Abstract
Objective: In a previous study (Brázdil et al., 2012), we revealed that the frequency of non-pathological déjà vu (DV) experience is related inversely to grey-matter (GM) volume throughout a diffuse set of cortical and subcortical brain structures. In this study, we set out to develop this finding by examining if patterns of GM decline occur in a co-ordinated fashion among these brain structures, and how patterns of GM co-variance relate to DV frequency.
Method: We compared GM covariance among 16 regions of interest emerging from our previous analyses, between three groups defined according to self-reported DV experience. To do so we employed partial least-squares (PLS). In contrast to voxel-based morphometry, this is a multivariate technique for structural co-variance mapping. To assess whether the emerging patterns of structural co-variance indexed functional networks, we compared our results to functional connectivity data supplied under the Human Connectome Project (HCP) database (http://www.humanconnectomeproject.org).
Results: Our PLS analyses revealed two patterns of GM co-variance associated with non-pathological DV frequency. The first revealed that grey-matter alterations co-varied increasingly with higher DV frequency among limbic structures and the caudate; the second identified decreasing co-variance in GM alterations with higher DV frequency among medial and lateral temporal structures. Assessing functional connectivity among the same set of brain structures comprising each structural covariance pattern indicated that they reflect two distinct brain networks.
Conclusions: We suggest non-pathological DV emerges as a result of specific alterations in patterns of neural connectivity within and between medial and lateral temporal cortical networks, leading to these distinct patterns of coordinated structural alterations. This goes some way towards reconciling the discrepant findings concerning the role of lateral temporal cortex in pathological DV.
Method: We compared GM covariance among 16 regions of interest emerging from our previous analyses, between three groups defined according to self-reported DV experience. To do so we employed partial least-squares (PLS). In contrast to voxel-based morphometry, this is a multivariate technique for structural co-variance mapping. To assess whether the emerging patterns of structural co-variance indexed functional networks, we compared our results to functional connectivity data supplied under the Human Connectome Project (HCP) database (http://www.humanconnectomeproject.org).
Results: Our PLS analyses revealed two patterns of GM co-variance associated with non-pathological DV frequency. The first revealed that grey-matter alterations co-varied increasingly with higher DV frequency among limbic structures and the caudate; the second identified decreasing co-variance in GM alterations with higher DV frequency among medial and lateral temporal structures. Assessing functional connectivity among the same set of brain structures comprising each structural covariance pattern indicated that they reflect two distinct brain networks.
Conclusions: We suggest non-pathological DV emerges as a result of specific alterations in patterns of neural connectivity within and between medial and lateral temporal cortical networks, leading to these distinct patterns of coordinated structural alterations. This goes some way towards reconciling the discrepant findings concerning the role of lateral temporal cortex in pathological DV.
Original language | English |
---|---|
Pages (from-to) | e38-e39 |
Journal | Clinical Neurophysiology |
Volume | 125 |
Issue number | 5 |
DOIs | |
Publication status | Published - May 2014 |