Search

Analytics for behavioral habit typically model one behavior at a time, despite the fact that habit often involves multiple cooccurring behaviors, such as food choices and physical activities, where interrelated behaviors are often equally recommended. We propose a novel Mixed-Effects Dynamic hAbit model (MEDA) to simultaneously model multiple related, habitual behaviors. As an illustrative example, MEDA was applied to real-time assessments of sun protection (sunscreen, shade, hat, and protective clothing) reported twice daily by first-degree relatives of melanoma patients who are themselves at an elevated risk of skin cancer. MEDA aims to explicate habits in sun protection under varying environmental cues (e.g., sunny and hot weather). We found consistent between-group differences (e.g., men responding to weather cues more consistently than women) and interactions between cooccurring behaviors (e.g., being in shade discourages sunscreen wearing, more so in men than women). Moreover, MEDA transcends conventional methods to address longstanding challenges—how cue to action and volitional choices differ by groups or even by individual persons. Such nuances in interrelated habitual behaviors are relevant in numerous other applications, such as recommended dietary or physical activity behaviors. These methods best inform personalized behavioral interventions targeting individual preferences for precision behavioral intervention.

Abstract: Recent findings show adult New Zealanders who do not speak te reo Māori (the Māori language, the indigenous language of New Zealand) nonetheless have impressive implicit lexical and phonotactic knowledge of the language. These findings have been interpreted as showing that regular ambient exposure to a non-native language develops an implicit “proto-lexicon”, a memory store of lexical forms in that language, without any meaning. However, what is not known is the timeframe over which this knowledge is acquired. Does the knowledge stem exclusively from implicit learning during childhood, or does it continue to grow based on exposure during adulthood? To investigate this question, we directly compare non-Māori-speaking school-aged adolescents and adults in New Zealand and explore how age affects the degree of observed knowledge. Our results show that ambient exposure leads to implicit knowledge both in childhood and adulthood, and that continuing exposure throughout the lifespan leads to increased knowledge.

Abstract: Previous studies report that exposure to the Māori language on a regular basis allows New Zealand adults who cannot speak Māori to build a proto‐lexicon of Māori—an implicit memory of word forms without detailed knowledge of meaning. How might this knowledge feed into explicit language learning? Is it possible to “awaken” the proto‐lexicon in the context of overt language learning? We investigate whether implicit linguistic knowledge represented in a proto‐lexicon gives any advantages for intentional language learning in a tertiary educational environment. We conducted a three‐task experiment which: (a) assessed participants’ Māori proto‐lexicon, (b) assessed their phonotactic knowledge, and (c) tested them on Māori vocabulary that they had been exposed to during the course at two time points. The results show that students with larger Māori proto‐lexicons learn more words in a classroom setting. This study shows that proto‐lexicon acquired from ambient exposure can lead to significant benefits in language learning.

We present an extension of the Morfessor Baseline model of unsupervised morphological segmentation (Creutz and Lagus, 2007) that incorporates abstract templates for reduplication, a typologically common but computationally underaddressed process. Through a detailed investigation that applies the model to Māori, the Indigenous language of Aotearoa New Zealand, we show that incorporating templates improves Morfessor’s ability to identify instances of reduplication, and does so most when there are multiple minimally-overlapping templates. We present an error analysis that reveals important factors to consider when applying the extended model and suggests useful future directions.

[This corrects the article DOI: 10.1371/journal.pone.0265254.].

Most people in New Zealand are exposed to the Māori language on a regular basis, but do not speak it. It has recently been claimed that this exposure leads them to create a large proto-lexicon, consisting of implicit memories of words and word parts, without semantic knowledge. This yields sophisticated phonotactic knowledge (Oh et al., 2020). This claim was supported by two tasks in which Non-Māori-Speaking New Zealanders: (i) Distinguished real words from phonotactically matched non-words, suggesting lexical knowledge; (ii) Gave wellformedness ratings of non-words almost indistinguishable from those of fluent Māori speakers, demonstrating phonotactic knowledge.Oh et al. (2020) ran these tasks on separate participants. While they hypothesised that phonotactic and lexical knowledge derived from the proto-lexicon, they did not establish a direct link between them. We replicate the two tasks, with improved stimuli, on the same set of participants. We find a statistically significant link between the tasks: Participants with a larger proto-lexicon (evidenced by performance in the Word Identification Task) show greater sensitivity to phonotactics in the Wellformedness Rating Task. This extends the previously reported results, increasing the evidence that exposure to a language you do not speak can lead to large-scale implicit knowledge about that language.

Public access to genetic information is increasing, and community dermatologists may progressively encounter patients interested in genetic testing for melanoma risk. Clarifying potential utility will help plan for this inevitability. We determined interest and uptake of genetic risk feedback based on melanocortin receptor gene (MC1R) variants, immediate (two weeks) responses to risk feedback, and test utility at three months in patients (age ≥ 18, with a history of nonmelanoma skin cancer). Participants (N = 50) completed a baseline survey and were invited to consider MC1R testing via the study website. Testing interest and uptake were assessed through registration of test decision, request of a saliva test kit, and kit return (all yes/no). Immediate responses to risk feedback included feedback-relevant thoughts, emotions, communication, and information seeking after result receipt; test utility outcomes included family and physician communication and information seeking. Results indicated good retention at both time points (76%; 74%). Half (48%) logged onto the study website, and of these, most (92%) chose testing and (95%) returned a saliva sample. After two weeks, most (94%) had read all the risk feedback information and distress was low (M = 8.81, 7-28, SD = 2.23). Many (69%) had talked with their family about the results. By three months, most had spoken with family (92%) and physicians (80%) about skin cancer risk. Physician communication was higher (70%) in those tested versus those not tested (40%, p = 0.02). The substantial interest and promising outcomes associated with MC1R genetic testing in dermatology patients inform intervention strategies to enhance benefits and minimize risks of skin cancer genetic testing.

Artificial intelligence and machine learning (AI/ML) is becoming increasingly more accessible to biomedical researchers with significant potential to transform biomedicine through optimization of highly-accurate predictive models and enabling better understanding of disease biology. Automated machine learning (AutoML) in particular is positioned to democratize artificial intelligence (AI) by reducing the amount of human input and ML expertise needed. However, successful translation of AI/ML in biomedicine requires moving beyond optimizing only for prediction accuracy and towards establishing reproducible clinical and biological inferences. This is especially challenging for clinical studies on rare disorders where the smaller patient cohorts and corresponding sample size is an obstacle for reproducible modeling results. Here, we present a model-agnostic framework to reinforce AutoML using strategies and tools of explainable and reproducible AI, including novel metrics to assess model reproducibility. The framework enables clinicians to interpret AutoML-generated models for clinical and biological verifiability and consequently integrate domain expertise during model development. We applied the framework towards spinal cord injury prognostication to optimize the intraoperative hemodynamic range during injury-related surgery and additionally identified a strong detrimental relationship between intraoperative hypertension and patient outcome. Furthermore, our analysis captured how evolving clinical practices such as faster time-to-surgery and blood pressure management affect clinical model development. Altogether, we illustrate how expert-augmented AutoML improves inferential reproducibility for biomedical discovery and can ultimately build trust in AI processes towards effective clinical integration.

Risk perception is an important construct in many health behavior theories. Smoking risk perceptions are thoughts and feelings about the harms associated with cigarette smoking. Wide variation in the terminology, definition, and assessment of this construct makes it difficult to draw conclusions about the associations of risk perceptions with smoking behaviors. To understand optimal methods of assessing adults' cigarette smoking risk perceptions (among both smokers and nonsmokers), we reviewed best practices from the tobacco control literature, and where gaps were identified, we looked more broadly to the research on risk perceptions in other health domains. Based on this review, we suggest assessments of risk perceptions (1) about multiple smoking-related health harms, (2) about harms over a specific timeframe, and (3) for the person affected by the harm. For the measurement of perceived likelihood in particular (ie, the perceived chance of harm from smoking based largely on deliberative thought), we suggest including (4) unconditional and conditional items (stipulating smoking behavior) and (5) absolute and comparative items and including (6) comparisons to specific populations through (7) direct and indirect assessments. We also suggest including (8) experiential (ostensibly automatic, somatic perceptions of vulnerability to a harm) and affective (emotional reactions to a potential harm) risk perception items. We also offer suggestions for (9) response options and (10) the assessment of risk perception at multiple time points. Researchers can use this resource to inform the selection, use, and future development of smoking risk perception measures.ImplicationsIncorporating the measurement suggestions for cigarette smoking risk perceptions that are presented will help researchers select items most appropriate for their research questions and will contribute to greater consistency in the assessment of smoking risk perceptions among adults.

Background: Avoiding health information is relatively common and is associated with lower knowledge of health risks and lower engagement in protective health behaviour. Health information avoidance likely limits the effectiveness of health communication interventions. Objective: To identify beliefs associated with avoiding health information. Design: Two cross-sectional studies. Setting: Two representative samples of adults residing in the USA. Method: We tested whether low health agency beliefs and low perceptions of threat underlie the tendency to avoid diabetes or colorectal cancer health information in two samples. Results: An exploratory factor analysis (EFA) of our variables indicated that beliefs could be grouped into two sets of constructs that mapped onto agency (locus of control, fate) and two that mapped onto threat perception (dread, perceived risk). Information avoidance was common. In the two samples, 30% and 34% indicated they preferred to avoid information about diabetes and 20% and 19% indicated they preferred to avoid colorectal cancer information. Results were largely consistent across studies and diseases. In final adjusted regression models, beliefs indicating lower health agency were consistently associated with more avoidance. Some threat perception variables (worry about getting the disease and having a family history of the disease) were associated with less avoidance; absolute and comparative perceived risk were not. Conclusion: Given that health information avoidance likely undermines a wide range of health communication and self-regulation strategies, future health communication efforts might be advanced by developing intervention approaches that involve enhancing perceived control over health prior to delivering health messages.