The Impact of Race, Ethnicity, and Sex on Fairness in Artificial Intelligence for Glaucoma Prediction Models.

Objective: Despite advances in artificial intelligence (AI) in glaucoma prediction, most works lack multicenter focus and do not consider fairness concerning sex, race, or ethnicity. This study aims to examine the impact of these sensitive attributes on developing fair AI models that predict glaucoma progression to necessitating incisional glaucoma surgery. Design: Database study. Participants: Thirty-nine thousand ninety patients with glaucoma, as identified by International Classification of Disease codes from 7 academic eye centers participating in the Sight OUtcomes Research Collaborative. Methods: We developed XGBoost models using 3 approaches: (1) excluding sensitive attributes as input features, (2) including them explicitly as input features, and (3) training separate models for each group. Model input features included demographic details, diagnosis codes, medications, and clinical information (intraocular pressure, visual acuity, etc.), from electronic health records. The models were trained on patients from 5 sites (N = 27 999) and evaluated on a held-out internal test set (N = 3499) and 2 external test sets consisting of N = 1550 and N = 2542 patients. Main Outcomes and Measures: Area under the receiver operating characteristic curve (AUROC) and equalized odds on the test set and external sites. Results: Six thousand six hundred eighty-two (17.1%) of 39 090 patients underwent glaucoma surgery with a mean age of 70.1 (standard deviation 14.6) years, 54.5% female, 62.3% White, 22.1% Black, and 4.7% Latinx/Hispanic. We found that not including the sensitive attributes led to better classification performance (AUROC: 0.77-0.82) but worsened fairness when evaluated on the internal test set. However, on external test sites, the opposite was true: including sensitive attributes resulted in better classification performance (AUROC: external #1 - [0.73-0.81], external #2 - [0.67-0.70]), but varying degrees of fairness for sex and race as measured by equalized odds. Conclusions: Artificial intelligence models predicting whether patients with glaucoma progress to surgery demonstrated bias with respect to sex, race, and ethnicity. The effect of sensitive attribute inclusion and exclusion on fairness and performance varied based on internal versus external test sets. Prior to deployment, AI models should be evaluated for fairness on the target population. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Comparative analysis of codon usage patterns in 16 chloroplast genomes of suborder Halimedineae.

The Halimedineae are marine green macroalgae that play crucial roles as primary producers in various habitats, including coral reefs, rocky shores, embayments, lagoons, and seagrass beds. Several tropical species have calcified thalli, which contribute significantly to the formation of coral reefs. In this study, we investigated the codon usage patterns and the main factors influencing codon usage bias in 16 chloroplast genomes of the suborder Halimedineae. Nucleotide composition analysis revealed that the codons of these species were enriched in A/U bases and preferred to end in A/U bases, and the distribution of GC content followed a trend of GC1 > GC2 > GC3. 30 optimal codons encoding 17 amino acids were identified, and most of the optimal codons and all of the over-expressed codons preferentially ended with A/U. The neutrality plot, effective number of codons (ENc) plot, and parity rule 2 (PR2) plot analysis indicated that natural selection played a major role in shaping codon usage bias of the most Halimedineae species. The genetic relationships based on their RSCU values and chloroplast protein-coding genes showed the closely related species have similar codon usage patterns. This study describes, for the first time, the codon usage patterns and characterization of Halimedineae chloroplast genomes, and provides new insights into the evolution of this suborder.

Comment on 'A PET reconstruction formulation that enforces non-negativity in projection space for bias reduction in Y-90 imaging'.

We read with great interest the paper by Limet al(2018Phys. Med. Biol.63035042) on bias reduction in Y-90 PET imaging. In particular, they proposed a new formulation of the tomographic reconstruction problem that enforces non-negativity in projection space as opposed to image space. We comment on the algorithm they derived from this formulation and bring some clarifications on the constraint that this algorithm respects.

Sampling Strategies for Assessing Male Clients of Female Sex Workers in Public Health Research: A Compilation of Global Evidence.

Clients of Female Sex Workers (FSWs) are major bridge population in HIV transmission. Any research among them remains challenging because they are hidden within society. The objective of this review was tocompile the global evidence on different sampling strategies used to access male clients of female sex workers for research purpose, the challenges faced during the sampling process andpossible sources of bias. Original articles and reports published globally in last 10 years, in English language and those with full text freely available online were included in this scoping review. A comprehensive search was carried out among the electronic peer-reviewed literature database (Pubmed and Web of Science) using a pre-designed peer reviewed search strategy. Narrative synthesis was applied out across all such articles. A total of 36 articles were finally included in this review. The common sampling techniques used include convenience sampling, referral by FSWs/ pimp/brothel manager/clients, time location cluster sampling, use of virtual network, anonymous telephone survey, referred by clinicians of STI clinic etc. Overall response rate varied between 35 and 90%. Major challenges in participant recruitment included non-response, feasibility issue specially to cater non-brothel-based clients, safety issue for investigators, over-representation of clients with lower socio-economic status, higher refusal rate for known HIV positive clients to provide biological sample etc. As different sampling techniques have comparable response rate, it can be recommended that a pilot study should be carried out in local context to finalise appropriate participant recruitment technique for a given population.

On the Unequal Burden of Obesity: Obesity's Adverse Consequences Are Contingent on Regional Obesity Prevalence.

Obesity has adverse consequences for those affected. We tested whether the association between obesity and its adverse consequences is reduced in regions in which obesity is prevalent and whether lower weight bias in high-obese regions can account for this reduction. Studies 1 and 2 used data from the United States (N = 2,846,132 adults across 2,546 counties) and United Kingdom (N = 180,615 adults across 380 districts) that assessed obesity's adverse consequences in diverse domains: close relationships, economic outcomes, and health. Both studies revealed that the association between obesity and its adverse consequences is reduced (or absent) in high-obese regions. Study 3 used another large-scale data set (N = 409,837 across 2,928 U.S. counties) and revealed that lower weight bias in high-obese regions seems to account for (i.e., mediate) the reduction in obesity's adverse consequences. Overall, our findings suggest that obesity's adverse consequences are partly social and, thus, not inevitable.

Leveraging Temporal Trends for Training Contextual Word Embeddings to Address Bias in Biomedical Applications: Development Study.

BACKGROUND: Women have been underrepresented in clinical trials for many years. Machine-learning models trained on clinical trial abstracts may capture and amplify biases in the data. Specifically, word embeddings are models that enable representing words as vectors and are the building block of most natural language processing systems. If word embeddings are trained on clinical trial abstracts, predictive models that use the embeddings will exhibit gender performance gaps. OBJECTIVE: We aim to capture temporal trends in clinical trials through temporal distribution matching on contextual word embeddings (specifically, BERT) and explore its effect on the bias manifested in downstream tasks. METHODS: We present TeDi-BERT, a method to harness the temporal trend of increasing women's inclusion in clinical trials to train contextual word embeddings. We implement temporal distribution matching through an adversarial classifier, trying to distinguish old from new clinical trial abstracts based on their embeddings. The temporal distribution matching acts as a form of domain adaptation from older to more recent clinical trials. We evaluate our model on 2 clinical tasks: prediction of unplanned readmission to the intensive care unit and hospital length of stay prediction. We also conduct an algorithmic analysis of the proposed method. RESULTS: In readmission prediction, TeDi-BERT achieved area under the receiver operating characteristic curve of 0.64 for female patients versus the baseline of 0.62 (P<.001), and 0.66 for male patients versus the baseline of 0.64 (P<.001). In the length of stay regression, TeDi-BERT achieved a mean absolute error of 4.56 (95% CI 4.44-4.68) for female patients versus 4.62 (95% CI 4.50-4.74, P<.001) and 4.54 (95% CI 4.44-4.65) for male patients versus 4.6 (95% CI 4.50-4.71, P<.001). CONCLUSIONS: In both clinical tasks, TeDi-BERT improved performance for female patients, as expected; but it also improved performance for male patients. Our results show that accuracy for one gender does not need to be exchanged for bias reduction, but rather that good science improves clinical results for all. Contextual word embedding models trained to capture temporal trends can help mitigate the effects of bias that changes over time in the training data.

The I2 Test for Selection Bias Risk Assessment in Single Trials: Recommended Simulated Comparator Trial (SCT) Settings.

Selection bias in clinical trials is a form of systematic error and may be detected using the I² test with a 0/>0% threshold (bias: I² > 0%, no bias: I² = 0%). The test operates on the premise that effective randomisation eliminates in-between study heterogeneity beyond the play of chance in a baseline variable meta-analysis of all the trial's baseline variables. Since the I² statistic was originally designed to measure in-between study heterogeneity in meta-analyses, the test requires the generation of at least two simulated comparator trials (SCTs). During this process, three parameters are set: SCT sample size (NSCT), the minimum-maximum range of random values (RSCT), and the number of generated SCTs to be used (SCTN). Each of these parameters influences the 0/>0% threshold of the resulting I² point estimate, thereby affecting the test's sensitivity in indicating a positive result. The purpose of this technical report is to highlight the effect that SCT parameters have on the test's accuracy and to recommend appropriate parameter settings.

Protocol for a Series of Systematic Reviews and Network Meta-analyses of Randomized Controlled Trials of Medications for Patients with Overactive Bladder Symptoms.

Multiple randomized controlled trials (RCTs) have examined first-line pharmacological agents such as anticholinergics and ß3 agonists for the management of overactive bladder symptoms (OAB). Although earlier systematic reviews and (network) meta-analyses aimed to summarize the evidence, a substantial number of trials were not included, so a comprehensive and methodologically rigorous evaluation of the comparative effectiveness of all first-line pharmacological treatments is lacking. We aim to conduct a series of systematic reviews and network meta-analyses (NMAs) for a comprehensive assessment of the effectiveness and safety of first-line pharmacological treatments for OAB. Eligible studies will include RCTs comparing anticholinergics and ß3 agonists to one another or to placebo in adults with OAB or detrusor overactivity. Pairs of reviewers with methodological training will independently evaluate candidate studies to determine eligibility and extract relevant data. We will incorporate patient-important outcomes, including urinary urgency episodes, urgency incontinence episodes, any type of incontinence episodes, urinary frequency, nocturia, and adverse events. We will conduct the NMAs using a frequentist framework and a graph theory model for each outcome. Analysis will follow rigorous methodologies, including handling of missing data and assessment of the risk of bias. We will conduct sensitivity and subgroup analyses and will apply the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach to rate evidence certainty. Our approach aims to address the knowledge gap in the treatment of OAB by synthesizing evidence from RCTs worldwide. We will employ robust statistical methods, including frequentist NMA, to general clinically relevant and patient-important insights. Sensitivity and subgroup analyses will enhance the robustness and generalizability of our findings. Our reviews strive to inform evidence-based decisions in the management of OAB, to ultimate improve patient outcomes. Our study results may guide health policy decisions, such as reimbursement policies, and future studies in functional urology. The protocol for the review series is registered on PROSPERO as CRD42023266915.

Assessing inter- and intra-rater reliability of movement scores and the effects of body-shape using a custom visualisation tool: an exploratory study.

BACKGROUND: The literature shows conflicting results regarding inter- and intra-rater reliability, even for the same movement screen. The purpose of this study was to assess inter- and intra-rater reliability of movement scores within and between sessions of expert assessors and the effects of body-shape on reliability during a movement screen using a custom online visualisation software. METHODS: Kinematic data from 542 athletes performing seven movement tasks were used to create animations (i.e., avatar representations) using motion and shape capture from sparse markers (MoSh). For each task, assessors viewed a total of 90 animations. Using a custom developed visualisation tool, expert assessors completed two identical sessions where they rated each animation on a scale of 1-10. The arithmetic mean of weighted Cohen's kappa for each task and day were calculated to test reliability. RESULTS: Across tasks, inter-rater reliability ranged from slight to fair agreement and intra-rater reliability had slightly better reliability with slight to moderate agreement. When looking at the average kappa values, intra-rater reliability within session with and without body manipulation and between sessions were 0.45, 0.37, and 0.35, respectively. CONCLUSIONS: Based on these results, supplementary or alternative methods should be explored and are likely required to increase scoring objectivity and reliability even within expert assessors. To help future research and practitioners, the custom visualisation software has been made available to the public.

Association of Double-Blind Reviews with Increases in Women as First Authors: An Initial Report from the Journal of Cardiac Failure.

INTRODUCTION: Women continue to remain under-represented in academic publishing in cardiology. Some evidence suggests that double-blind peer reviews may mitigate the impact of gender bias. In July 2021, the Journal of Cardiac Failure implemented a process for the conduct of double-blind reviews after previously utilizing single-blind reviews with the aim of improving author diversity. The purpose of the current manuscript was to examine the association between changes in authorship characteristics and implementation of double-blind reviews. METHODS: Manuscripts were stratified into 3 eras: March - September 2021 (Era 1 - prior to double blind reviews), March - September 2022 (Era 2), and March - September 2023 (Era 3). All article types except invited editorials were included. Data were abstracted, including names, genders, ranks, and discipline of first and senior authors. RESULTS: A total of 310 manuscripts were included in the analysis. The proportion of women first authors increased from 24% in Era 1 to 34% in Era 2 to 39% in Era 3 while the percentage of women authors serving in a senior authorship role remained fairly stable over time around 21-22%. Even after adjusting for region, article type, first author discipline, and last author gender, there was an increase in female first author over time (p= 0.015). Manuscripts with a female senior author were significantly more likely to have a female first author. CONCLUSIONS: Our findings suggest that double-blind peer review may contribute to increased gender diversity of first authors and highlight areas for future improvement for JCF and academic publishing.

On the move: Influence of animal movements on count error during drone surveys.

The use of remote sensing to monitor animal populations has greatly expanded during the last decade. Drones (i.e., Unoccupied Aircraft Systems or UAS) provide a cost- and time-efficient remote sensing option to survey animals in various landscapes and sampling conditions. However, drone-based surveys may also introduce counting errors, especially when monitoring mobile animals. Using an agent-based model simulation approach, we evaluated the error associated with counting a single animal across various drone flight patterns under three animal movement strategies (random, directional persistence, and biased toward a resource) among five animal speeds (2, 4, 6, 8, 10 m/s). Flight patterns represented increasing spatial independence (ranging from lawnmower pattern with image overlap to systematic point counts). Simulation results indicated that flight pattern was the most important variable influencing count accuracy, followed by the type of animal movement pattern, and then animal speed. A  awnmower pattern with 0% overlap produced the most accurate count of a solitary, moving animal on a landscape (average count of 1.1 ± 0.6) regardless of the animal's movement pattern and speed. Image overlap flight patterns were more likely to result in multiple counts even when accounting for mosaicking. Based on our simulations, we recommend using a lawnmower pattern with 0% image overlap to minimize error and augment drone efficacy for animal surveys. Our work highlights the importance of understanding interactions between animal movements and drone survey design on count accuracy to inform the development of broad applications among diverse species and ecosystems.

On some problems of Bayesian region construction with guaranteed coverages.

The general problem of constructing regions that have a guaranteed coverage probability for an arbitrary parameter of interest ψ ∈ Ψ is considered. The regions developed are Bayesian in nature and the coverage probabilities can be considered as Bayesian confidences with respect to the model obtained by integrating out the nuisance parameters using the conditional prior given ψ . Both the prior coverage probability and the prior probability of covering a false value (the accuracy) can be controlled by setting the sample size. These coverage probabilities are considered as a priori figures of merit concerning the reliability of a study while the inferences quoted are Bayesian. Several problems are considered where obtaining confidence regions with desirable properties have proven difficult to obtain. For example, it is shown that the approach discussed never leads to improper regions which has proven to be an issue for some confidence regions.

Modeling the impact of MRI acquisition bias on structural connectomes: Harmonizing structural connectomes.

One way to increase the statistical power and generalizability of neuroimaging studies is to collect data at multiple sites or merge multiple cohorts. However, this usually comes with site-related biases due to the heterogeneity of scanners and acquisition parameters, negatively impacting sensitivity. Brain structural connectomes are not an exception: Being derived from T1-weighted and diffusion-weighted magnetic resonance images, structural connectivity is impacted by differences in imaging protocol. Beyond minimizing acquisition parameter differences, removing bias with postprocessing is essential. In this work we create, from the exhaustive Human Connectome Project Young Adult dataset, a resampled dataset of different b-values and spatial resolutions, modeling a cohort scanned across multiple sites. After demonstrating the statistical impact of acquisition parameters on connectivity, we propose a linear regression with explicit modeling of b-value and spatial resolution, and validate its performance on separate datasets. We show that b-value and spatial resolution affect connectivity in different ways and that acquisition bias can be reduced using a linear regression informed by the acquisition parameters while retaining interindividual differences and hence boosting fingerprinting performance. We also demonstrate the generative potential of our model, and its generalization capability in an independent dataset reflective of typical acquisition practices in clinical settings.

One of the main roadblocks to using multisite neuroimaging data is the effect of acquisition bias due to the heterogeneity of acquisition parameters associated with various sites. This can negatively impact the sensitivity of machine learning models employed in neuroscience. Thus, it is extremely important to model the effect of this bias. In this work, we address this issue at the level of brain structural connectivity, an important biomarker for various brain disorders. We propose a simple linear regression model to minimize this effect using high-quality data from the Human Connectome Project, and show its generalizability to a clinical dataset.

New Insight into the Role of Vitamin D in the Stroke Risk: A Meta-Analysis of Stratified Data by 25(OH)D Levels.

Mendelian Randomization (MR) studies have emerged as a powerful tool for investigating causal relationships between modifiable risk factors and clinical outcomes, using genetic variants as instrumental variables. In the context of vitamin D research, MR is a promising approach to elucidate the effects of vitamin D on various health outcomes, including adverse cardiovascular events. However, the validity of MR analyses relies heavily on the strength of the genetic associations found. "Weak instrument bias", arising from instruments with low explanatory power for the exposure of interest, can lead to biased estimates and compromise causal inference. We have, herein, briefly reviewed the challenges posed by weak instrument bias in a large MR study on vitamin D [25(OH)D] and stroke, exploring implications for the study's validity and reliability of findings. We have then added an original meta-analysis stratified by 25(OH)D levels. By using aggregated data from a recent MR study, an original meta-analysis stratified by population mean levels of 25(OH)D has indicated that interventions based on vitamin D supplementations in population mean levels ranging from 50 to 70 nmol/L are likely to translate into a 13% reduction of stroke risk (pooled odds ratio=0.873, 95% CI: 0.764-0.997, p-value=0.04). MR studies are a valuable approach for discerning causal relationships between exposures, such as vitamin D, and health outcomes. However, the effectiveness of MR analyses depends on the robustness of the genetic instruments employed. By recognizing and addressing weak instrument bias in MR studies of vitamin D, researchers can enhance the credibility and utility of causal inference in understanding the health effects of this essential nutrient. A metaanalysis stratified by population mean levels of 25(OH)D has revealed the potential benefits of targeted interventions with vitamin D supplementations for stroke.

How context influences the processing of relevant information and judgment accuracy-the role of information restriction in judgment processes in diagnosing misconceptions.

To adapt teaching to the prerequisites of students, teachers have various options at their disposal to gather and process information as the basis to form a judgment, such as carrying out tests, talking to and observing the behavior of students, or administering tasks. The complexity of such a judgment arises from the multitude of observations and their different possible explanations. This complexity might be reduced when teachers focus on one hypothesis instead of considering multiple hypotheses, interpret information in a confirmatory way, and not collect diagnostically relevant information. However, in this way, they run the risk of undesirable biased judgments. It therefore seems important to improve diagnostic judgments by selecting and processing information in a more reflective way. Research indicates that if information on a student is not easily available but restricted (e.g., by time pressure, difficult access to the student or high effort), a teacher who wants to make a careful decision is forced to rely on more reflective processes in the selection of tasks and in the interpretation of solutions. The present experimental study therefore investigates how the restricted availability of information in a specific diagnostic situation-when diagnostically inexperienced prospective mathematics teachers determine misconceptions in decimal fractions-influences the underlying cognitive processes. We assume that restricting the availability of information on student behavior augments the attentional focus and therefore reduces cognitive biases. Such more reflective processing can be observed by an increased time spent per piece of information, which should lead to the processing of relevant information and further increase judgment accuracy. To investigate these hypotheses, prospective teachers without prior knowledge in diagnosing misconceptions (N = 81) were asked to diagnose misconceptions on decimal fractions of virtual students by collecting information on students' solutions. Data concerning the effects of restricting the availability of information on teachers' cognitive processes were analyzed. The results show that with restricted information, participants indeed select a greater proportion of diagnostically relevant tasks, which positively influences judgment accuracy. These results are discussed with respect to their significance for framing teacher training and for further research.

Comparing self-other distinction across motor, cognitive and affective domains.

The self-other distinction (SOD) is a process by which humans disentangle self from other-related mental representations. This online study investigated two unresolved questions: (i) whether partially the same processes underpin SOD for motor, cognitive and affective representations, and (ii) whether SOD overlaps with domain-general cognitive control processes. Participants (N = 243) performed three SOD tasks (motor: automatic imitation inhibition (AIT); cognitive: visual perspective-taking (VPT); affective: emotional egocentricity bias (av-EEB) tasks) and two cognitive control tasks (Stroop and stop-signal reaction time (SSRT) tasks). Correlation analyses showed no associations among the motor, cognitive and affective SOD indexes. Similarly, distinct SOD clusters emerged in the hierarchical clustering dendrogram, indicating clear separations among SODs. However, the results of multidimensional scaling suggested a tendency towards two clusters, as evidenced by the proximity of AIT and VPT indexes in relation to EEB indexes. AIT spatial laterality and Stroop domain-general cognitive control confounded AIT and VPT indexes, albeit slightly differently depending on the analysis method used. SSRT showed neither associations with SODs nor with other domain-general indexes. These findings underscore the complexity of SOD processes and have notable implications for basic and applied research, e.g. in the domain of clinical disorders affected by deficiencies in SOD.

Late Talkers can generalise trained labels by object shape similarities, but not unfamiliar labels.

Late talkers (LTs) exhibit delayed vocabulary development, which might stem from a lack of a typical word learning strategy to generalise object labels by shape, called the 'shape bias'. We investigated whether LTs can acquire a shape bias and whether this accelerates vocabulary learning. Fourteen LTs were randomly allocated to either a shape training group (Mage = 2.76 years, 6 males), which was taught that objects similar in shape have the same name, or a control group (Mage = 2.61 years, 4 males), which was taught real words without any focus on object shape. After seven training sessions, children in the shape training group generalised trained labels by shape (d = 1.28), but not unfamiliar labels. Children in the control group extended all labels randomly. Training did not affect expressive vocabulary.

Survival Analysis of Conversion Surgery in Borderline Resectable and Locally Advanced Unresectable Pancreatic Ductal Adenocarcinoma Addressing Selection and Immortal Time Bias: A Retrospective Single-Center Study.

BACKGROUND: The purpose of this study was to provide a detailed evaluation of the oncological advantages of surgery following neoadjuvant chemotherapy (NAC) for patients with borderline resectable (BR) or unresectable (UR) pancreatic ductal adenocarcinoma (PDAC), with a focus on minimizing biases. Recently, NAC has become the standard care for BR or UR locally advanced (UR-LA) PDAC, however, many studies have assessed survival benefits and favorable variables without consideration for biases, particularly immortal time bias. PATIENTS AND METHODS: This study included patients diagnosed with BR or UR-LA PDAC at Juntendo University Hospital from 2019 to 2022. To mitigate bias, we applied methods such as propensity score matching (PSM), time-dependent covariate Cox proportional hazard regression analysis (TDC), landmark analysis, and multivariable Cox proportional hazards regression model. RESULTS: The study analyzed 124 patients, dividing them into a surgery group (n = 57) and a chemotherapy-only group (n = 67). After PSM, there were 21 matched pairs. Survival analysis using TDC analysis showed that the surgery group had significantly better overall survival compared with the chemotherapy-only group in both the entire cohort and the matched pairs. Cox regression analysis of the entire cohort also revealed a similar superiority of surgery, while the landmark analysis showed varying results depending on the landmark setting. CONCLUSIONS: After careful adjustment for selection and immortal time biases, surgery following NAC appears to significantly extend survival in patients with BR or UR PDAC.

Gender and age biases in the assessment of speech accuracy: A study of speech-language clinicians' ratings of /s/ accuracy.

PURPOSE: Typically developing children assigned male at birth (AMAB) and children assigned female at birth (AFAB) produce the fricative /s/ differently: AFAB children produce /s/ with a higher spectral peak frequency. This study examined whether implicit knowledge of these differences affects speech-language pathologists'/speech and language therapists' (SLPs'/SLTs') ratings of /s/ accuracy, by comparing ratings made in conditions where SLPs/SLTs were blind to children's sex assigned at birth (SAB) to conditions in which they were told this information. METHODS: SLPs (n = 95) varying in clinical experience rated the accuracy of word-initial /s/ productions (n = 87) of eight children with speech sound disorder in one of four conditions: one in which no information about the children was revealed, one in which children's SAB was revealed, one in which children's age was revealed, and one in which both were revealed. RESULTS: Despite there being no statistically significant differences between AFAB and AMAB children's /s/ production in researcher-determined accuracy or in one acoustic characteristic, spectral centroid, SLPs in all four conditions judged the /s/ productions of AFAB children as more accurate than AMAB children. Listeners were significantly less likely to judge the productions of AMAB children to be inaccurate in the conditions in which age or age and SAB were revealed. These effects were consistent across SLPs with greatly varying levels of clinical experience. CONCLUSION: Knowing or imputing children's age and SAB can affect ratings of /s/ accuracy. Clinicians should be mindful of these potential effects. Future research should understand how expectations about sociolinguistic variation in speech affect appraisals of their speech and language. WHAT THIS PAPER ADDS: What is already known on the subject Adult men and women produce /s/ differently. A consensus is that these differences reflect sociolinguistic gender marking, rather than being the passive consequence of vocal-tract differences. Recent studies have shown that children assigned female at birth (AFAB) and those assigned male at birth (AMAB) produce /s/ differently in ways that mirror the differences between adult men and women, and which presumably reflect gender marking. What this paper adds to existing knowledge We asked whether US-based speech-language pathologists' (SLPs) ratings of the accuracy of /s/ differ depending on whether they are rating an AFAB or an AMAB child, and whether these differences are greater in conditions in which people are told the sex assigned at birth of the child being rated. We found that SLPs were more likely to judge AFAB children's /s/ productions to be more accurate than AMAB children's, even though the productions from the AMAB and AFAB children that were used as stimuli were matched for accuracy as determined by trained researchers. What are the clinical implications of this work? SLPs/speech-language therapists should be sensitive to the influence of social variables when assessing /s/. SLPs/speech-language therapists might rate children's productions differently depending on whether they believe they are rating an AFAB or an AMAB child.

Assessment of Reduction in Stimulus Generalization of Ethanol-Seeking During Recovery: A Rapid Procedure.

Previously, we developed a procedure which showed that longer histories of reinforced alternative behavior decrease the risk of relapse caused by a range of stimuli which had previously occasioned drinking. The decrease in relapse risk was likely due to a decrease in attention to the stimuli over the course of repeated engagement in the alternative behavior. However, this previous procedure was time consuming and did not mirror the procedure we used to observe changes in relapse risk. This study aimed at replicating the previous relationship between the duration of engaging in an alternative behavior and shift in stimulus generalization for drinking using a procedure that allows longitudinal analysis over time and is consistent with other procedures we have developed. Rats were trained to respond for ethanol in the presence of one stimulus (16kHz tone; food Fixed Ratio (FR)150 and ethanol FR5), and for food in the under another stimulus (8kHz tone; food and ethanol FR5). Then, recovery-like sessions with food predominant responding occurred in the presence of only the low-cost food stimulus. During these sessions, rats were exposed to non-reinforced graded stimuli alternation from 8 to 16kHz alternating with the reinforced low-cost food stimulus. The number of responses on each (food and ethanol) lever before completing 5 responses on either lever was the main measure. Consistent with the earlier procedure, the current procedure showed that graded variation of tone from 8 to 16kHz produced a graded increase in responding for ethanol compared to responding for food. In addition, longer periods of engaging in recovery-like responding shift the generalization function downwards. This procedure confirms the earlier pattern of stimulus generalization over longer periods of behavior consistent with recovery. This strengthens our hypothesis that shifts in attention to alcohol-related stimuli are important to the development of relapse resistance during recovery.