However, using hemodynamic responses derived from real data, Schippers et al. (2011) demonstrated DAPT that GCA identified causal influences in group studies with good sensitivity and specificity. When effects are observed using random-effects analysis in which the effect to interests is compared with variance between

subjects, the detection of a significant group effect implies the occurrence of a systematic delay in neural and/or hemodynamic response. The results obtained by Schippers et al. (2011) indicate that the effects are most likely to be neural. This conclusion is supported by the fact that the regions involved are served by different arteries and therefore group effects due to hemodynamic delay would only be expected if there were differences in arterial transmission times that were consistent across subjects. However, any such systematic differences would be expected to be similar in the two hemispheres, yet neither the effects reported by Sridharan et al. (2008) nor those that we report are symmetrical across the hemispheres. Furthermore, examination of the timing of regional neural activity using magnetoencephalography (Brookes et al., 2012) demonstrates appreciable neural delays between occipital cortex and insula during various visual tasks, consistent with our present findings that occipital cortex exerts a Granger causal influence on insula. An additional issue

raised by Smith et al. (2011) is the possibility that in a Granger causality analysis, findings might be distorted by zero-lag correlations “bleeding into” the time-lagged

relationships. We have demonstrated JQ1 mouse that significant zero-lag correlations between insula and other brain regions occur at different locations from the Granger causal effects of insula on other brain regions. To our knowledge, this is the first study to examine time-directed neural primacy effects during task-free resting state in schizophrenia. Our findings extend the neuronal network level models informing the pathophysiology of this illness. Effective cognitive control requires successful suppression of distractors (e.g., spontaneous internal thoughts) but at the same time must be responsive to unexpected stimuli, which though irrelevant to the task are salient for our homeostatic however defense (Su et al., 2011). The concept of “proximal salience” refers to the switching between brain states (e.g., task-focused, resting or internally focused, and sensory-processing states) brought on by a momentary state of neural activity within the salience processing system, anchored in the rAI and the dACC (Palaniyappan and Liddle, 2012). We infer that the breakdown of the causal influence to and from the salience processing system in schizophrenia amounts to a failure of proximal salience mechanism. The present study highlights the importance of studying the pathways of failed interaction between large-scale networks in the pathophysiology of schizophrenia.