Chatbot responses suggest that hypothetical biology questions are harder than realistic ones

The biology education literature includes compelling assertions that unfamiliar problems are especially useful for revealing students' true understanding of biology. However, there is only limited evidence that such novel problems have different cognitive requirements than more familiar problems. Here, we sought additional evidence by using chatbots based on large language models as models of biology students. For human physiology and cell biology, we developed sets of realistic and hypothetical problems matched to the same lesson learning objectives (LLOs). Problems were considered hypothetical if (i) known biological entities (molecules and organs) were given atypical or counterfactual properties (redefinition) or (ii) fictitious biological entities were introduced (invention). Several chatbots scored significantly worse on hypothetical problems than on realistic problems, with scores declining by an average of 13%. Among hypothetical questions, redefinition questions appeared especially difficult, with many chatbots scoring as if guessing randomly. These results suggest that, for a given LLO, hypothetical problems may have different cognitive demands than realistic problems and may more accurately reveal students' ability to apply biology core concepts to diverse contexts. The Test Question Templates (TQT) framework, which explicitly connects LLOs with examples of assessment questions, can help educators generate problems that are challenging (due to their novelty), yet fair (due to their alignment with pre-specified LLOs). Finally, ChatGPT's rapid improvement toward expert-level answers suggests that future educators cannot reasonably expect to ignore or outwit chatbots but must do what we can to make assessments fair and equitable.

Making the case for prophylactic use of betaine to promote brain health in young (15-24 year old) athletes at risk for concussion

Betaine supplementation in the context of human nutrition, athletic performance, and clinical therapy demonstrate that the osmolyte and methyl donor, betaine, is cytoprotective and beneficial to human health. These studies also demonstrate that betaine supplementation in healthy humans is straight-forward with no reported adverse effects. Here, we explore betaine uptake in the central nervous system (CNS) and contribute to evidence that betaine may be uniquely protective to the brain. We specifically describe the therapeutic potential of betaine and explore the potential implications of betaine on inhibition mediated by GABA and glycine neurotransmission. The influence of betaine on neurophysiology complement betaine's role as an osmolyte and metabolite and is consistent with clinical evidence of betaine-mediated improvements to cognitive function (reported in elderly populations) and its anti-convulsant properties. Betaine's therapeutic potential in neurological disorders including epilepsy and neurodegenerative diseases combined with benefits of betaine supplementation on athletic performance support the unique application of betaine as a prophylaxis to concussion. As an example, we identify young athletes (15-24 years old), especially females, for prophylactic betaine supplementation to promote brain health and resilience in a cohort at high risk for concussion and for developing Alzheimer's disease.

Inhibition and interneuron distribution in the dentate gyrus of p35 knockout mice

The p35 knockout (p35-/-) mouse is an animal model of temporal lobe epilepsy that recapitulates key neuroanatomic abnormalities-granule cell dispersion and mossy fiber sprouting-observed in the hippocampal formation of humans, as well as spontaneous seizure activity. It is a useful model in which to study the relationship between the abnormal neuronal structure and seizure activity to further our understanding of cortical dysplasia in epileptogenesis. Our previous work using this mouse model characterized the anatomic features of the dentate granule cells and the functional implications of these abnormalities on increased recurrent excitation. These data also suggested that there might be compromised inhibition in this animal model. We pursued this possibility, focusing our investigation on inhibitory circuitry. In preliminary investigations using neuroanatomic tools (immunocytochemistry, camera lucida reconstructions of individually labeled interneurons, and electron microscopy) combined with intracellular electrophysiology, we observed no significant reduction in the number of symmetric versus asymmetric synaptic contacts on dentate granule cell somata, and no statistically significant changes in evoked early or late inhibition. Although there were some abnormalities in the morphology/distribution of inhibitory interneurons (as well as a larger population of dentate granule cells) of the dentate gyrus, overall inhibition in the p35 knockout mouse appeared to be largely intact.

Comparison of binding specificity and the function of two human IgM anti-lipid A monoclonal antibodies

The interactions of two anti-lipid A monoclonal antibodies (mAb)--HA-1A and SdJ5-1.17.15--with their antigenic sites on lipid A, were compared using a dot-blot assay and lipid A structural analogues, as well as lipid A-high-density lipoprotein (HDL) complexes. The reactivities of both mAb were affected by the type of fatty acid side chains and by the phosphate group on the glucosamine residue II; however, the interaction of SdJ5-1.17.15 appeared to be more markedly affected by the fatty acid side chains. A determination of the biological significance of these antigenic differences was made. Human peripheral blood mononuclear cells (hPBMC) challenged with Escherichia coli 055:B5 lipopolysaccharide (LPS) pre-incubated with SdJ5-1.17.15 released significantly less tumor necrosis factor-alpha (TNF-alpha) and interleukin-1 beta (IL-1 beta), compared to hPBMC exposed to vehicle preincubated LPS. HA-1A did not attenuate the in vitro release of either cytokine. The ability of both mAb to neutralize the in vivo toxicity of LPS was also evaluated. Rats administered E. coli 055:B5 pre-incubated with SdJ5-1.17.15 had a significantly reduced 24-hr mortality rate compared to vehicle controls. HA-1A did not attenuate the in vivo mortality rate. Therefore, the reactivity of anti-lipid A mAb with the antigen is preferentially affected by different residues on the lipid A moiety. Thus, the differences in biological activity seen with SdJ5-1.17.15 and HA-1A may be due in part to differences in their recognition sites on lipid A.

Inhibition of bacterial motility with human antiflagellar monoclonal antibodies attenuates Pseudomonas aeruginosa-induced pneumonia in the immunocompetent rat

Two human monoclonal antibodies, directed against the type a and type b flagellar proteins of Pseudomonas aeruginosa, inhibited bacterial motility in vitro specifically and in a concentration-dependent manner. In order to determine if this decreased bacterial motility was associated with a decreased pathogenicity, the ability of these human antiflagellar monoclonal antibodies to attenuate P. aeruginosa-induced pneumonia in the rat was assessed. Incubation of P. aeruginosa with a 1:1 mixture of the human antiflagellar monoclonal antibodies prior to pulmonary instillation significantly (P < 0.05) ameliorated the bacterium-induced decrease in arterial blood oxygen pressure, blunted the increase in respiratory rate, and markedly reduced the area of pulmonary inflammation. Similarly, intravenous administration of the human antiflagellar monoclonal antibodies 1 h after pulmonary instillation of the bacteria also reduced the in vivo pathogenicity of P. aeruginosa. Therefore, human antiflagellar monoclonal antibodies can decrease the in vitro motility of P. aeruginosa and can reduce its in vivo pathogenicity when administered either before or after bacterial challenge.

Inhibition of lipopolysaccharide-associated endotoxin activities in vitro and in vivo by the human anti-lipid A monoclonal antibody SdJ5-1.17.15

The present study evaluated the effect of a novel anti-lipid A monoclonal antibody, termed SdJ5, on the in vitro production of tumor necrosis factor alpha (TNF-alpha) and interleukin-1 beta by endotoxin- or lipopolysaccharide (LPS)-challenged human peripheral blood mononuclear cells (hPBMC). In addition, the present study determined whether SdJ5 could neutralize the in vivo toxicity of LPS. SdJ5, at a concentration equal to or greater than 3 micrograms/ml, specifically inhibited TNF-alpha and interleukin-1 beta production by hPBMC stimulated with every type of LPS and lipid A assessed. SdJ5 also showed a significantly greater inhibition of cytokine production than a nonrelevant human immunoglobulin M myeloma control. The SdJ5-mediated inhibition of TNF-alpha production was rapid, as the simultaneous addition of the SdJ5 and LPS still resulted in a marked decrease in hPBMC cytokine synthesis. The ability of SdJ5 to neutralize in vivo toxicity was also determined by using LPS from four different strains of gram-negative bacteria. LPS, when preincubated with SdJ5, resulted in a significant decrease in the 24-h mortality rate compared with that for the control. These studies show that the anti-lipid A monoclonal antibody SdJ5 can modulate LPS-induced cytokine production in vitro and increase the survival rate of rats challenged with lethal doses of LPS.