Percepción de Exigencias de la educación superior a media

The detection of multiple nodes of these networks requires measurement from locations other than just the prefrontal areas accessible through the forehead. This underscores the need for the development of headgear that can efficiently and reliably be used across the whole head. By monitoring networks across the brain instead of single locations of activity, the brain state in a

28

given time period is truly being monitored, as opposed to only monitoring activity over time at one location, which may not be functionally specific for the state of interest (LaConte et al., 2007). For example, fNIRS measurements made on the forehead only and averaged broadly across each hemisphere may be sensitive to workload while not indicating specific causal states (Izzetoglu et al., 2004). Such measurements should be interpreted with care, given the role of the medial frontal gyrus in the DMN. Separately measuring ATN and DMN activations may allow state sensing with improved accuracy by providing additional information to improve state differentiation. This research question is addressed in Chapter 3.

Detailed coordinates of specific surface optode locations are given below and plotted in Figure 3. These coordinates are used to measure the selected volumetric regions of interest. The Brodman Area (BA) is given, along with the depth of the cortex of interest and the Talairach to Ten-Twenty (TT) coordinates. TT coordinates are generated by projecting a line from a volume of cortex in the brain along the normal axis to the scalp surface (Steinstrater et al., 2001). Cz is the origin at the center, with Fz at (0, 1) and C4 at (1, 0). The two dimensional coordinate system and relevant volume to surface projections are also shown in Figure 3. The depth of the tissue is identified by the color bar on the right. The selected regions of interest are bilateral.

Based on fMRI literature (Kelly et al., 2008; Fox et al., 2005; Weissman et al., 2006), we predicted these task negative / resting state network sub-components would show decreasing activations during engagement:

 Medial Frontal Gyrus (BA 10, TT(±0.2,1.5), depth 18mm) (marked in Figure 3

bilaterally with blue circles between Fp and Fpz; the center of the forehead)

 Angular Gyrus (BA 39, TT(±1.3,-1.0), depth 15mm) (marked in Figure 3 with

blue circles and by projection to the area between P4 and T6)

29

Figure 3. Tissue projections with respect to the 10-20 measurement locations

Eight measurement locations (source-detector pair midpoints) are shown schematically with cortex to be interrogated beneath the marked circles. The color scale indicates depth. The tissue projection tool is available at http://wwwneuro03.uni-muenster.de/ger/t2tconv/conv3d.html .

Similarly, we predicted these task-positive / attentional network sub-components would show increasing activations during task engagement, and together be anti-correlated with the task- negative network activations:

 DLPFC (BA 46, TT(1.3,1.0), depth 10 to 13mm) (marked in Figure 3 with red

circles and by projection to the area between F4 and F8; just into the hairline from the temple)

 Supplementary Motor (BA 6, TT(0.9,0.3), depth 15mm) (marked in Figure 3 with

red circles and by projection to the area anterior to the line between C3 and C4)

 Inferior Parietal Lobule (BA 40, TT(1.3,-0.5), depth 19mm) (not shown in Figure) We have chosen to focus initially on the following two locations. The dorsolateral PFC (DLPFC) is of particular interest to us for monitoring task engagement and sustained attention (vigilance) (Posner, 2004), and is also part of the TPN. The Medial Frontal Gyrus (MFG), as a

30

node of the DMN (Greicius, et al. 2003; Raichle, et al. 2001), is of particular interest for indicating rest or a low level of task engagement, and is also part of the TNN. (Kelly et al., 2008; Fox et al., 2005; Weissman et al., 2006)

Details of the implementation of classification algorithms to predict operator state (engaged or not engaged in a task) based on measures of activity in these regions will be discussed in section 2.4. First, we present methods for inducing the task engagement we wish to quantify.

31 2.3 Behavioral Methods

To facilitate our ability to make this distinction, we asked participants to perform the multi-source interference task (MSIT) (Bush and Shin, 2006; Stins et al., 2005). The MSIT is a selective attention task in which optimal performance requires participants to suppress multiple sources of interference. Stroop interference may be due to any semantic meaning of the prompt, Eriksen interference is present in non-congruent trials due to flanking distractors, and Simon interference is present in non-congruent trials due to conflict between the placement of the target numbers and the finger used for the correct response. Thus, it reliably and robustly activates the TPN, even in individual task blocks from the same participant (Bush and Shin, 2006). Given these characteristics, we reasoned that the MSIT would provide a strong signal with which to monitor task engagement. Consistent with this reasoning, in the present study we were successful at distinguishing between relatively high and relatively low periods of task engagement.

The variability of the cognitive functions themselves, the task execution strategy, the presence of internally-guided or task-unrelated thoughts, and the physiological state of the operator, are inherent confounding factors (Arenth et al., 2007) that change over time. These were not corrected for in this study, beyond controlling for these variances by collecting the fNIRS and fMRI data simultaneously in part two of the study. Results are presented in sections 3.4 and 4.2. All statistical tests on behavioral results were one-tailed, with comparisons paired by participant.