Hudson Fernandes Golino

Predicting Increased Blood Pressure Using Machine Learning

The present study investigates the prediction of increased blood pressure by body mass index (BMI), waist (WC) and hip circumference (HC), and waist hip ratio (WHR) using a machine learning technique named classification tree. Data were collected from 400 college students (56.3% women) from 16 to 63 years old. Fifteen trees were calculated in the training group for each sex, using different numbers and combinations of predictors. The result shows that for women BMI, WC, and WHR are the combination that produces the best prediction, since it has the lowest deviance (87.42), misclassification (.19), and the higher pseudo R 2 (.43). This model presented a sensitivity of 80.86% and specificity of 81.22% in the training set and, respectively, 45.65% and 65.15% in the test sample. For men BMI, WC, HC, and WHC showed the best prediction with the lowest deviance (57.25), misclassification (.16), and the higher pseudo R 2 (.46). This model had a sensitivity of 72% and specificity of 86.25% in the training set and, respectively, 58.38% and 69.70% in the test set. Finally, the result from the classification tree analysis was compared with traditional logistic regression, indicating that the former outperformed the latter in terms of predictive power.

Validação da Escala de Abordagens de Aprendizagem (EABAP) em uma amostra Brasileira

The main objective of the current study is to build and validate the Learning Approach Scale (LAS), which was developed to assess the deep and surface dimensions in a Brazilian sample. Seven hundred sixteen participants from junior and high schools in the city of Belo Horizonte took part in the research. The exploratory factor analysis was used to identify those dimensions and to select the best items. Only 17 items were kept from the 27 original ones. By the Confirmatory Factor Analysis it was possible to identify the adequacy of one general and two specific factors, the latter representing the deep and surface approaches (χ2=281.50; gl=116; CFI=0.94; GFI=0.95 and RMSEA=0.05). The implications of the results for the theory of approaches and the application of the scale in the Educational Psychology field are discussed.

c-Fos expression predicts long-term social memory retrieval in mice.

The way the rodent brain generally processes socially relevant information is rather well understood. How social information is stored into long-term social memory, however, is still under debate. Here, brain c-Fos expression was measured after adult mice were exposed to familiar or novel juveniles and expression was compared in several memory and socially relevant brain areas. Machine Learning algorithm Random Forest was then used to predict the social interaction category of adult mice based on c-Fos expression in these areas. Interaction with a familiar co-specific altered brain activation in the olfactory bulb, amygdala, hippocampus, lateral septum and medial prefrontal cortex. Remarkably, Random Forest was able to predict interaction with a familiar juvenile with 100% accuracy. Activity in the olfactory bulb, amygdala, hippocampus and the medial prefrontal cortex were crucial to this prediction. From our results, we suggest long-term social memory depends on initial social olfactory processing in the medial amygdala and its output connections synergistically with non-social contextual integration by the hippocampus and medial prefrontal cortex top-down modulation of primary olfactory structures.

Validade incremental da Escala de Abordagens de Aprendizagem (EABAP)

This paper investigates the incremental validity of learning approach in academic achievement. Participants were 684 junior and high school students from a private school in Belo Horizonte, Minas Gerais, Brazil. Intelligence is measured by fluid intelligence items of the Higher-Order Cognitive Factors Kit. Learning approach is measured by the Learning Approaches Scale. Academic achievement is measured by annual grades in Mathematics, Portuguese (native language), Geography, and History. Three hypotheses about the relation among intelligence, learning approach and academic achievement are tested through structural equation modeling. The direct relation model was the most adequate and showed good fit. Intelligence and learning approach show direct effect in academic achievement. Learning approach has incremental validity of individual differences in academic achievement, independently of intelligence.

Relações hierárquicas entre os traços amplos do Big Five

The Big Five model sustains that human personality is composed by dozens of specific factors. Despite of diversity, specific factors are integrated in five broad traits that are in the same hierarchical level. The current study presents an alternative hypothesis arguing that there are hierarchical levels between the broad traits of the model. Six hundred and eighty-four junior and high school level students from 10 to 18 years old (M = 13.71 and SD= 2.11) of a private school in the city of Belo Horizonte, Minas Gerais, Brazil participated in the study. The Big Five was measured by an Inventory of Personality Traits, initially named as Personality Adjective Inventory, elaborated by Pinheiro, Gomes and Braga (2009). This instrument measures eight polarities of the ten presented in the Big Five Model. Two models were compared through path analysis: a four-level hierarchical model and a non-hierarchical one. The hierarchical model showed adequate data fit, pointing to its superiority in relation to the non-hierarchical model, which did not present it. Implications to the Big Five Model are discussed.

Is Small Still Beautiful for the Strengths and Difficulties Questionnaire? Novel Findings Using Exploratory Structural Equation Modeling

During the present decade a large body of research has employed confirmatory factor analysis (CFA) to evaluate the factor structure of the Strengths and Difficulties Questionnaire (SDQ) across multiple languages and cultures. However, because CFA can produce strongly biased estimations when the population cross-loadings differ meaningfully from zero, it may not be the most appropriate framework to model the SDQ responses. With this in mind, the current study sought to assess the factorial structure of the SDQ using the more flexible exploratory structural equation modeling approach. Using a large-scale Spanish sample composed of 67,253 youths aged between 10 and 18 years (M = 14.16, SD = 1.07), the results showed that CFA provided a severely biased and overly optimistic assessment of the underlying structure of the SDQ. In contrast, exploratory structural equation modeling revealed a generally weak factorial structure, including questionable indicators with large cross-loadings, multiple error correlations, and significant wording variance. A subsequent Monte Carlo study showed that sample sizes greater than 4,000 would be needed to adequately recover the SDQ loading structure. The findings from this study prevent recommending the SDQ as a screening tool and suggest caution when interpreting previous results in the literature based on CFA modeling.

Random forest as an imputation method for education and psychology research: its impact on item fit and difficulty of the Rasch model

ABSTRACT This paper presents a non-parametric imputation technique, named random forest, from the machine learning field. The random forest procedure has two main tuning parameters: the number of trees grown in the prediction and the number of predictors used. Fifty experimental conditions were created in the imputation procedure, with different combinations of predictors (from 1 to 10) and number of trees (10, 50, 100, 500 and 1000). We examined how each experimental condition affected the items fit and the difficulty of an inductive reasoning test to the dichotomous Rasch model. The results point that using random forest to impute missing values is a reliable technique to be used in psychological researches, since it led to statistically significant differences in the infit’s median only in 4% of the experimental conditions investigated, compared to the original data set result. However, researchers should be aware that in 32% of the experimental conditions, the imputation procedure significantly increased the median of the estimated items’ difficulty, compared to the original data set.

Mining concepts of health responsibility using text mining and exploratory graph analysis

Abstract Background: Occupational therapists need to know about people’s beliefs about personal responsibility for health to help them pursue everyday activities. Aim: The study aims to employ state-of-the-art quantitative approaches to understand people’s views of health and responsibility at different ages. Methods: A mixed method approach was adopted, using text mining to extract information from 233 interviews with participants aged 5 to 96 years, and then exploratory graph analysis to estimate the number of latent variables. The fit of the structure estimated via the exploratory graph analysis was verified using confirmatory factor analysis. Results: Exploratory graph analysis estimated three dimensions of health responsibility: (1) creating good health habits and feeling good; (2) thinking about one’s own health and wanting to improve it; and 3) adopting explicitly normative attitudes to take care of one’s health. The comparison between the three dimensions among age groups showed, in general, that children and adolescents, as well as the old elderly (>73 years old) expressed ideas about personal responsibility for health less than young adults, adults and young elderly. Conclusions: Occupational therapists’ knowledge of the concepts of health responsibility is of value when working with a patient’s health, but an identified challenge is how to engage children and older persons.

Formal-Logic Development Program: Effects on Fluid Intelligence and on Inductive Reasoning Stages

The Formal-Logic Development Program (FLDP) is an intervention that focuses in the argumentative abilities and aims to change and increase the intelligence level of people. This study showed the first empirical analysis of the effect of the program on fluid intelligence and on the inductive reasoning stages. A 15 years old Brazilian teenager, enrolled in an education program for youth and adults, participated in the intervention (case) and the control group (N=12) was composed by teenagers enrolled in the same educational program. Four fluid intelligence tests were administered at pre and posttest. The intervention lasted approximately 4 months, 2 hours per session. The fluid intelligence score was generated using the Rasch model and the participants answers to the items of the four tests used. Regarding the adjustment of the items to the model, on average the