Spatial Inhibition of Return Affected by Self-Prioritization Effect in Three-Dimensional Space

Abstract

Spatial inhibition of return (IOR) being affected by the self-prioritization effect (SPE) in a two-dimensional plane has been well documented. However, it remains unknown how the spatial IOR interacts with the SPE in three-dimensional (3D) space. By constructing a virtual 3D environment, Posner’s classically two-dimensional cue-target paradigm was applied to a 3D space. Participants first associated labels for themselves, their best friends, and strangers with geometric shapes in a shape-label matching task, then performed Experiment 1 (referential information appeared as the cue label) and Experiment 2 (referential information appeared as the target label) to investigate whether the IOR effect could be influenced by the SPE in 3D space. This study showed that when the cue was temporarily established with a self-referential shape and appeared in far space, the IOR effect was the smallest. When the target was temporarily established with a self-referential shape and appeared in near space, the IOR effect disappeared. This study suggests that the IOR effect was affected by the SPE when attention was oriented or reoriented in 3D space and that the IOR effect disappeared or decreased when affected by the SPE in 3D space.

Keywords

three-dimensional space self-prioritization effect attention orientation/reorientation inhibition of return

Introduction

To avoid constantly checking previously attended locations, the attention system tends to inhibit attention returning to the original location, facilitate attention toward a novel spatial location, and then respond more slowly and less accurately to target stimuli at previously attended spatial locations (Klein, 2000; Posner & Cohen, 1984). This phenomenon is identified as a special visual search promotion mechanism known as inhibition of return (IOR; Klein & MacInnes, 1999; Roggeveen et al., 2005; Z. G. Wang et al., 2010). Posner and Cohen (1984) first found the IOR effect in their research, whereby detection latencies for visual targets on the same side as the preceding cue were slower than those on the opposite side, after approximately 300 milliseconds had elapsed since the cue (Posner & Cohen, 1984).

This effect has been extensively studied in the two-dimensional (2D) plane, as in the early characterization of the IOR effect, the reasons for the IOR effect, and the neural basis of the IOR effect (Dodd et al., 2003; Eng et al., 2018; Hilchey et al., 2014; Klein, 2000; Martín-Arévalo et al., 2013; Maylor & Hockey, 1985; Posner & Cohen, 1984; Taylor & Klein, 2000). However, we live in a Three-dimensional (3D) space and use binocular and monocular cues, and the human brain can reconstruct a 3D space from 2D retinal images. This 3D space we live in is characterized by depth (Durand et al., 2007; Georgieva et al., 2009; Sereno et al., 2002). Attention needs to be oriented or reoriented to a specific depth location in real 3D space during daily activities. As a result, the IOR effect, which is affected by attention orientation/reorientation, has begun to be studied in 3D space (Bourke et al., 2006; Theeuwes & Pratt, 2003; A. J. Wang, Li, et al., 2015; A. J. Wang et al., 2016; A. J. Wang, Yue, et al., 2015; A. J. Wang & Zhang, 2015). Different from the 2D plane, studies of 3D space focus on depth in the real world.

Our recent studies in 3D space showed that attention orientation/reorientation from near to far locations was different from that from far to near locations, and the spatial IOR effect was different when the target appeared in near space versus far space (A. J. Wang, Li, et al., 2015; A. J. Wang et al., 2016, 2017; A. J. Wang, Yue, et al., 2015). These results agreed with the attentional gradient theory and the theory of view-centered spatial representation. According to attentional gradient theory, attentional intensity radiates from the focus (Gawryszewski et al., 1987), and the theory of view-centered spatial representation believes that the distribution of attention resources decreases as the distance from the observer’s nearest position increases (Andersen, 1990; Andersen & Kramer, 1993). The reaction time/response time (RT) increases from near space to far space. In addition, the results also agreed with an ecological view because the distance in the depth of a potentially threatening or rewarding object relative to the observer is crucial for its evaluation in the real world; some objects that suddenly appear around us demand an instantaneous shift in attention toward them to guarantee survival, but those far away from us are generally not noticeable (Chen et al., 2012; Gawryszewski et al., 1987; Graziano & Cooke, 2006; Previc, 1998). Therefore, under the condition (invalid) that the cue and target are in different positions, when the target appears unexpectedly in near space, participants have to reorient their attention from far space to near space, which accelerates their responses and increases the IOR; when the target appears unexpectedly in far space, participants have to reorient their attention from near space to far space, compared with the attention required for a target in near space. Also, far space has no advantage in visual processing, which causes a slower reaction and a smaller IOR (A. J. Wang et al., 2016).

In addition to differences in the processing of visual information in 3D space, individuals also have a bias for processing self-referential information; they respond faster and more accurately to self-referential stimuli, which is known as the self-prioritization effect (SPE) or self-referential bias (Gao et al., 2020; Rogers et al., 1977; Sui & Gu, 2017; Sui et al., 2012; Truong et al., 2017). In previous studies of self-referential information and attentional orientation/reorientation, researchers found that self-referential cues accelerate an individual’s response to target stimuli appearing lateral to the cue, that is, they cause a faster redirection of attention to the previously searched location, which leads to a reduction in the IOR (Sui et al., 2009; Zhang & Wu, 2013). Chen et al. (2012) also showed that attentional reorientation in 3D space may involve self-referential bias with objects that are unexpectedly close to the observer having higher self-relevance than objects that are far from the observer, and self-referential targets can be well processed in near space. Moreover, when objects are unexpectedly present in near space, the default-mode network of the human brain, including the posterior cingulate cortex (PCC), orbital prefrontal cortex (OPFC), and left angular gyrus (AG), will be activated, which has been shown to exhibit enhanced neural activity when a person is assessing self-referential information (Chen et al., 2012; Corbetta et al., 2008; Fox et al., 2005; Gusnard & Raichle, 2001; Qin & Northoff, 2011; Raichle et al., 2001).

Therefore, the differences in attention orientation/reorientation from near to far locations and from far to near locations in 3D space are perhaps self-related, and the spatial effects of attention orientation/reorientation at different spaces may also be self-related. Then, when attention orients or reorients to the target that is self-related in the 3D space, the IOR effect may be effected by the self-target.

Previous studies have investigated how self-referential information alters voluntary attention orientation in 2D space (Liu et al., 2012; Sui et al., 2009). Similar to the Posner exogenous cueing paradigm, self-referential information was used as an endogenous cue to shift participants’ spatial attention to target location in the left or right visual field. Researchers found that self-referential information could affect the efficiency of the cue in guiding endogenous attention: Self-referential cues were more efficient in capturing reflexive attention at the early stage of perceptual processing and shifting voluntary attention at the later stage of perceptual processing (Sui et al., 2009). Moreover, Sui et al. (2009) used temporarily established self-referential shapes to avoid the confounding impact of familiarity for self-face stimuli or names. Then, Sui et al. (2012) further developed this novel associative learning approach, having participants form new associations between geometric shapes and labels referring to the self, a familiar other, or an unfamiliar other. Participants then had to judge whether shape–label pairs matched the originally presented result (e.g., square–friend, circle–stranger, triangle–self; Sui et al., 2012). The associative learning paradigm was subsequently extended to the discrimination task, and by using the discrimination task, Sui, Liu, et al. (2013) were also able to successfully establish associations between labels and figures (Sui, Liu, et al., 2013). H. X. Wang et al. (2016) also used the discrimination task. They believed that the discrimination task can maximize the number of matching trials as no mismatch trials were presented, thus creating a link between labels and figures in a few trials (H. X. Wang, et al., 2016). Furthermore, Zhao et al. (2015) used this associative learning approach, in which the self-referential information was used as a nonpredictive cue, and found that reflexive attention was also modulated by self-referential information. Compared with other-referential information, the participants were more accurate and made faster responses to self-referential information. This phenomenon is referred to as the SPE (Sui & Gu, 2017; Sui et al., 2012). Self-referential information regulated the facilitation effect in the facilitated stage as well as the inhibitory effect in the inhibition stage (Qian et al., 2016; Zhang & Wu, 2013). A previous study used a paradigm similar to that used by Sui et al. (2012) in which participants formed new associations between a color and the self or a familiar other. The results showed that the IOR effect did not appear in the temporarily established self-referential exogenous cue condition (Zhang & Wu, 2013). Similarly, Qian et al. (2016) used the self-face or an unfamiliar face as an endogenous cue and found that the IOR effect decreased significantly. Then, it was considered that when the self-face or temporarily established self-referential cue was used as an exogenous cue, the IOR effect may decrease or disappear, perhaps due to the SPE.

Affected by the differences in attention orientation/reorientation in 3D space, the spatial IOR effect was different when the target appeared in near space versus far space (A. J. Wang, Li, et al., 2015; A. J. Wang et al., 2016, 2017; A. J. Wang, Yue, et al., 2015). More importantly, previous studies found that attention orientation/reorientation in depth perhaps relates to self-relevancy processing in the following manner: Objects that unexpectedly approach the observer have higher self-relevancy, whereas unexpected objects farther away involve fewer self-referential thoughts (Chen et al., 2012; Corbetta et al., 2008; Fox et al., 2005; Gusnard & Raichle, 2001; Qin & Northoff, 2011; Raichle et al., 2001). Furthermore, they showed that ecologically important behavior specifically engaged a network of areas, including the PCC, OPFC, and left AG; also, neural mechanisms of the network would be activated only when the target unexpectedly appeared near the observer, which could accelerate the process of attention orientation/reorientation (Chen et al., 2012; Corbetta et al., 2008; Fox et al., 2005; Gusnard & Raichle, 2001; Qin & Northoff, 2011; Raichle et al., 2001). Some studies also indicated that in the processing of near and far spaces, there were processing advantages in near space (Cardin & Smith, 2011; Chen et al., 2012; Quinlan & Culham, 2007; A. J. Wang et al., 2016). Therefore, we assumed that there would be an interaction between the SPE and 3D spatial attention orientation/reorientation.

In addition, some studies have found that the IOR effect is influenced by biologically meaningful stimuli (Chao, 2010; Gobel et al., 2017; Stoyanova et al., 2007). Due to the different attentional processing mechanisms elicited by biologically meaningful stimuli as cues and targets, previous studies have used stimuli as cues and targets (Berdica et al., 2017; Jia et al., 2019; Qian et al., 2016; Zhang & Wu, 2013). When using a biologically meaningful stimulus as a cue, the stimulus is able to induce attention and keep it at the cued location.

Self-referential information first appeared as a cue predicting the target location in Experiment 1, and we assumed that as the processing of stimuli appearing in far space would not be advantageous (Cardin & Smith, 2011; Quinlan & Culham, 2007), the SPE can work to accelerate the response to the target at the self-referential label (valid condition), thereby reducing the IOR. Because of the advantage of processing in near space (Cardin & Smith, 2011; Quinlan & Culham, 2007), the processing of all stimuli that appear in near space is faster; thus, the SPE of self-referential cues in near space is no longer obvious, and self-referential cues do not reduce the IOR compared with the IOR of other labels.

The manipulation of target features may be a more meaningful approach than the manipulation of cue features because the task requires the participant to respond to the target, and the effect of biologically meaningful stimuli on the IOR will be reflected in the target processing response (J. X. Wang et al., 2013). Therefore, in Experiment 2, we placed biologically meaningful stimuli at the target location so that we could reveal its interaction with the IOR. We assumed that because the processing of far-space targets does not involve self-relevant processing (Chen et al., 2012; Corbetta et al., 2008; Fox et al., 2005; Gusnard & Raichle, 2001; Qin & Northoff, 2011; Raichle et al., 2001), the SPE produced by far-space targets would not be better at accelerating responses to inhibitory (valid) locations than noninhibitory (invalid) locations, thereby increasing the IOR. We also assumed that because the processing of near-space targets involves self-relevant processing, self-referential targets can be well processed in near space (Chen et al., 2012; Corbetta et al., 2008; Fox et al., 2005; Gusnard & Raichle, 2001; Qin & Northoff, 2011; Raichle et al., 2001), so the resulting SPE is well-equipped to counteract the IOR and better accelerates responses at inhibitory (valid) locations compared with noninhibitory (invalid) locations, thereby reducing the IOR.

Experiment 1: Self-Referential Cues Affect the IOR Effect

Method

Participants

On one hand, the sample size was based on previous studies of the IOR in 3D (Chen et al., 2012; A. J. Wang et al., 2016; A. J. Wang, Yue, et al., 2015); on the other hand, G*Power 3.1.9.2 (Faul et al., 2007) was used to estimate the sample size for a one-way repeated measures analysis of variance (ANOVA) with three within-subject measurements (estimated effect size f = 0.3, alpha = .05, power = 0.8), suggesting a sample of 20 participants. Twenty-three subjects (13 females) participated in Experiment 1. The participants were between 18 and 22 years old (M = 19.1, standard deviation [SD] = 2.20). All participants were right-handed, with normal vision or corrected-to-normal vision and color vision. None of the participants had a history of neurological or psychiatric disorders, and they each provided written informed consent before participating. After the experiment, all participants were paid according to the Helsinki Declaration. The study was approved by the Academic Committee of the Department of Psychology, Suzhou University, China.

Apparatus, Stimuli, and Experimental Setup

In Experiment 1, all stimuli were presented on 27-in. screen. The ASUS 3D monitor was driven by the NVIDIA GeForce FX 5200 graphics card. A pair of NVIDIA 3D shutter glasses synchronized with the monitor provided two independent stereoscopic displays for each eye with a resolution of 800 (Horizontal) ×600 (Vertical) pixels and a refresh rate of 60 Hz. All 3D objects were presented on a black background customized by computer scripts (demonstration packages; Neurobehavioral systems, Albany, California).

In the first phase, geometric shapes (3.8° × 3.8°) were presented above a fixed cross on a white background (0.8° × 0.8°). Text (names of self, friends, or strangers) was displayed below the fixation cross. The shape center was 3.2° from the fixed point, and the text was randomly assigned to a set location (left, middle, or right) for each participant. All stimuli were shown on a gray background.

In the second phase, the default virtual display consisted of three white placeholders in near (sagittal) space, three white placeholders in mid space, and three white placeholders in far space; therefore, nine spatial locations were marked in the virtual 3D space (see Figure 2). The 3D objects in far space were displayed behind the monitor screen, while the 3D objects in near space pop up from the monitor screen. The objects in mid space were equal to those in near and far spaces and appeared to be on the surface of the screen. The binocular disparity in far and near spaces was ±52.40 minutes of arc relative to the fusion plane (zero disparity), which presented the central location of mid space. Different target distances were simulated by adjusting the binocular disparity. Participants reported that when they fixed the central location in mid space, they could clearly perceive far space and near space. The distance between the retinal space locations matched far and near spaces, and the horizontal space distance (along the x-axis) between the left and right locations in far and near spaces matched the angle of view (16.85°). The 3D objects in the two spaces had the same perceptual size, which resulted in slightly different retinal sizes of near and far stimuli: near and far stimuli were 1.98° and 1.52°, respectively. To avoid distal objects from being occluded by proximal objects, the stimuli in far and near spaces moved slightly, making the vertical distance between far and near objects on the retina 2.36° of the view. The target was a blue sphere. The cue was one of the three geometric shapes (triangle, square, or circle), and the central cue consisted of transient flashes of one of the central placeholders.

Experimental Procedures and Design

There were two phases of this experiment. Participants had to first learn shape-person label associations and then perform an association matching task (in Phase 1); then, the second phase was a spatial cue task (in Phase 2) that used learned shapes from Phase 1.

In the first phase, shapes were randomly assigned to three person labels (self, friend, and stranger). The order of the tasks was counterbalanced across participants. For example, participants could be told that a triangle represented themselves, a square represented their best gender-matched friends (named by participants before the experiment), and a circle represented a gender-matched stranger (selected by participants from a common list of names). After the association instruction, the participants had to select which of the three labels matches the shape on the monitor. In this task, the shape appeared above the fixation point, the three labels were displayed later, the order of their positions was changed randomly from one trial to the next to avoid a mental set, and the participants had to press one of the three keys to select the label that matched the shape. After six consecutive correct responses for shapes, the task terminated. After completing the association matching task, participants performed the spatial cue task in the second phrase (cited from Sui, Liu, et al., 2013; H. X. Wang et al., 2016; see Figure 1).

Figure 1.

Participants Are Instructed to Associate Three Geometric Shapes With Three People and Then Complete a Shape-Label Matching Task, Choosing Which Label Matches the Shape (Quoted and Adapted From ; Sui & Humphreys, 2017; Sui, Liu, et al., 2013; H. X. Wang et al., 2016).

In the second phase, participants needed to complete a detection task. At the beginning of each trial, a peripheral location in near or far space was prompted for 300 milliseconds, followed by a 200 milliseconds interstimulus interval. Subsequently, in the same hemisphere, a location in mid space was prompted for 300 milliseconds to attract the attention from the previous location. After another period of 150 milliseconds, the target was presented for 450 milliseconds at one of the same hemispace peripheral locations in either the valid or the invalid space, with equal probabilities (see Figure 2). These cues were valid in 50% of the trials. Throughout the experiment, participants were required to fixate on a central fixed location on mid space (zero disparity) and respond to the target as quickly and accurately as possible by pressing a button on the keyboard (PXN-9603) with their right index finger. Before the formal test, each participant had a practice run in which the composition of the trial types were the same as those in the formal experiment.

Figure 2.

Front View of Exemplary Trials in the Experimental Paradigm. The default visual scene consisted of nine white placeholders (three in near space, three in mid space, and three in far space). The spatial distances between the three occupants on near, mid, and far spaces had the same visual angle, and the same 3D objects were displayed in all three spaces. All participants reported seeing objects of the same size in three spaces because of the size–distance constancy effect (Boring, 1964). To prevent distal objects from being occluded by proximal objects, the vertical distance between the locations of near and far spaces was slightly shifted by 2.36°. Geometric shapes represented the cues, and blue balls represented the targets. ISI = interstimulus interval.

Figure 3.

IOR Size, Shown as a Function of the Experimental Conditions in Experiment 1. The error bars represent standared errors of the mean. IOR = inhibition of return; n.s. = not statistically significant.*p<.05. **p<.01.

Figure 4.

IOR Size, Shown as a Function of the Experimental Conditions in Experiment 2. The error bars represent standared errors of the mean. IOR = inhibition of return; n.s. = not statistically significant.*p<.05.

Therefore, in Experiment 1, there was a 2 (Cue Depth [near vs. far]) × 3 (Cue Label [self vs. friend vs. stranger]) × 2 (Cue Validity [valid vs. invalid]) design, resulting in 12 experimental conditions in total. The experiment consisted of 900 trials, of which 864 (72 in each case) were experimental trials and 36 were catch trials (in which there were only cues and no targets to prevent participants from reacting before the target appeared), and lasted for approximately 60 minutes.

Results

Reaction Times

Incorrect answers (miss trials) and RTs exceeding three SDs were discarded. The errors in all conditions were less than 1% of trials, and no further analysis was performed. Table 1 summarizes the average RTs and standard errors in all conditions in Experiment 1. Resulting in 12 experimental conditions, correct RTs were analyzed with a 2 (Cue Depth [near vs. far]) × 3 (Cue Label [self vs. friend vs. stranger]) × 2 (Cue Validity [valid vs. invalid]) repeated measures ANOVA. The results showed that the main effect of cue validity was significant, F(1, 22) = 108.67, p < .001, η² = .83, suggesting that RTs to the target at the valid location (356 milliseconds) were longer than RTs to the target at the invalid location (328 milliseconds). However, the main effect of cue depth was not significant, F(1, 22) = 2.08, p > .05; the main effect of cue label was not significant, F(2, 21) = 0.17, p > .05. The interaction between cue depth and cue validity was significant, F(1, 22) = 6.35, p < .05, η² = .22. No other significant effects of the two- or three-factor interactions were found; all ps > .05.

Table 1.

Mean Reaction Times (RT) and Standard Errors (SEs) of the Means in the Experiment 1 (Presented in the Format RT ± SE).

	Far cue			Near cue
	Valid	Invalid	Valid–invalid	Valid	Invalid	Valid–invalid
Self	348 ± 22	330 ± 22	18***	360 ± 21	327 ± 22	33***
Friend	352 ± 22	330 ± 22	22***	359 ± 21	327 ± 23	32***
Stranger	356 ± 22	327 ± 22	29***	360 ± 22	324 ± 23	35***

***p < .001.

To further test the potential interaction between cue validity and cue label, a 2 (Cue Validity [valid vs. invalid]) × 3 (Cue Label [friend vs. self vs. stranger]) repeated measures ANOVA was conducted. The main effect of cue validity on cues in far space was significant. F(1, 22) = 41.71, p < .001, η² = .66, indicating that RTs to the target in the valid location (352 milliseconds) are longer than those to the target in the invalid location (329 milliseconds). The main effect of cue label was not significant, F(2, 21) = 1.25, p > .05. Furthermore, the interaction between cue validity and cue label was also significant, F(2, 21) = 5.13, p < .05, η² = .33. Simple effect analysis suggested that RTs to the target in the valid location (348 milliseconds) were significantly longer than RTs to the target in the invalid location (330 milliseconds) when the cue label was self, t(22) = 5.06, p < .001. The RTs to the target in the valid location (352 milliseconds) were significantly longer than RTs to the target in the invalid location (330 milliseconds) when the cue label was friend, t(22) = 5.27, p < .001. The RTs to the target in the valid location (356 milliseconds) were significantly longer than RTs to the target in the invalid location (327 milliseconds) when the cue label was stranger, t(22) = 7.04, p < .001. For cues in near space, the main effect of cue validity was significant, F(1, 22) = 104.88, p < .001, η² = .83, suggesting that RTs to the target in the valid location (360 milliseconds) were longer than RTs to the target in the invalid location (326 milliseconds). The main effect of cue label was not significant, F(2, 21) = 0.54, p > .05. Furthermore, the interaction between cue validity and cue label was not significant, F(2, 21) = 0.36, p > .05.

IOR Effect Size

To further explore the potential relation between cue label and the IOR effect, we compared the IOR effect size (RTs for valid minus invalid locations) for cue labels. As Experiment 1 investigated the effect of self-referential stimuli used as cues on the IOR in 3D space, we needed to control for cues being presented in far and near spaces. The IOR effect size in far space = RT in the valid condition (cuing locations in far space and targets appearing at the same locations) − RT in the invalid condition (cuing locations in far space and targets appearing at the locations in near space); the IOR effect size in near space = RT in the valid condition (cuing locations in near space and targets appearing at the same locations) − RT in the invalid condition (cuing locations in near space and targets appearing at the locations in far space). When the cue was in far space, the IOR effect size for the self-referential cue (18 milliseconds) was significantly smaller than that for the stranger-referential cue (29 milliseconds), t(22) = 3.29, p < .005, and the IOR effect size was significantly smaller for the friend-referential cue (22 milliseconds) than for the stranger-referential cue (29 milliseconds), t(22) = 2.14, p < .05. However, there was no significant difference between the IOR effect size for the self-referential cue and the friend-referential cue, t(22) = 1.77 p > .05. When the cue was in near space, there were no significant differences among the cue labels; all ps > .05 (see Figure 3).

The result suggests that self-referential cues presented in far space can produce the SPE that accelerates responses to cued locations, thereby reducing the IOR, which is consistent with previous 2D research, that is, self-referential cues can continually draw covert attention back to previous locations (Qian et al., 2016; Zhang & Wu, 2013), which can reduce the IOR. The self-referential cues presented in near space did not reflect the SPE, which may be because of the advantage of processing in near space (Cardin & Smith, 2011; Quinlan & Culham, 2007).

In addition to being influenced by cue properties, the IOR will also be influenced by target properties, and the effect of biologically meaningful stimuli on the IOR will be more directly reflected in the target processing response. Next, we modified the target properties to explore whether self-referential targets could similarly reduce the IOR.

Experiment 2: Self-Referential Targets Affect the IOR Effect

Method

Participants

On one hand, the sample size was based on previous studies of the IOR in 3D (Chen et al., 2012; A. J. Wang et al., 2016; A. J. Wang, Yue, et al., 2015); on the other hand, G*Power 3.1.9.2 (Faul et al., 2007) was used to estimate the sample size for a one-way repeated measures ANOVA with three within-subject measurements (estimated effect size f = 0.3, alpha = .05, power = 0.8), suggesting a sample of 20 participants. Twenty-one participants (12 females) took part in Experiment 2, and the age of the participants ranged from 18 to 22 years (M =19.1, SD = 2.20). All participants were right-handed and had normal or corrected-to-normal vision and color vision. None of the participants had a history of neurological or psychiatric disorders. All participants were paid for their participation in the experiment in accordance with the Helsinki Declaration. The Academic Committee of the Department of Psychology, Soochow University, China, approved this study.

Apparatus, Stimuli, and Experimental Setup

In Experiment 2, the target was one of the three geometric shapes (triangle, square, or circle), and the cue and the central cue consisted of transient flashes of one of the peripheral placeholders and one of the central placeholders, respectively. The other apparatuses and stimuli and the experimental setup matched those in Experiment 1.

Experimental Procedures and Design

In Experiment 2, the experimental steps were the same as those described in Experiment 1; however, two independent variables were changed: the target depth replaced the cue depth in Experiment 1 and the target label replaced the cue label in Experiment 1. Therefore, this experiment has a 2 (Target Depth [near vs. far]) × 3 (Target Label [self vs. friend vs. stranger]) × 2 (Cue Validity [valid vs. invalid]) design. Experiment 2 was a discrimination task: for example, pressing the Number 1 button on the keyboard with their right index finger when the target was self, pressing the Number 2 button when the target was friend, and pressing the Number 3 button when the target was stranger; the order of key assignments was counterbalanced across participants.

Results

Reaction Times

Incorrect responses (miss trials) and RTs beyond three SDs were discarded. Errors were below 5% across all of the conditions, and no further analysis was performed. Table 2 summarizes the average RTs and standard errors for all conditions in Experiment 2. Resulting in 12 experimental conditions, correct RTs were analyzed with a 2 (Target Depth [near vs. far]) × 3 (Target Label [self vs. friend vs. stranger]) × 2 (Cue Validity [valid vs. invalid]) repeated measures ANOVA. The results showed that the main effect of cue validity was significant, F(1, 20) = 13.42, p < .01, η² = .40, suggesting that RTs to the target at the valid location (668 milliseconds) were longer than RTs to the target at the invalid location (658 milliseconds). However, the main effect of target depth was not significant, F(1, 20) = 3.50, p > .05; the main effect of target label was not significant, F(2, 19) = 1.90, p > .05. The three-way interaction between target depth, target label, and cue validity was significant, F(2, 19) = 5.10, p < .05, η² = .35. No other significant effects of the two-factor interactions were found; all ps > .05.

Table 2.

Mean Reaction Times (RT) and Standard Errors (SEs) of the Means in the Experiment 2 (Presented in the Format RT ± SE).

	Far target			Near target
	Valid	Invalid	Valid–invalid	Valid	Invalid	Valid–invalid
Self	656 ± 19	644 ± 17	12*	665 ± 15	657 ± 16	8
Friend	672 ± 19	666 ± 17	6	688 ± 18	668 ± 17	20***
Stranger	663 ± 18	655 ± 17	8	664 ± 16	656 ± 17	10

*p < .05. ***p < .001.

To further test the potential interaction between cue validity and target label, we conducted a repeated measures ANOVA with a 2 (Cue Validity [valid vs. invalid]) ×3 (Target Label [self vs. friend vs. stranger]) framework. For the target in far space, the main effect of cue validity was significant, F(1, 20) = 6.34, p <.05, η² = .24, suggesting that RTs to the target at the valid location (664 milliseconds) were longer than RTs to the target at the invalid location (655 milliseconds). The main effect of target label was not significant, F(2, 19) = 1.53, p > .05. The interaction between cue validity and target label was not significant, F(2, 19) = 0.66, p > .05. For the target in near space, the main effect of cue validity was significant, F(1, 20) = 11.91, p < .005, η² = .37, suggesting that RTs to the target at the valid location (672 milliseconds) were longer than RTs to the target at the invalid location (660 milliseconds). The main effect of target label was not significant, F(2, 19) = 2.14, p > .05. Furthermore, the interaction between cue validity and target label was not significant, F(2, 19) = 2.31, p > .05.

IOR Effect Size

To further explore the potential relationship between target label and IOR effect, we compared the IOR effect size of target labels (valid minus invalid). As Experiment 2 investigated the effect of self-referential stimuli used as targets on the IOR in 3D space, we needed to control for targets being presented in far and near spaces. The IOR effect size in far space = RT in the valid condition (cuing locations in far space and targets appearing at the same locations) − RT in the invalid condition (cuing locations in near space and targets appearing at the locations in far space); the IOR effect size in near space = RT in the valid condition (cuing locations in near space and targets appearing at the same locations) − RT in the invalid condition (cuing locations in far space and targets appearing at the locations in near space). Due to the differences in what was explored in Experiments 1 and 2, the amount of the IOR effect size in far space and near space were calculated differently, and in the far-space invalid condition of Experiment 1, attention was oriented from far space to the fixation and then to near space; in the near-space invalid condition of Experiment 1, attention was oriented from near space to the fixation and then to far space. In the far-space invalid condition of Experiment 2, attention was oriented from near space to the fixation and then to far space; in the near-space invalid condition of Experiment 2, attention was oriented from far space to the fixation and then to near space. When the target appeared in near space, the IOR effect size of the self-referential target (8 milliseconds) was significantly smaller than that of the friend-referential target (20 milliseconds), t(21) = 2.24, p < .05. There were no other significant differences among the cue labels; all ps > .05. When the target appeared in far space, there were no significant differences among the cue labels; all ps > .05 (see Figure 4).

General Discussion

By integrating the Posner exogenous cue paradigm into virtual 3D space and introducing self-referential information into the 3D experimental paradigm, we studied whether self-referential information in 3D space can regulate the spatial IOR effect. The results showed that when the cue was a temporary self-referential cue that appeared in far space, it had the smallest IOR effect. When the temporarily established self-referential target appeared in near space, the IOR effect disappeared. The results showed that the SPE could adjust the IOR effect in 3D space.

Previous studies have found that when cue labels are self-referential information in the 2D plane, the IOR effect decreases or disappears (Qian et al., 2016; Sui et al., 2009; Zhang & Wu, 2013). The reason is that self-referential cues constantly attract the implicit attention of individuals and promote the quick reorientation of attention to the previously attended locations (Qian et al., 2016; Sui et al., 2009; Zhang & Wu, 2013). After self-referential cues appear, the RT for targets at the valid location is faster than the RT for other referential cues. However, as attentional orientation/reorientation differs in far and near spaces, the IORs in far and near spaces are different (A. J. Wang et al., 2016), and there are advantages in near-space processing (Cardin & Smith, 2011; Quinlan & Culham, 2007), which can lead to different effects of self-referential cues appearing in far and near spaces on the IOR.

In Experiment 1, when social-referential cues appeared in far space, individuals responded faster to targets on the same side (valid condition) of self-referential cues than to targets on the same side of friend- and stranger-referential cues. Also, there were no differences in the responses to targets on the opposite side (invalid condition) of self-referential cues than to those to targets on the opposite side of friend- and stranger-referential cues. This result suggested that self-referential cues produced an SPE effect in far space that consistently attracted attention and led to a faster reorientation of attention to previously searched locations (valid locations), which is consistent with previous studies in 2D in which self-referential cues were able to overcome the inhibition effect in cognitive processes, allowing participants to actively attend to previously attended locations, thereby reducing the IOR (Qian et al., 2016; Sui et al., 2009; Zhang & Wu, 2013). When social-referential cues appeared in near space, individuals did not differ in their responses to targets on the same side (valid condition) of self-, friend- and stranger-referential cues, nor did they differ in their responses to targets on the opposite side (invalid condition) of self-, friend-, and stranger-referential cues. The result suggests that self-referential cues in near space may not produce the SPE, which may be because when stimuli are presented in near space, they are prioritized for processing (Cardin & Smith, 2011; Quinlan & Culham, 2007).

Thus, cues appearing in near space are likely to be treated as self-relevant, and there is no difference between valid and invalid conditions when responding to targets that appear after different reference cues, resulting in a lack of difference in the IOR elicited by different reference cues. In addition, from a biological/ecological point of view, previous studies have suggested that to survive safely, our attention must instantly focus on objects that suddenly appear around us (Gawryszewski et al., 1987; Graziano & Cooke, 2006; Previc, 1998). Because the cues in our research were referential to social-referential information (self, friends, and strangers), they also attract attention when they appear in near space. Therefore, the SPE can influence the IOR effect in 3D space when self-referential cues appear in far space rather than in near space.

In Experiment 2, when social-referential targets appeared in far space, individuals responded faster to self-referential targets than to friend- and stranger-referential targets in both valid and invalid conditions, suggesting that self-referential targets in far space can produce the SPE. Unlike self-referential cues that continue to attract attention, self-referential targets that appear in far space do not reorient attention quickly to previously searched locations (valid locations); thus, the gains of self-referenced cues in valid conditions are less than those in invalid conditions, thus increasing the IOR. When social-referential targets appeared in near space, individuals responded faster to self-referenced targets than to friend-referential targets in both valid and invalid conditions, which suggests that self-referential targets can orient attention much quicker and reorient attention more quickly to previously searched locations than other new locations. This means that self-referential targets can produce the SPE in near space, and the gains of the SPE in near space are larger in valid locations than in invalid locations. The reason may be related to the default-mode network of the brain, including the PCC, OPFC, and left AG, which is known to exhibit enhanced neural activity when evaluating self-referential information from the body and the world. Targets appearing in near space have higher self-relevance (Chen et al., 2012; Corbetta et al., 2008; Fox et al., 2005; Gusnard & Raichle, 2001; Raichle et al., 2001; Qin & Northoff, 2011). Self-referential targets appearing in near space may be better processed by the default-mode network of the brain, thereby accelerating responses at the slowest valid location and reducing the IOR.

Surprisingly, individuals also react quickly to stranger-referential targets in near space at valid and invalid locations, which may be related to their survival habits. To survive safely, our attention must instantly focus on objects that suddenly appear around us (Gawryszewski et al., 1987; Graziano & Cooke, 2006; Previc, 1998). Stranger-referential targets may be perceived by individuals as negative survival-related threat stimuli, and J. X. Wang et al. (2013) believed that if the stimuli was related with survival, although the IOR has suppressed the location, individuals will preferentially select and process the biologically meaningful information of stimuli for better survival than if the IOR requires “blind” promotion of the new location (J. X. Wang et al., 2013, p. 338). Thus, the presence of stranger-referential targets in near space enables faster reorientation of attention to previously searched locations and faster responses under valid conditions, thereby reducing the IOR.

Experimentally, we found that the IOR affected by SPE differed in far and near spaces, which may have involved differences in the processing of SPE in far and near spaces; there is a more significant SPE on self-referential cues in far space compared with that in near space, thereby accelerating the reorientation to previously searched locations (valid locations) and reducing the IOR. However, this result may also be because stimuli in near space are perceived as self-relevant regardless of their label, all of which have prioritization effects; self-referential targets in near space are better processed for self-relevance than those in far space, especially for the processing of attentional inhibition locations (valid conditions). As a result, reorientation to previously searched locations (valid locations) is accelerated, and IOR is reduced.

Together, these results from Experiments 1 and 2 show that the IOR affected by SPE differs in far and near spaces, which may involve differences in the processing of SPE in far and near spaces; there is a more significant SPE on self-referential cues in far space compared with that in near space, thereby reducing the IOR. However, this outcome may also be because stimuli are perceived as self-relevant regardless of their label, all of which have prioritization effects; self-referential targets in near space are better processed for self-relevance than those in far space, especially for the processing of attentional inhibition locations (valid conditions). All of these results suggest that there is an advantage of self-referential processing in near space and that self-referential processing also prefers near space. In addition, this research provides evidence for the conclusion that the properties of cues and targets influence the IOR, which previous studies have assumed to be a stabilizing effect, but can in fact be influenced by altering the social-referential information of cues and targets. It has also been shown that biologically meaningful stimuli act as cues and targets that play distinct roles in visual processing for the IOR. Finally, combining the IOR and the SPE has higher ecological validity in 3D than in 2D and provides guidance for future similar studies in 3D.

The research is actually an indirect study of the SPE, that is, an exploration of geometric figures or pictures of unfamiliar faces with self-referential information established through an associative learning paradigm (Sui & Humphreys, 2015, 2017; Sui, Liu, et al., 2013; Sui, Rotshtein, et al., 2013; Wozniak & Knoblich, 2019). In addition to the indirect study of the SPE, some researchers have conducted direct studies of the SPE by using stimulus materials that are part of us, such as the ego voice and the ego face. Therefore, it is debatable whether there is a difference between direct and indirect studies of SPE on the IOR in 3D. In addition, it may be possible in the future to build on this foundation by using virtual reality immersive experience techniques with higher ecological efficacy to explore the impact of the SPE on the IOR.

Footnotes

Authors’ Contributions

A. W., L. W., and M. Z. designed the research. X. L. and A. W. performed the research. X. L., Q. Q., and L. W. analyzed the data. X. L., Q. Q., and A. W. wrote the manuscript text. A. W. and M. Z. reviewed the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the National Natural Science Foundation of China (31700939 and 31871092). A. W. was also supported by the Natural Science Foundation of Jiangsu Province (BK20170333), and Ministry of Education Project of Humanities and Social Sciences (17YJC190024).

Ethical Approval

After the experiment, all participants were paid according to the Helsinki Declaration. The study was approved by the Academic Committee of the Department of Psychology, Suzhou University, China.

Informed Consent

Informed consent was obtained from all individual participants included in the study.

ORCID iDs

Xiaoyuan Liu

Aijun Wang

References

Andersen

G. J.

(1990). Focused attention in three-dimensional space. Perception & Psychophysics, 47, 112–120.

Andersen

G. J.

Kramer

A. F.

(1993). Limits of focused attention in three-dimensional space. Perception & Psychophysics, 53, 658–667.

Berdica

Gerdes

A. B. M.

Alpers

G. W.

(2017). A comprehensive look at phobic fear in inhibition of return: Phobia-related spiders as cues and targets. Journal of Behavior Therapy and Experimental Psychiatry, 54, 158–164.

Boring

E. G.

(1964). Size-constancy in a picture. The American Journal of Psychology, 77, 494–498.

Bourke

P. A.

Partridge

Pollux

P. M. J.

(2006). Additive effects of inhibiting attention to objects and locations in three-dimensional displays. Visual Cognition, 13, 643–654.

Cardin

Smith

A. T.

(2011). Sensitivity of human visual cortical area V6 to stereoscopic depth gradients associated with self-motion. Journal of Neurophysiology, 106, 1240–1249.

Chao

H. F.

(2010). Inhibition of return to negative emotion: Evidence from an emotional expression detection task. Emotion, 10, 272–277.

Chen

Weidner

Vossel

Weiss

P. H.

Fink

G. R.

(2012). Neural mechanisms of attentional reorienting in three-dimensional space. Journal of Neuroscience, 32, 13352–13362.

Corbetta

Patel

Shulman

G. L.

(2008). The reorienting system of the human brain: From environment to theory of mind. Neuron, 58, 306–324.

10.

Dodd

M. D.

Castel

A. D.

Pratt

(2003). Inhibition of return with rapid serial shifts of attention: Implications for memory and visual search. Perception & Psychophysics, 65, 1126–1135.

11.

Durand

J. B.

Nelissen

Joly

Wardak

Todd

J. T.

Norman

J. F.

Orban

G. A.

(2007). Anterior regions of monkey parietal cortex process visual 3D shape. Neuron, 55, 493–505.

12.

Eng

Lim

Janssen

S. M. J.

Satel

(2018). Time course of inhibition of return in a spatial cueing paradigm with distractors. Acta Psychologica, 183, 51–57.

13.

Faul

Erdfelder

Lang

A. G.

Buchner

(2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175–191.

14.

Fox

M. D.

Snyder

A. Z.

Vincent

J. L.

Corbetta

Van Essen

D. C.

Raichle

M. E.

(2005). The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proceedings of the National Academy of Sciences, 102, 9673–9678.

15.

Gao

Sui

Chang

R. S.

(2020). Research progress and unsolved problems of self-predominance effect. Journal of Psychological Science, 43, 918–924.

16.

Gawryszewski

L. G.

Riggio

Rizzolatti

Umiltà

(1987). Movements of attention in the three spatial dimensions and the meaning of “neutral” cues. Neuropsychologia, 25, 19–29.

17.

Georgieva

Peeters

Kolster

Todd

J. T.

Orban

G. A.

(2009). The processing of three-dimensional shape from disparity in the human brain. Journal of Neuroscience, 29, 727–742.

18.

Gobel

M. S.

Tufft

M. R. A.

Richardson

D. C.

(2017). Social beliefs and visual attention: How the social relevance of a cue influences spatial orienting. Cognitive Science, 42, 1–25.

19.

Graziano

M. S. A.

Cooke

D. F.

(2006). Parieto-frontal interactions, personal space, and defensive behavior. Neuropsychologia, 44, 2621–2635.

20.

Gusnard

D. A.

Raichle

M. E.

(2001). Searching for a baseline: Functional imaging and the resting human brain. Nature Reviews Neuroscience, 2, 685–694.

21.

Hilchey

M. D.

Klein

R. M.

Satel

(2014). Returning to “inhibition of return” by dissociating long-term oculomotor IOR from short-term sensory adaptation and other nonoculomotor “inhibitory” cueing effects. Journal of Experimental Psychology: Human Perception & Performance, 40, 1603–1616.

22.

Jia

L. P.

Wang

J. X.

Zhang

H. F.

Sun

H. J.

(2019). Do emotional faces affect inhibition of return? An ERP study. Frontiers in Psychology, 10, 1–8.

23.

Klein

R. M.

(2000). Inhibition of return. Trends in Cognitive Sciences, 4, 138–147.

24.

Klein

R. M.

MacInnes

W. J.

(1999). Inhibition of return is a foraging facilitator in visual search. Psychological Science, 10, 346–352.

25.

Liu

M. H.

Wang

L. Y.

Sui

Zhang

(2012). Modulation of self-face for visual spatial attention: Evidence from a Posner’s cueing paradigm. Journal of Psychological Science, 35, 24–29.

26.

Martín-Arévalo

Kingstone

Lupiáñez

(2013). Is “inhibition of return” due to the inhibition of the return of attention? The Quarterly Journal of Experimental Psychology, 66, 347–359.

27.

Maylor

E. A.

Hockey

(1985). Inhibitory component of externally controlled covert orienting in visual space. Journal of Experimental Psychology: Human Perception & Performance, 11, 777–787.

28.

Posner

M. I.

Cohen

(1984). Components of visual orienting. Attention & Performance, 32, 531–556.

29.

Previc

F. H.

(1998). The neuropsychology of 3-D space. Psychological Bulletin, 124, 123–164.

30.

Qian

H. Y.

Gao

X. P.

(2016). The influence of self-face on the inhibition of return: Insights from response time ex-Gaussian distribution. Journal of Psychological Science, 39, 1051–1057.

31.

Qin

P. M.

Northoff

(2011). How is our self related to midline regions and the default-mode network? NeuroImage, 57, 1221–1233.

32.

Quinlan

D. J.

Culham

J. C.

(2007). fMRI reveals a preference for near viewing in the human parieto-occipital cortex. NeuroImage, 36, 167–187.

33.

Raichle

M. E.

MacLeod

A. M.

Snyder

A. Z.

Powers

W. J.

Gusnard

D. A.

Shulman

G. L.

(2001). A default mode of brain function. Proceedings of the National Academy of Sciences, 98, 676–682.

34.

Rogers

T. B.

Kuiper

N. A.

Kirker

W. S.

(1977). Self-reference and the encoding of personal information. Journal of Personality and Social Psychology, 35, 677–688.

35.

Roggeveen

A. B.

Prime

D. J.

Ward

L. M.

(2005). Inhibition of return and response repetition within and between modalities. Experimental Brain Research, 167, 86–94.

36.

Sereno

M. E.

Trinath

Augath

Logothetis

N. K.

(2002). Three-dimensional shape representation in monkey cortex. Neuron, 33, 635–652.

37.

Stoyanova

R. S.

Pratt

Anderson

A. K.

(2007). Inhibition of return to social signals of fear. Emotion, 7, 49–56.

38.

Sui

X. S.

(2017). Self as object: Emerging trends in self research. Trends in Neurosciences, 40, 643–653.

39.

Sui

Humphreys

G. W.

(2012). Perceptual effects of social salience: Evidence from self-prioritization effects on perceptual matching. Journal of Experimental Psychology: Human Perception & Performance, 38, 1105

40.

Sui

Humphreys

G. W.

(2015). More of me! Distinguishing self and reward bias using redundancy gains. Attention, Perception, & Psychophysics, 77, 2549–2561.

41.

Sui

Humphreys

G. W.

(2017). Aging enhances cognitive biases to friends but not the self. Psychonomic Bulletin & Review, 24, 2021–2030.

42.

Sui

Liu

C. H.

Wang

L. Y.

Han

S. H.

(2009). Attentional orientation induced by temporarily established self-referential cues. The Quarterly Journal of Experimental Psychology, 62, 844–849.

43.

Sui

Liu

Mevorach

Humphreys

G. W.

(2013). The salient self: The left intraparietal sulcus responds to social as well as perceptual-salience after self-association. Cerebral Cortex, 25, 1060–1068.

44.

Sui

Rotshtein

Humphreys

G. W.

(2013). Coupling social attention to the self forms a network for personal significance. Proceedings of the National Academy of Sciences, 110, 7607–7612.

45.

Taylor

T. L.

Klein

R. M.

(2000). Visual and motor effects in inhibition of return. Journal of Experimental Psychology: Human Perception & Performance, 26, 1639–1656.

46.

Theeuwes

Pratt

(2003). Inhibition of return spreads across 3-D space. Psychonomic Bulletin & Review, 10, 616–620.

47.

Truong

Roberts

K. H.

Todd

R. M.

(2017). I saw mine first: A prior-entry effect for newly acquired ownership. Journal of Experimental Psychology: Human Perception & Performance, 43, 192–205.

48.

Wang

A. J.

B. Q.

Zhang

(2015a). Location-based inhibition of return along depth plane in three-dimensional space. Acta Psychologica Sinica, 47, 859–868.

49.

Wang

A. J.

Liu

X. L.

Chen

Zhang

(2016). Effect of different directions of attentional shift on inhibition of return in three-dimensional space. Attention, Perception, & Psychophysics, 78, 838–847.

50.

Wang

A. J.

Liu

X. L.

Tang

X. Y.

Zhang

(2017). Inhibition of return at different eccentricities in visual field under three-dimensional (3D) world. Acta Psychologica Sinica, 49, 723–732.

51.

Wang

A. J.

Yue

Z. Z.

Zhang

Chen

(2015b). Interaction between spatial inhibition of return (IOR) and executive control in three-dimensional space. Experimental Brain Research, 233, 3059–3071.

52.

Wang

A. J.

Zhang

(2015). Influence of visuospatial inhibition of return on Simon effect in three-dimensional space. Journal of Psychological Science, 38, 792–800.

53.

Wang

H. X.

Humphreys

Sui

(2016). Expanding and retracting from the self: Gains and costs in switching self-associations. Journal of Experimental Psychology: Human Perception & Performance, 42, 247–256.

54.

Wang

J. X.

Jia

L. P.

Zhang

(2013). Influence of emotional faces for different location on inhibition of return. Journal of Psychological Science, 36, 335–339.

55.

Wang

Z. G.

Zhang

Klein

R. M.

(2010). Inhibition of return in static but not necessarily in dynamic search. Attention, Perception, & Psychophysics, 72, 76–85.

56.

Wozniak

Knoblich

(2019). Self-prioritization of fully unfamiliar stimuli. The Quarterly Journal of Experimental Psychology, 72, 2110–2120.

57.

Zhang

T. Y.

Y. H.

(2013). Effects of temporarily established self-referential cues on inhibition of return. Journal of Psychological Science, 36, 265–270.

58.

Zhao

Uono

Yoshimura

Toichi

(2015). Self make-up: The influence of self-referential processing on attention orienting. Scientific Reports, 5, 14169.