Right Parietal rTMS Induces Bidirectional Effects of Selective Attention upon Object Integration

Abstract

Background/Objectives: Part-to-whole object completion and search guidance by salient, integrated objects has been proposed to require attentional resources, as shown by studies of neglect patients suffering from right-parietal brain damage. The current study was performed to provide further causal evidence for the link between attention and object integration. Methods: Healthy observers detected targets in the left and/or right hemifields, and these targets were in turn embedded in various Kanizsa-type configurations that systematically varied in the extent to which individual items could be integrated into a complete, whole object. Moreover, repetitive transcranial magnetic stimulation (rTMS) was applied over the right intraparietal sulcus (IPS) and compared to both active and passive baseline conditions. Results: The results showed that target detection was substantially facilitated when the to-be detected item(s) were fully embedded in a salient, grouped Kanizsa figure, either a unilateral triangle or a bilateral diamond. However, object groupings in one hemifield did not facilitate target detection to the same extent when there were bilateral targets, one inside the (triangle) grouping and the other outside of the grouped object. These results extend previous findings from neglect patients. Moreover, a subgroup of observers was found to be particularly sensitive to IPS stimulation, revealing neglect-like extinction behavior with the single-hemifield triangle groupings and bilateral targets. Conversely, a second subgroup showed the opposite effect, namely an overall, IPS-dependent improvement in performance. Conclusions: These explorative analyses show that the parietal cortex, in particular IPS, seems to modulate the processing of object groupings by up- and downregulating the deployment of attention to spatial regions were to-be-grouped items necessitate attentional resources for object completion.

Keywords: intraparietal sulcus; object integration; parietal cortex; perceptual grouping; rTMS; visual attention; visual extinction.

Figure 1

(A) Example trial sequence. First, a fixation cross was shown for 1000 ms, followed by a premask display presented for 2000 ms. Next, participants saw a Kanizsa-type configuration which was shown for 150 ms with quarter segments removed from the top and bottom, and from either the left side (unilateral left), the right side (unilateral right), both sides (bilateral), or no side (catch), as depicted in the example search displays from top to bottom, respectively. Finally, a postmask display (with nine big and four small disks arranged in random orientation) was presented until a response was given. In the example trial sequence, search displays present possible variants of a diamond configuration. (B) Examples of the four different types of object groupings presented in bilateral trials (i.e., displays containing target cut-out segments in both hemifields): In the diamond configuration, a complete illusory figure spanning across both hemifields was visible (right panel). The right triangle condition (middle-right panel) presented an illusory triangle confined to only the right hemifield, and in the left triangle condition (middle-left panel) an illusory triangle emerged in only the left hemifield. The ungrouped configuration (left panel), which did not lead to the emergence of any illusory figure, served as a baseline.

Figure 2

Individual, MRI-guided TMS target sites for all participants. (A) Target site in the IPS (shown in red in the figures) with the associated mean MNI coordinates [20, −69, 44]. (B) Target site M1 (shown in green in the figures) as used in the active baseline condition, with the associated mean MNI coordinates [41, −10, 51].

Figure 3

Mean percentages of correct detections (with associated within-subject 95% confidence intervals) as a function of object configuration (ungrouped, left triangle, right triangle, diamond) for (A) unilateral left, (B) unilateral right, and (C) bilateral targets.

Figure 4

Mean percentages of correct detections (and associated 95% confidence intervals) as a function of object configuration (ungrouped, triangle, diamond) in the IPS and M1 TMS stimulation conditions. The results are depicted for the “IPS-cost” (A) and “IPS-benefit” (B) subgroups for bilateral target displays (left panels) and for unilateral target displays (right panels).