Abstract
We have previously demonstrated a horizontal shift advantage for visual object-based attention (OBA; Barnas & Greenberg, 2016, 2019). Furthermore, we demonstrated that this shift direction anisotropy (SDA) is modulated by whether targets cross the visual field meridians. We hypothesized that this was due to greater OBA horizontal shift efficiency of neural mechanisms during interhemispheric (horizontal) versus intrahemispheric (vertical) orienting. We used fMRI to investigate this hypothesis in early visual cortex. Subjects completed a target detection task as described by Barnas and Greenberg (2016, 2019), briefly described below. A single ‘L’-shaped object appeared with its vertex positioned in either the upper-left or lower-right quadrant, and a central arrow cue always pointed to the object’s vertex (60% valid, 20% invalid, 20% catch). A target then appeared at one of five possible locations. On valid trials, the target (“T”) appeared at the object’s vertex. On invalid trials, the target appeared horizontally (invalid-horizontal), or vertically (invalid-vertical), displaced from the vertex by either a long (beyond the meridian) or short (nearer than the meridian) distance. Non-targets (“L”) served as placeholders at locations not occupied by the target. Analysis of behavioral data revealed a significant SDA within a 95% confidence interval for the upper-left object but not the lower-right object, suggesting that the SDA was modulated by object location. Additionally, we found a significant correlation (p<0.001) between changes in horizontal efficiency and SDA over the course of the experiment, suggesting that horizontal efficiency influenced the SDA. Neural correlates subserving this behavior were derived from event-related averages and were largely in agreement with behavioral results. Taken together, these results suggest that changes in horizontal efficiency modulated the SDA effect, indicating that interhemispheric shifts of OBA were supported by attentional resources from both hemispheres.