present) and its function (what genes are active and what metabolites are being produced). The stool samples were processed to extract DNA and RNA, allowing the researchers to examine the genetic activity of the microbiome and identify specific microbial functions related to resilience.
In addition to resilience, the study measured several other psychological and behavioral traits using standardized questionnaires. These included measures of anxiety, depression, mindfulness, cognitive abilities (such as memory and attention), and perceived stress. By examining these traits alongside the biological data, the researchers aimed to identify correlations that could shed light on the mechanisms underlying resilience.
The brain imaging data revealed that individuals with high resilience showed greater activity in brain regions associated with emotional regulation and cognitive functioning. Specifically, they had more activity in the anterior cingulate cortex and prefrontal cortex. These regions help control the “fight or flight” response, preventing it from becoming overwhelming and allowing for more adaptive responses to stress.
Moreover, the high-resilience group exhibited different patterns of brain connectivity. They showed stronger connections within the brain’s default mode network (DMN), which is involved in self-referential thinking and daydreaming, but also plays a role in recovering from stress. The DMN’s connections with other brain regions, such as those involved in emotional processing and the brainstem, were more robust in the high-resilience group. This suggests that resilient individuals may have a more coordinated brain network that helps them manage stress more effectively.
The analysis of the gut microbiome revealed significant differences between the high-resilience and low-resilience groups. The microbiomes of the high-resilience group were more likely to produce metabolites that reduce inflammation and strengthen the gut barrier. These metabolites, such as short-chain fatty acids, are known to have anti-inflammatory properties and play a crucial role in maintaining the integrity of the gut barrier—a critical factor in preventing harmful substances from entering the bloodstream and triggering systemic inflammation.
Additionally, the high-resilience group had a gut microbiome that exhibited increased activity in genes related to environmental adaptation, genetic propagation, and metabolism. These findings suggest that a resilient individual’s microbiome is better equipped to handle changes and stressors, potentially contributing to their overall ability to cope with stress. The researchers also noted that the high-resilience group’s microbiome was associated with better gut health, as indicated by the production of metabolites that support a strong and healthy gut barrier.
One of the key strengths of this study is how it integrated the brain and gut data to provide a comprehensive picture of resilience. By using advanced statistical techniques, the researchers were able to identify specific patterns of brain activity, brain structure, and gut microbiome function that distinguished the high-resilience group from the low-resilience group.
For example, they found that certain brain regions’ connectivity patterns were linked to specific gut microbiome activities. This supports the idea of a bidirectional relationship between the brain and the gut, where the gut microbiome can influence brain function and vice versa. The study’s findings suggest that resilience is not just a psychological trait but also a biological one, involving a complex interaction between the brain and gut microbiome.
The study also found that the high-resilience group scored better on several psychological measures. They were less likely to experience anxiety and depression, scored higher on measures of mindfulness, and performed better on cognitive tasks related to memory and attention. These psychological traits were correlated with the biological markers identified in the study, further reinforcing the idea that resilience is a multifaceted phenomenon involving both the mind and body.
“Resilience truly is a whole-body phenomenon that not only affects your brain but also your microbiome and what metabolites that it is producing,” Church said. “When you’re stressed, your brain sends signals to your gut. This can upset the balance of your gut bacteria, leading to digestive issues like stomachaches or changes in your bowel movements. But it goes beyond that—the gut bacteria also send signals back to your brain, which can make you feel more anxious or stressed, creating a loop.”
“This connection is why managing stress is so important. If you’re constantly stressed, it can keep this cycle going, leading to chronic inflammation, weakening your immune system, and increasing your risk of long-term illnesses like heart disease, diabetes, and even mental health conditions like depression. By managing stress, you help maintain a healthy balance in your gut, which in turn supports your overall physical and mental well-being.”
Despite the groundbreaking nature of the study, the researchers acknowledge some limitations. One major limitation is that the study was cross-sectional, meaning it looked at a single point in time rather than following participants over time. This makes it difficult to determine whether the brain and gut differences observed in the study cause resilience or are a result of it. Future studies will need to take a longitudinal approach, tracking changes in resilience, brain function, and gut microbiome activity over time to better understand these relationships.
Another limitation is the study’s focus on healthy adults. While this was important for establishing a baseline understanding of resilience, future research should also include people with mental health conditions like depression and post-traumatic stress disorder. Comparing these groups could help identify how the brain and gut microbiome might differ in individuals who struggle with resilience.
Looking ahead, the researchers are excited about the potential for developing new treatments based on their findings. By targeting both the brain and the gut, it may be possible to create interventions that enhance resilience and prevent mental health conditions before they start. This could include dietary changes, probiotics, or even more advanced therapies like fecal transplants.
“Resilience truly is a whole-body phenomenon,” Church said. “We have this whole community of microbes in our gut that exudes therapeutic properties and biochemicals, so I’m looking forward to building upon this research.” The team plans to explore whether interventions designed to increase resilience can change both brain and gut microbiome activity, potentially offering a new way to boost mental health and well-being.
“While we wait for the development of targeted resilience treatments, eating diets rich in diverse fruits and vegetables is the best way to support a healthy gut microbiome and intern healthy brain,” Church explained. “I talk about the ‘ABC’s’ (always be counting), if we can count and get 30 different vegetables and fruits per week we will be helping maintain a healthy microbiome. Therefore, by making mindful dietary choices, we can help maintain a healthy gut microbiome and support optimal brain function and mental well-being.”
The study, “(https://www.nature.com/articles/s44220-024-00266-6) Stress-resilience impacts psychological wellbeing as evidenced by brain–gut microbiome interactions,” was authored by Eric An, Desiree R. Delgadillo, Jennifer Yang, Rishabh Agarwal, Jennifer S. Labus, Shrey Pawar, Madelaine Leitman, Lisa A. Kilpatrick, Ravi R. Bhatt, Priten Vora, Allison Vaughan, Tien S. Dong, and Arpana Gupta (Church).

(https://www.psypost.org/new-research-shows-dogs-can-smell-your-stress-and-it-affects-their-behavior/) New research shows dogs can smell your stress — and it affects their behavior
Aug 27th 2024, 14:00

Have you ever wondered if your stress affects your dog? A new study published in (https://doi.org/10.1038/s41598-024-66147-1) Scientific Reports suggests that it might. The study found that the scent of human stress can influence dogs’ behavior, potentially making them more cautious and less likely to take risks, even when it comes to something as simple as approaching a food bowl.
Dogs and humans have shared a close bond for thousands of years, with dogs often being called “man’s best friend.” This deep relationship has led dogs to become incredibly attuned to human emotions, reading our body language, facial expressions, and even our tone of voice. However, researchers wanted to explore another, less obvious way dogs might be picking up on our emotions—through their sense of smell. Specifically, they sought to determine whether dogs could detect the smell of human stress and how it might affect their behavior.
The idea of “emotional contagion,” where individuals in a group can pick up on and mirror the emotions of others, is well-documented in humans. But how does this concept apply to the human-dog relationship? Researchers theorized that if dogs could detect stress through smell, it might influence their mood and decision-making. This study is the first to test how the odor of human stress affects dogs’ learning and emotional state.
“Most dog owners will tell you their dogs can sense when they’re stressed or upset. When interpreting our emotions, dogs likely use a combination of cues, such as our body language, voice tone, etc., but we didn’t know if they responded only to scent, without behavior or vocal cues present,” said study author Zoe Parr-Cortes, a qualified veterinarian and PhD student in clinical veterinary science at the University of Bristol and Cardiff University.
“We also know that dogs can be trained to detect changes in cortisol in humans. For example, dogs trained by Medical Detection Dogs in the UK can detect when a person’s cortisol levels in people with Addison’s disease drop too low. So, since cortisol tends to increase during stress, we wondered if dogs could detect cortisol during stress and further how the smell affects dogs’ own emotions.”
“What’s interesting about this study is that these odors were from unfamiliar people, so it suggests a common ‘stress’ smell among individuals that dogs can identify, rather than just learning the smell of their stressed owner from previous experience.”
The researchers conducted a two-phase study involving both human and canine participants. The first phase focused on collecting odor samples from human volunteers, while the second phase involved testing dogs with these odors to see how they would react.
Eleven volunteers were recruited to provide odor samples under two conditions: stress and relaxation. During the stress condition, participants were subjected to a well-known stress-inducing test, which involved public speaking and mental arithmetic in front of evaluators. In the relaxation condition, they watched calming videos of nature scenes. Throughout both sessions, the participants wore cotton cloths under their arms to absorb sweat and exhaled breath, which were later used as odor samples.
The second phase involved 18 dogs of various breeds and ages. These dogs were put through a cognitive bias test, a type of test that assesses an animal’s emotional state by measuring their response to ambiguous stimuli. In this case, the ambiguous stimuli were bowls placed at different locations in a testing room, some of which previously contained food (positive location) and others that did not (negative location). The key question was how quickly and eagerly the dogs would approach the bowls when they were placed in new, intermediate locations—closer to the positive, negative, or somewhere in between.
Each dog was exposed to three conditions over three separate sessions: a stress odor from a human participant, a relaxed odor from the same participant, and a control condition with no added odor. The researchers measured how long it took for the dogs to approach the bowls and whether they approached them at all, particularly focusing on the bowls placed in ambiguous locations.
The study revealed that dogs were less likely to approach a bowl placed in a location near where they had previously found no food (near-negative location) when exposed to the stress odor compared to when there was no odor or a relaxed odor. This suggests that the smell of human stress made the dogs more cautious or risk-averse, even though the bowl might have contained food.
“While we found that being stressed around your dog may affect their mood and how they learn and respond to training, we found that the relaxed odor did not have this effect,” Parr-Cortes told PsyPost. “So being aware of this and keeping calm or even doing a relaxing activity before training your dog could reduce this effect if you’re stressed.”
“We also know that using positive training with lots of rewards is good for your relationship with your dog and can have positive effects on their welfare. So, keeping your relationship with your dog based on positive reinforcement and fun engagement is the best way to keep your dog happy and look after their well-being.”
Interestingly, this effect was most pronounced during the third session when the stress odor was introduced after the dogs had already been exposed to the relaxed odor. This finding hints at a possible interaction between the odor and the learning process, as the dogs seemed to become increasingly cautious as the testing progressed.
“Originally, we were just looking at how the smell of stress affected the dog’s cognitive bias (optimism/pessimism),” Parr-Cortes said. “So, when we found an effect of the stress odor on dogs’ learning and how they perceive rewards as well, it was an interesting additional finding.”
The researchers also found that dogs were quicker to approach the positive location (where they expected to find food) and slower to approach the negative location (where they expected no food) as the sessions went on. This suggests that dogs were not only influenced by the odor but also by their previous experiences during the testing, learning to differentiate between the likely and unlikely rewards.
But the study, like all research, does have some limitations. First, the sample size of both human and dog participants was relatively small, which means that the results may not be fully generalizable to all dogs. Future studies with larger groups could help confirm and expand on these findings.
“As with many first studies, the sample size could always be bigger,” Parr-Cortes said. “So, it would be great to repeat this study with more dogs. Nonetheless we’re extremely pleased we were able to show differences with the odors we tested. But it opens the door to so many new questions and avenues for future research.”
Additionally, the study focused on unfamiliar human odors, so it remains unclear how dogs might react to stress odors from their owners, with whom they have a closer bond. Future research could explore how familiar versus unfamiliar stress odors affect dogs differently.
The findings open up several interesting avenues for future research. “Moving forward, we would like to look at the effect of other emotions, such as happiness or deep relaxation, and also look at the effect of repeated exposure to odors of stress and other emotions rather than just a one-off test,” Parr-Cortes explained.
“While the findings may seem negative, I think it’s an important reminder of how important dogs are in our lives, as well as how important we are in our dog’s lives,” she added. “This research should make us consider how we interact with and act around dogs and should encourage people to be more compassionate with the dogs they live or work with. It’s an opportunity to see things from the dog’s perspective and ensure our interactions with our dogs are as positive as possible so we can minimize any negative effects our stress or emotions have on our dogs.”
The study, “(https://www.nature.com/articles/s41598-024-66147-1) The odour of an unfamiliar stressed or relaxed person affects dogs’ responses to a cognitive bias test,” was authored by Z. Parr-Cortes, C. T. Müller, L. Talas, M. Mendl, C. Guest, and N. J. Rooney.

(https://www.psypost.org/intersectional-hallucinations-the-ai-flaw-that-could-lead-to-dangerous-misinformation/) “Intersectional hallucinations”: The AI flaw that could lead to dangerous misinformation
Aug 27th 2024, 12:00

When you go to the hospital and get a blood test, the results are put in a dataset and compared with other patients’ results and population data. This lets doctors compare you (your blood, age, sex, health history, scans, etc) to other patients’ results and histories, allowing them to predict, manage and develop new treatments.
For centuries, this has been the bedrock of scientific research: identify a problem, gather data, look for patterns, and build a model to solve it. The hope is that Artificial Intelligence (AI) – the kind called (https://en.wikipedia.org/wiki/Machine_learning) Machine Learning that makes models from data – will be able to do this far more quickly, effectively and accurately than humans.
However, training these AI models needs a LOT of data, so much that some of it has to be synthetic – not real data from real people, but data that reproduces existing patterns. Most synthetic datasets are themselves generated by Machine Learning AI.
Wild inaccuracies from (https://theconversation.com/if-ai-image-generators-are-so-smart-why-do-they-struggle-to-write-and-count-208485) image generators and (https://theconversation.com/eat-a-rock-a-day-put-glue-on-your-pizza-how-googles-ai-is-losing-touch-with-reality-230953) chatbots are easy to spot, but synthetic data also produces hallucinations – results that are unlikely, biased, or plain impossible. As with images and text, they can be amusing, but the widespread use of these systems in (https://theconversation.com/ai-bias-the-organised-struggle-against-automated-discrimination-223988) all areas of public life means that the potential for harm is massive.
What is synthetic data?
AI models need much more data than the real world can offer. Synthetic data provides a solution – generative AI examines the statistical distributions in a real dataset and creates a new, (https://research.ibm.com/blog/what-is-synthetic-data) synthetic one to train other AI models.
This synthetic ‘pseudo’ data is similar but not identical to the original, meaning it can also ensure privacy, skirt data regulations, and be freely shared or distributed.
Synthetic data can also supplement real datasets, making them big enough to train an AI system. Or, if a real dataset is biased (has too few women, for example, or over-represents cardigans instead of pullovers), synthetic data can balance it out. There is ongoing debate around how far synthetic data can stray from the original.
Glaring omissions
Without proper curation, the tools that make synthetic data will always over-represent things that are already dominant in a dataset and (https://arxiv.org/abs/2109.11429) under-represent (or even omit) less common ‘edge-cases’.
This was what initially sparked my interest in synthetic data. (https://press.uchicago.edu/ucp/books/book/chicago/I/bo5414954.html) Medical research already under-represents women and other minorities, and I was concerned that synthetic data would exacerbate this problem. So, I teamed up with a machine learning scientist, (https://liu.se/en/employee/sagha08) Dr Saghi Hajisharif, to explore the phenomenon of disappearing edge-cases.
In (https://link.springer.com/article/10.1007/s00146-024-02017-8) our research, we used (https://theconversation.com/a-brief-history-of-ai-how-we-got-here-and-where-we-are-going-233482) a type of AI called a GAN to create synthetic versions of 1990 US adult census data. As expected, edge-cases were missing in the synthetic datasets. In the original data we had 40 countries of origin, but in a synthetic version, there were only 31 – the synthetic data left out immigrants from 9 countries.
Once we knew about this error, we were able to tweak our methods and include them in a new synthetic dataset. It was possible, but only with careful curation.
‘Intersectional hallucinations’ – AI creates impossible data
We then started noticing something else in the data – (https://link.springer.com/article/10.1007/s00146-024-02017-8) intersectional hallucinations.
(https://chicagounbound.uchicago.edu/uclf/vol1989/iss1/8/) Intersectionality is a concept in gender studies. It describes (https://www.jstor.org/stable/10.1086/669608) power dynamics that produce discrimination and privilege for different people in different ways. It looks not just at gender, but also at age, race, class, disability, and so on, and how these elements ‘intersect’ in any situation.
This can inform how we analyse synthetic data – all data, not just population data – as the intersecting aspects of a dataset produce complex combinations of (https://www.dukeupress.edu/the-body-multiple) whatever that data is describing.
In our synthetic dataset, the statistical representation of separate categories was quite good. Age distribution, for example, was similar in the synthetic data to the original. Not identical, but close. This is good because synthetic data should be similar to the original, not reproduce it exactly.
Then we analysed our synthetic data for intersections. Some of the more complex intersections were being reproduced, too. For example, in our synthetic dataset, the intersection of age-income-gender was reproduced quite accurately. We called this accuracy ‘intersectional fidelity’.
But we also noticed the synthetic data had 333 datapoints labelled “husband/wife and single” – an intersectional hallucination. The AI had not learned (or been told) that this is impossible. Of these, over 100 datapoints were “never-married-husbands earning under 50,000 USD a year”, an intersectional hallucination that did not exist in the original data.
On the other hand, the original data included multiple “widowed females working in tech support”, but they were completely absent from the synthetic version.
This means that our synthetic dataset could be used for research on age-income-gender questions (where there was intersectional fidelity) but not if one were interested in “widowed females working in tech support”. And one should watch out for “never-married-husbands” in the results.
The big question is: where does this stop? These hallucinations are 2-part and 3-part intersections, but what about 4-part intersections? Or 5-part? At what point (and for what purposes) would the synthetic data become irrelevant, misleading, useless or dangerous?
Embracing intersectional hallucinations
Structured datasets exist because the relationships between the columns on a spreadsheet tell us something useful. Remember the blood test. Doctors want to know how your blood compares to normal blood, and to other diseases and treatment outcomes. That is why we organise data in the first place, and have done for centuries.
However, when we use synthetic data, intersectional hallucinations are always going to happen because the synthetic data must be slightly different to the original, otherwise it would simply be a copy of the original data. Synthetic data therefore requires hallucinations, but only the right kind – ones that amplify or expand the dataset, but do not create something impossible, misleading or biased.
The existence of intersectional hallucinations means that one synthetic dataset cannot work for lots of different uses. Each use-case will need bespoke synthetic datasets with labelled hallucinations, and this needs a recognised system.
Building reliable AI systems
For AI to be trustworthy, we have to know which intersectional hallucinations exist in its training data, especially when it is used to predict how people will act, or to regulate, govern, treat or police us. We need to ensure they are not trained on dangerous or misleading intersectional hallucinations – like a 6 year old medical doctor receiving pension payments.
But what happens when synthetic datasets are used carelessly? Right now there is no standard way to mark them, and they are often mixed up with real data. When a dataset is shared for others to use, it is impossible to know if it can be trusted, and to know what is a hallucination and what is not. We need clear, universally recognisable ways to identify synthetic data.
Intersectional hallucinations may not be as amusing as (https://theconversation.com/if-ai-image-generators-are-so-smart-why-do-they-struggle-to-write-and-count-208485) a hand with 15 fingers, or recommendations to put (https://theconversation.com/eat-a-rock-a-day-put-glue-on-your-pizza-how-googles-ai-is-losing-touch-with-reality-230953) glue on a pizza. They are boring, unsexy numbers and statistics, but they will affect us all – sooner or later, synthetic data is going to spread everywhere, and it will always, by its very nature, contain intersectional hallucinations. Some we want, some we don’t, but the problem is telling them apart. We need to make this possible before it is too late.
 
This article is republished from (https://theconversation.com) The Conversation under a Creative Commons license. Read the (https://theconversation.com/intersectional-hallucinations-why-ai-struggles-to-understand-that-a-six-year-old-cant-be-a-doctor-or-claim-a-pension-234879) original article.

Forwarded by:
Michael Reeder LCPC
Baltimore, MD

This information is taken from free public RSS feeds published by each organization for the purpose of public distribution. Readers are linked back to the article content on each organization's website. This email is an unaffiliated unofficial redistribution of this freely provided content from the publishers. 

 

(#) unsubscribe from this feed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.clinicians-exchange.org/pipermail/article-digests-clinicians-exchange.org/attachments/20240828/a1072a52/attachment.htm>