Correction
Psychology, Mathematical
Chi-Shing Tse, Yuen-Lai Chan, Melvin J. Yap, Ho Chung Tsang
BEHAVIOR RESEARCH METHODS
(2023)
Article
Psychology, Mathematical
Benjamin W. Roop, Benjamin Parrell, Adam C. Lammert
Summary: This study demonstrates that compressive sensing can alleviate the methodological barrier of the reverse correlation method and improve the accuracy and efficiency of reconstructing cognitive representations. Simulations show that compressive sensing can dramatically reduce the required number of stimulus-response pairs, and this finding is further validated using human subject data.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Psychology, Educational
Andreas Frey, Christoph Koenig, Aron Fink
Summary: The highly adaptive testing (HAT) design is introduced as an alternative test design for PISA. HAT combines established methods from computerized adaptive testing and outperforms the PISA 2018 multistage design (MST) in terms of test information, RMSE, and constraint management. However, it shows slightly weaker item exposure. Increasing the response probability to .62 is a viable option to enhance students' test-taking experience with HAT.
JOURNAL OF EDUCATIONAL MEASUREMENT
(2023)
Article
Mathematics, Interdisciplinary Applications
Sun-Joo Cho, Hao Wu, Matthew Naveiras
Summary: Popular statistical software provides the Bayesian information criterion (BIC) for multi-level models or linear mixed models. However, the combination of statistical literature and software documentation has led to discrepancies in the formulas of the BIC and uncertainties as to the proper use of the BIC in selecting a multi-level model with respect to level-specific fixed and random effects. In this study, new versions of BIC, called BICE1 and BICE2, are derived for level-specific fixed- and random-effect selection in a two-level nested design. These new versions perform at least as well as existing BIC variations and are demonstrated to be the best global selection criterion across various multi-level conditions.
BRITISH JOURNAL OF MATHEMATICAL & STATISTICAL PSYCHOLOGY
(2023)
Review
Psychology, Mathematical
Marie-Ange Lecerf, Severine Casalis, Eva Commissaire
Summary: In the past ten years, research on bilingual visual word recognition has led to a new line of study focusing on a sublexical orthographic variable called orthographic markedness. This variable is derived from comparing the two orthotactic distributions known by bilingual readers. Orthographic markers have been shown to speed up language decisions and also modulate the nonselectivity of language during lexical access. This review summarizes the available literature on orthographic markedness and its effects on language membership detection and lexical access. The review also discusses theoretical extensions to bilingual interactive activation models and proposes future research directions.
PSYCHONOMIC BULLETIN & REVIEW
(2023)
Article
Mathematics, Interdisciplinary Applications
Keivan Mallahi-Karai, Adele Diederich
Summary: In this paper, a dynamical model for the best-worst choice task is proposed, which explains the sequential choices made in multiple episodes using a multivariate Wiener process.
JOURNAL OF MATHEMATICAL PSYCHOLOGY
(2023)
Article
Mathematics, Interdisciplinary Applications
Dominik R. Bach
Summary: This paper discusses the measurement of validity in psychological theories, focusing on how to maximize information gathering and minimize the sample variance of retrodictive validity estimators in calibration experiments. Through analyzing different distribution features and numerical simulations, recommendations for the distribution of predicted values and resource investment are provided, and the issues of misspecified theories are highlighted.
JOURNAL OF MATHEMATICAL PSYCHOLOGY
(2023)
Article
Mathematics, Interdisciplinary Applications
V. G. Bardakhchyan, A. E. Allahverdyan
Summary: This article studies a general regret criterion for choosing between two probabilistic lotteries. The criterion is consistent with stochastic dominance for independent lotteries and can be made transitive by selecting a unique regret function. In addition, the criterion resolves Allais' paradox, including cases where the paradox disappears and choices align with expected utility. The superadditivity property of the criterion is employed for establishing consistency between regret and stochastic dominance for dependent lotteries.
JOURNAL OF MATHEMATICAL PSYCHOLOGY
(2023)
Article
Mathematics, Interdisciplinary Applications
Ram Kumar Dhurkari
Summary: A better method called MAGL (Multi-Attribute Gain Loss) is proposed to predict consumers' choices in a multi-attribute setting. The MAGL method incorporates prospect theory, Kauffman's complexity theory, norm theory, and context-dependent choice theory to model and predict the context-dependent choice behavior of consumers. The predictions of the MAGL method are valuable for marketing/product managers in the design of new products.
JOURNAL OF MATHEMATICAL PSYCHOLOGY
(2023)
Article
Psychology, Educational
Sijia Huang, Jinwen (Jevan) Luo, Li Cai
Summary: This study introduces an explanatory multidimensional random item effects rating scale model, which allows for flexible inclusion of person-related and item-related covariates and utilizes the MH-RM algorithm for parameter estimation.
EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT
(2023)
Article
Psychology, Educational
Tong Wu, Stella Y. Kim, Carl Westine
Summary: This study evaluates the effects of six different approaches to handling missing data on the accuracy of scale linking based on item response theory. The results show that imputing with a response function, multiple imputation, and full information likelihood produce less errors in scale linking accuracy, while listwise deletion is associated with the most errors.
EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT
(2023)
Article
Mathematics, Interdisciplinary Applications
Maria M. Robinson, Isabella C. Destefano, Edward Vul, Timothy F. Brady
Summary: In decision tasks, the problem of using latent beliefs and preferences to make a single choice is addressed by models such as signal detection theory and softmax function. However, the connection between these two models is rarely explored. This study shows that the signal detection model provides more generalizable predictions across changes in task structure, suggesting the parametric assumptions of this model may reflect something real about human decision-making.
JOURNAL OF MATHEMATICAL PSYCHOLOGY
(2023)
Article
Mathematics, Interdisciplinary Applications
Joshua Calder-Travis, Rafal Bogacz, Nick Yeung
Summary: This study introduces a new approach to modeling decision confidence, which allows computationally cheap predictions by taking into account trial-by-trial variability in stochastically fluctuating stimuli. The authors derive expressions for the probability distribution over confidence reports using the drift diffusion model of decision making, time-dependent thresholds, and the idea of a Bayesian confidence readout. The derived expressions provide a feasible way to model confidence data in tasks involving stochastically fluctuating stimuli.
JOURNAL OF MATHEMATICAL PSYCHOLOGY
(2023)
Article
Mathematics, Interdisciplinary Applications
Bochi Xu, Jinjin Li, Wen Sun, Bo Wang
Summary: This paper introduces the concepts of forward-graded and backward-graded knowledge structures as well as fuzzy skills, and extends the related theory to explore the conditions for using fuzzy skill maps to delineate different knowledge structures. Additionally, it introduces the competence-based local independence model with fuzzy skills and discusses its unidentifiability.
JOURNAL OF MATHEMATICAL PSYCHOLOGY
(2023)
Article
Psychology, Educational
Karl Schweizer, Andreas Gold, Dorothea Krampen
Summary: This study extends the modeling missing data approach to tetrachoric correlations and explores the consequences of switching between models with free and fixed factor loadings. Results show that an additional missing data latent variable can recover the degree-of-model fit characterizing complete data when tetrachoric correlations serve as input, but the comparative fit index may overestimate this degree-of-model fit. Therefore, modeling missing data with fixed factor loadings is recommended.
EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT
(2023)
Article
Mathematics, Interdisciplinary Applications
Ippei Takasawa, Kensuke Tanioka, Hiroshi Yadohisa
Summary: Joint analysis with clustering and structural equation modeling is a popular approach for analyzing heterogeneous data. However, difficulties in interpreting coefficients and the inability to handle different path diagrams for each cluster are common problems. To address these issues, we propose two methods that simplify the path structure and improve interpretation by estimating different forms of path diagrams for each cluster using sparse estimation. Numerical simulations and real data examples demonstrate that the proposed methods outperform existing methods in terms of fitting and interpretation.
JOURNAL OF CLASSIFICATION
(2023)
Article
Education & Educational Research
Jordan M. Wheeler, Allan S. Cohen, Shiyu Wang
Summary: Topic models are mathematical and statistical models used to analyze textual data and gain information about the latent semantic space. LSA and LDA are popular topic models used in educational measurement research for algorithmic scoring and analyzing students' responses.
JOURNAL OF EDUCATIONAL AND BEHAVIORAL STATISTICS
(2023)
Article
Psychology, Mathematical
Brianna E. Cairney, Stanley H. West, Eileen Haebig, Christopher R. Cox, Heather D. Lucas
Summary: This study created a database of 162 gesture videos and investigated the differences in processing and comprehension of co-speech gestures between individuals with autism spectrum disorder (ASD) and neurotypical individuals. The results showed a high level of agreement between raters with and without ASD in terms of meaningfulness ratings. Autistic raters produced a more diverse set of verbal labels for each gesture, but there was no significant difference in within-gesture semantic similarity between the two groups.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Education & Educational Research
George Leckie, Richard Parker, Harvey Goldstein, Kate Tilling
Summary: School value-added models are important for studying and monitoring school differences in student learning. This article argues that studying the variance in student achievement within each school can provide further insights, including identifying schools with unusually high or low achievement, even after accounting for differences in student intakes.
JOURNAL OF EDUCATIONAL AND BEHAVIORAL STATISTICS
(2023)
Article
Mathematics, Interdisciplinary Applications
Mans Karlsson, Ola Hossjer
Summary: In this paper, the authors review and unify several methods of Bayesian set-valued classification and explore the effects of reward functions and set sizes on classification results. They provide general expressions for Bayesian classifiers applicable to different hypothesis blocks and discuss the application of well-known classification methods within this framework. The set-valued classification is illustrated using an ornithological dataset.
JOURNAL OF CLASSIFICATION
(2023)
