Fronto-parietal networks activation during the contingent negative variation period

Abstract

The preparation for stimuli and responses in which the position and required finger to respond are cued, produces the preparatory activation of the specific neural resources that are going to be needed for the completion of the task. The focus of the present report is to evaluate if the fronto-parietal networks activated in fMRI studies during endogenous attention are also activated during the CNV period using EEG recording. The behavioural responses and 64 EEG channels were recorded during an S1–S2 paradigm similar to Posner central cue paradigms. The LORETA analysis based in the averaging of the z-LORETA values showed that the Brodmann's areas with the highest activation during the CNV period were in the medial and superior frontal areas, fronto-parietal lateral areas (including the premotor cortex) and extrastriate visual cortex. These results suggest that in addition to the previously described activation in premotor-motor, posterior sensory and superior and medial frontal areas, the activation of fronto-parietal networks is a main contributor to the CNV, indicating the endogenous attentional effort during the CNV period.

Introduction

The preparation by a central spatial cue to respond to an S2 produces a decrease in the RT when the S2 is validly cued when compared with an invalidly cued target [22]. The contingent negative variation (CNV) is generated when the preparation for an imperative stimulus (S2) is induced by a warning stimulus (S1) [27], [25]. Moreover, the experiments in which S2 discrimination is required and the S1 conveys specific information about the features of the S2, the activation of the neural resources needed for the sensory analysis and required responses to the S2 is observed during the CNV period [5], [8], [11], [14], [13], [17]. During the CNV period there is also a considerable reorganization of brain rhythms that includes desynchronization but also an increase of coherence between different cortical regions [1], [2], [15].

The CNV represents then the activation of specific resources, but it is also a correlate of the endogenous attentional effort during the expectancy period [7], [17]. On the other hand, there is a growing consensus on the importancy of the fronto-parietal networks for inducing endogenous attention [10], [12]. In the review of Cabeza and Nyberg [9], the fronto-parietal networks appeared activated in most of the referred contrasts. In an study using central and peripheral cues, and comparing the bold response of cued trials with neutral trials, a neural network including the inferior frontal gyrus (BA 9), the medial frontal cortex including supplementary motor area and anterior cingulated cortex (BA 6, 32), the bilateral premotor cortex including frontal eye fields (BA 6), the right temporoparietal junction (BA 22/40), the bilateral precuneus (BA 19) and the cerebellum was activated [21]. On a more general ground, a fronto-parietal network including the frontal eye-fields and the intraparietal sulcus has been defined to control spontaneous and externally cued spatial orienting [10], [12].

Until which extent the CNV represents the activation of fronto-parietal networks has only been scarcely investigated. Brunia and Damen [7] found a sustained negativity over the prefrontal cortex, accompanied by a sharply increasing negativity over the parietal cortex, during the waiting period for feedback after a time estimation task, the so-called stimulus preceding negativity (SPN). This component tries to isolate the sensory anticipatory contributions from the motor component, given that in the CNV both components are intermingled. Dipole localization yielded a bilateral fronto-temporal dipole, possibly in the insula Reili [6]. Gathering these results, it has been suggested a fronto-parietal network and the insula Reili as a possible source of the SPN. A PET study confirmed the dipole localization suggestions [8]. In another study using LORETA, the left prefrontal cortex and the right inferior parietal cortex were activated [14] during the CNV period in a non-cued task. The posterior parietal cortex was also activated during the contingent magnetic variation in a spatially cued task [13]. Recently the implication of frontal and parietal networks in the generation of CNV has also been proposed by means of intracerebral recordings [3]. All these studies suggest the involvement of a fronto-parietal network activation during the CNV period. The aim of present report was to localize, by means of a distributed source model (LORETA), the sources of the CNV, in order to test the hypothesis that, in fact, CNV represents the activation of the fronto-parietal networks that are activated during endogenous attentional experiments, as measured by hemodynamics techniques [9], [10], [12]. For that objective, the late period of the CNV will be analyzed by means of LORETA, using a thresholding approach. This analysis would allow to remark the attentional functionality of the CNV based on an anatomical argument.