Temporal dynamics of brain activation during a working memory task

Abstract

Gamma band oscillations participate in the temporal binding needed to synchronize cortical networks, involved in early sensory and short term memory processes. In earlier studies, alterations of these neurophysiological parameters have been found in psychotic disorders. To date no study has explored the temporal dynamics and signal complexity of gamma band oscillations in first episode psychosis (FEP). To address this issue, gamma band analysis was performed in 15 FEP patients and 18 healthy controls who successfully performed an adapted 2-back working memory task. Multiple linear and logistic regression models were computed to explore the relationship between the cognitive status and gamma oscillation changes over time. Based on regression model results, phase diagrams were constructed and their complexity was estimated using fractal dimension, a mathematical tool that describes shapes as numeric values. When adjusted for gamma values at time lags-3 to-4 ms and-15 to-16 ms, FEP patients displayed significantly higher time-dependent changes than controls, independently of the nature of the task. The present results are consistent with a discoordination of the activity of cortical generators engaged by the stimulus apparition in FEP patients, leading to a global binding deficit. In addition, fractal analysis showing higher complexity of gamma signal, confirmed this deficit. Our results provide evidence for recruitment of supplementary cortical generators as compensating mechanisms and yield further understanding for the pathophysiology cognitive impairments in FEP.

Tuning down the emotional brain: An fMRI study of the effects of cognitive load on the processing of affective images

“…Consistent with previous work (Cohen et al, 1997;Prabhakaran et al, 2000), high task load further resulted in an increase in activity in the regions implicated in the arithmetic task (right dorsolateral frontal cortex, right superior parietal cortex, dorsal occipital cortex). More importantly, task load also resulted in a decrease in brain regions involved in emotion processing (bilateral amygdalae, right insula).…”

Brain imaging of the central executive component of working memory

“…For example, the middle prefrontal gyrus (BA 9/46) was found in manipulation [30,40,98] and updating [103,124] tasks as well as in dual-task coordination [39], inhibition processes [26,31] and shifting processes [102]. Furthermore, the activity of that region was directly linked to the memory load, as demonstrated by Braver et al [17] and Cohen et al [28] using parametric design studies. Similarly, activity of the frontopolar cortex (BA 10) was found not only in updating tasks [103,124] but also in verbal span tasks in which the number of words to recall was superior to the subject's span [13], or in tasks requiring retrieval processes in episodic memory [126].…”

Optimization of experimental design in fMRI: a general framework using a genetic algorithm

“…Designs with no HRF shape estimation power confound these types of effects. Another case where HRF shape estimation is particularly useful is in estimating differences in HRF delay onset between events (Miezin et al, 2000) and in cases where the hypothesis depends on the shape of activation, as in studies that examine maintenance of information in working memory over short delay intervals (Cohen et al, 1997). Friston et al (1999) gave a formulation for the standard error of a contrast across parameter estimates for different predictors, allowing for generalization of efficiency to contrasts among psychological conditions.…”