Item

Abstract

This article has been written as a comment to Dr Thomas and Dr Baker's article “Teaching an adult brain new tricks: A critical review of evidence for training-dependent structural plasticity in humans”. We deliberately expand on the key question about the biological substrates underlying use-dependent brain plasticity rather than reiterating the authors' main points of criticism already addressed in more general way by previous publications in the field. The focus here is on the following main issues: i) controversial brain plasticity findings in voxel-based morphometry studies are partially due to the strong dependency of the widely used T1-weighted imaging protocol on varying magnetic resonance contrast contributions; ii) novel concepts in statistical analysis allow one to directly infer topological specificity of structural brain changes associated with plasticity. We conclude that iii) voxel-based quantification of relaxometry derived parameter maps could provide a new perspective on use-dependent plasticity by characterisation of brain tissue property changes beyond the estimation of volume and cortical thickness changes. In the relevant sections we respond to the concerns raised by Dr Thomas and Dr Baker from the perspective of the proposed data acquisition and analysis strategy.

In their article “Teaching an adult brain new tricks: A critical review of evidence for training-dependent structural plasticity in humans” Dr Thomas and Dr Baker emphasise the importance of two key issues in brain plasticity studies using magnetic resonance (MR) imaging: first—the question about reliability of evidence for use-dependent changes and second—the biological substrates of MR measures. Aiming at constructive scientific dialogue we expand on the second key point concerning the main biological contributors to MR contrast, which was only sketchily covered by the critical review. We also consider the fact that the major points of Dr Thomas and Dr Baker's criticism come as no surprise to researchers in the field of neuroimaging accustomed to factorial design and were already brought to the readership's attention in previous studies (Kriegeskorte et al., 2009 and Nieuwenhuis et al., 2011). Further, we propose a complementary approach for data acquisition and analysis based on quantitative assessment of brain tissue properties, which is particularly advantageous for longitudinal study design with multiple scan time points. This strategy has also the potential to provide more straightforward interpretation of plasticity associated structural findings by decomposing the MR contrast to its main contributors—myelin, iron and water protons bound to macromolecules.