Abstract / Description of output
Background: Early visual information processing and higher-order cognitive processes like working memory and attention are considerably intertwined. This is exemplified by the interplay of visual areas with the bottom-up attention network during target detection. Hence, studying functional interactions between early visual areas and regions closely involved in higher order cognitive processes are of particular importance. However, the underlying functional connectivity patterns are not fully understood. Therefore, we aimed to investigate the interplay between early visual and higher-order brain areas during a target-detection task.
Methods: 50 healthy participants underwent fMRI in a 3 T Siemens Trio scanner. A target-detection task consisting of a series of flickering black-and-white-colored round shaped checkerboard stimuli subtending 0.3 degrees of visual angle was employed. Every two seconds, the checkerboards position switched among one of the possible quadrants. Each of the positions were located at equidistant points on an imaginary circle surrounding a central fixation cross. Thus, checkerboards appeared at homologous positions in each visual quadrant. The whole array subtended 2.4 degrees of visual angle. At random intervals, two centrally located squares of the checkerboards changed their color to yellow (target trials). Participants had to indicate by button press the detection of these targets. Throughout the task participants had to fixate a black, x-shaped fixation cross at the center of the screen. Overall, task duration was 5.5 minutes. Data analysis in BrainVoyager included standard data pre-processing. Additionally, we used a multiscale curvature driven cortex-based alignment procedure mitigating interindividual macro-anatomical correspondence. Functional data were analyzed using a random-effects multi-subject general linear model (GLM). We assessed functional connectivity using the instantaneous term of Granger causality mapping (GCM).
Results: Subjects’ target detection rate was approximately 95%. For each visual quadrant, our GLM yielded clear circumscribed activations at the stimulus position across early visual areas. During target trials widespread activity was observed in a network comprising ventrolateral and dorsolateral prefrontal cortex (DLPFC), supplementary motor area and the lower right visual quadrant. In addition, GCM revealed consistent patterns of functional connectivity between bilateral anterior insula, somatosensory regions, right motor cortex and each visual quadrant during target presentation at that quadrant. Moreover, GCM revealed asymmetries regarding connectivity patterns in the form of a rightward bias. Here, the strongest connectivity pattern was observed between higher-order target detection networks like the DLPFC and temporoparietal junction and regions processing stimuli in the lower right visual quadrant.
Discussion: Our results confirm the central involvement of the insula in modulating attentional control and its close interactions with early visual areas. Furthermore, we found tentative evidence for differences regarding interactions between visual hemifields and higher-order cognitive areas in the form of a rightward bias. Our paradigm appears to be suitable to study interactions between early visual areas and regions supporting higher-order cognitive processes as well as disturbances of these interactions in neuropsychiatric disorders such as schizophrenia and autism spectrum disorder. Simultaneously, our paradigm functions as a time-efficient localizer to map regions in early visual areas for the analysis of more complex cognitive tasks using machine-learning algorithms.
Methods: 50 healthy participants underwent fMRI in a 3 T Siemens Trio scanner. A target-detection task consisting of a series of flickering black-and-white-colored round shaped checkerboard stimuli subtending 0.3 degrees of visual angle was employed. Every two seconds, the checkerboards position switched among one of the possible quadrants. Each of the positions were located at equidistant points on an imaginary circle surrounding a central fixation cross. Thus, checkerboards appeared at homologous positions in each visual quadrant. The whole array subtended 2.4 degrees of visual angle. At random intervals, two centrally located squares of the checkerboards changed their color to yellow (target trials). Participants had to indicate by button press the detection of these targets. Throughout the task participants had to fixate a black, x-shaped fixation cross at the center of the screen. Overall, task duration was 5.5 minutes. Data analysis in BrainVoyager included standard data pre-processing. Additionally, we used a multiscale curvature driven cortex-based alignment procedure mitigating interindividual macro-anatomical correspondence. Functional data were analyzed using a random-effects multi-subject general linear model (GLM). We assessed functional connectivity using the instantaneous term of Granger causality mapping (GCM).
Results: Subjects’ target detection rate was approximately 95%. For each visual quadrant, our GLM yielded clear circumscribed activations at the stimulus position across early visual areas. During target trials widespread activity was observed in a network comprising ventrolateral and dorsolateral prefrontal cortex (DLPFC), supplementary motor area and the lower right visual quadrant. In addition, GCM revealed consistent patterns of functional connectivity between bilateral anterior insula, somatosensory regions, right motor cortex and each visual quadrant during target presentation at that quadrant. Moreover, GCM revealed asymmetries regarding connectivity patterns in the form of a rightward bias. Here, the strongest connectivity pattern was observed between higher-order target detection networks like the DLPFC and temporoparietal junction and regions processing stimuli in the lower right visual quadrant.
Discussion: Our results confirm the central involvement of the insula in modulating attentional control and its close interactions with early visual areas. Furthermore, we found tentative evidence for differences regarding interactions between visual hemifields and higher-order cognitive areas in the form of a rightward bias. Our paradigm appears to be suitable to study interactions between early visual areas and regions supporting higher-order cognitive processes as well as disturbances of these interactions in neuropsychiatric disorders such as schizophrenia and autism spectrum disorder. Simultaneously, our paradigm functions as a time-efficient localizer to map regions in early visual areas for the analysis of more complex cognitive tasks using machine-learning algorithms.
| Original language | English |
|---|---|
| Article number | 100278 |
| Journal | Neuroscience Applied |
| Volume | 1 |
| Issue number | Supplement 2 |
| DOIs | |
| Publication status | Published - 24 Dec 2022 |