Electrophysiological Explorations of the Cause and Effect of Inhibition of Return in a Cue–Target Paradigm

Abstract

Facilitation and inhibition of return (IOR) are, respectively, faster and slower responses to a peripherally cued target. In a spatially uninformative peripheral cueing task, facilitation is normally observed when the interval between the cue and target stimulus, the stimulus onset asynchrony (SOA), is shorter than 250 ms, while IOR is normally observed when an SOA greater than 250 ms is used. Since Posner and Cohen’s (Attention and performance X, 1984) seminal study, IOR has become an actively investigated component of orienting. In this study, using ERPs and the source localization algorithm, LORETA, we seek to examine the brain mechanisms involved in IOR by localizing the different stages of processing after the appearance of a cue that captures attention exogenously. Unlike previous ERP investigations of IOR, this study analyzes the neural activity (via EEG) produced in response to the cue, prior to the appearance of the target. Neural activations were approximately divided into three stages. In the early stage (110–240 ms), involved activations are in the prefrontal cortex, the bilateral intraparietal cortex, and the contralateral occipito-temporal cortex. In the middle stage (240–350 ms), activations are primarily found in the frontal cortex and the parietal cortex. In the late stage (350–650 ms), the main activations are in the occipito-parietal cortex, but unlike in the early stage, the activation areas have shifted to the hemisphere ipsilateral to the cued location. These findings indicate that IOR is related to both attentional and motor response processes and suggest that the time course of initial facilitation and IOR is concurrent and mediated by two neural networks. Building upon our results, electrophysiological, electroencephalographic, and behavioral results in the literature and extending previous spatial theories of IOR, we propose here a spatio-temporal theory of IOR based upon post-cue dynamics.

Appendix

Calculation of the Amplitude Weight Centre (AWC)

In order to compare the differences among the scalp topologies of different moments or different subjects, we utilized the amplitude weight centre (AWC) technique (Yao et al. 2005). In this paper, we define a local AWC of an electrode with its N neighbor electrodes to characterize the topology distribution in the region of interest (ROI). In the current work, ROI is defined as the scalp area enclosing the electrodes with the positive/negative maximum potential.

The local AWC at moment t is defined as follows:

$$ X_{C} \left( t \right) = {\frac{{\sum\nolimits_{n = 1}^{N} {x_{n} A_{n} \left( t \right)} }}{{\sum\nolimits_{n = 1}^{N} {A_{n} \left( t \right)} }}},\quad Y_{C} \left( t \right) = {\frac{{\sum\nolimits_{n = 1}^{N} {y_{n} A_{n} \left( t \right)} }}{{\sum\nolimits_{n = 1}^{N} {A_{n} \left( t \right)} }}},\quad Z_{C} \left( t \right) = {\frac{{\sum\nolimits_{n = 1}^{N} {z_{n} A_{n} \left( t \right)} }}{{\sum\nolimits_{n = 1}^{N} {A_{n} \left( t \right)} }}} $$

(1)

where A _n (t) is the potential at the nth electrode in the ROI ($x_{n} ,y_{n} ,z_{n}$) represents the coordinates of the nth electrode, and N is the number of electrodes consisting of the ROI (in this work, N = 11).