The immediate trigger for Owen’s intervention is a recent Cochrane systematic review, led by Nonino and colleagues, which concluded that amyloid-beta-targeting monoclonal antibodies offer no clinically meaningful benefit to people with mild cognitive impairment or mild dementia due to Alzheimer’s disease. The review’s publication provoked sharp criticism from within the Alzheimer’s research community, but Owen observes that much of that criticism focused on the review’s methodology, on questions of statistical pooling, trial selection and effect-size calculation, rather than on the underlying clinical case for the drugs themselves. In his view, this deflection represents a missed opportunity. Rather than interrogating whether the treatments help patients in ways that matter in daily life, the debate has circled technical questions that leave patients, families and clinicians without a clear answer.

At the heart of the controversy lies a deceptively simple question: what does a ‘modest’ or ‘moderate’ benefit actually mean for a person living with early Alzheimer’s disease? The pivotal trials of lecanemab, reported in the New England Journal of Medicine by van Dyck and colleagues in 2023, and of donanemab, reported in JAMA by Sims and colleagues in the TRAILBLAZER-ALZ 2 trial, both demonstrated statistically significant slowing of decline on the Clinical Dementia Rating Scale Sum of Boxes, a standard instrument for staging dementia that was first described by Hughes and colleagues in 1982. But statistical significance is not the same as clinical meaning. A change on a rating scale, even one that reaches conventional thresholds for significance, may translate into differences in daily functioning that are imperceptible to patients and their caregivers, or it may not. Owen’s Comment presses precisely on this gap between the numbers reported in trial publications and the lived reality of the disease.

The problem of defining clinically meaningful outcomes is not unique to Alzheimer’s research, but it is unusually acute there. A 2025 review by Stoeckel and colleagues examined what constitutes clinically meaningful outcomes in Alzheimer’s disease and related dementia trials, and a 2024 rapid review by Muir and colleagues tackled the related concept of the minimal clinically important difference in the disease. Both underline a persistent difficulty: the field has lacked consensus on how large an effect must be, and on which measures, before it can honestly be described as beneficial to a patient. Without such consensus, terms like ‘modest’ and ‘moderate’ float free of any anchor, allowing advocates and skeptics to describe the same trial results in dramatically different terms. Owen argues that this ambiguity is not an accident of imprecise language but a structural feature of how the field has communicated its results, and one that has gone largely unexamined.

One way to calibrate expectations is to compare the anti-amyloid drugs against other interventions whose effects are better understood. Cognitive training and cognitive stimulation therapies, which aim to maintain or improve thinking skills through structured mental activity, have been studied for decades with rigorous randomised controlled designs. Work by Huntley, Hampshire and Owen himself in 2017 tested adaptive working memory strategy training in early Alzheimer’s disease, and broader meta-analyses by Zhang and colleagues in 2019 and Chan and colleagues in 2024 have synthesised the effects of computerised cognitive training in mild cognitive impairment. Cochrane reviews by Bahar-Fuchs and colleagues and meta-analyses of cognitive stimulation therapy by Desai and colleagues and by Spector and colleagues provide further benchmarks. These interventions produce effects on cognitive measures that are, in many cases, comparable in magnitude to those reported for anti-amyloid antibodies, yet they are rarely described in the triumphant language that has accompanied the new drugs. That asymmetry, Owen suggests, is itself informative about how the field weighs evidence.

The history of the amyloid programme also includes sobering episodes that complicate the narrative of steady progress. Aducanumab, the first anti-amyloid antibody to receive accelerated approval, was discontinued as an Alzheimer’s treatment, a decision acknowledged by the Alzheimer’s Association, after a trajectory marked by contested trial results and contentious regulatory review. A 2022 analysis by Kim and colleagues catalogued key insights from two decades of clinical trial failures in Alzheimer’s disease, a record that spans far more than amyloid-targeting agents and reflects the extraordinary difficulty of intervening in a neurodegenerative process that begins years before symptoms appear. Against that backdrop, the licensing of lecanemab and donanemab, with the United States Food and Drug Administration converting one novel treatment to traditional approval in September 2025, represents genuine scientific progress in the narrow sense that amyloid can now be reliably cleared from the brain. The open question is whether plaque removal, on its own, changes the course of the disease in ways patients can feel.

Owen’s Comment also situates the current debate within the broader architecture of Alzheimer’s diagnosis and drug development. Revised criteria for the diagnosis and staging of Alzheimer’s disease, published by Jack and colleagues in 2024 under the auspices of the Alzheimer’s Association Workgroup, have moved the field toward a biological definition of the disease, in which biomarkers of amyloid and tau pathology can establish a diagnosis even in the absence of symptoms. That shift has been celebrated as a way to intervene earlier, before irreversible neuronal loss occurs, but it also raises the stakes of the benefit question: if treatments are to be offered to ever larger populations identified by biomarkers rather than symptoms, the size of the benefit each patient can expect becomes a matter of profound practical importance. The drug development pipeline, surveyed by Cummings and colleagues in 2025, remains crowded with agents, and the proportion of the population carrying risk variants such as APOE4, estimated in pooled analyses of nearly 389,000 community-dwelling individuals by Wang and colleagues, underscores the scale of the population that stands to be affected by these decisions.

What makes Owen’s argument distinctive is not a claim that the trials were flawed or that the drugs are useless, but a call for honesty about the magnitude of their effects and for a vocabulary that patients and clinicians can actually use. The pivotal trials enrolled people with mild cognitive impairment or mild dementia, the earliest symptomatic stages of the disease, precisely because that is where intervention is thought to hold the most promise. In those populations, the reported slowing of decline, while statistically robust, must be weighed against practical realities: the drugs require regular intravenous infusions over extended periods, monitoring with magnetic resonance imaging for a serious side effect known as amyloid-related imaging abnormalities, and careful patient selection. For a patient and family deciding whether to commit to that regimen, the relevant question is not whether a p-value crossed a threshold but whether the treatment will meaningfully extend the time they can live independently, recognise loved ones, or continue the activities that give their lives structure and meaning. Owen contends that the field has not done the work of answering that question in those terms.

The Comment also highlights a striking feature of the public response to the Cochrane review: the speed and intensity with which it was challenged by patient organisations and leading experts, including a public response from the Alzheimer’s Society defending the anti-amyloid drugs as effective. In a field that has endured repeated disappointments, the desire for treatments that finally work is entirely understandable, and patient advocacy has historically been a powerful force for accelerating research and improving access to care. But Owen’s concern is that advocacy, however well intentioned, can short-circuit the kind of critical self-examination that science requires. If the conversation about clinical meaningfulness is framed as an attack on hope rather than as a legitimate scientific question, then the field loses the opportunity to establish, once and for all, what its therapies can and cannot do, and to design the next generation of trials around outcomes that matter.

The stakes of this unresolved conversation extend well beyond the current generation of antibodies. If the field cannot agree on what constitutes a clinically meaningful benefit, then even genuinely transformative therapies of the future may struggle to demonstrate their value, and patients may continue to face difficult decisions armed only with vague descriptors and contested statistics. Owen’s contribution, published as a Neurological Update in the Journal of Neurology and supported by the Canadian Institutes of Health Research, is ultimately an appeal for clarity: for researchers, regulators and clinicians to specify, in concrete and patient-centred terms, what ‘modest’ and ‘moderate’ benefit mean, and to have that discussion openly rather than allowing it to remain the conversation the Alzheimer’s field isn’t having. Whether the field takes up that challenge will shape not only how the current drugs are used, but how the next decades of Alzheimer’s research are judged by the people who matter most, the patients and families living with the disease.

Subject of Research: Clinical meaningfulness of anti-amyloid monoclonal antibody therapies in Alzheimer's disease

Article Title: What have trials targeting amyloid in Alzheimer’s told us to date? The conversation the Alzheimer’s field isn’t having

Article References: Owen, A. M. (2026). What have trials targeting amyloid in Alzheimer’s told us to date? The conversation the Alzheimer’s field isn’t having. Journal of Neurology, 273(10), Article 617. https://doi.org/10.1007/s00415-026-14147-8

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14147-8

Keywords: Alzheimer's disease, anti-amyloid therapies, lecanemab, donanemab, Cochrane review, clinical trials, clinical meaningfulness, amyloid-beta, dementia, cognitive training, biomarkers, drug development

At the heart of the initiative is a simple but consequential premise: brain health is not fixed, and proactive training in how we think can change trajectories at any stage of life. Rather than treating cognitive performance as an innate trait that students either possess or lack, the program treats it as a trainable capacity, one that can be deliberately developed through structured practice, assessment, and feedback. This philosophy reflects the Center for BrainHealth’s broader research mission, which has long focused on translating laboratory findings about neuroplasticity, attention, and reasoning into practical tools that people can apply in classrooms, workplaces, and everyday life.

The practical architecture of the program is deliberately layered. Faculty and staff will receive training on core brain health principles and strategies that they can then reinforce with students, with the added benefit of applying the same techniques to their own cognitive lives. This train-the-trainer model is significant from a research standpoint: it embeds brain health concepts into the daily interactions of the campus rather than confining them to occasional seminars. Students, meanwhile, will take part in tailored workshops targeting three higher-order cognitive capacities that the Center for BrainHealth has studied extensively: strategic attention, integrated reasoning, and innovation. These are the same domains emphasized in the center’s SMART brain health training, a strategy-based toolkit developed and tested by researchers over roughly thirty years.

Beyond the core cognitive curriculum, the initiative addresses the affective and social dimensions of student life. Topic-driven sessions will tackle stress management, mindfulness, and social connection, recognizing that emotional regulation and interpersonal bonds are not peripheral to academic achievement but tightly intertwined with it. Decades of research in cognitive neuroscience and psychology have linked chronic stress to impaired working memory and reduced prefrontal efficiency, while social isolation has been associated with measurable declines in cognitive health across the lifespan. By folding these topics into a single campus-wide program, the partnership acknowledges that thinking well depends on living well, and that a student’s capacity for complex reasoning cannot be separated from their emotional and social circumstances.

Measurement is another defining feature of the initiative. Participants will have access to regular brain health assessments, allowing them to track changes in cognitive performance over time rather than relying on subjective impressions. This longitudinal approach mirrors the methodology of The BrainHealth Project, the center’s landmark study launched in 2020, which aims to identify actionable strategies for optimizing brain health throughout life and to clarify the dynamic relationships among lifestyle factors, biological markers, brain training, and cognitive performance. The center’s proprietary BrainHealth Index, a measure designed to chart an individual’s upward or downward brain health trajectory from whatever their starting point, exemplifies this philosophy of treating brain fitness as a dynamic, modifiable trajectory rather than a static snapshot.

To ensure the program outlives its initial rollout, the partnership includes ongoing access to self-directed online training and application tools through the BrainHealth Platform. Sustainability is a persistent challenge for campus wellness initiatives, which often flourish briefly and then fade when grant cycles end or institutional priorities shift. By building digital, self-paced resources into the program’s foundation, the collaborators are attempting to convert a structured intervention into a durable part of the campus culture, one that students and employees can return to throughout their time at the college and beyond.

Leaders on both sides have framed the partnership in terms that go well beyond test scores. Lori Cook, PhD, CCC-SLP, director of clinical research at the Center for BrainHealth, said the initiative builds on years of collaboration with Dallas College and reflects more than three decades of cognitive neuroscience research demonstrating that it is never too soon to become proactive about the brain’s fitness and performance. By integrating evidence-based tools and strategies into the campus experience, she noted, the program equips students as well as faculty and staff with actionable ways to become more strategic and intentional about how they think, manage challenges, and thrive in an increasingly complex world.

Dallas College administrators have emphasized the whole-student dimension of the effort. Thaddeus Mantaro, PhD, dean of student health and wellness, said the college is deeply committed to supporting students academically, emotionally, and socially, and described the partnership as bringing innovative, science-backed resources directly to the community, with optimism about its potential to enhance student well-being, strengthen academic success, and contribute to improved retention. Carlos Cruz, EdD, associate vice chancellor of the college, added that Dallas College recognizes students as more than learners, understanding that academic success is closely linked to personal growth, emotional well-being, and overall health. Retention, in particular, is a high-stakes metric for community colleges, where students often balance coursework with work and family obligations, and where even modest improvements in persistence can translate into substantially better life outcomes.

The scale of the institution involved gives the initiative considerable reach. Founded in 1965, Dallas College operates seven campuses—Brookhaven, Cedar Valley, Eastfield, El Centro, Mountain View, North Lake, and Richland—along with a dozen centers across Dallas County. As one of the largest community colleges in the United States, it serves more than 133,000 credit, workforce, and continuing education students annually, offering associate degrees and career and technical certificates in more than 100 areas of study, as well as bachelor’s degrees in education, nursing, software development, and management. It is also the largest provider of dual credit in Texas, serving 30,000 high school students through 64 dual credit programs. That breadth means the BrainHealthy Campus framework will be tested across an unusually diverse population, from teenagers earning college credit to working adults retraining mid-career.

Financial support for the initiative comes from philanthropic sources, with grants from The Meadows Foundation and The Hoglund Foundation fueling the program at Dallas College. The collaboration also arrives at a moment when the Center for BrainHealth is broadening its regional footprint, having recently appointed new members to its 2026-27 Advisory Board, drawing leaders from business, health care, science, technology, real estate, and philanthropy who share a commitment to advancing brain health across North Texas and beyond. For researchers, the partnership offers a real-world laboratory for studying how proactive cognitive training performs at scale in a post-secondary setting, a context that differs markedly from the controlled cohorts of typical laboratory studies. For the tens of thousands of students who pass through Dallas College each year, it offers something rarer: a systematic, science-driven attempt to treat the brain itself as the most important asset a learner brings to campus, and one worth training as deliberately as any subject on the syllabus.

Subject of Research: A research partnership applying cognitive neuroscience-based brain health training and assessment to improve student well-being and academic outcomes at a large community college.

Article Title: New BrainHealthy Campus initiative aims to strengthen community brain health and enhance academic outcomes

Article References: New BrainHealthy Campus initiative aims to strengthen community brain health and enhance academic outcomes. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: brain health, cognitive training, Center for BrainHealth, Dallas College, higher education, student well-being, academic outcomes, neuroscience, SMART training, BrainHealth Index, retention, community college

The concern motivating the research is well founded. Generative AI systems now produce text, images, and videos that closely mimic human work, and these outputs have already been exploited in disinformation campaigns, AI-written phishing emails, deceptive voice cloning, and fabricated hotel reviews. Prior studies have found that human evaluators perform at or near chance when judging whether scientific abstracts, faces, or videos are authentic. Some evidence suggests that expertise helps: professional writers detect AI-authored essays better than lay readers, and people judge general-interest news more accurately than scientific news, presumably because they lack domain knowledge to spot subtle errors. Multimodal information, such as combining transcripts with audio and video, also improves detection, hinting that a real, detectable signal separates human from machine output even when average performance is poor.

The Temple team hypothesized that individual differences in attitudes toward AI, and in the history of engagement with AI tools, might explain why some people discern better than others. Two competing predictions framed the question of AI use. On one hand, frequent use of a platform could act as practice, sharpening sensitivity to that system’s stylistic fingerprints; one recent study found that heavy ChatGPT users were indeed better detectors of ChatGPT-written text. On the other hand, theories of extended cognition and cognitive offloading suggest that habitual reliance on external tools to support reasoning may blur the boundary between information arising inside one’s own mind and information arriving from outside. Under that view, heavy AI engagement could desensitize users to the very cues that distinguish synthetic from human content.

To test these ideas, the researchers recruited 117 adults aged 18 to 34 through the Prolific platform, all fluent English speakers located in the United States. The stimuli were 176 vacation rental listings presented in an Airbnb-like format. Half were genuine: descriptions extracted from real Airbnb listings with at least ten public reviews and ratings of 4.5 stars or higher, verified as human-written with an AI detector, and paired where appropriate with a real photograph of the property and a real human face drawn from the Flickr-Faces-HQ dataset. The other half were synthetic: texts generated by ChatGPT 4.0 and Claude 3.5 Sonnet describing fictional rentals, matched to the human texts in word count, and accompanied by AI-generated property images from Mage.space and Gemini plus AI faces produced by StyleGAN2. Notably, the AI texts were actually grammatically superior to the human ones, ruling out simple error-spotting as a strategy.

Participants first completed a baseline assessment, judging 32 listings as human or AI generated, with half presented as text only and half as text plus images. They then moved through two intervention or control stages, with assessments interleaved, before finishing with questionnaires measuring AI attitudes on the ATTARI-12 scale and AI engagement on a newly developed survey called the Socioaffective and Cognitive Artificial Intelligence Engagement Survey, or SCAIES. At baseline, participants were 57.4 percent accurate on average, only modestly above chance, but individual scores ranged from 34 to 90 percent, a spread that invited explanation.

The explanation that emerged was striking. Attitudes toward AI, which were generally positive in this sample, did not predict detection performance at all. But SCAIES engagement scores did: heavier overall engagement with AI was associated with significantly weaker discernment, a correlation of negative 0.29. The same negative relationship held separately for socioaffective uses of AI, such as social and emotional purposes, and for cognitive uses, such as outsourcing effortful mental tasks. Interestingly, an objective measure of ChatGPT usage, drawn from participants’ logged session counts over the prior 30 days, did not predict discernment, even though participants slightly underestimated their actual use. This dissociation suggests that it is not raw frequency of use but the qualitative breadth and motivational depth of AI integration into daily thinking that erodes sensitivity to the difference between synthetic and human content, consistent with the idea that deeply enmeshing AI into one’s cognitive life makes machine output feel less foreign.

The intervention half of the study offered a counterweight to this bleak picture. Sixty participants were assigned to an experimental group and 57 to a control group, with the two groups performing identically at baseline, around 57 percent accuracy. The experimental group first viewed 40 accurately labeled human and AI exemplars, half text-only and half multifeatured, with a minimum ten-second viewing period per item. The control group viewed the same number of unlabeled listings for the same duration. In the second stage, experimental participants judged 40 new listings while receiving immediate Correct or Incorrect feedback and earning 25 cents per correct answer, with cumulative earnings displayed on screen. Control participants received the same monetary incentive but only delayed feedback and no running tally. The results were decisive: the experimental group improved by 9.4 percentage points from baseline to final assessment, while the control group changed by a negligible negative 0.6 points.

Most of the gain, 6.7 percentage points, came after the labeled exemplar stage alone, indicating that detection failure stems less from an absence of diagnostic cues than from uncertainty about which cues matter. Explicit labels appear to recalibrate attention toward informative properties. The feedback and incentive stage added a further, non-significant 2.7 points. Reaction time analyses ruled out a speed-accuracy tradeoff, showing that improved performance reflected more efficient use of cues rather than slower deliberation. Encouragingly, participants with the lowest baseline scores gained the most, and prior AI engagement, which predicted poor baseline performance, did not predict resistance to training. Any desensitization caused by habitual AI use appears reversible with structured exposure and feedback.

Item-level analyses revealed what cues successful detectors may have exploited. Using Google’s Universal Sentence Encoder to compute semantic similarity between all pairs of texts, the researchers found that AI-generated listings were significantly more similar to one another than human-generated listings were, with intra-class similarity of 0.50 for AI texts versus 0.45 for human texts. In other words, AI outputs converge toward prototypical patterns while human writing displays greater idiosyncratic variety. Human texts that stood out as most semantically distinct were judged most accurately at baseline, suggesting evaluators are implicitly sensitive to this distributional signature. Parallel image analyses using a ResNet-50 model showed the same pattern of greater homogenization among AI-generated images. Participants were also better at identifying AI texts in the text-only condition but better at identifying human listings when images were present, implying a shift in cue use across modalities.

The broader implications reach into education, digital literacy, and information integrity. As generative AI embeds itself in everyday tasks, spontaneous sensitivity to its distinguishing qualities may quietly diminish, a paradox the authors describe as normalization. If intuitive detection erodes, intentional training frameworks and accurate labeling of AI content may become essential safeguards. The study also carries a caveat: current generative systems still leave detectable statistical regularities, but as models evolve and diversify, those signatures may fade. The stimulus set was limited to promotional vacation rental language, the sample was restricted to young, digitally fluent adults, and the durability of training gains remains unknown. Still, the central message stands: human discernment of AI content is neither obsolete nor fixed. It varies widely across individuals, is dulled by habitual AI engagement, and can be measurably restored through brief, well-designed training, offering a practical path toward keeping human judgment sharp in a marketplace increasingly saturated with synthetic content.

Subject of Research: How experience with AI tools and targeted training influence human ability to distinguish AI-generated from human-created online content

Article Title: Human discernment of artificial intelligence in online markets can be shaped by experience and training

Article References: Human discernment of artificial intelligence in online markets can be shaped by experience and training. (n.d.). https://doi.org/10.1007/s00146-026-03373-3

Image Credits: AI Generated

DOI: 10.1007/s00146-026-03373-3

Keywords: artificial intelligence, AI detection, generative AI, digital literacy, cognitive training, human-AI interaction, misinformation, cognitive offloading, AI & Society, Temple University, synthetic content, extended cognition