Mechanisms of hemispheric lateralization: A replication study

Abstract

It has been shown, using functional magnetic resonance imaging (fMRI), that hemispheric lateralization of brain activity depends on the requirements of the cognitive task performed during the processing of a sensory stimulus rather than on the intrinsic characteristics of that stimulus [Stephan et al., 2003, Science 301 (5631): 384–6]. Task-dependent increase in the coupling of the anterior cingulate cortex (ACC), a region involved in cognitive control, and brain areas in the left prefrontal and right parietal cortex, respectively, regions involved in task execution, was proposed as the mechanism underlying this task-dependency of hemispheric lateralization. The aim of the present study was two-fold: First, we aimed for a conceptual replication of these findings in an independent sample of subjects. Second, we investigated the test–retest reliability of the imaging paradigm to assess whether the task can be used to capture reliable measures of inter-individual differences in hemispheric lateralization. We were able to confirm previous findings showing that hemispheric lateralization depends on the nature of the cognitive task rather than on the nature of the processed stimuli. The task-related brain activation patterns were highly reliable across sessions (as indicated by intra-class correlation coefficients – ICCs, ≥.51). We could, however, not replicate previous results proposing task-dependent changes in the coupling between ACC and brain regions for task execution as the mechanism underlying hemispheric lateralization. This re-opens the question which mechanisms could determine the task-dependent functional asymmetries that were observed previously and replicated in this study.

Introduction

Functional asymmetries between the hemispheres have been known since the middle of the 19th century. The underlying mechanisms however are still not completely understood. Imaging techniques such as functional magnetic resonance imaging (fMRI) can help to develop and test theories on these mechanisms. Newer approaches emphasize on the one hand that hemispheric lateralization should not only be operationalized by hemispheric asymmetries in local structure or function, but also in terms of asymmetries of intra- and interhemispheric brain connectivity (Frässle et al., 2016a, Frässle et al., 2016b, Seghier et al., 2011, Stephan et al., 2005, Stephan et al., 2007a, Stephan et al., 2007b). On the other hand, comprehensive models of hemispheric lateralization must also consider the interaction of several lateralized brain functions such as language, spatial attention or memory, since this interaction can lead to tremendous heterogeneity across participants with regard to hemispheric lateralization (Axmacher et al., 2009, Jansen et al., 2006, Jansen et al., 2005, Weber et al., 2007, Willems et al., 2014, Willems et al., 2010). We therefore need a battery of imaging tasks that (1) are able to assess the lateralization of several cognitive functions, (2) allow the calculation of intra- as well as interhemispheric connectivity measures, (3) yield reliable results at the individual subject level.

In the present study, we explored the utility of an imaging task previously described by Stephan et al. (2003). In this paradigm, subjects processed stimuli containing both verbal and spatial information; critically, task instructions were varied (subjects had to perform either a verbal or a spatial task) while stimuli were identical across the tasks. In a first analysis that directly contrasted the two tasks, the authors showed that processing of verbal information led to left-lateralized brain activity, processing of spatial information to right-hemispheric activity. Since the same stimuli were used throughout the experiment, the authors could show that hemispheric lateralization depends on a top-down regulated mechanism, i.e., that it is task-dependent rather than stimulus-dependent. In a second analysis, Stephan et al. further investigated potential mechanisms that were responsible for the differences in the lateralization of brain activation. They showed that the anterior cingulate cortex (ACC) selectively increased its coupling with either left prefrontal brain regions or right parietal brain areas depending on the task (for a more comprehensive summary of the study, see Supplementary material S1).

The purpose of the present study was two-fold: The first aim was to replicate the results of Stephan and colleagues in an independent sample of subjects [“constructive type replication”, (Lykken 1968)]. Replication of previous results is an important, but often neglected issue in neuroscientific studies. As stated by Bennett and Miller (2010) in a recent review on the reliability of fMRI results, “if results do not generalize from one set of subjects to another (…), then the findings are of little value scientifically”. The second aim was to assess the test–retest reliability of the paradigm, as a crucial prerequisite to decide whether this paradigm can be used to determine stable brain imaging markers characterizing hemispheric lateralization differences (both with regard to brain activation and brain connectivity) of individual subjects [“internal replication”, (Lykken 1968)].