We examine a highly cited randomized controlled trial on dance-movement therapy with adolescent girls with mild depression and examine its treatment in 14 evidence reviews and meta-analyses of dance research. We demonstrate substantial limitations in the trial which seriously undermine the conclusions reached regarding the effectiveness of dance movement therapy in reducing depression. We also show that the dance research reviews vary substantially in their treatment of the study. Some reviews provide a positive assessment of the study and take its findings at face value without critical commentary. Others are critical of the study, identifying significant limitations, but showing marked differences in Cochrane Risk of Bias assessments. Drawing on recent criticisms of systematic reviewing and meta-analysis, we consider how reviews can be so variable and discuss what is needed to improve the quality of primary studies, systematic reviews, and meta-analyses in the field of creative arts and health.

Globally, dietary intake of calcium is often insufficient, and it is unclear if adequacy could be achieved by promoting calcium-rich local foods. This study used linear programming and household consumption data from Uganda, Bangladesh, and Guatemala to assess whether local foods could meet calcium population reference intakes (Ca PRIs). The most promising food-based approaches to promote dietary calcium adequacy were identified for 12- to 23-month-old breastfed children, 4- to 6-year-old children, 10- to 14-year-old girls, and nonpregnant and nonbreastfeeding (NPNB) women of reproductive age living in two regions of each country. Calcium-optimized diets achieved 75-253% of the Ca PRI, depending on the population, and were <100% for 4- to 6-year-olds in one region of each country and 10- to 14-year-old girls in Sylhet, Bangladesh. The best food sources of calcium were green leafy vegetables and milk, across geographic locations, and species of small fish, nixtamalized (lime-treated) maize products, sesame seeds, and bean varieties, where consumed. Food-based recommendations (FBRs) achieving the minimum calcium threshold were identified for 12- to 23-month-olds and NPNB women across geographic locations, and for 4- to 6-year-olds and 10-to 14-year-old girls in Uganda. However, for 4- to 6-year-olds and 10- to 14-year-old girls in Bangladesh and Guatemala, calcium-adequate FBRs could not be identified, indicating a need for alternative calcium sources or increased access to and consumption of local calcium-rich foods.

Language models (LMs) like GPT-3, PaLM, and ChatGPT are the foundation for almost all major language technologies, but their capabilities, limitations, and risks are not well understood. We present Holistic Evaluation of Language Models (HELM) to improve the transparency of LMs. LMs can serve many purposes and their behavior should satisfy many desiderata. To navigate the vast space of potential scenarios and metrics, we taxonomize the space and select representative subsets. We evaluate models on 16 core scenarios and 7 metrics, exposing important trade-offs. We supplement our core evaluation with seven targeted evaluations to deeply analyze specific aspects (including world knowledge, reasoning, regurgitation of copyrighted content, and generation of disinformation). We benchmark 30 LMs, from OpenAI, Microsoft, Google, Meta, Cohere, AI21 Labs, and others. Prior to HELM, models were evaluated on just 17.9% of the core HELM scenarios, with some prominent models not sharing a single scenario in common. We improve this to 96.0%: all 30 models are now benchmarked under the same standardized conditions. Our evaluation surfaces 25 top-level findings. For full transparency, we release all raw model prompts and completions publicly. HELM is a living benchmark for the community, continuously updated with new scenarios, metrics, and models https://crfm.stanford.edu/helm/latest/.

Astrocytes are the most abundant glial cell type in the central nervous system and are essential to the development, plasticity, and maintenance of neural circuits. Astrocytes are heterogeneous, with their diversity rooted in developmental programs modulated by the local brain environment. Astrocytes play integral roles in regulating and coordinating neural activity extending far beyond their metabolic support of neurons and other brain cell phenotypes. Both gray and white matter astrocytes occupy critical functional niches capable of modulating brain physiology on time scales slower than synaptic activity but faster than those adaptive responses requiring a structural change or adaptive myelination. Given their many associations and functional roles, it is not surprising that astrocytic dysfunction has been causally implicated in a broad set of neurodegenerative and neuropsychiatric disorders. In this review, we focus on recent discoveries concerning the contributions of astrocytes to the function of neural networks, with a dual focus on the contribution of astrocytes to synaptic development and maturation, and on their role in supporting myelin integrity, and hence conduction and its regulation. We then address the emerging roles of astrocytic dysfunction in disease pathogenesis and on potential strategies for targeting these cells for therapeutic purposes.

Biomolecular communication demands that interactions between parts of a molecular system act as scaffolds for message transmission. It also requires an organized system of signs-a communicative agency-for creating and transmitting meaning. The emergence of agency, the capacity to act in a given context and generate end-directed behaviors, has baffled evolutionary biologists for centuries. Here, I explore its emergence with knowledge grounded in over two decades of evolutionary genomic and bioinformatic exploration. Biphasic processes of growth and diversification exist that generate hierarchy and modularity in biological systems at widely ranging time scales. Similarly, a biphasic process exists in communication that constructs a message before it can be transmitted for interpretation. Transmission dissipates matter-energy and information and involves computation. Agency emerges when molecular machinery generates hierarchical layers of vocabularies in an entangled communication network clustered around the universal Turing machine of the ribosome. Computations canalize biological systems to perform biological functions in a dissipative quest to structure long-lived occurrents. This occurs within the confines of a "triangle of persistence" that maximizes invariance with trade-offs between economy, flexibility, and robustness. Thus, learning from previous historical and circumstantial experiences unifies modules in a hierarchy that expands the agency of systems.

Seasonal changes in food intake and adiposity in many animal species are triggered by changes in the photoperiod. These latter changes are faithfully transduced into a biochemical signal by melatonin secreted by the pineal gland. Seasonal variations, encoded by melatonin, are integrated by third ventricular tanycytes of the mediobasal hypothalamus through the detection of the thyroid-stimulating hormone (TSH) released from the pars tuberalis. The mediobasal hypothalamus is a critical brain region that maintains energy homeostasis by acting as an interface between the neural networks of the central nervous system and the periphery to control metabolic functions, including ingestive behavior, energy homeostasis, and reproduction. Among the cells involved in the regulation of energy balance and the blood-hypothalamus barrier (BHB) plasticity are tanycytes. Increasing evidence suggests that anterior pituitary hormones, specifically TSH, traditionally considered to have unitary functions in targeting single endocrine sites, display actions on multiple somatic tissues and central neurons. Notably, modulation of tanycytic TSH receptors seems critical for BHB plasticity in relation to energy homeostasis, but this needs to be proven.

Anemia is a major global public health concern with a complex etiology. The main determinants are nutritional factors, infection and inflammation, inherited blood disorders, and women's reproductive biology, but the relative role of each varies between settings. Effective anemia programming, therefore, requires evidence-based, data-driven, contextualized multisectoral strategies, with coordinated implementation. Priority population groups are preschool children, adolescent girls, and pregnant and nonpregnant women of reproductive age. Opportunities for comprehensive anemia programming include: (i) bundling interventions through shared delivery platforms, including antenatal care, community-based platforms, schools, and workplaces; (ii) integrating delivery platforms to extend reach; (iii) integrating anemia and malaria programs in endemic areas; and (iv) integrating anemia programming across the life course. Major barriers to effective anemia programming include weak delivery systems, lack of data or poor use of data, lack of financial and human resources, and poor coordination. Systems strengthening and implementation research approaches are needed to address critical gaps, explore promising platforms, and identify solutions to persistent barriers to high intervention coverage. Immediate priorities are to close the gap between access to service delivery platforms and coverage of anemia interventions, reduce subnational coverage disparities, and improve the collection and use of data to inform anemia strategies and programming.

Emotion regulation is essential to survive in a world full of challenges with rapidly changing contextual demands. The ability to flexibly shift between different emotional control strategies is critical to successfully deal with these demands. Recently, decision neuroscience has shown the importance of monitoring alternative control strategies. However, this insight has not been incorporated into current neurocognitive models of emotional control. Here, we integrate insights from decision and affective sciences into a novel viewpoint on emotion control, the Flexible Emotion Control Theory (FECT). This theory explains how an individual can flexibly change emotion-regulatory behavior to adapt to varying goals and contextual demands. Crucially, FECT proposes that rapid switching between alternative emotional control strategies requires concurrent evaluation of current as well as alternative (unchosen) options. The neural implementation of FECT relies on the involvement of distinct prefrontal structures, including the lateral frontal pole (FPl) and its connections with other cortical (prefrontal, parietal, motor) and subcortical systems. This novel account of emotion control integrates insights from decision sciences, clinical research, as well as meta-analytic evidence for the consistent FPl involvement during emotional control when monitoring of alternative emotional control strategies is required. Moreover, it provides novel, neurocognitively grounded starting points for interventions to improve emotion control in affective disorders, such as anxiety and aggression.

Nongenetic information implies all the forms of biological information not related to genes and DNA in general. Despite the deep scientific relevance of the concept, we currently lack reliable knowledge about its carriers and origins; hence, we still do not understand its true nature. Given that genes are the targets of nongenetic information, it appears that a parsimonious approach to find the ultimate source of that information is to trace back the sequential steps of the causal chain upstream of the target genes up to the ultimate link as the source of the nongenetic information. From this perspective, I examine seven nongenetically determined phenomena: placement of locus-specific epigenetic marks on DNA and histones, changes in snRNA expression patterns, neural induction of gene expression, site-specific alternative gene splicing, predator-induced morphological changes, and cultural inheritance. Based on the available evidence, I propose a general model of the common neural origin of all these forms of nongenetic information in eumetazoans.

A functional interplay of bottom-up and top-down processing allows an individual to appropriately respond to the dynamic environment around them. These processing modalities can be represented as attractor states using a dynamical systems model of the brain. The transition probability to move from one attractor state to another is dependent on the stability, depth, neuromodulatory tone, and tonic changes in plasticity. However, how does the relationship between these states change in disease states, such as anxiety or depression? We describe bottom-up and top-down processing from Marr's computational-algorithmic-implementation perspective to understand depressive and anxious disease states. We illustrate examples of bottom-up processing as basolateral amygdala signaling and projections and top-down processing as medial prefrontal cortex internal signaling and projections. Understanding these internal processing dynamics can help us better model the multifaceted elements of anxiety and depression.