Limits...
Cortical dynamics and subcortical signatures of motor-language coupling in Parkinson's disease.

Melloni M, Sedeño L, Hesse E, García-Cordero I, Mikulan E, Plastino A, Marcotti A, López JD, Bustamante C, Lopera F, Pineda D, García AM, Manes F, Trujillo N, Ibáñez A - Sci Rep (2015)

Bottom Line: The action-sentence compatibility effect (ACE) paradigm has revealed that EPD involves deficits to integrate action-verb processing and ongoing motor actions.These results corroborate that motor-language coupling is mainly subserved by a cortico-subcortical network including the BG as a key hub.Our findings suggest that research on the relationship between language and motor domains is crucial to develop models of motor cognition as well as diagnostic and intervention strategies.

View Article: PubMed Central - PubMed

Affiliation: 1] Laboratory of Experimental Psychology and Neuroscience (LPEN), Institute of Cognitive Neurology (INECO), Favaloro University, Buenos Aires, 1854, Argentina [2] National Scientific and Technical Research Council (CONICET), Buenos Aires, 1033 Argentina [3] UDP-INECO Foundation Core on Neuroscience (UIFCoN), Faculty of Psychology, Diego Portales University, Santiago, 8370076, Chile.

ABSTRACT
Impairments of action language have been documented in early stage Parkinson's disease (EPD). The action-sentence compatibility effect (ACE) paradigm has revealed that EPD involves deficits to integrate action-verb processing and ongoing motor actions. Recent studies suggest that an abolished ACE in EPD reflects a cortico-subcortical disruption, and recent neurocognitive models highlight the role of the basal ganglia (BG) in motor-language coupling. Building on such breakthroughs, we report the first exploration of convergent cortical and subcortical signatures of ACE in EPD patients and matched controls. Specifically, we combined cortical recordings of the motor potential, functional connectivity measures, and structural analysis of the BG through voxel-based morphometry. Relative to controls, EPD patients exhibited an impaired ACE, a reduced motor potential, and aberrant frontotemporal connectivity. Furthermore, motor potential abnormalities during the ACE task were predicted by overall BG volume and atrophy. These results corroborate that motor-language coupling is mainly subserved by a cortico-subcortical network including the BG as a key hub. They also evince that action-verb processing may constitute a neurocognitive marker of EPD. Our findings suggest that research on the relationship between language and motor domains is crucial to develop models of motor cognition as well as diagnostic and intervention strategies.

No MeSH data available.


Related in: MedlinePlus

Functional connectivity during the ACE task.Global broadcasting of information across distant cortical regions (wSMI). Two tail t-test on the wSMI matrices of each group were obtained by subtracting the correlation matrices of the incongruent and congruent conditions. The Frequency Specificity Graph shows the sensitivity of wSMI to pure-frequency signals. The value of τ makes the wSMI measure sensitive to different frequency ranges (τ = 4 ms is specific for frequencies among 11–40 Hz and τ = 32 ms is specific for frequencies ranging between 1 and 11 Hz). (A) Analysis for tau 4 ms (>11 Hz): (i) histogram showing the number of occurrences (y axis) of the t values (x axis); the distribution of these values exhibits a positive trend, indicating that information sharing is larger for controls than EPD patients; (ii) correlation matrix of raw T value; (iii) masked correlation matrix: T values were corrected with an alpha level set at p < 0.01; non-significant values were assigned a 0; (iv) connectivity map of significant connections only across the scalp indicating that controls presented higher information sharing at frontotemporal regions. (B) Analysis for tau 32 ms (specific for 1–11 Hz): (i) histogram showing the number of occurrences (y axis) of the t values (x axis); the distribution of these values exhibits a positive trend, indicating that information sharing is larger for controls than EPD patients; (ii) correlation matrix of raw T value; (iii) masked correlation matrix: T values were corrected with an alpha level set at p < 0.001; non-significant values were assigned a 0; (iv) connectivity map of significant connections only across the scalp indicating that controls presented higher information sharing mainly at bilateral temporal regions.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4495549&req=5

f2: Functional connectivity during the ACE task.Global broadcasting of information across distant cortical regions (wSMI). Two tail t-test on the wSMI matrices of each group were obtained by subtracting the correlation matrices of the incongruent and congruent conditions. The Frequency Specificity Graph shows the sensitivity of wSMI to pure-frequency signals. The value of τ makes the wSMI measure sensitive to different frequency ranges (τ = 4 ms is specific for frequencies among 11–40 Hz and τ = 32 ms is specific for frequencies ranging between 1 and 11 Hz). (A) Analysis for tau 4 ms (>11 Hz): (i) histogram showing the number of occurrences (y axis) of the t values (x axis); the distribution of these values exhibits a positive trend, indicating that information sharing is larger for controls than EPD patients; (ii) correlation matrix of raw T value; (iii) masked correlation matrix: T values were corrected with an alpha level set at p < 0.01; non-significant values were assigned a 0; (iv) connectivity map of significant connections only across the scalp indicating that controls presented higher information sharing at frontotemporal regions. (B) Analysis for tau 32 ms (specific for 1–11 Hz): (i) histogram showing the number of occurrences (y axis) of the t values (x axis); the distribution of these values exhibits a positive trend, indicating that information sharing is larger for controls than EPD patients; (ii) correlation matrix of raw T value; (iii) masked correlation matrix: T values were corrected with an alpha level set at p < 0.001; non-significant values were assigned a 0; (iv) connectivity map of significant connections only across the scalp indicating that controls presented higher information sharing mainly at bilateral temporal regions.

Mentions: The ACE is hypothesized to require integration of information among frontotemporal regions313. We explored this conjecture using a novel measure of integration and global broadcasting of information across distant cortical regions, called Weighted Symbolic Mutual Information (wSMI) (for details, see Supplementary Data, Connectivity subsection of Material and Methods, and31). The wSMI measure assesses the extent to which two signals present nonrandom joint fluctuations (sharing information), characterized by (a) fast and robust estimation of the signals’ entropies, (b) detection of nonlinear coupling, and (c) absence of spurious correlations between EEG signals arising from common sources31. EEG signals were first transformed into a series of discrete symbols defined by the ordering of k time samples segregated by a temporal separation τ. Analysis was restricted to a fixed symbol size (k = 3) and two different values of τ (τ = 4 and 32 ms between time samples). We focused our analysis on τ = 4 and 32 ms because they are specific for two different frequency ranges: slow oscillations related to MP (assessed with τ = 32 ms, 1–11 Hz) and a broad band of higher frequencies (τ = 4 ms, 11–40 Hz). Previous studies on PD have reported affectations in the former21 and aberrant oscillations in the latter3233 (see Supplementary Data, Connectivity subsection of Material and Methods)31. To calculate wSMI for each pair of transformed EEG signals, we estimated the joint probability of each pair of symbols. To reduce spurious correlations between signals, the joint probability matrix was multiplied by binary weights. The weights were set to zero for pairs of identical symbols, which could be elicited by a unique common source, and for opposed symbols, which could reflect the two sides of a single electric dipole. We used the increment of sharing information between ACE conditions by subtracting the correlation matrix of the incongruent condition from that of the congruent condition. This measure was used to investigate connectivity differences between both groups (EPD and controls, Fig. 2). Also, two-tail t tests were performed between correlation matrices of each group. Alpha levels were set at p < 0.01.


Cortical dynamics and subcortical signatures of motor-language coupling in Parkinson's disease.

Melloni M, Sedeño L, Hesse E, García-Cordero I, Mikulan E, Plastino A, Marcotti A, López JD, Bustamante C, Lopera F, Pineda D, García AM, Manes F, Trujillo N, Ibáñez A - Sci Rep (2015)

Functional connectivity during the ACE task.Global broadcasting of information across distant cortical regions (wSMI). Two tail t-test on the wSMI matrices of each group were obtained by subtracting the correlation matrices of the incongruent and congruent conditions. The Frequency Specificity Graph shows the sensitivity of wSMI to pure-frequency signals. The value of τ makes the wSMI measure sensitive to different frequency ranges (τ = 4 ms is specific for frequencies among 11–40 Hz and τ = 32 ms is specific for frequencies ranging between 1 and 11 Hz). (A) Analysis for tau 4 ms (>11 Hz): (i) histogram showing the number of occurrences (y axis) of the t values (x axis); the distribution of these values exhibits a positive trend, indicating that information sharing is larger for controls than EPD patients; (ii) correlation matrix of raw T value; (iii) masked correlation matrix: T values were corrected with an alpha level set at p < 0.01; non-significant values were assigned a 0; (iv) connectivity map of significant connections only across the scalp indicating that controls presented higher information sharing at frontotemporal regions. (B) Analysis for tau 32 ms (specific for 1–11 Hz): (i) histogram showing the number of occurrences (y axis) of the t values (x axis); the distribution of these values exhibits a positive trend, indicating that information sharing is larger for controls than EPD patients; (ii) correlation matrix of raw T value; (iii) masked correlation matrix: T values were corrected with an alpha level set at p < 0.001; non-significant values were assigned a 0; (iv) connectivity map of significant connections only across the scalp indicating that controls presented higher information sharing mainly at bilateral temporal regions.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4495549&req=5

f2: Functional connectivity during the ACE task.Global broadcasting of information across distant cortical regions (wSMI). Two tail t-test on the wSMI matrices of each group were obtained by subtracting the correlation matrices of the incongruent and congruent conditions. The Frequency Specificity Graph shows the sensitivity of wSMI to pure-frequency signals. The value of τ makes the wSMI measure sensitive to different frequency ranges (τ = 4 ms is specific for frequencies among 11–40 Hz and τ = 32 ms is specific for frequencies ranging between 1 and 11 Hz). (A) Analysis for tau 4 ms (>11 Hz): (i) histogram showing the number of occurrences (y axis) of the t values (x axis); the distribution of these values exhibits a positive trend, indicating that information sharing is larger for controls than EPD patients; (ii) correlation matrix of raw T value; (iii) masked correlation matrix: T values were corrected with an alpha level set at p < 0.01; non-significant values were assigned a 0; (iv) connectivity map of significant connections only across the scalp indicating that controls presented higher information sharing at frontotemporal regions. (B) Analysis for tau 32 ms (specific for 1–11 Hz): (i) histogram showing the number of occurrences (y axis) of the t values (x axis); the distribution of these values exhibits a positive trend, indicating that information sharing is larger for controls than EPD patients; (ii) correlation matrix of raw T value; (iii) masked correlation matrix: T values were corrected with an alpha level set at p < 0.001; non-significant values were assigned a 0; (iv) connectivity map of significant connections only across the scalp indicating that controls presented higher information sharing mainly at bilateral temporal regions.
Mentions: The ACE is hypothesized to require integration of information among frontotemporal regions313. We explored this conjecture using a novel measure of integration and global broadcasting of information across distant cortical regions, called Weighted Symbolic Mutual Information (wSMI) (for details, see Supplementary Data, Connectivity subsection of Material and Methods, and31). The wSMI measure assesses the extent to which two signals present nonrandom joint fluctuations (sharing information), characterized by (a) fast and robust estimation of the signals’ entropies, (b) detection of nonlinear coupling, and (c) absence of spurious correlations between EEG signals arising from common sources31. EEG signals were first transformed into a series of discrete symbols defined by the ordering of k time samples segregated by a temporal separation τ. Analysis was restricted to a fixed symbol size (k = 3) and two different values of τ (τ = 4 and 32 ms between time samples). We focused our analysis on τ = 4 and 32 ms because they are specific for two different frequency ranges: slow oscillations related to MP (assessed with τ = 32 ms, 1–11 Hz) and a broad band of higher frequencies (τ = 4 ms, 11–40 Hz). Previous studies on PD have reported affectations in the former21 and aberrant oscillations in the latter3233 (see Supplementary Data, Connectivity subsection of Material and Methods)31. To calculate wSMI for each pair of transformed EEG signals, we estimated the joint probability of each pair of symbols. To reduce spurious correlations between signals, the joint probability matrix was multiplied by binary weights. The weights were set to zero for pairs of identical symbols, which could be elicited by a unique common source, and for opposed symbols, which could reflect the two sides of a single electric dipole. We used the increment of sharing information between ACE conditions by subtracting the correlation matrix of the incongruent condition from that of the congruent condition. This measure was used to investigate connectivity differences between both groups (EPD and controls, Fig. 2). Also, two-tail t tests were performed between correlation matrices of each group. Alpha levels were set at p < 0.01.

Bottom Line: The action-sentence compatibility effect (ACE) paradigm has revealed that EPD involves deficits to integrate action-verb processing and ongoing motor actions.These results corroborate that motor-language coupling is mainly subserved by a cortico-subcortical network including the BG as a key hub.Our findings suggest that research on the relationship between language and motor domains is crucial to develop models of motor cognition as well as diagnostic and intervention strategies.

View Article: PubMed Central - PubMed

Affiliation: 1] Laboratory of Experimental Psychology and Neuroscience (LPEN), Institute of Cognitive Neurology (INECO), Favaloro University, Buenos Aires, 1854, Argentina [2] National Scientific and Technical Research Council (CONICET), Buenos Aires, 1033 Argentina [3] UDP-INECO Foundation Core on Neuroscience (UIFCoN), Faculty of Psychology, Diego Portales University, Santiago, 8370076, Chile.

ABSTRACT
Impairments of action language have been documented in early stage Parkinson's disease (EPD). The action-sentence compatibility effect (ACE) paradigm has revealed that EPD involves deficits to integrate action-verb processing and ongoing motor actions. Recent studies suggest that an abolished ACE in EPD reflects a cortico-subcortical disruption, and recent neurocognitive models highlight the role of the basal ganglia (BG) in motor-language coupling. Building on such breakthroughs, we report the first exploration of convergent cortical and subcortical signatures of ACE in EPD patients and matched controls. Specifically, we combined cortical recordings of the motor potential, functional connectivity measures, and structural analysis of the BG through voxel-based morphometry. Relative to controls, EPD patients exhibited an impaired ACE, a reduced motor potential, and aberrant frontotemporal connectivity. Furthermore, motor potential abnormalities during the ACE task were predicted by overall BG volume and atrophy. These results corroborate that motor-language coupling is mainly subserved by a cortico-subcortical network including the BG as a key hub. They also evince that action-verb processing may constitute a neurocognitive marker of EPD. Our findings suggest that research on the relationship between language and motor domains is crucial to develop models of motor cognition as well as diagnostic and intervention strategies.

No MeSH data available.


Related in: MedlinePlus