Almutairi, J. S., Alshewier, S. A., & Alkathiry, A. A. (2025). Ranking the difficulty of the cognitive tasks in Dual-Tasks during walking in healthy adults. Neurosciences Journal, 30(2), 124-130.

 What They Did

The researchers recruited 26 participants for a study of the effect of cognitive tasks on postural sway while walking. (Postural sway refers to the subtle movements used to maintain balance while standing or moving. Everyone has some sway, but a large amount can indicate problems with balance.) Participants first walked 10 meters in a straight line at a comfortable pace to get a baseline for their speed and sway using smartphone accelerometers; this was repeated three times for each participant.

In the next phase, participants were randomly assigned cognitive tasks while walking. The tasks included: listing the months in reverse order, naming category members that begin with a given letter, participating in typical question-answer conversation, counting backwards from 100 by 3’s, spelling five-letter words backwards, and indicating whether a number is larger or smaller than 50. Participants also rated the difficulty of each task on a scale of 0 to 10. On average, the conversation task was perceived by participants as the easiest, and the tasks of counting or spelling backwards were the most difficult. The other three tasks were in-between and were all about the same difficulty.

The researchers did not find a strong overall correlation between task or perceived difficulty and sway. Although the relationships between perceived difficulty and measurement of sway were significantly correlated for some tasks, the direction of correlation varied. Furthermore, each task showed a significant correlation between difficulty and sway for only one of four sway measurements, and the correlated measurement varied across tasks.

 

Further Exploration

Although the correlations between difficulty and sway were minimal, the researchers noted that the difficulty ranking could have clinical applications. When patients have rehabilitation for walking, clinicians increase the challenge by asking them to perform cognitive tasks. The study suggests that basic conversations should be the first challenge, followed by listing the months backwards, listing items that start with a given letter, or determining whether a number is greater or less than 50. Later, the clinician can challenge the patient to count or spell backwards. The researchers also note that, because the participants had no difficulty walking, the effects of cognitive demands on sway may have been too small to detect. A sample of people with at least minor mobility difficulties might be a better source of information.

The difficulty ratings for the cognitive tasks generally match my own. I’d definitely find the conversation the easiest, probably followed by the listing task, then the classification of a number as over or under fifty, followed by reciting the months backwards, then counting backwards by 3’s, then finally spelling words backwards.

It feels like the difficulty correlates with how much I have to visualize to complete the task. Conversation and list making don’t require any visualization, but spelling a word backwards means holding an image of the word in mind and “reading off” the letters in reverse order. I’m curious about how my sense of the reasons for task difficulty aligns with empirical evidence, but that’s a rabbit hole for another day!

Image credit: Ducky2315

https://commons.wikimedia.org/wiki/File:Walking_gait_cycle.png

Morejón-Arrojo, Ramón D., Florian Lüskow, Alfredo Fernández-Alías, Humberto Ramírez, and Aldo Cróquer. "First Record of a Cannonball Jellyfish Bloom (Stomolophus sp.) in Venezuelan Waters." Journal of Marine Science and Engineering 13, no. 4 (2025): 689.

 What They Did

While on a diving trip off the coast of Venezuela, one of the researchers encountered several unusual jellyfish. These were photographed and identified as cannonball jellyfish, genus Stomolophus, and were found at eight other locations over 120 km of the coast. Observed population densities were about 3 jellyfish/m³. Based on information from local fishers and divers, the researchers determined that the population bloom had begun on March 5, 2024, and it continued through late April of that year.

They downloaded environmental data from monitoring organizations for the period from September 2023 to August 2024 and examined environmental conditions at the time the bloom occurred and the likely time of stroblilation (the process by which one or more immature free-living jellyfish bud off of a stationary polyp). The researchers found that the period of likely strobilation was correlated with decreasing water temperature, high chlorophyll a in the water, high rainfall, and low salinity. Cooler temperatures are known to trigger strobiliation, and the large amount of chlorophyll likely meant that food was abundant for the immature jellyfish. The large amount of rain could have increased nutrient runoff into the water, further increasing food availability, and also decreased the salinity.

It therefore seems likely that strobilation occurred between November 2023 and January 2024, as would be predicted from the time of the bloom.  The authors note, however, that polyps for this genus have not been observed in the wild, so it is also possible that the immature jellyfish may have migrated from elsewhere.

 

Further Exploration

I was surprised to learn that the polyps haven’t been observed in the wild. Apparently, members of the genus are bred from polyps in captivity (see https://www.raisingpetjellyfish.com/blue-cannonball), so we know they exist. Cannonball jellyfish aren’t typically found on the coast of Venezuela, however, being more common along the Gulf of Mexico and the southern Atlantic coast of the United States.

They’re frequently eaten in Asia and are becoming an income source for fishers during the off-season for vertebrate fish species. The jellyfish are typically sold partially dried, then soaked and rinsed before eating. They’re described as both crunchy and chewy in texture but lacking much flavor, instead absorbing the taste of any added sauces (see https://www.seaaroundus.org/magazines/2011/INFOFISHInternational_AreJellyfishTheFoodOfTheFuture.pdf).

The researchers suggest that the range of cannonball jellyfish might be increasing, possibly due to heating of the ocean: the Caribbean has warmed by 0.5 °C every decade since the 1980s, which would mean an increase of about 2.5 °C by now. That’s a scary statistic considering the U.N. goal of keeping atmospheric warming to 1.5 °C (see https://www.un.org/en/climatechange/science/climate-issues/degrees-matter.) Water has a higher specific heat than air, meaning it takes more energy to heat water the same number of degrees. The ocean has been absorbing a lot of the heat from the greenhouse effect (see https://unric.org/en/global-warming-90-of-emissions-heat-absorbed-by-the-ocean/), but it can’t do that forever, and doing so affects the ocean ecosystems. I’m not clear on how increasing the ocean temperature would expand the range of the jellyfish from cooler to warmer areas, but that’s a rabbit hole for another day!   

Image Credit: Cape Hatteras National Seashore, public domain

https://commons.wikimedia.org/wiki/File:Cannonball_jellyfish_washed_ashore_near_Ramp_59._It%27s_not_common_to_see_Cannonball_Jellyfish_around_this_time_of_year_-_this_one_may_have_been_cold-stunned._(51759006613).jpg

Doru, B., Maier, C., Busse, J. S., Lücke, T., Schönhoff, J., Enax-Krumova, E., ... & Tokic, M. (2025). Detecting artificial intelligence–generated versus human-written medical student essays: Semirandomized controlled study. JMIR Medical Education, 11, e62779.

 What They Did

The researchers had medical and humanities experts identify which of a set of German-language term papers were written by medical students and which were generated by ChatGPT 3.5, as well as rate several aspects of each paper. Each participant had a week to make their decisions, though they were instructed not to discuss the project. Participants identified student or AI authorship of the papers 70% of the time, with no significant differences between the medical or humanities experts, nor correlation with participant traits such as experience in academia, experience with ChatGPT, or knowledge of the subject matter.

For papers that were correctly identified, the medical experts rated the student papers as having better language use, logic, and scientific approaches, while the humanities experts gave the correctly-identified student papers better ratings on scientific approaches but not other traits. Medical experts also gave student papers better ratings on citation of sources, even when they incorrectly identified them as AI-generated. When papers were correctly identified, medical experts rated student papers better at suggesting new research directions, but when they were incorrectly identified, the AI papers were rated better.

In follow-up interviews, participants frequently indicated that they identified papers as AI-generated because they were redundant, repetitive, or lacked a sense of coherence. The researchers point out that although the participants relied heavily on linguistic style to distinguish between the student and AI-generated papers, it is not clear how effective they would be if they didn’t already know that one of the papers was AI-generated.

Further Exploration

Difficulty distinguishing between papers written by students and those generated by AI poses a major challenge to education, and advances in AI will likely make it even harder. AI-generated books have appeared for sale on Amazon, including a mushroom-foraging guide that could pose a real danger with inaccurate information (see https://www.theguardian.com/technology/2023/sep/01/mushroom-pickers-urged-to-avoid-foraging-books-on-amazon-that-appear-to-be-written-by-ai.) At the same time, students may be falsely accused of cheating with AI and have trouble proving otherwise (see https://odsc.medium.com/ai-detectors-wrongly-accuse-students-of-cheating-sparking-controversy-7afb2ea7edc8).

The prompts the researchers used for generating papers with ChatGPT simply told the program to write sections of the paper one at a time with citations. I imagine that a tech-savvy student could edit such a paper to sound more human without having to know much about the topic. At the same time, using natural langue to prompt an AI can make it more effective than a traditional thesaurus, and students can also use AI to help organize and structure their own ideas. These uses of AI seem akin to the use of a calculator in a math class.

More complex is the question of what constitutes fair use of written material in building an AI. On one hand, the major AIs have been built using copyright material without permission (see https://authorsguild.org/advocacy/artificial-intelligence/faq/.) On the other hand, my own human “brain soup” is full of fragments of books I’ve read and ideas I’ve absorbed without the ability to truly credit all of the sources. Figuring out what’s ethical in this sense is a huge question, but that’s a rabbit hole for another day!

Image credit: Salino01

https://commons.wikimedia.org/wiki/File:Literatursuche_mit_KI_(3).jpg
 

Kolff, Kayla, and Simone Pika. "Turn-taking in grooming interactions of chimpanzees (Pan troglodytes schweinfurthii) in the wild: the role of demographic and social factors." Animal Cognition 28, no. 1 (2025): 26.

 What They Did

The researchers observed and filmed several hundred grooming-based interactions among 42 male chimpanzees to examine the impact of age, dominance, and closeness of relationship on turn-taking during grooming sessions. Closeness of social relationships was quantified based on the relative frequency with which pairs of individuals engaged in grooming together or were in close physical proximity. To determine the dominance hierarchy, the researchers analyzed the frequency with which individuals had clearly lost fights with other chimpanzees and had engaged in submissive behaviors.

Turn-taking transitions were classified into four types: action-action, in which A grooms B, then B grooms A; action-signal, in which A grooms B, then B responds with a gesture; signal-action, in which A makes a gesture and B responds by grooming; and signal-signal, in which A makes a gesture and B responds with another gesture.

They found that turn-taking transitions were most likely when the individual initiating the transition was older or when the responding individual was younger or lower ranked. Younger chimpanzees were more likely to respond to grooming by grooming the other chimpanzee in return. Transitions in which grooming by one chimpanzee was followed by a gesture from another were most likely to occur when the individual that initiated grooming was younger. Lower ranked individuals were the most likely to respond to a gesture by grooming the initiator; this interaction pattern was also more likely between individuals with weak social bonds. Finally, older chimpanzees were the most likely to respond to gestures with gestures of their own.

 

Further Exploration

I was surprised that the chimpanzees were more likely to respond to a gesture requesting grooming when they had a weaker social bond with the initiator. I’m more likely to respond to requests from people I’m close to than from people I don’t know, and I don’t think I’m terribly unusual in that regard. On the other hand, the chimpanzee groups only consist of 100 to 200 individuals. In a group that small, no one is exactly a stranger.

The researchers also suggest that the process of requesting and receiving grooming may strengthen the social bonds, allowing for future benefits, and that reciprocal grooming may occur more naturally in strongly bonded pairs, without the need for a gesture of request. Both of these potential explanations make sense to me. Trying to expand my social circle with new friends involves more tentative bids for attention than starting up a conversation with someone I’m already close to.

It’s interesting, though, that the data didn’t show a higher rate of action-action transitions (i.e. A grooms B, then B grooms A) for more closely bonded pairs, since that would seem to reflect the more natural occurrence of reciprocal grooming. One possibility is that the reciprocity in grooming among closely bonded pairs is more spread out in time: perhaps A grooms B for a while, then a few hours or a day later, B grooms A. I’m sure one could design a study to explore that hypothesis, but that’s a rabbit hole for another day!

Image credit: Craig Hamnett

https://commons.wikimedia.org/wiki/File:Chimps_Grooming_(2394016065).jpg

Winiger, F., Schneider, G., Goldzycher, J., Neuhold, D., & Peng-Keller, S. (2025). The ‘Spiritual’ and the ‘Religious’ in the Twittersphere: A Topic Model and Semantic Map. Journal of Religion, Media and Digital Culture, 14(1), 1-22.

 What They Did

The researchers collected 138,000 Twitter posts containing the words “religion” or “spirituality” between March and April of 2021. They used language analysis software to generate seven clusters of words associated with “religion” or “spirituality,” label each post as positive or negative in sentiment, and generate a concept map of 200 words frequently appearing in the data. Three of the authors (an anthropologist, a church historian, and a theologian) separately examined the clusters and suggested names to capture the overall theme of each cluster. The final cluster names were theistic/Western, mystical/subjective/practices, Islam, Christian/Evangelical, politics, conduct of life, and Indian.

The mystical/subjective/practices cluster and the conduct of life cluster were associated with positive sentiment, and the Islam cluster also had a slight positive association. The other clusters were associated with negative sentiment. The theistic/Western cluster and the Indian cluster were associated with “spirituality” and “religion” at near equal strengths. The Christian/Evangelical cluster and the conduct of life cluster were also associated with both terms but had a stronger association with “spirituality.” The mystical/subjective/practices cluster and the Islam cluster were only associated with “spirituality,” and the politics cluster was only associated with “religion.”

The researchers conclude that their data is consistent with a linguistic differentiation between religion as negative, morally prescriptive, and public versus spirituality as positive, life-enhancing, and private. They note, however, that the methodology used only provides a broad-stroke picture of the data and that more granular examination of the Twitter posts reveals varied and sometimes contradictory usage of both terms.

 

Further Exploration

The concept map generated for this paper was made from the most frequent 200 words in the data set, excluding English “stop words.” That was a new term for me, and it refers to words that appear frequently but don’t carry very much meaning (see https://smltar.com/stopwords.) Obviously, leaving those out allows for generation of a much more interesting concept map!

On the concept map itself, I noticed that the words closest to “spirituality” were “healing,” “connection,” and “soul,” while the words closest to “religion” were “freedom,” “politics,” and “race.” This seems consistent with the sense of “spirituality” as more private and “religion” as more public. At the same time, “freedom of religion” is a common phrase describing the U.S. first-amendment right, and “politics and religion” is a common phrase grouping both topics as inappropriate for general conversation, so I wonder if that could’ve skewed the results.

As far as I gathered from the paper, the Twitter posts specifically used the words “spirituality” or “religion,” not any related words like “spiritual” or “religious.” That means the phrase “spiritual but not religious” wouldn’t show up. It also seems that the words aren’t exactly grammatically equivalent. I might refer to “a religion” but not usually to “a spirituality” in casual conversation. Instead, I would refer to a spiritual “path,” “practice,” or “tradition.” I could also refer to a religious path, practice, or tradition. That makes me wonder if using the adjective forms would give different results, but that’s a rabbit hole for another day!

Image credit: Nancystodd

https://commons.wikimedia.org/wiki/File:Icon-religion.svg

Nong, Junqing, Ruijing Xu, Songpo Wei, Shaohui Fan, Quan Qiu, Yiju Li, and Guanglu Liu. "Geographical variation and the role of climate and soil on phenotypic traits of Calamus rhabdocladus across provenances in China." BMC Plant Biology 25, no. 1 (2025): 365.

 What They Did

The researchers collected samples of Calamus rhabdocladus, a woody vine in the palm family, from ten locations across China. They measured 15 plant traits and examined how they correlated with climate and soil conditions, as well as the degree to which each trait varied among locations. They found that the dry weight of the leaves, stems, and roots, as well as the volume of the flagellum (a whiplike climbing structure) varied the most, suggesting that the plant adapts to its environment by adjusting the relative growth of the different organs.

Plants from areas with lower temperatures had more root growth. Because lower temperatures can limit the growth rate of above-ground plant parts, the plant may need to put more resources into root growth as a survival adaptation. The chlorophyll content was higher in areas with more precipitation, possibly as an adaptation to the lower light conditions associated with rain.

In areas with more organic material in the soil, the specific leaf area (one-side area of leaf / leaf dry weight) was larger, suggesting a faster growth rate. The flagellum, however, was smaller, possibly because the plant has less need to climb over other plants when soil fertility is high. Higher soil nitrogen content was correlated with higher dry matter percentages in the above-ground parts of the plant, perhaps because the availability of nutrients provides an opportunity for the plant to generate additional tissue. Overall, however, the climate variables affected the differences in plant traits more than the soil variables did.

 

Further Exploration

In this paper, the researchers used two measures to compare the degree of variation in each plant trait: coefficient of variation and plasticity index. The coefficient of variation is equal to the standard deviation of each trait divided by its mean, times 100%. The plasticity index is the difference between the largest and smallest value for each trait, divided by the largest value. The difference in the kind of information you get from these two measures isn’t immediately obvious to me.

It seems like the plasticity index is only concerned with the total magnitude of variation observed; it doesn’t matter how many individuals are closer to the largest or smallest values, just how large the difference is. If there was no variation, the plasticity index would be zero, and the larger the difference gets, the closer the value gets to 1. Since we’re dividing by the largest value, the plasticity index won’t get any larger than 1.

The coefficient of variation, however, does depend on the total amount of variation. If all values are the same, the coefficient of variation is zero, just like the plasticity index would be. But if there’s enough variation, the standard deviation can be greater than the mean, so the coefficient of variation doesn’t have the upper limit that the plasticity index has. My vague sense is that the plasticity index references how much variation is possible, and the coefficient of variation references how often variation occurs, but that’s a rabbit hole for another day!

Image credit: Obsidian Soul

https://commons.wikimedia.org/wiki/File:Daemonorops_mollis_(rattan_palm)_-_Bukidnon_Philippines.jpg

Note: This plant is in the same genus as the one described in the paper, but it's a different species. I couldn't find a Creative Commons image of the correct species.

van Rijswijk, J., & Curșeu, P. L. (2025). Formal Versus Self-Identified Neurodivergence: A Comparative Study in Work Environments. Behavioral Sciences, 15(4), 420.

 What They Did

The researchers recruited 357 participants from 19 workplaces in 7 industries and had them report formal diagnoses and self-identification of neurodivergent conditions, including autism, ADHD, ADD, DCD (developmental coordination disorder), dyslexia, dyscalculia, and high giftedness. About 12% of participants had a formal diagnosis of at least one neurodivergent condition, while about 27% answered that they believed they had at least one  neurodivergent condition. All participants completed assessments of decision-making styles, verbal vs. visual cognitive styles, and deliberate reasoning.

The researchers found that including self-identification in linear regression models increased the predictive power of the models for several combinations of neurodivergent conditions and thinking patterns. In particular, inclusion of self-identification increased the predictive power of the positive correlations between ADD and avoidant decision making (e.g. putting off making decisions as long as possible), between ADHD and spontaneous decision making, between dyslexia and visual cognitive style, and between high giftedness and deliberate reasoning. Self-identification also increased the predictive power of the negative correlations between autism and intuitive decision making and between dyslexia and verbal cognitive style.

The researchers conclude that self-identification and formal diagnosis can be correlated with the same thinking patterns and that inclusion of self-identified individuals in the study of neurodivergent thinking patterns may sometimes be appropriate. They note that self-identification rates may more accurately reflect the prevalence of neurodivergent conditions: formal diagnosis is often unavailable or is not sought because of concern over the stigma attached to neurodivergent conditions or because formal diagnosis is sometimes less accurate than self-identification.

Further Exploration

Based on my personal experience and casual reading, I’m not surprised that people who self-identified with neurodivergent conditions had variations in thinking patterns along the same lines as those with formal diagnoses. In particular, self-diagnosis of autism is accepted by many members of the autistic community, and it’s not uncommon for clinicians to rule out diagnosis for people who have strong communication skills, are married, work full-time, are Black or Latinx, or are female or trans (see https://www.verywellmind.com/self-diagnosis-according-to-an-autistic-psychologist-8650879.)

At the same time, I have some uncertainty about the questionnaires used to assess thinking patterns. I found the decision-making style assessment online (see https://www.google.com/url?sa=i&url=https%3A%2F%2Fs57e3f70e36f4dc05.jimcontent.com%2Fdownload%2Fversion%2F1561360698%2Fmodule%2F9817879385%2Fname%2FGeneral%2520Decision%2520Making%2520Style%2520Questionnaire.pdf&psig=AOvVaw0v2gIyS8w7tgPNXRquHxqD&ust=1743681032881000&source=images&cd=vfe&opi=89978449&ved=0CAQQn5wMahcKEwiAm_KVpLmMAxUAAAAAHQAAAAAQBA.) When I tried it myself, I experienced tension between my desire to answer accurately and my sense of what the assessment is looking for.

 The first question reads “I make sure that I have all the facts before I make a decision.” I realize that they don’t mean every fact in the universe, and they probably don’t mean facts that I might care about but that would be really hard to find out, but I’m still really unsure what counts as “all the facts.” Given that level of ambiguity, I’d choose the answer “The statement is sometimes true. It depends on the situation,” but I’d be inclined to give that answer for almost all the questions because it always depends on the situation. I wonder how much research has been done on neurodivergent information processing for this kind of psychological assessments, but that’s a rabbit hole for another day!

Image credit: Arafat Uddin

https://commons.wikimedia.org/wiki/File:Noun_questions_1325510.svg