An RMSEA of less than .08 is indicative of a close fit of the sample (empirical) covariance selleck compound matrix to the population matrix (Browne & Cudeck, 1993). Goodness of fit of the

overall model was determined using descriptive statistics such as the likelihood ratio chi-square statistic, χ2, (models with a χ2 of zero indicate a theoretical model that fits the data perfectly), p-value (high p-values indicate a model is unlikely to be refuted in other independent samples), and a root mean square error of approximation (RMSEA) index of less than .08 indicating minimal discrepancy between the empirical or sample covariance matrix and the population. The class of models evaluated in this study was nonrecursive. In nonrecursive SEMs the presence of bidirectional feedback loops creates the possibility of a non-stable system resulting in biased parameter estimates. In our models, stability of the nonrecursive system was evaluated using the stability index based on the work of Bentler and Freeman (1983). In all models the stability index was between −1.0 and 1.0 verifying that the nonrecursive models were stable. Separate nonrecursive models were created for the shift and no shift conditions. The no shift

condition revealed connectivity associated with vocalization without error with a chi square fit index of 31.411, RMSEA = .071. Not surprisingly, we found that there are many connections IWR-1 datasheet between and within hemispheres. Connections presented in the left hemisphere include left Forskolin manufacturer M1 to left PMC, left STG, and left IFG which emphasizes the extent of connectivity necessary with the motor cortex to execute

speech accurately. Left IFG showed coupling with left PMC regions commonly associated with the voice and speech network and contributors to speech articulation and retrieval of speech sounds. Left STG showed a relationship with left IFG and likely contributes to voice perception and processing. Right hemisphere connections include right M1 to right IFG and right PMC. A negative connection from right IFG to right M1 was also observed. The connections in the right hemisphere contribute to pitch processing. Cross hemisphere connections include, left STG to right M1, left IFG to right M1, left STG to right STG, and left IFG to right PMC. Lastly, a negative connection is visible from right PMC to left IFG. These cross-hemisphere connections indicate that vocalization requires crosstalk from both hemispheres to ensure accurate vocalization. No shift connectivity is shown in black (Fig. 1). The shift condition consisted of rapid 200 ms shifts presented to the subject. These quick deviations from the subjects’ intended vocal output were likely processed as errors. Therefore, changes in connectivity between the no shift and shift conditions are likely due to this detected error and the processes associated with error correction. Here we present the resulting shift model which yielded a chi square fit index of 32.302, and RMSEA = .072.