Limits...
Construct validation of a DCM for resting state fMRI.

Razi A, Kahan J, Rees G, Friston KJ - Neuroimage (2014)

Bottom Line: Dynamic causal modelling (DCM) is a framework that allows for the identification of the causal (directed) connections among neuronal systems--known as effective connectivity.We also simulated group differences and compared the ability of spectral and stochastic DCMs to identify these differences.We show that spectral DCM was not only more accurate but also more sensitive to group differences.

View Article: PubMed Central - PubMed

Affiliation: The Wellcome Trust Centre for Neuroimaging, University College London, 12 Queen Square, London WC1N 3BG, UK; Department of Electronic Engineering, NED University of Engineering and Technology, Karachi, Pakistan. Electronic address: a.razi@ucl.ac.uk.

Show MeSH

Related in: MedlinePlus

This figure reports the results of a simulated group comparison study of two groups of 24 participants (with 512 scans per participant). The upper row shows the Bayesian parameter averages of the differences using the same format as previous figures. For the spectral DCM (left panel) it can be seen that increases in the extrinsic forward connections from the second to the third region (seventh parameter) has been estimated accurately. Similarly, the decrease in the backward connection from the fourth to the third region is also estimated accurately. For the stochastic DCM (right panel), the estimation of the differences in the two parameter sets is not as accurate — although the direction is detected correctly. The equivalent classical inference — based upon the t-statistic is shown on lower row. Here the posterior means from each of 48 subjects were used as summary statistics and entered into a series of univariate t-tests to assess differences in group means. The red lines correspond to significance thresholds at a nominal false-positive rate of p = 0.05 corrected (solid lines) and uncorrected (broken lines). Clearly the connections with differences survive the corrected threshold for spectral DCM (left panel) whereas for the stochastic DCM (right panel) few other connections are also above threshold.
© Copyright Policy - CC BY
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4295921&req=5

f0030: This figure reports the results of a simulated group comparison study of two groups of 24 participants (with 512 scans per participant). The upper row shows the Bayesian parameter averages of the differences using the same format as previous figures. For the spectral DCM (left panel) it can be seen that increases in the extrinsic forward connections from the second to the third region (seventh parameter) has been estimated accurately. Similarly, the decrease in the backward connection from the fourth to the third region is also estimated accurately. For the stochastic DCM (right panel), the estimation of the differences in the two parameter sets is not as accurate — although the direction is detected correctly. The equivalent classical inference — based upon the t-statistic is shown on lower row. Here the posterior means from each of 48 subjects were used as summary statistics and entered into a series of univariate t-tests to assess differences in group means. The red lines correspond to significance thresholds at a nominal false-positive rate of p = 0.05 corrected (solid lines) and uncorrected (broken lines). Clearly the connections with differences survive the corrected threshold for spectral DCM (left panel) whereas for the stochastic DCM (right panel) few other connections are also above threshold.

Mentions: The Bayesian parameter averages of the group differences for each of the coupling estimates are presented in Fig. 6. The spectral scheme produces both highly accurate and precise estimates of the differences at the two connections manipulated. Any changes in coupling strengths that were not changed were < 0.05 Hz, and can thus be considered a small (and standardised) effect size in quantitative terms. Regarding stochastic DCM, group differences were less accurate but remain qualitatively similar (i.e., they are still in the correct direction). Estimates of differences in other parameters again remained < 0.05 Hz. When applying classical inference (as appears routine in such studies) to the Bayesian parameter averages, both schemes detected significant changes in the two manipulated parameters. The dotted red line corresponds to the thresholds for a nominal level of p = 0.05 uncorrected, whereas the broken red line corresponds to the corrected thresholds for the 16 tests. We used the self-connections for the correction but do not report the (non-significant) differences. For spectral DCM, both forward and backward connections are highly significant, even after correcting for multiple comparisons, and no other connection reached uncorrected threshold. Regarding stochastic DCM, although both backward and forward connections are still detected as the two most statistically significant connections, small changes estimated in a few other connections were found to be significant at corrected statistical thresholds, suggesting that sDCM is sensitive to group differences, but perhaps not as specific as spDCM.


Construct validation of a DCM for resting state fMRI.

Razi A, Kahan J, Rees G, Friston KJ - Neuroimage (2014)

This figure reports the results of a simulated group comparison study of two groups of 24 participants (with 512 scans per participant). The upper row shows the Bayesian parameter averages of the differences using the same format as previous figures. For the spectral DCM (left panel) it can be seen that increases in the extrinsic forward connections from the second to the third region (seventh parameter) has been estimated accurately. Similarly, the decrease in the backward connection from the fourth to the third region is also estimated accurately. For the stochastic DCM (right panel), the estimation of the differences in the two parameter sets is not as accurate — although the direction is detected correctly. The equivalent classical inference — based upon the t-statistic is shown on lower row. Here the posterior means from each of 48 subjects were used as summary statistics and entered into a series of univariate t-tests to assess differences in group means. The red lines correspond to significance thresholds at a nominal false-positive rate of p = 0.05 corrected (solid lines) and uncorrected (broken lines). Clearly the connections with differences survive the corrected threshold for spectral DCM (left panel) whereas for the stochastic DCM (right panel) few other connections are also above threshold.
© Copyright Policy - CC BY
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4295921&req=5

f0030: This figure reports the results of a simulated group comparison study of two groups of 24 participants (with 512 scans per participant). The upper row shows the Bayesian parameter averages of the differences using the same format as previous figures. For the spectral DCM (left panel) it can be seen that increases in the extrinsic forward connections from the second to the third region (seventh parameter) has been estimated accurately. Similarly, the decrease in the backward connection from the fourth to the third region is also estimated accurately. For the stochastic DCM (right panel), the estimation of the differences in the two parameter sets is not as accurate — although the direction is detected correctly. The equivalent classical inference — based upon the t-statistic is shown on lower row. Here the posterior means from each of 48 subjects were used as summary statistics and entered into a series of univariate t-tests to assess differences in group means. The red lines correspond to significance thresholds at a nominal false-positive rate of p = 0.05 corrected (solid lines) and uncorrected (broken lines). Clearly the connections with differences survive the corrected threshold for spectral DCM (left panel) whereas for the stochastic DCM (right panel) few other connections are also above threshold.
Mentions: The Bayesian parameter averages of the group differences for each of the coupling estimates are presented in Fig. 6. The spectral scheme produces both highly accurate and precise estimates of the differences at the two connections manipulated. Any changes in coupling strengths that were not changed were < 0.05 Hz, and can thus be considered a small (and standardised) effect size in quantitative terms. Regarding stochastic DCM, group differences were less accurate but remain qualitatively similar (i.e., they are still in the correct direction). Estimates of differences in other parameters again remained < 0.05 Hz. When applying classical inference (as appears routine in such studies) to the Bayesian parameter averages, both schemes detected significant changes in the two manipulated parameters. The dotted red line corresponds to the thresholds for a nominal level of p = 0.05 uncorrected, whereas the broken red line corresponds to the corrected thresholds for the 16 tests. We used the self-connections for the correction but do not report the (non-significant) differences. For spectral DCM, both forward and backward connections are highly significant, even after correcting for multiple comparisons, and no other connection reached uncorrected threshold. Regarding stochastic DCM, although both backward and forward connections are still detected as the two most statistically significant connections, small changes estimated in a few other connections were found to be significant at corrected statistical thresholds, suggesting that sDCM is sensitive to group differences, but perhaps not as specific as spDCM.

Bottom Line: Dynamic causal modelling (DCM) is a framework that allows for the identification of the causal (directed) connections among neuronal systems--known as effective connectivity.We also simulated group differences and compared the ability of spectral and stochastic DCMs to identify these differences.We show that spectral DCM was not only more accurate but also more sensitive to group differences.

View Article: PubMed Central - PubMed

Affiliation: The Wellcome Trust Centre for Neuroimaging, University College London, 12 Queen Square, London WC1N 3BG, UK; Department of Electronic Engineering, NED University of Engineering and Technology, Karachi, Pakistan. Electronic address: a.razi@ucl.ac.uk.

Show MeSH
Related in: MedlinePlus