Inhibition of return is at the midpoint of simultaneous cues

Abstract

When multiple cues are presented simultaneously, Klein, Christie, and Morris (Psychonomic Bulletin & Review 12:295-300, 2005) found a gradient of inhibition (of return, IOR), with the slowest simple manual detection responses occurring to targets in the direction of the center of gravity of the cues. Here, we explored the possibility of extending this finding to the saccade response modality, using methods of data analysis that allowed us to consider the relative contributions of the distance from the target to the center of gravity of the array of cues and the nearest element in the cue array. We discovered that the bulk of the IOR effect with multiple cues, in both the previous and present studies, can be explained by the distance between the target and the center of gravity of the cue array. The present results are consistent with the proposal advanced by Klein et al., (2005) suggesting that this IOR effect is due to population coding in the oculomotor pathways (e.g., the superior colliculus) driving the eye movement system toward the center of gravity of the cued array.

Multiple cueing dissociates location- and feature-based repetition effects

“…In multiple-cue IOR studies, more than one initial cue is presented (thus, cueing more than one location), and IOR is measured at multiple locations. Prior studies have shown that if the multiple cue locations are distributed evenly around a midpoint, location-based IOR is not observed, consistent with the idea that inhibitory effects are spread evenly (Christie, Hilchey & Klein, 2013; Klein, Christie & Morris., 2005). In the current study, we test whether these conditions also eliminate feature-based inhibition.…”

“…Recently, Christie, Hilchey and Klein (2013) continued this line of argument, and proposed that IOR occurs at the center of gravity of simultaneously presented cues. Collectively, the results from Klein et al (2005) and Christie et al (2013) demonstrated that if the center of gravity of the cue array was at the fixation point, reaction times to targets at peripherally cued and uncued locations were statistically indistinguishable. In other words, there is no location-based IOR when the center of gravity of a multiple-cue event is at the fixation point.…”

“…Based on Klein et al (2005) and Christie et al (2013), we should find that simultaneous multiple cueing, with the cues centered on fixation, will eliminate location-based IOR in these displays. Our central question is whether this manipulation will similarly eliminate the feature-based inhibitory effect, or if instead the two types of inhibition will dissociate, with feature-based inhibition remaining in the absence of location-based IOR.…”

“…For instance, researchers have extended the IOR effect to different dependent variables and tasks (e.g., Posner & Cohen, 1984: manual keypress; Abrams & Dobkein, 1994, Christie et al, 2013, and Taylor & Klein, 1998: eye movement latency; Maylor, 1985 and Pratt, 1995: target detection and localization; Hu & Samuel, 2011, and Lupiáñez et al, 1997: target discrimination). There have also been a number of studies that demonstrate the inhibitory effect under conditions that do not involve a withdrawal of attention, suggesting that attentional allocation and removal are not necessary (e.g., Berlucchi, Chelazzi, & Tassinari, 2000; Rafal, Davies, & Lauder, 2006).…”

Returning to “inhibition of return” by dissociating long-term oculomotor IOR from short-term sensory adaptation and other nonoculomotor “inhibitory” cueing effects.

“…Despite this, because the ICE was longlasting and, following an eye movement, was represented in an environmental rather than retinotopic reference frame, Posner and Cohen also reasoned that this ICE plays a "role in directing future covert and overt" (p. 550) orienting and that it "evolved to max imize sampling of the visual environment" (p. 550). These attri butions are represented in the left-most column of Table 1. Posner and colleagues (Posner, Rafal, Choate, & Vaughan, 1985) later demonstrated that a degenerative neurological condi tion (progressive nuclear palsy) impairing normal oculomotor 1 A finding that was not replicated by Maylor (1985) and that has been called into question by the finding that ICEs often appear along the average vector or near the midpoint (or "center of gravity") of simultaneously presented visual inputs (Klein, Christie, & Morris, 2005;Langley, Gayzur, Saville, Morlock, & Bagne, 2011;Christie, Hilchey, & Klein, 2013). Table 1 The Defining concepts Posner et al (1985) Any visual stimulus Cause Oculomotor activation Reduces efficiency of target detection Effect Motor bias…”