Model-based clustering of meta-analytic functional Imaging data

Max-Planck-Institute for Human Cognitive and Brain Sciences, Stephanstrasse 1a, Leipzig, Germany.
Human Brain Mapping (Impact Factor: 5.97). 02/2008; 29(2):177-92. DOI: 10.1002/hbm.20380
Source: PubMed


We present a method for the analysis of meta-analytic functional imaging data. It is based on Activation Likelihood Estimation (ALE) and subsequent model-based clustering using Gaussian mixture models, expectation-maximization (EM) for model fitting, and the Bayesian Information Criterion (BIC) for model selection. Our method facilitates the clustering of activation maxima from previously performed imaging experiments in a hierarchical fashion. Regions with a high concentration of activation coordinates are first identified using ALE. Activation coordinates within these regions are then subjected to model-based clustering for a more detailed cluster analysis. We demonstrate the usefulness of the method in a meta-analysis of 26 fMRI studies investigating the well-known Stroop paradigm.

Download full-text


Available from: D. Y von Cramon,
  • Source
    • "Hierarchical clustering algorithms with linkage functions (such as complete, average, ward, and so on) are based on geometrical ideas where the distances between new clusters and old ones are computed by a linear combination of the distance of their members, which are not meaningful for clustering time series since these linear combinations do not have a meaning in terms of the spectral densities. Clustering brain data based on linear Gaussian models have been used, [5] [16]. However, our main interest is spectral synchronicity and small changes in coefficients of the linear Gaussian models can produce big changes in the spectral density. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Brain activity following stimulus presentation and during resting state are often the result of highly coordinated responses of large numbers of neurons both locally and globally. Coordinated activity of neurons can give rise to oscillations which are captured by electroencephalograms (EEG). In this paper, we examine EEGs as this is the primary data being used by our collaborators who are studying coordination of neuronal response during the execution of tasks such as learning, and memory formation, retention and retrieval. In this paper, we develop the spectral merger clustering (SMC) method that identifies synchronized brain regions during resting state in a sense that these regions share similar oscillations or waveforms. The SMC method, produces clusters of EEGs which serve as a proxy for segmenting the brain cortical surface since the EEGs capture neuronal activity over a locally distributed region on the cortical surface. The extent of desynchronicity between a pair of EEGs is measured using the total variation distance (TVD) which gives the largest possible difference between the spectral densities of the pair of EEGs. We considered the spectral merger algorithm for clustering EEGs, which updates the spectral estimate of the cluster from a weighted average of the spectral estimate obtained from each EEG in the cluster. Numerical experiments suggest that the SMC method performs very well in producing the correct clusters. When applied to resting state EEG data, the method showed how some regions, though not contiguously connected on the cortical surface, are spectrally synchronized during resting state. Moreover, the method demonstrates that brain organization, as expressed in cluster formation, evolves over resting state.
  • Source
    • "Cognitive control refers to a set of processes that enable humans to flexibly shape thoughts and behavior in order to accomplish internal goals (Miller and Cohen 2001). Young adults typically recruit a distributed network of frontoparietal regions during the performance of cognitive control tasks (Badre and Wagner 2006; Cole and Schneider 2007; Neumann et al. 2008; Kim et al. 2012). While a variety of age-related alterations in brain activation during cognitive control tasks have been reported, among the most common appears to be increased activation in frontal brain regions (Drag and Bieliauskas 2010; Reuter-Lorenz and Park 2010). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Increased frontal cortex activation during cognitive task performance is common in aging but remains poorly understood. Here we explored patterns of age-related frontal brain activations under multiple task performance conditions and their relationship to white matter (WM) microstructure. Groups of younger (N = 28) and older (N = 33) participants completed a task-switching paradigm while functional magnetic resonance imaging (fMRI) was performed, and rested while diffusion tensor imaging was performed. Results from fMRI analyses indicated age-related increases in frontal brain activations under conditions of poorer performance in the older group (the nonswitch and switch conditions) and for a contrast in which behavioral performance was equated (older group nonswitch condition vs. younger group switch condition). Within the older adult group, higher frontal activation was associated with poorer behavioral performance under all task conditions. In 2 regions in right frontal cortex, blood oxygen level–dependent (BOLD) magnitudes were negatively correlated with WM integrity in tracts connecting these structures with other task-relevant frontoparietal and striatal regions. Our results link age-related declines in the efficiency of frontal cortex functioning with lower WM integrity in aging.
    Cerebral Cortex 08/2013; 25(1). DOI:10.1093/cercor/bht212 · 8.67 Impact Factor
  • Source
    • "Counterfactuals inherently contain an aspect of conflict, as the counterfactual event contradicts the factual past event. We assume that focusing on the distinction between the counterfactual and factual outcome engages conflict detection (posterior medial frontal cortex; Botvinick et al., 2004) and eliminating the undesirable elements of the past and constructing an alternative course of action strongly requires executive functionality involving performance monitoring, adaptive control, conflict resolution and physical causal reasoning (bilateral PFC; Ridderinkhof et al., 2004; Neumann et al., 2008; Van Overwalle, 2009; Barbey and Patterson, 2011). Counterfactuals not only generate cognitive conflict, but also their emotional counterpart in the form of feelings of regret or relief. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Recent studies suggest that a brain network mainly associated with episodic memory has a more general function in imagining oneself in another time, place or perspective (e.g. episodic future thought, theory of mind, default mode). If this is true, counterfactual thinking (e.g. 'If I had left the office earlier, I wouldn't have missed my train.') should also activate this network. Present functional magnetic resonance imaging (fMRI) study explores the common and distinct neural activity of counterfactual and episodic thinking by directly comparing the imagining of upward counterfactuals (creating better outcomes for negative past events) with the re-experiencing of negative past events and the imagining of positive future events. Results confirm that episodic and counterfactual thinking share a common brain network, involving a core memory network (hippocampal area, temporal lobes, midline, and lateral parietal lobes) and prefrontal areas that might be related to mentalizing (medial prefrontal cortex) and performance monitoring (right prefrontal cortex). In contrast to episodic past and future thinking, counterfactual thinking recruits some of these areas more strongly and extensively, and additionally activates the bilateral inferior parietal lobe and posterior medial frontal cortex. We discuss these findings in view of recent fMRI evidence on the working of episodic memory and theory of mind.
    Social Cognitive and Affective Neuroscience 06/2013; 8:556-564. DOI:10.1093/scan/nss031 · 7.37 Impact Factor
Show more