Dynamic diversity is the answer to proxy failure.

We argue that a diverse and dynamic pool of agents mitigates proxy failure. Proxy modularity plays a key role in the ongoing production of diversity. We review examples from a range of scales.

Impact of Depression and Anxiety on Patient Reported Outcomes Measures after Lumbar Fusion.

BACKGROUND: Depression and anxiety are common in patients undergoing spinal surgery and might negatively impact outcomes. This study investigates the possible effect of these diagnoses on patient reported outcomes following lumbar fusion. METHODS: Retrospective review of a registry containing prospectively collected data of lumbar fusion procedures at a single institution was performed from May 23, 2012 to June 15, 2022. Patients with a minimum of two year follow-up were included. Demographic information, diagnoses, medications, patient-reported outcomes measures (PROMs), and complications data at preoperative, three months, six months, 1 year, and two years postoperative were collected. Statistical analysis was performed using Student's t-tests, χ2, binomial correlation, odds ratios, logistic regression, and mean clinically important difference. RESULTS: A total of 156 patients were included (60 males, 96 females) with mean age 62.6 ± 11.1 years at surgery. Thirty-nine (25%) had depression and/or anxiety (DA). Baseline Oswestry Disability Index (ODI) and EuroQol Group 5D questionnaire (EQ5D) scores were significantly worse in the DA cohort compared to controls (ODI 51.1 ± 18.3 vs. 42.9 ± 15.8; P = 0.010, EQ5D 0.46 ± 0.21 vs. 0.57 ± 0.21; P = 0.005). Both cohorts experienced similar relative improvement at two years (delta ODI -18.2 ± 27.9 vs. -17.8 ± 22.1; P = 0.924, EQ5D 6.8 ± 33.8 vs. 8.1 ± 32.9; P = 0.830). Absolute outcome scores were worse in the DA cohort at all intervals. DA were not independently predictive of changes in PROMs (delta ODI mean difference 4.49, r2 = 0.36, P = 0.924). CONCLUSIONS: The present study showed similar improvement in PROMs following lumbar fusion for patients with anxiety and depression compared to healthy controls. These data suggest these patients are no less likely to benefit from appropriately planned lumbar fusion.

Framework-based qualitative analysis of free responses of Large Language Models: Algorithmic fidelity.

Today, with the advent of Large-scale generative Language Models (LLMs) it is now possible to simulate free responses to interview questions such as those traditionally analyzed using qualitative research methods. Qualitative methodology encompasses a broad family of techniques involving manual analysis of open-ended interviews or conversations conducted freely in natural language. Here we consider whether artificial "silicon participants" generated by LLMs may be productively studied using qualitative analysis methods in such a way as to generate insights that could generalize to real human populations. The key concept in our analysis is algorithmic fidelity, a validity concept capturing the degree to which LLM-generated outputs mirror human sub-populations' beliefs and attitudes. By definition, high algorithmic fidelity suggests that latent beliefs elicited from LLMs may generalize to real humans, whereas low algorithmic fidelity renders such research invalid. Here we used an LLM to generate interviews with "silicon participants" matching specific demographic characteristics one-for-one with a set of human participants. Using framework-based qualitative analysis, we showed the key themes obtained from both human and silicon participants were strikingly similar. However, when we analyzed the structure and tone of the interviews we found even more striking differences. We also found evidence of a hyper-accuracy distortion. We conclude that the LLM we tested (GPT-3.5) does not have sufficient algorithmic fidelity to expect in silico research on it to generalize to real human populations. However, rapid advances in artificial intelligence raise the possibility that algorithmic fidelity may improve in the future. Thus we stress the need to establish epistemic norms now around how to assess the validity of LLM-based qualitative research, especially concerning the need to ensure the representation of heterogeneous lived experiences.

Modeling the Role in pH on Contaminant Sequestration by Zerovalent Metals: Chromate Reduction by Zerovalent Magnesium.

The role of pH in sequestration of Cr(VI) by zerovalent magnesium (ZVMg) was characterized by global fitting of a kinetic model to time-series data from unbuffered batch experiments with varying initial pH values. At initial pH values ranging from 2.0 to 6.8, ZVMg (0.5 g/L) completely reduced Cr(VI) (18.1 µM) within 24 h, during which time pH rapidly increased to a plateau value of ∼10. Time-series correlation analysis of the pH and aqueous Cr(VI), Cr(III), and Mg(II) concentration data suggested that these conditions are controlled by combinations of reactions (involving Mg0 oxidative dissolution and Cr(VI) sequestration) that evolve over the time course of each experiment. Since this is also likely to occur during any engineering applications of ZVMg for remediation, we developed a kinetic model for dynamic pH changes coupled with ZVMg corrosion processes. Using this model, the synchronous changes in Cr(VI) and Mg(II) concentrations were fully predicted based on the Langmuir-Hinshelwood kinetics and transition-state theory, respectively. The reactivity of ZVMg was different in two pH regimes that were pH-dependent at pH < 4 and pH-independent at the higher pH. This contrasting pH effect could be ascribed to the shift of the primary oxidant of ZVMg from H+ to H2O at the lower and higher pH regimes, respectively.

Emotions and courtship help bonded pairs cooperate, but emotional agents are vulnerable to deceit.

Coordinated pair bonds are common in birds and also occur in many other taxa. How do animals solve the social dilemmas they face in coordinating with a partner? We developed an evolutionary model to explore this question, based on observations that a) neuroendocrine feedback provides emotional bookkeeping which is thought to play a key role in vertebrate social bonds and b) these bonds are developed and maintained via courtship interactions that include low-stakes social dilemmas. Using agent-based simulation, we found that emotional bookkeeping and courtship sustained cooperation in the iterated prisoner's dilemma in noisy environments, especially when combined. However, when deceitful defection was possible at low cost, courtship often increased cooperation, whereas emotional bookkeeping decreased it.

CASE FATALITY RATE OF CHILDHOOD TUBERCULOSIS: EXPERIENCE FROM FEDERAL TEACHING HOSPITAL GOMBE: A TWO DECADE REVIEW.

Introduction: Tuberculosis (TB) was the leading cause of death from an infectious agent worldwide, until the Coronavirus (COVID-19) pandemic, ranking above HIV/AIDS. Nigeria ranks 6th among the 30 TB high-burden countries (TB, TB/HIV, DRTB) and 1st in Africa. The estimated case fatality rate (CFR) of TB in Sub-Sahara Africa (SSA) is 15%. Objective: To review the Tuberculosis case fatality rate (TCFR) in children diagnosed with TB from 2000-2019 in Federal Teaching Hospital Gombe. Methodology: All cases of Tuberculosis (TB) diagnosed in children using ICD 10 classification were retrieved and analyzed. These included deaths from TB. The mainstay of TB diagnosis was clinical using TB Score (81%), Gene Xpert was 7%, and AFB was 10%. Results: 26,716 children were admitted; 383 had TB out of which 208(54.3%) were males and 175 (45.7%) females. TB constituted 1.4% of Paediatric admissions. Children 0 -5 years constituted 46.7% (179/383) of cases and 11 - 18 years were 31.3% (120/383). Fulani, Hausa, and Tangale constituted 43.6% (167), 21.1% (81), and 6.8% (26) of TB cases respectively. TB admissions were highest between 2015 and 2019 (31.8%). TB adenitis was the most common extrapulmonary TB. Tuberculosis/HIV co-infection accounted for 103(27%), out of which 74% (44) died. Overall TCFR was 15.6%; TCFR was 16.3% in males and 14.8% in females. The TCFR was 46.7% in 0-5yrs; 15% in 6-9yrs and 38.3% in 10-18yrs.Fulani had the highest CFR (11.9%). Tuberculosis CFR was highest between 2010-2014 (30.0%) and lowest in 2005-2009 (21.6%). Conclusion: The Tuberculosis CFR is comparable to SSA CFR.

REVEALING THE VARIED REALITY OF MALARIA-RELATED DEATHS IN CHILDREN ACROSS GENDER, AGE, AND ETHNICITY IN FEDERAL TEACHING HOSPITAL GOMBE OVER 20 YEARS: AN ESSENTIAL EPIDEMIOLOGIC IMPERATIVE.

Introduction: Nigeria recorded 31% of 619,000 malaria deaths globally and accounts for 25-30% of all childhood mortality in the country. Few studies in Nigeria, have reported malaria's case fatality rate over a long period. Objective: To determine Malaria Case Fatality Rate among Children admitted from 2000-2019. Methodology: All severe malaria cases and deaths amongst children aged 0-18 over the last two decades were analysed using ICD-10. The diagnosis was based on clinical and microscopic findings. Results: 26,716 children were admitted, 2494 (9.3%) were diagnosed with malaria and 209 died. Malaria constituted 5.3% (209/3956) of all childhood mortality. Males constituted 58.9 % (1468/2494) while 65% (1642/2494) were aged 0-5 years. Of the malaria admissions, Fulani and Hausa constituted 948(38%) and 438(17.6%) respectively. Admissions were highest in October (15%) and in 2012 (9.6%). The overall malaria CFR was 8.3%; 8.8% in Females (91/1026) and 8.03% in Males P-value <0.05 (X2=54.735); 8.6% in children aged 0-5years, 8.2% in 6-10 years and 7.4% in 11-18 years, P-value <0.05 (X2=893.164). CFR was highest in April (11.4%)and lowest in November (5.2%). Kanuri and Igbo had CFR of 70% and 38.4% respectively while it was lowest in Tera tribe (4.3%), P-value<0.05. The CFR was highest in the year 2004 (22%), 3.5% in 2000 and 2006. Over the years, case fatality rate was 15.9% between 2000-2004, 6.1% from 2005-2009. Between 2010-2015, it was 7.3% and 8.5% from 2016-2019. Conclusion: This study revealed the deadly reality of severe malaria with increased CFR among females, aged 0-5 and the Kanuri tribe.

EPIDEMIOLOGIC TREND AND CASE FATALITY OF CHILDHOOD TETANUS IN FEDERAL TEACHING HOSPITAL GOMBE; A 20YEARS REVIEW.

Introduction: Tetanus is a vaccine-preventable disease, it remains a significant cause of morbidity and mortality in both neonatal and post-neonatal periods, especially in developing countries with limited health facilities and inadequate vaccination. The overall case fatality rate (CFR) is 13.2% globally, highest in the neonatal period and in sub-Saharan Africa. CFR is 64%, 47%, and 43% in Nigeria, Uganda, and Tanzania respectively. Objectives: To determine the Case Fatality Rate of Childhood tetanus in FTHG from 2000-2019. Methodology: All cases and deaths from tetanus amongst children aged 0-18 years in paediatric medical ward of FTHG over the last two decades diagnosed clinically and classified using ICD-10 were analysed. Results: 95 cases of tetanus out of 26,716 total admissions constituting 0.004%. There were 49 tetanus deaths out of 3956 total childhood deaths (0.012%) over the study period. Males constituted 66% (63/95). 30% (28/95) were aged 0-28 days; 23.1% (22/95) were adolescents. Fulani and Hausa constituted 37% (34/95) and 31% (29/95) respectively. Admission was highest in the dry season 52% (50/95 %). The overall tetanus CFR was 51.6%; 78% of deaths were in males (38/49), 30% in neonates, and 23% in adolescents. CFR was highest during the dry season (67.3%). Hausa and Fulani had CFR of 51% and 40% respectively. P-value <0.05 The CFR was 88% between 2000-2004, 72% from 2005-2009, 71% between 2010-2014 and 33% from 2015-2019. Conclusion: Tetanus CFR is still high among neonates and adolescents. Maternal tetanus vaccine and booster doses in children need strengthening.

Machine culture.

The ability of humans to create and disseminate culture is often credited as the single most important factor of our success as a species. In this Perspective, we explore the notion of 'machine culture', culture mediated or generated by machines. We argue that intelligent machines simultaneously transform the cultural evolutionary processes of variation, transmission and selection. Recommender algorithms are altering social learning dynamics. Chatbots are forming a new mode of cultural transmission, serving as cultural models. Furthermore, intelligent machines are evolving as contributors in generating cultural traits-from game strategies and visual art to scientific results. We provide a conceptual framework for studying the present and anticipated future impact of machines on cultural evolution, and present a research agenda for the study of machine culture.

IN-PATIENT PNEUMONIA BURDEN AND CASE FATALITY RATES IN CHILDREN OVER TWO DECADES IN FEDERAL TEACHING HOSPITAL, GOMBE (FTHG).

Introduction: Pneumonia is the leading cause of death among children globally accounting for an estimated 1.2 million (18%) total deaths annually. The number of childhood-related deaths from pneumonia is approximately 2000-fold higher in developing than in developed countries. Nigeria contributes the highest of pneumonia-related deaths globally. Objectives: To determine the case fatality rates (CFR) of pneumonia from 2000-2019 in paediatric ward, FTHG. Methodology: All cases of pneumonia admissions and deaths in patients aged 0-18 years, using ICD-10 classification, were retrieved and analysed. The mainstay of diagnosis is clinical and/or radiographic features. Results: A total of 26,716 children were admitted during this period, 1151 had pneumonia (4.3%) and 118 died. Males constituted 647 (56.2%) and females 43.8% of the total pneumonia admissions. Children aged 0-5 years had the highest pneumonia admissions, followed by 6-9 years. Admissions were highest in the wet than the dry season. Pneumonia CFR was 10.2%; 10.9% in females and 9.7% in males. Under-5 constituted 84% (969/1151) of pneumonia admission with a CFR of 9.3%. CFR were 10.3% and 21% in 6-10 years, and 11-18 years respectively. The CFR between2000-2004 was 14.1%, 2005-2009:21.1%, 2010-2014:10.2% and 2015-2019:7.2%. Kanuri had the highest CFR of 56.2%.(P <0.05) Other ethnic groups were 29.4% in Waja, 25% in Tula, 21.4% in Igbo, 16.6% in Yoruba, 12.1% in Tangale, 10.2% in Hausa, 8.8%in Bolewa and 8.3% in Fulani. The CFR was highest in February20.2%. Conclusion: Pneumonia Case fatality is high.

The emergence of division of labour through decentralized social sanctioning.

Human ecological success relies on our characteristic ability to flexibly self-organize into cooperative social groups, the most successful of which employ substantial specialization and division of labour. Unlike most other animals, humans learn by trial and error during their lives what role to take on. However, when some critical roles are more attractive than others, and individuals are self-interested, then there is a social dilemma: each individual would prefer others take on the critical but unremunerative roles so they may remain free to take one that pays better. But disaster occurs if all act thus and a critical role goes unfilled. In such situations learning an optimum role distribution may not be possible. Consequently, a fundamental question is: how can division of labour emerge in groups of self-interested lifetime-learning individuals? Here, we show that by introducing a model of social norms, which we regard as emergent patterns of decentralized social sanctioning, it becomes possible for groups of self-interested individuals to learn a productive division of labour involving all critical roles. Such social norms work by redistributing rewards within the population to disincentivize antisocial roles while incentivizing prosocial roles that do not intrinsically pay as well as others.

The Puzzle of Evaluating Moral Cognition in Artificial Agents.

In developing artificial intelligence (AI), researchers often benchmark against human performance as a measure of progress. Is this kind of comparison possible for moral cognition? Given that human moral judgment often hinges on intangible properties like "intention" which may have no natural analog in artificial agents, it may prove difficult to design a "like-for-like" comparison between the moral behavior of artificial and human agents. What would a measure of moral behavior for both humans and AI look like? We unravel the complexity of this question by discussing examples within reinforcement learning and generative AI, and we examine how the puzzle of evaluating artificial agents' moral cognition remains open for further investigation within cognitive science.

Identification of scaffold proteins for improved endogenous engineering of extracellular vesicles.

Extracellular vesicles (EVs) are gaining ground as next-generation drug delivery modalities. Genetic fusion of the protein of interest to a scaffold protein with high EV-sorting ability represents a robust cargo loading strategy. To address the paucity of such scaffold proteins, we leverage a simple and reliable assay that can distinguish intravesicular cargo proteins from surface- as well as non-vesicular proteins and compare the EV-sorting potential of 244 candidate proteins. We identify 24 proteins with conserved EV-sorting abilities across five types of producer cells. TSPAN2 and TSPAN3 emerge as lead candidates and outperform the well-studied CD63 scaffold. Importantly, these engineered EVs show promise as delivery vehicles in cell cultures and mice as demonstrated by efficient transfer of luminal cargo proteins as well as surface display of different functional entities. The discovery of these scaffolds provides a platform for EV-based engineering.

Rethink reporting of evaluation results in AI.

Aggregate metrics and lack of access to results limit understanding.

REsCue trial: Randomized controlled clinical trial with extended-release calcifediol in symptomatic COVID-19 outpatients.

OBJECTIVES: This double-blind randomized controlled trial investigated raising serum 25-hydroxyvitamin D (25D) with extended-release calcifediol (ERC) on time to symptom resolution in patients with mild to moderate COVID-19. METHODS: COVID-19 outpatients received oral ERC (300 mcg on days 1-3 and 60 mcg on days 4-27) or placebo (NCT04551911). Symptoms were self-reported daily. Primary end points were raising 25D to ≥50 ng/mL and decreasing resolution time for five aggregated symptoms (three respiratory). RESULTS: In all, 171 patients were randomized, 160 treated and 134 (65 ERC, 69 placebo) retained. The average age was 43 y (range 18-71), 59% were women. The mean baseline 25D was 37 ± 1 (SE) ng/mL. In the full analysis set (FAS), 81% of patients in the ERC group achieved 25D levels of ≥50 ng/mL versus 15% in the placebo group (P < 0.0001). In the per-protocol (PP) population, mean 25D increased with ERC to 82 ± 4 (SE) ng/mL (P < 0.0001) by day 7; the placebo group trended lower. Symptom resolution time was unchanged in the FAS by ERC (hazard ratio [HR], 0.983; 95% confidence interval [CI], 0.695-1.390; P = 0.922). In the PP population, respiratory symptoms resolved 4 d faster when 25D was elevated above baseline level at both days 7 and 14 (median 6.5 versus 10.5 d; HR, 1.372; 95% CI, 0.945-1.991; P = 0.0962; Wilcoxon P = 0.0386). Symptoms resolved in both treatment groups to a similar extent by study end. Safety concerns including hypercalcemia were absent with ERC treatment. CONCLUSION: ERC safely raised serum 25D to ≥50 ng/mL in outpatients with COVID-19, possibly accelerating resolution of respiratory symptoms and mitigating the risk for pneumonia. These findings warrant further study.

What is the simplest model that can account for high-fidelity imitation?

What inductive biases must be incorporated into multi-agent artificial intelligence models to get them to capture high-fidelity imitation? We think very little is needed. In the right environments, both instrumental- and ritual-stance imitation can emerge from generic learning mechanisms operating on non-deliberative decision architectures. In this view, imitation emerges from trial-and-error learning and does not require explicit deliberation.

Enabling Aromatic Hydroxylation in a Cytochrome P450 Monooxygenase Enzyme through Protein Engineering.

The cytochrome P450 (CYP) family of heme monooxygenases catalyse the selective oxidation of C-H bonds under ambient conditions. The CYP199A4 enzyme from Rhodopseudomonas palustris catalyses aliphatic oxidation of 4-cyclohexylbenzoic acid but not the aromatic oxidation of 4-phenylbenzoic acid, due to the distinct mechanisms of aliphatic and aromatic oxidation. The aromatic substrates 4-benzyl-, 4-phenoxy- and 4-benzoyl-benzoic acid and methoxy-substituted phenylbenzoic acids were assessed to see if they could achieve an orientation more amenable to aromatic oxidation. CYP199A4 could catalyse the efficient benzylic oxidation of 4-benzylbenzoic acid. The methoxy-substituted phenylbenzoic acids were oxidatively demethylated with low activity. However, no aromatic oxidation was observed with any of these substrates. Crystal structures of CYP199A4 with 4-(3'-methoxyphenyl)benzoic acid demonstrated that the substrate binding mode was like that of 4-phenylbenzoic acid. 4-Phenoxy- and 4-benzoyl-benzoic acid bound with the ether or ketone oxygen atom hydrogen-bonded to the heme aqua ligand. We also investigated whether the substitution of phenylalanine residues in the active site could permit aromatic hydroxylation. Mutagenesis of the F298 residue to a valine did not significantly alter the substrate binding position or enable the aromatic oxidation of 4-phenylbenzoic acid; however the F182L mutant was able to catalyse 4-phenylbenzoic acid oxidation generating 2'-hydroxy-, 3'-hydroxy- and 4'-hydroxy metabolites in a 83 : 9 : 8 ratio, respectively. Molecular dynamics simulations, in which the distance and angle of attack were considered, demonstrated that in the F182L variant, in contrast to the wild-type enzyme, the phenyl ring of 4-phenylbenzoic acid attained a productive geometry for aromatic oxidation to occur.

The impact of expanded endoscopic approaches on oncologic and functional outcomes for clival malignancies:a case series.

INTRODUCTION: Clival malignancies pose particular surgical challenges due to complex skull base anatomy and the involvement of vital neurovascular structures. While endoscopic endonasal approached are widely used, the outcomes for clival malignancies remain poorly understood. In this study we assessed the impact of endoscopic and open surgical approaches on PFS, time to initiation of radiotherapy, KPS, and GTR rates for clival malignancies. METHODS: A retrospective case series for clival malignancies operated between 1993 and 2019 was conducted. Inclusion criteria were age over 18 and a follow-up of at least a 6 months. Statistical analyses were conducted using STATA version 15 statistical software package StataCorp. RESULTS: For the whole cohort (113 patients), and for upper and middle lesions, open surgical approaches increased odds of disease progression, compared to EEA (HR 2.10 to HR 2.43), p < 0.05. EEA had a shorter time interval from surgery to initiation of radiotherapy. No difference in 6 and 12 month KPS was found between surgical groups. Patients undergoing open surgery were less likely to achieve GTR for upper clival lesions. CONCLUSIONS: EEA was found to be associated with increased PFS, for upper and middle clival malignancies. The time to initiation of radiotherapy was shorter for patients undergoing EEA compared to open surgery for patients with middle clival involvement. GTR rates were found to be significantly better with EEA for patients with upper clival malignancies.

Learning agents that acquire representations of social groups.

Humans are learning agents that acquire social group representations from experience. Here, we discuss how to construct artificial agents capable of this feat. One approach, based on deep reinforcement learning, allows the necessary representations to self-organize. This minimizes the need for hand-engineering, improving robustness and scalability. It also enables "virtual neuroscience" research on the learned representations.