The effect of volatility in linguistic input on prediction behavior in autistic toddlers.

Domain-general prediction differences have been posited as underlying many aspects of the cognitive-behavioral profile in autism. An interesting potential implication of such differences is hyperplasticity of learning-the idea that autistic individuals may privilege more recent input over the accumulation of prior learning. Because real world language input is highly variable, hyperplasticity could have serious ramifications for language learning. To investigate potential hyperplasticity during a language processing task, we administered an experimental anticipatory eye movement (AEM) task to 2- to 3-year-old autistic children and neurotypical (NT) peers. Autistic children's change in anticipation from before to after a switch in contingencies did not significantly differ from NT counterparts, failing to support claims of hyperplasticity in the linguistic domain. Analysis of individual differences among autistic children revealed that cognitive ability was associated with prediction of the initial, stable contingencies, but neither age nor receptive language related to task performance. Results are discussed in terms of clinical implications and the broader context of research investigating prediction differences in autism.

Clinical Prediction Models in Neurocritical Care: An Overview of the Literature and Example Application to Prediction of Hospital Mortality in Traumatic Brain Injury.

Clinical prediction models serve as valuable instruments for assessing the risk of crucial outcomes and facilitating decision-making in clinical settings. Constructing these models requires nuanced analytical decisions and expertise informed by the current statistical literature. Access and thorough understanding of such literature may be limited for neurocritical care physicians, which may hinder the interpretation of existing predictive models. The present emphasis is on narrowing this knowledge gap by providing neurocritical care specialists with methodological guidance for interpreting predictive models in neurocritical care. Presented are the statistical learning principles integral to constructing a model predicting hospital mortality (nonsurvival during hospitalization) in patients with moderate and severe blunt traumatic brain injury using components of the IMPACT-Core model. Discussion encompasses critical elements such as model flexibility, hyperparameter selection, data imbalance, cross-validation, model assessment (discrimination and calibration), prediction instability, and probability thresholds. The intricate interplay among these components, the data set, and the clincal context of neurocritical care is elaborated. Leveraging this comprehensive exploration of statistical learning can enhance comprehension of articles encompassing model generation, tailored clinical care, and, ultimately, better interpretation and clinical applicability of predictive models.

Peripheral vision contributes to implicit attentional learning: Findings from the "mouse-eye" paradigm.

The central visual field is essential for activities like reading and face recognition. However, the impact of peripheral vision loss on daily activities is profound. While the importance of central vision is well established, the contribution of peripheral vision to spatial attention is less clear. In this study, we introduced a "mouse-eye" method as an alternative to traditional gaze-contingent eye tracking. We found that even in tasks requiring central vision, peripheral vision contributes to implicit attentional learning. Participants searched for a T among Ls, with the T appearing more often in one visual quadrant. Earlier studies showed that participants' awareness of the T location probability was not essential for their ability to learn. When we limited the visible area around the mouse cursor, only participants aware of the target's location probability showed learning; those unaware did not. Adding placeholders in the periphery did not restore implicit attentional learning. A control experiment showed that when participants were allowed to see all items while searching and moving the mouse to reveal the target's color, both aware and unaware participants acquired location probability learning. Our results underscore the importance of peripheral vision in implicitly guided attention. Without peripheral vision, only explicit, but not implicit, attentional learning prevails.

Risk factors of early spontaneous preterm birth despite carrying a cervical pessary in singleton pregnancies with a short cervix: Development of a risk prediction model.

Introduction: We aimed to identify the incidence and risk factors of spontaneous preterm birth in pessary carriers with singleton pregnancies and a short cervix in the mid-trimester of pregnancy. Material and Methods: Patient data were obtained from the PECEP Trial. We analyzed singleton pregnancies in pessary carriers with a short cervix (≤25 mm) between 18 and 22 gestational weeks. Demographics and obstetric history were compared to identify risk factors for spontaneous preterm birth < 34 gestational weeks. Each demographic and obstetric variable was compared between spontaneous preterm birth < 34 and ≥ 34 weeks of gestation.Regression analysis was used to identify risk factors. A risk score model was generated using the odds ratio for significant factors. The risk score model and spontaneous preterm birth risk were assessed using the receiver operating characteristic curve. Perinatal outcomes were compared by risk score. Results: Among 190 pregnant individuals, 12 (6.3%) had spontaneous preterm birth < 34 gestational weeks. In the bivariate analysis, statistically significant differences between those with and without spontaneous preterm birth were only observed for mean cervical length at diagnosis and mean cervical length after pessary placement. By multiple logistic regression analysis, maternal age (OR 0.818; 95% CI 0.69-0.97; P 0.020), cervical length at diagnosis (OR 0.560; 95% CI 0.43-0.73; P < 0.001) and smoking status (OR 7.276; 95% CI 1.02-51.80; P 0.048) remained significantly associated with spontaneous preterm birth.The ROC curve from the multiple logistic regression analysis, including cervical length, maternal age and smoking status, had an area under the curve (AUC) of 0.952 (P < 0.001). The ROC curve for the risk score model incorporating all three variables had an AUC of 0.864 (95% CI 0.77-0.96; P < 0.001). A high-risk score was predictive of spontaneous preterm birth with a sensitivity of 75%, specificity of 84%, positive predictive value of 24%, and negative predictive value of 98%.Women with a high-risk score had a significantly reduced latency to delivery and poorer neonatal outcomes than those with a low-risk score. Conclusions: Patients at a high risk for spontaneous preterm birth despite pessary therapy may be identified using cervical length at diagnosis added to maternal age and smoking status.

Attenuation of spatial bias with target template variation.

This study investigated the impact of target template variation or consistency on attentional bias in location probability learning. Participants conducted a visual search task to find a heterogeneous shape among a homogeneous set of distractors. The target and distractor shapes were either fixed throughout the experiment (target-consistent group) or unpredictably varied on each trial (target-variant group). The target was often presented in one possible search region, unbeknownst to the participants. When the target template was consistent throughout the biased visual search, spatial attention was persistently biased toward the frequent target location. However, when the target template was inconsistent and varied during the biased search, the spatial bias was attenuated so that attention was less prioritized to a frequent target location. The results suggest that the alternative use of target templates may interfere with the emergence of a persistent spatial bias. The regularity-based spatial bias depends on the number of attentional shifts to the frequent target location, but also on search-relevant contexts.

Identification of Atrial Fibrillation With Single-Lead Mobile ECG During Normal Sinus Rhythm Using Deep Learning.

BACKGROUND: The acquisition of single-lead electrocardiogram (ECG) from mobile devices offers a more practical approach to arrhythmia detection. Using artificial intelligence for atrial fibrillation (AF) identification enhances screening efficiency. However, the potential of single-lead ECG for AF identification during normal sinus rhythm (NSR) remains under-explored. This study introduces a method to identify AF using single-lead mobile ECG during NSR. METHODS: We employed three deep learning models: recurrent neural network (RNN), long short-term memory (LSTM), and residual neural networks (ResNet50). From a dataset comprising 13,509 ECGs from 6,719 patients, 10,287 NSR ECGs from 5,170 patients were selected. Single-lead mobile ECGs underwent noise filtering and segmentation into 10-second intervals. A random under-sampling was applied to reduce bias from data imbalance. The final analysis involved 31,767 ECG segments, including 15,157 labeled as masked AF and 16,610 as Healthy. RESULTS: ResNet50 outperformed the other models, achieving a recall of 79.3%, precision of 65.8%, F1-score of 71.9%, accuracy of 70.5%, and an area under the receiver operating characteristic curve (AUC) of 0.79 in identifying AF from NSR ECGs. Comparative performance scores for RNN and LSTM were 0.75 and 0.74, respectively. In an external validation set, ResNet50 attained an F1-score of 64.1%, recall of 68.9%, precision of 60.0%, accuracy of 63.4%, and AUC of 0.68. CONCLUSION: The deep learning model using single-lead mobile ECG during NSR effectively identified AF at risk in future. However, further research is needed to enhance the performance of deep learning models for clinical application.

Contribution of peripheral vision to attentional learning.

Attention is tuned towards locations that frequently contain a visual search target (location probability learning; LPL). Peripheral vision, covering a larger field than the fovea, often receives information about the target. Yet what is the role of peripheral vision in attentional learning? Using gaze-contingent eye tracking, we examined the impact of simulated peripheral vision loss on location probability learning. Participants searched for a target T among distractor Ls. Unbeknownst to them, the T appeared disproportionately often in one quadrant. Participants searched with either intact vision or "tunnel vision," restricting the visible search items to the central 6.7º (in diameter) of the current gaze. When trained with tunnel vision, participants in Experiment 1 acquired LPL, but only if they became explicitly aware of the target's location probability. The unaware participants were not faster finding the target in high-probability than in low-probability locations. When trained with intact vision, participants in Experiment 2 successfully acquired LPL, regardless of whether they were aware of the target's location probability. Thus, whereas explicit learning may proceed with central vision alone, implicit LPL is strengthened by peripheral vision. Consistent with Guided Search (Wolfe, 2021), peripheral vision supports a nonselective pathway to guide visual search.

Stability of individual differences in implicitly guided attention.

Daily activities often occur in familiar environments, affording us an opportunity to learn. Laboratory studies have shown that people readily acquire an implicit spatial preference for locations that frequently contained a search target in the past. These studies, however, have focused on group characteristics, downplaying the significance of individual differences. In a pre-registered study, we examined the stability of individual differences in two variants of an implicit location probability learning (LPL) task. We tested the possibility that individual differences were stable in variants that shared the same search process, but not in variants involving different search processes. In Experiment 1, participants performed alternating blocks of T-among-Ls and 5-among-2s search tasks. Unbeknownst to them, the search target appeared disproportionately often in one region of space; the high-probability regions differed between the two tasks. LPL transferred between the two tasks. In addition, individuals who showed greater LPL in the T-task also did so in the 5-task and vice versa. In Experiment 2, participants searched for either a camouflaged-T against background noise or a well-segmented T among well-segmented Ls. These two tasks produced task-specific learning that did not transfer between tasks. Moreover, individual differences in learning did not correlate between tasks. Thus, LPL is associated with stable individual differences across variants, but only when the variants share common search processes.

Looking away to see: The acquisition of a search habit away from the saccade direction.

Growing evidence has shown that attention can be habit-like, unconsciously and persistently directed toward locations that have frequently contained search targets in the past. The attentional preference typically arises when the eye gaze aligns with the attended location. Here we tested whether this spatial alignment is necessary for the acquisition of a search habit. To divert eye movements away from an attended location, we used gaze-contingent eye tracking, restricting the visible portion of the screen to an area opposite to the current gaze. Participants searched for a T target amidst a circular array of L distractors. Unbeknownst to them, the target appeared more frequently in one screen quadrant. Despite fixating on a location diametrically opposite to the visible, attended region, participants acquired probability cuing, producing quicker responses when the target appeared in the high-probability quadrant. They also showed a speed advantage in the diagonal quadrant. The attentional preference for the high-probability quadrant persisted during a testing phase in which the target's location was unbiased, but only when participants continued to search with the restricted view. These results indicate that a search habit can be acquired even when participants are required to look away from the high-probability locations. The finding suggests that the learned search habit is not solely a result of oculomotor learning.

A machine learning model to assess potential misdiagnosed dengue hospitalization.

Dengue, like other arboviruses with broad clinical spectra, can easily be misdiagnosed as other infectious diseases due to the overlap of signs and symptoms. During large outbreaks, severe dengue cases have the potential to overwhelm the health care system and understanding the burden of dengue hospitalizations is therefore important to better allocate medical care and public health resources. A machine learning model that used data from the Brazilian public healthcare system database and the National Institute of Meteorology (INMET) was developed to estimate potential misdiagnosed dengue hospitalizations in Brazil. The data was modeled into a hospitalization level linked dataset. Then, Random Forest, Logistic Regression and Support Vector Machine algorithms were assessed. The algorithms were trained by dividing the dataset in training/test set and performing a cross validation to select the best hyperparameters in each algorithm tested. The evaluation was done based on accuracy, precision, recall, F1 score, sensitivity, and specificity. The best model developed was Random Forest with an accuracy of 85% on the final reviewed test. This model shows that 3.4% (13,608) of all hospitalizations in the public healthcare system from 2014 to 2020 could have been dengue misdiagnosed as other diseases. The model was helpful in finding potentially misdiagnosed dengue and might be a useful tool to help public health decision makers in planning resource allocation.

Prediction of homologous recombination deficiency from cancer gene expression data.

OBJECTIVE: Homologous recombination deficiency (HRD) is the main mechanism of tumorigenesis in some cancers. HRD causes abnormal double-strand break repair, resulting in genomic scars. Some scoring HRD tests have been approved as companion diagnostics of polyadenosine diphosphate-ribose polymerase (PARP) inhibitor treatment. This study aimed to build an HRD prediction model using gene expression data from various cancer types. METHODS: The cancer genome atlas data were used for HRD prediction modeling. A total of 10,567 cases of 33 cancer types were included, and expression data from 5128 out of 20,502 genes were included as predictors. A penalized logistic regression model was chosen as a modeling technique. RESULTS: The area under the curve of the receiver operating characteristic curve of HRD status prediction was 0.98 for the training set and 0.93 for the test set. The accuracy of HRD status prediction was 0.93 for the training set and 0.88 for the test set. CONCLUSIONS: Our study suggests that the HRD prediction model based on penalized logistic regression using gene expression data can be used to select patients for treatment with PARP inhibitors.

Interaction of phonological biases and frequency in learning a probabilistic language pattern.

We examine how learning a phonological rule in an artificial language interacts with morphological and lexical learning. We exposed adult participants to an artificial language in which noun plurals were marked by one of two prefix forms (ba- or ni-), one of which also triggered a velar palatalization rule (e.g., singular kimu, plural ni-chimu). In some conditions, the rule additionally created homophony. We also manipulated the relative frequency of the two prefix variants. The results showed that participants shifted away from using the rule-triggering prefix (ni-), but only when it was already the less frequent prefix. We attribute this effect to a paradigm uniformity bias leading participants to avoid phonological alternations (particularly in the stem). When the rule created homophony between lexical items, participants were less able to learn the rule, but it did not affect their choice of prefix. We attribute this effect to homophony avoidance interfering with participants' ability to extract the phonological generalization.

Prediction of preterm birth using artificial intelligence: a systematic review.

Preterm birth is the leading cause of neonatal death. It is challenging to predict preterm birth. We elucidated the state of artificial intelligence research on the prediction of preterm birth, clarifying the predictive values and accuracy. We performed a systematic review using three databases (PubMed, Web of Science, and Scopus) in August 2020, with keywords as 'artificial intelligence,' 'deep learning,' 'machine learning,' and 'neural network' combined with 'preterm birth'. We included 22 publications between 2010 and 2020. Regarding the predictive values, electrohysterogram images were mostly used, followed by the biological profiles, the metabolic panel in amniotic fluid or maternal blood, and the cervical images on the ultrasound examination. The size of dataset in most studies was hundred cases and too small for learning, although only three studies used the medical database over a hundred thousand cases. The accuracy was better in the studies using the metabolic panel and electrohysterogram images. Impact statementWhat is already known on this subject? Preterm birth is the leading cause of newborn morbidity and mortality. Presently, the prediction of preterm birth in individual cases is still challenging.What the results of this study add? Using artificial intelligence such as deep learning and machine learning models, clinical data could lead to accurate prediction of preterm birth.What the implications are of these findings for clinical practice and/or further research? The size of the datasets was too small for the models using artificial intelligence in the previous studies. Big data should be prepared for the future studies.

Habit-like attentional bias is unlike goal-driven attentional bias against spatial updating.

Statistical knowledge of a target's location may benefit visual search, and rapidly understanding the changes in regularity would increase the adaptability in visual search situations where fast and accurate performance is required. The current study tested the sources of statistical knowledge-explicitly-given instruction or experience-driven learning-and whether they affect the speed and location spatial attention is guided. Participants performed a visual search task with a statistical regularity to bias one quadrant ("old-rich" condition) in the training phase, followed by another quadrant ("new-rich" condition) in the switching phase. The "instruction" group was explicitly instructed on the regularity, whereas the "no-instruction" group was not. It was expected that the instruction group would rely on goal-driven attention (using regularities with explicit top-down knowledge), and the no-instruction group would rely on habit-like attention (learning regularities through repetitive experiences) in visual search. Compared with the no-instruction group, the instruction group readjusted spatial attention following the regularity switch more rapidly. The instruction group showed greater attentional bias toward the new-rich quadrant than the old-rich quadrant; however, the no-instruction group showed a similar extent of attentional bias to two rich quadrants. The current study suggests that the source of statistical knowledge can affect attentional allocation. Moreover, habit-like attention, a different type of attentional source than goal-driven attention, is relatively implicit and inflexible.

Independence of implicitly guided attention from goal-driven oculomotor control.

Location probability learning-the acquisition of an attentional bias toward locations that frequently contained a search target-shows many characteristics of a search habit. To what degree does it depend on oculomotor control, as might be expected if habit-like attention is grounded in eye movements? Here, we examined the impact of a spatially incompatible oculomotor signal on location probability learning (LPL). On each trial of a visual search task, participants first saccaded toward a unique C-shape, whose orientation determined whether participants should continue searching for a T target among L distractors. The C-shape often appeared in one, "C-rich" quadrant that differed from where the T was frequently located. Experiment 1 showed that participants acquired LPL toward the high-probability, "T-rich" quadrant, an effect that persisted in an unbiased testing phase. Participants were also faster finding the target in the vicinity of the C-shape, but this effect did not persist after the C-shape was removed. Experiment 2 found that the C-shape affected search only when it was task-relevant. Experiment 3 replicated and extended the findings of Experiment 1 using eye tracking. Thus, location probability learning is robust in the face of a spatially incompatible saccade, demonstrating partial independence between experience-guided attention and goal-driven oculomotor control. The findings are in line with the modular view of attention, which conceptualizes the search habit as a high-level process abstracted from eye movements.

Pupil dilation and response slowing distinguish deliberate explorative choices in the probabilistic learning task.

This study examined whether pupil size and response time would distinguish directed exploration from random exploration and exploitation. Eighty-nine participants performed the two-choice probabilistic learning task while their pupil size and response time were continuously recorded. Using LMM analysis, we estimated differences in the pupil size and response time between the advantageous and disadvantageous choices as a function of learning success, i.e., whether or not a participant has learned the probabilistic contingency between choices and their outcomes. We proposed that before a true value of each choice became known to a decision-maker, both advantageous and disadvantageous choices represented a random exploration of the two options with an equally uncertain outcome, whereas the same choices after learning manifested exploitation and direct exploration strategies, respectively. We found that disadvantageous choices were associated with increases both in response time and pupil size, but only after the participants had learned the choice-reward contingencies. For the pupil size, this effect was strongly amplified for those disadvantageous choices that immediately followed gains as compared to losses in the preceding choice. Pupil size modulations were evident during the behavioral choice rather than during the pretrial baseline. These findings suggest that occasional disadvantageous choices, which violate the acquired internal utility model, represent directed exploration. This exploratory strategy shifts choice priorities in favor of information seeking and its autonomic and behavioral concomitants are mainly driven by the conflict between the behavioral plan of the intended exploratory choice and its strong alternative, which has already proven to be more rewarding.

Spatial enhancement due to statistical learning tracks the estimated spatial probability.

It is well known that attentional selection is sensitive to the regularities presented in the display. In the current study we employed the additional singleton paradigm and systematically manipulated the probability that the target would be presented in one particular location within the display (probabilities of 30%, 40%, 50%, 60%, 70%, 80%, and 90%). The results showed the higher the target probability, the larger the performance benefit for high- relative to low-probability locations both when a distractor was present and when it was absent. We also showed that when the difference between high- and low-probability conditions was relatively small (30%) participants were not able to learn the contingencies. The distractor presented at a high-probability target location caused more interference than when presented at a low-probability target location. Overall, the results suggest that attentional biases are optimized to the regularities presented in the display tracking the experienced probabilities of the locations that were most likely to contain a target. We argue that this effect is not strategic in nature nor the result of repetition priming. Instead, we assume that through statistical learning the weights within the spatial priority map are adjusted optimally, generating the efficient selection priorities.

Space, feature, and risk sensitivity in homing pigeons (Columba livia): Broadening the conversation on the role of the avian hippocampus in memory.

David Sherry has been a pioneer in investigating the avian hippocampal formation (HF) and spatial memory. Following on his work and observations that HF is sensitive to the occurrence of reward (food), we were interested in carrying out an exploratory study to investigate possible HF involvement in the representation goal value and risk. Control sham-lesioned and hippocampal-lesioned pigeons were trained in an open field to locate one food bowl containing a constant two food pellets on all trials, and two variable bowls with one containing five pellets on 75% (High Variable) and another on 25% (Low Variable) of their respective trials (High-Variable and Low-Variable bowls were never presented together). One pairing of pigeons learned bowl locations (space); another bowl colors (feature). Trained to color, hippocampal-lesioned pigeons performed as rational agents in their bowl choices and were indistinguishable from the control pigeons, a result consistent with HF regarded as unimportant for non-spatial memory. By contrast, when trained to location, hippocampal-lesioned pigeons differed from the control pigeons. They made more first-choice errors to bowls that never contained food, consistent with a role of HF in spatial memory. Intriguingly, the hippocampal-lesioned pigeons also made fewer first choices to both variable bowls, suggesting that hippocampal lesions resulted in the pigeons becoming more risk averse. Acknowledging that the results are preliminary and further research is needed, the data nonetheless support the general hypothesis that HF-dependent memory representations of space capture properties of reward value and risk, properties that contribute to decision making when confronted with a choice.

The Accuracy of Causal Learning Over Long Timeframes: An Ecological Momentary Experiment Approach.

The ability to learn cause-effect relations from experience is critical for humans to behave adaptively - to choose causes that bring about desired effects. However, traditional experiments on experience-based learning involve events that are artificially compressed in time so that all learning occurs over the course of minutes. These paradigms therefore exclusively rely upon working memory. In contrast, in real-world situations we need to be able to learn cause-effect relations over days and weeks, which necessitates long-term memory. 413 participants completed a smartphone study, which compared learning a cause-effect relation one trial per day for 24 days versus the traditional paradigm of 24 trials back- to- back. Surprisingly, we found few differences between the short versus long timeframes. Subjects were able to accurately detect generative and preventive causal relations, and they exhibited illusory correlations in both the short and long timeframe tasks. These results provide initial evidence that experience-based learning over long timeframes exhibits similar strengths and weaknesses as in short timeframes. However, learning over long timeframes may become more impaired with more complex tasks.

Toward Dynamic Risk Prediction of Outcomes After Coronary Artery Bypass Graft: Improving Risk Prediction With Intraoperative Events Using Gradient Boosting.

BACKGROUND: Intraoperative data may improve models predicting postoperative events. We evaluated the effect of incorporating intraoperative variables to the existing preoperative model on the predictive performance of the model for coronary artery bypass graft. METHODS: We analyzed 378 572 isolated coronary artery bypass graft cases performed across 1083 centers, using the national Society of Thoracic Surgeons Adult Cardiac Surgery Database between 2014 and 2016. Outcomes were operative mortality, 5 postoperative complications, and composite representation of all events. We fitted models by logistic regression or extreme gradient boosting (XGBoost). For each modeling approach, we used preoperative only, intraoperative only, or pre+intraoperative variables. We developed 84 models with unique combinations of the 3 variable sets, 2 variable selection methods, 2 modeling approaches, and 7 outcomes. Each model was tested in 20 iterations of 70:30 stratified random splitting into development/testing samples. Model performances were evaluated on the testing dataset using the C statistic, area under the precision-recall curve, and calibration metrics, including the Brier score. RESULTS: The mean patient age was 65.3 years, and 24.7% were women. Operative mortality, excluding intraoperative death, occurred in 1.9%. In all outcomes, models that considered pre+intraoperative variables demonstrated significantly improved Brier score and area under the precision-recall curve compared with models considering pre or intraoperative variables alone. XGBoost without external variable selection had the best C statistics, Brier score, and area under the precision-recall curve values in 4 of the 7 outcomes (mortality, renal failure, prolonged ventilation, and composite) compared with logistic regression models with or without variable selection. Based on the calibration plots, risk restratification for mortality showed that the logistic regression model underestimated the risk in 11 114 patients (9.8%) and overestimated in 12 005 patients (10.6%). In contrast, the XGBoost model underestimated the risk in 7218 patients (6.4%) and overestimated in 0 patients (0%). CONCLUSIONS: In isolated coronary artery bypass graft, adding intraoperative variables to preoperative variables resulted in improved predictions of all 7 outcomes. Risk models based on XGBoost may provide a better prediction of adverse events to guide clinical care.