Is it worth generating rules from neural network ensembles?

Although many authors generated comprehensible models from individual networks, much less work has been done in the explanation of ensembles. DIMLP is a special neural network model from which rules are generated at the level of a single network and also at the level of an ensemble of networks. We applied ensembles of 25 DIMLP networks to several datasets of the public domain and a classification problem related to post-translational modifications of proteins. For the classification problems of the public domain, the average predictive accuracy of rulesets extracted from ensembles of neural networks was significantly better than the average predictive accuracy of rulesets generated from ensembles of decision trees. By varying the architectures of DIMLP networks we found that the average predictive accuracy of rules, as well as their complexity were quite stable. The comparison to other rule extraction techniques applied to neural networks showed that rules generated from DIMLP ensembles gave very good results. In the last problem related to bioinformatics, the best result obtained by ensembles of DIMLP networks was also significantly better than the best result obtained by ensembles of decision trees. Thus, although neural networks take much longer to train than decision trees and also rules are generated at a greater computational cost (however, still polynomial), at least for several classification problems it was worth using neural network ensembles, as extracted rules were more accurate, on average. The DIMLP software is available for PC-Linux under http://us.expasy.org/people/Guido.Bologna.html.     

The Future of the Christian Past: Marcel Gauchet and Charles Taylor on the Essence of Religion and its Evolution

This article explores the differences between Marcel Gauchet and Charles Taylor with respect to their theories of secularization. It starts by looking at their resemblances; it continues by distinguishing a two‐fold difference in their approach. The variation within their similar methodologies is examined, and then the consequences of these divergent definitions of religion are investigated. We focus on four themes: the role of the Axial religions, the significance of Incarnation and Reformation, the significance of Christianity as the ‘religion of the departure from religion’, and the possibility of religious ‘conversion’. Taylor's and Gauchet's views on the future of religion diverge as a function of their different interpretations of ‘fulfilment’ and ‘hunger for meaning’.     

Positivity: The Dispositional Basis of Happiness

The present study examined in a stringent developmental model whether positivity, conceptualized as a pervasive mode of appraising, viewing, and perceiving life from a positive stance, predicts chronic positive affectivity across time or vice versa. Participants [263 participants (47 % males)], aged 15.5 at the beginning of the study and 23.5 at the end, reported on measures of positivity and dispositional positive affect four times. Longitudinal findings corroborated the posited paths of relations, with positivity significantly predicting positive affectivity across time rather than vice versa.     

Smiles make it easier and so do frowns: masked affective stimuli influence mental effort

Two experiments tested the hypothesis that exposure to masked emotional expressions during the performance of cognitive tasks influences effort mobilization. In support of the predictions, participants who processed masked sad faces during task performance under "do your best" instructions showed stronger sympathetic nervous system discharge to the heart (shorter pre-ejection period, higher systolic blood pressure) than participants who were exposed to masked smiling faces or angry faces. Assessed task appraisals suggest that these effects on effort-related cardiovascular reactivity occurred because the masked emotional stimuli influenced the level of experienced task difficulty. The findings are compatible with the effects of consciously experienced affect on effort-related cardiovascular response.     

Linear mapping of numbers onto space requires attention

Mapping of number onto space is fundamental to mathematics and measurement. Previous research suggests that while typical adults with mathematical schooling map numbers veridically onto a linear scale, pre-school children and adults without formal mathematics training, as well as individuals with dyscalculia, show strong compressive, logarithmic-like non-linearities when mapping both symbolic and non-symbolic numbers onto the numberline. Here we show that the use of the linear scale is dependent on attentional resources. We asked typical adults to position clouds of dots on a numberline of various lengths. In agreement with previous research, they did so veridically under normal conditions, but when asked to perform a concurrent attentionally-demanding conjunction task, the mapping followed a compressive, non-linear function. We model the non-linearity both by the commonly assumed logarithmic transform, and also with a Bayesian model of central tendency. These results suggest that veridical representation numerosity requires attentional mechanisms.     

Implicit happiness and sadness are associated with ease and difficulty: evidence from sequential priming

Psychological Research - Three experiments tested the hypothesis of implicit associations between happiness and the performance ease concept and between sadness and the performance difficulty...     

Optimal training design for procedural motor skills: a review and application to laparoscopic surgery

This literature review covers the choices to consider in training complex procedural, perceptual and motor skills. In particular, we focus on laparoscopic surgery. An overview is provided of important training factors modulating the acquisition, durability, transfer, and efficiency of trained skills. We summarize empirical studies and their theoretical background on the topic of training complex cognitive and motor skills that are pertinent to proficiency in laparoscopic surgery. The overview pertains to surgical simulation training for laparoscopy, but also to training in other demanding procedural and dexterous tasks, such as aviation, managing complex systems and sports. Evidence-based recommendations are provided for facilitating efficiency in laparoscopic motor skill training such as session spacing, adaptive training, task variability, part-task training, mental imagery and deliberate practice.     

Using Latent Class Analysis to Model Temperament Types

Mixture models are appropriate for data that arise from a set of qualitatively different subpopulations. In this study, latent class analysis was applied to observational data from a laboratory assessment of infant temperament at four months of age. The EM algorithm was used to fit the models, and the Bayesian method of posterior predictive checks was used for model selection. Results show at least three types of infant temperament, with patterns consistent with those identified by previous researchers who classified the infants using a theoretically based system. Multiple imputation of group memberships is proposed as an alternative to assigning subjects to the latent class with maximum posterior probability in order to reflect variance due to uncertainty in the parameter estimation. Latent class membership at four months of age predicted longitudinal outcomes at four years of age. The example illustrates issues relevant to all mixture models, including estimation, multi-modality, model selection, and comparisons based on the latent group indicators.