Haeike Josephy

Preschool predictors of mathematics in first grade children with autism spectrum disorder

Up till now, research evidence on the mathematical abilities of children with autism spectrum disorder (ASD) has been scarce and provided mixed results. The current study examined the predictive value of five early numerical competencies for four domains of mathematics in first grade. Thirty-three high-functioning children with ASD were followed up from preschool to first grade and compared with 54 typically developing children, as well as with normed samples in first grade. Five early numerical competencies were tested in preschool (5-6 years): verbal subitizing, counting, magnitude comparison, estimation, and arithmetic operations. Four domains of mathematics were used as outcome variables in first grade (6-7 years): procedural calculation, number fact retrieval, word/language problems, and time-related competences. Children with ASD showed similar early numerical competencies at preschool age as typically developing children. Moreover, they scored average on number fact retrieval and time-related competences and higher on procedural calculation and word/language problems compared to the normed population in first grade. When predicting first grade mathematics performance in children with ASD, both verbal subitizing and counting seemed to be important to evaluate at preschool age. Verbal subitizing had a higher predictive value in children with ASD than in typically developing children. Whereas verbal subitizing was predictive for procedural calculation, number fact retrieval, and word/language problems, counting was predictive for procedural calculation and, to a lesser extent, number fact retrieval. Implications and directions for future research are discussed.

Within-Subject Mediation Analysis in AB/BA Crossover Designs

Abstract Crossover trials are widely used to assess the effect of a reversible exposure on an outcome of interest. To gain further insight into the underlying mechanisms of this effect, researchers may be interested in exploring whether or not it runs through a specific intermediate variable: the mediator. Mediation analysis in crossover designs has received scant attention so far and is mostly confined to the traditional Baron and Kenny approach. We aim to tackle mediation analysis within the counterfactual framework and elucidate the assumptions under which the direct and indirect effects can be identified in AB/BA crossover studies. Notably, we show that both effects are identifiable in certain statistical models, even in the presence of unmeasured time-independent (or upper-level) confounding of the mediator–outcome relation. Employing the mediation formula, we derive expressions for the direct and indirect effects in within-subject designs for continuous outcomes that lend themselves to linear modelling, under a large variety of settings. We discuss an estimation approach based on regressing differences in outcomes on differences in mediators and show how to allow for period effects as well as different types of moderation. The performance of this approach is compared to other existing methods through simulations and is illustrated with data from a neurobehavioural study. Lastly, we demonstrate how a sensitivity analysis can be performed that is able to assess the robustness of both the direct and indirect effect against violation of the “no unmeasured lower-level mediator–outcome confounding” assumption.

More Precise Estimation of Lower-Level Interaction Effects in Multilevel Models

ABSTRACT In hierarchical data, the effect of a lower-level predictor on a lower-level outcome may often be confounded by an (un)measured upper-level factor. When such confounding is left unaddressed, the effect of the lower-level predictor is estimated with bias. Separating this effect into a within- and between-component removes such bias in a linear random intercept model under a specific set of assumptions for the confounder. When the effect of the lower-level predictor is additionally moderated by another lower-level predictor, an interaction between both lower-level predictors is included into the model. To address unmeasured upper-level confounding, this interaction term ought to be decomposed into a within- and between-component as well. This can be achieved by first multiplying both predictors and centering that product term next, or vice versa. We show that while both approaches, on average, yield the same estimates of the interaction effect in linear models, the former decomposition is much more precise and robust against misspecification of the effects of cross-level and upper-level terms, compared to the latter.

Corrigendum : a review of R-packages for random-intercept probit regression in small clusters

A corrigendum on: A Review of R-packages for Random-Intercept Probit Regression in Small Clusters by Josephy, H., Loeys, T., and Rosseel, Y. (2016). Front. Appl. Math. Stat. 2, 1–13. doi: 10.3389/fams.2016.00018

Anodal tDCS over the right dorsolateral prefrontal cortex modulates cognitive processing of emotional information as a function of trait rumination in healthy volunteers

Healthy individuals reporting higher (as compared to lower) levels of trait rumination recruit more neural activity in dorso-cortical regions (mostly in the right hemisphere) when inhibiting negative information, possibly to compensate their difficulty to disengage from it. In the present study, we investigated whether these latter neural correlates are causally implicated in cognitive control in these individuals. We administered the Cued Emotional Control Task, a measure of cognitive control indexed by cognitive costs for inhibiting versus providing a habitual response for emotional information, in thirty-five healthy volunteers reporting a broad range of trait rumination levels. Participants completed the task after receiving both real and sham-placebo (counterbalanced order) anodal transcranial Direct Current Stimulation (tDCS) over the right dorsolateral prefrontal cortex (DLPFC). Results reveal that the tDCS induced effects on cognitive costs for emotional information were associated with individual differences in trait rumination: the higher the trait rumination level, the less cognitive costs following real neuromodulation of the right DLPFC. Interestingly, these effects were observed for both positive and negative stimuli, and not only negative information as hypothesized. Overall, the data suggest that the right DLPFC is causally involved in the alteration of cognitive control in healthy individuals who tend to ruminate, possibly by helping them to disengage from emotional material.