Jurnal Online STTKD (Sekolah Tinggi Teknologi Kedirgantaraan)
Not a member yet
77141 research outputs found
Sort by
Meaning creation in novel noun-noun compounds: humans and language models
No full textThe interpretation of novel noun-noun compounds (NNCs, e.g. “devil salary”) requires the combination of nouns in the absence of syntactic cues, an interesting facet of complex meaning creation. Here we examine unconstrained interpretations of a large set of novel NNCs, to investigate how NNC constituents are combined into novel complex meanings. The data show that words’ lexical-semantic features (e.g. material, agentivity, imageability, semantic similarity) differentially contribute to the grammatical relations and the semantics of NNC interpretations. Further, we demonstrate that passive interpretations incur higher processing cost (longer interpretation times and more eye-movements) than active interpretations. Finally, we show that large language models (GPT-2, BERT, RoBERTa) can predict whether a NNC is interpretable by human participants and estimate differences in processing cost, but do not exhibit sensitivity to more subtle grammatical differences. The experiments illuminate how humans can use lexical-semantic features to interpret NNCs in the absence of explicit syntactic information
Bayesian regularized SEM: Current capabilities and constraints
No full textAn important challenge in statistical modeling is to balance how well our model explains the phenomenon under investigation with the parsimony of this explanation. In structural equation modeling (SEM), penalization approaches that add a penalty term to the estimation procedure have been proposed to achieve this balance. An alternative to the classical penalization approach is Bayesian regularized SEM in which the prior distribution serves as the penalty function. Many different shrinkage priors exist, enabling great flexibility in terms of shrinkage behavior. As a result, different types of shrinkage priors have been proposed for use in a wide variety of SEMs. However, the lack of a general framework and the technical details of these shrinkage methods can make it difficult for researchers outside the field of (Bayesian) regularized SEM to understand and apply these methods in their own work. Therefore, the aim of this paper is to provide an overview of Bayesian regularized SEM, with a focus on theoretical developments as well as available software implemen- tations. Through an empirical example, various open-source software packages for (Bayesian) regularized SEM are illustrated and all code is made available online to aid researchers in applying these methods. Finally, reviewing the current capabilities and constraints of Bayesian regularized SEM identifies several directions for future research
Algorithm aversion or appreciation? Three randomized field trials of personalized risk-communication nudges to encourage flu vaccination
No full textArtificial intelligence (AI) is increasingly used in healthcare to identify high-risk patients needing intervention. Communicating AI-derived personalized risk information is an untested health behavior change intervention; laboratory studies are mixed regarding patient favorability toward AI use. We ran a field test comprising three large-scale preregistered randomized controlled trials with over 90,000 unique patients. Nudge messages encouraged influenza vaccination in patients identified by a previously validated machine-learning algorithm to be at high risk for flu and flu-related complications. For patients informed they were at high risk, vaccination was 1.1–1.4 percentage points (3.3%–5.4%) higher vs. those not told they were at high risk, and 1.7–3.5 percentage points (3.3%–14.7%) higher vs. non-messaged patients. Therefore, informing patients of personalized risk may effectively encourage their acting on AI-derived information. Vaccination was similar across message arms that did vs. did not disclose the “algorithm,” indicating patients are neither averse to nor appreciative of recommendations that cite an algorithm
La Transición Neolítica en el Sur de Centroamérica
No full textSouthern Central America played a key role in the continent-wide transition to the Neolithic and the transmission of its culture between South and Mesoamerica. Nevertheless, there has been a notable decline in linking processes within this region to contemporary developments in the rest of the continent. Here, there are early indicators of the Neolithic, besides evidence of the movement of cultigens and likely information about pottery manufacturing towards the other continent. Using a co-adaptive framework, I examine the social and ecological changes during the Archaic and Early Formative American Neotropics and adjacent regions, highlighting surprising networks that moved plants, goods and cultural systems during these periods. I focus mainly on the behaviors that link humans with their natural environments, beyond mere social changes occurring at the level of the producer. Finally, I present a Bayesian analysis of 161 radiocarbon dates from 39 sites in Costa Rica and Panama from the Late Archaic through the Middle Formative. These methods can identify anomalous dates and generate probability distributions regarding chronological questions about eight different material culture complexes: In Tilaran-Arenal, Fortuna* and Tronadora; in the Costa Rican Atlantic Watershed, La Montaña, in Gran Chiriquí, Talamanca*, Boquete* and Black Creek; and in Gran Coclé, the Preceramic B tradition and Monagrillo (archaic assemblages denoted with an asterisk*). I argue that there exists evidence for the early presence of ceramic from 4840-4144 BC cal, representing a correction to the established chronology of 1000-3000 years. Early ceramic complexes commonly interpreted as Formative coexisted with Archaic-period assemblages for hundreds, if not thousands of years before the abandonment of archaic-style complexes. Therefore, the presence of ceramics is not a sufficient marker of social processes commonly associated with the Formative. Resumen: El Sur de Centroamérica jugó un papel crucial en la transición al Neolítico y en la transmisión de la cultura neolítica entre Sudamérica y Mesoamérica. A pesar de ello, se ha disminuido la relación de procesos en esta región con el resto del continente. Aquí se encuentran tempranos indicadores del Neolítico, evidencia del tránsito de cultígenos y probable información sobre alfarería hacia el otro continente. Examinando los procesos sociales y ecológicos durante el Arcaico y Formativo temprano en el Neotrópico Americano y regiones aledañas desde una perspectiva co-adaptiva, destaco la sorprendente movilidad de plantas, bienes y sistemas culturales durante estos periodos. Me enfoco en los comportamientos que vinculan a los humanos con sus ambientes, más allá de los procesos sociales a nivel productivo. Un análisis Bayesiano de 161 determinaciones radiométricas de 39 sitios en Costa Rica y Panamá, desde el Arcaico tardío hasta el Formativo medio, permite identificar las determinaciones aberrantes y generar distribuciones de probabilidad sobre cuestiones cronológicas para ocho diferentes complejos: en Tilarán-Arenal, Fortuna* y Tronadora; en el Vertiente Atlántico Costarricense, La Montaña; en el Gran Chiriquí, Talamanca*, Boquete*, y Black Creek; y en el Gran Coclé, Preceramic B* y Monagrillo (complejos arcaicos marcados con asterisco*). Argumento que existe evidencia de la presencia temprana de alfarería desde 4840 - 4144 a.C. cal, representando una corrección en la cronología establecida de 1000-3000 años. Los complejos cerámicos frecuentemente interpretados como Formativos coexistieron con los complejos Arcaicos durante cientos, sino miles, de años antes del abandono de complejos asociados con la caza-recolección. Así, la presencia plena de alfarería no puede considerarse como un marcador de los procesos sociales asociados con el periodo Formativo
Predicting Survey Nonresponse with Registry Data in Sweden between 1992 to 2023: Cohort Replacement or a Deteriorating Survey Climate?
No full textAn updated version is published in the journal Survey Research Methods https://ojs.ub.uni-konstanz.de/srm/article/view/8278.
Declining response rates have remained a major worry for survey research in the 21st century. In
the past decades, it has become harder to convince people to participate in surveys in virtually
all Western nations. Worrisome, declining willingness to participate in surveys (i.e., response
propensities) may increase the risk of extensive nonresponse bias. Therefore, a better understanding
of which factors are associated with survey nonresponse and its impact on nonresponse
bias is paramount for any survey researcher interested in accurate statistical inferences. Knowing
which factors relate to low response propensities enables appropriate models of nonresponse
weights and aids in identifying which groups to tailor efforts for turning nonrespondents into
respondents. This manuscript draws on previous theories and research on nonresponse and investigates
the risk of nonresponse bias, both cross-sectionally and over time, in two time series
cross-sectional studies administered in Sweden (the National SOM Surveys 1993-2023 and the
Swedish National Election Study 2022). Capitalizing on available registry data on all sampled
persons and their corresponding neighborhood-level contextual data, a meta-analytical analysis
of nine years of data collection finds that educational attainment, age, and country of birth
are among the strongest predictors of response propensities. However, contextual factors—such
as living in socially disadvantaged neighborhoods—also predict willingness to participate in
surveys. Furthermore, utilizing the three decades of data, the growing nonresponse could be
identified to be wholly attributable to a deteriorating survey climate rather than birth cohort
replacement or immigration patterns
Great Expectations: Anticipating a Reminder Influences Prospective Memory Encoding and Unaided Retrieval
No full textResearch in the retrospective memory domain has shown that individuals encode information less effortfully when expecting a reminder system (i.e., external store) to be available at test. Critically, this expectation leads to worse memory performance when the reminder is unexpectedly removed. The current study examined whether these findings extend to prospective memory (PM) intentions, which are thought to maintain a privileged status in memory and therefore may be less sensitive to expectancy effects. Participants formed the intention to make a special PM response to target items across four ongoing task blocks. Study duration (Exp. 1 and 3), pupil size (Exp. 2), and self-report (Exp. 1-3) indexed encoding effort while learning these targets. Participants had reminders available across the first three blocks (i.e., targets listed at the top of the screen), but not on the fourth. Critically, only one condition was informed that they would not have a reminder prior to encoding targets in the fourth block. Results showed that expecting a reminder lowered objective (Exp. 1 and 3) and subjective (Exp. 1-3) encoding effort and reduced unaided PM retrieval (Exp. 1-3) in the fourth block, independent of memory load (Exp. 3). Objective (Exp. 1 and 3) and subjective (Exp. 1-3) effort also partially mediated the influence of expectations on unaided PM retrieval. These findings suggest PM and retrospective memory encoding operate similarly and that participants can alter learning to more effectively commit PM targets to memory when reminders are not expected
A transdiagnostic, dimensional classification of anxiety shows improved parsimony and predictive noninferiority to DSM.
No full textThe current classification of anxiety is cumbersome; does not align with evidence that anxiety problems cut across disorder categories; and fails to acknowledge that severity of anxiety matters, even at low levels. We developed a new classification that distills key features of anxiety – intensity, avoidance, pervasiveness, and onset – across disorders, allowing any individual to be located along a gradient from none to severe for each feature. This transdiagnostic dimensional approach is much simpler than the current DSM approach to anxiety, incorporates information about severity, and performs just as well as DSM diagnoses in predicting important clinical outcomes
Effect of geometric complexity on intuitive model selection - experiment on Pavlovia
No full textA psychophysics study of intuitive model selection
Preregistration: Revisiting the logic in language - Experiment 3
No full textThe results of Experiment 1 more-or-less seemed to replicate the pattern observed in Feiman and Snedeker (2016; Experiment 2): Priming of logical representations was stronger when prime and target contained the same quantifiers compared to when prime and target involved different quantifiers. In Experiment 2, we repeated Experiment 1, but in a fully within-subjects design. The results of this experiment revealed comparable effects of priming in all three prime quantifier conditions, which indicates that priming of logical representations is possible between quantifiers as well as within quantifiers. We hypothesis that the discrepancy between Experiment 1 and 2 might be due to effects of bias adaptation: Priming of logical representations is constrained as people first need to become aware that the target sentences allow two possible interpretations. This bias adaptation emerges in a within-quantifier condition. We will test this hypothesis in Experiment 3 (described here), in which prime quantifier is manipulated in a blocked design