Attention attenuates metacontrast masking☆
Introduction
The ability of humans to process a visual scene is a seemingly effortless task, however, it requires numerous cognitive resources. The allocation of attention is a vital component in enabling an individual to selectively process components of a visual field while ignoring others. How attention influences the processing of stimuli has been thoroughly investigated, but questions still remain as to the influence it has on temporal and spatial representations of the world around us. The temporal and spatial influences of attention on visual awareness were examined here in the context of a metacontrast masking paradigm.
Metacontrast masking occurs when a temporally proximal but spatially non-overlapping stimulus (mask) impairs detection of a preceding target stimulus (Breitmeyer, 1984). The stimulus onset asynchrony (SOA) between the target and the mask is varied, with detection of the target showing a U-shaped function. Target detection is high at very short target-to-mask SOAs but drops dramatically during the optimal masking SOAs, typically between 40 and 60 ms, and then gradually recovers to very high levels. Low-level visual processing theories have proposed that during the optimal masking time window, the trailing mask interferes with the ongoing early visual processing of the preceding target and can thus inhibit the target from entering conscious awareness (Breitmeyer, 1984, Breitmeyer and Ganz, 1976, Breitmeyer and Ogmen, 2000). However, recent evidence suggests that theories of low-level visual processing may not provide a complete picture of the mechanisms operating to produce metacontrast masking. Rather, there seems to be a complex interaction between high-level visual processes, such as visual selective attention, and masking.
Attention may play a role in the effectiveness of metacontrast masking and may serve to bring a target stimulus into awareness. For example, Tata (2002) examined the allocation of attention across a circular visual array containing either 1, 2, 4, or 8 distractors in a metacontrast masking paradigm. The typical U-shaped function with set sizes of 4 and 8 was found, with a steep decrease in performance with set size 8 that was larger and longer lasting than set size 4. However, there was no masking present at a set size of 1 or 2. This is consistent with earlier reports by Spencer and Shuntich (1970) who found that the masking function was extended under conditions of high attentional load (12 stimuli) compared to that of low load (one stimulus), suggesting that attentional load may interact with masking effects in such a way that high load produces more stimulus masking (cf. Lavie, 1995, Lavie and Cox, 1997).
Tata (2002) found in one of his experiments that exogenously orienting attention also modulates the masking function. Exogenous orienting was induced with a peripheral cue (a brief flash), with no predictive value, that reflexively oriented the subject’s attention to one position in space (Posner & Cohen, 1984). When a cue was presented at the target location (valid cue) at least 50 ms prior to the array, Tata found that masking was reduced relative to when the cue was presented in the location of a distractor (invalid cue). These results indicate that an exogenous valid cue can aid in target detection (reduce masking) by orienting reflexive attention to the proper location. However, no difference was found between invalid trials and trials in which no cue was present, suggesting only a benefit without any costs of exogenous attention on metacontrast masking.
In another study, Shelley-Tremblay and Mack (1999) used stimuli that have been shown to be detected without attention to test the influence of salient, attention capturing stimuli on metacontrast masking. Studies of inattentional blindness, which occurs when one fails to detect a new or salient visual stimulus when attention is focused on another visual stimulus, have demonstrated that a happy face or a person’s own name can avoid inattentional blindness and capture attention (Mack & Rock, 1998). Shelley-Tremblay and Mack found that detection of a happy face target followed by a metacontrast mask was greater (more resistant to masking) across all SOAs compared to inverted or scrambled faces. Also, detection of the person’s own name was greater than the scrambled variant of their name or the word “TIME”. They also examined the effectiveness of salient stimuli as masks, and found that target detection was worse when the target had been masked by the person’s name than its scramble. Taken together it can be concluded that attention interacts with the mechanisms of metacontrast masking, and can facilitate performance (also see Ramachandran & Cobb, 1995).
The allocation of attention over spatial distances and across spatial resolutions has also been examined with respect to metacontrast masking. Such experiments explored the consequences of a spatial variance between a target and mask set. In a study by Enns and DiLollo (1997), for example, the location of the target/mask set was varied and positioned at fixation (foveally) or 3° to the left or right of fixation (peripherally). The masking effect produced by a metacontrast mask was optimal at an SOA of 45 ms, but increased substantially when the stimulus set appeared in a peripheral location as compared to when it was presented foveally. More interestingly, a mask consisting of just four dots presented around the target failed to produce any masking at the foveal location, but produced large masking effects when presented peripherally, with the optimal masking SOA of 45 ms mimicking that of the metacontrast mask at the foveal location. Their results suggest that a spatial separation of target and mask (four dot mask) can decrease the masking effect when presented foveally (increase conscious awareness of it), but fails to do so when presented peripherally. Furthermore, when stimuli appeared outside the focus of attention, masking was more effective at preventing stimulus entry into conscious awareness.
The current study varied the temporal and spatial properties between a circular target and an annulus mask and employed valid and invalid endogenous cues to assess the influence of spatial voluntary attention on target visibility using a metacontrast masking paradigm. If the allocation of endogenous attention can increase target detection, then the cueing of spatial attention to the proper location of a target/mask set should act in reducing the masking effect across time (Experiment 1). Furthermore, attention acts differentially on items outside its focus, then by varying the spatial distance between the target and mask peripherally, we can assess the greatest spatial separation sufficient to produce a masking effect and can determine how that distance changes with the allocation of attention (Experiment 2). Due to the heightened effectiveness of masking outside the focus of attention, it is predicted that more masking will occur between spatially distant items predominantly in the invalidly as compared to the validly cued condition.
Section snippets
Participants
Eighteen undergraduate students from Rice University participated in this experiment for partial fulfillment of a course requirement. There were two left handed participants; all other subjects were right handed. Five males and 13 females, ranging in age from 18 to 21 (mean = 19) participated after informed consent, which along with this study was approved by the Institutional Review Board at Rice University. All subjects had normal or corrected to normal vision.
Stimuli, apparatus, and procedure
The stimuli were presented on a
Participants
Twenty-four undergraduate students from Rice University participated in this experiment as partial fulfillment of a course requirement. All participants were right handed with normal or corrected to normal vision. There were 14 males and 10 females, ranging in age from 17 to 22, with a mean age of 20. All participated after informed consent.
Stimuli, apparatus, and procedure
The stimuli were the same size and shape as those used in Experiment 1, and were presented using the same computer. The target/mask set presented in the
General discussion
This study sought to determine the influence of attentional allocation on target visibility using a metacontrast masking paradigm. It was hypothesized that allocating attention through the use of an endogenous cue would modulate the masking effect. The overall effectiveness of the cues in allocating attention was demonstrated via faster response times and higher target detection rates when the cue was valid than when it was invalid. Furthermore, when attention was allocated to the target,
References (28)
- et al.
The distinct modes of vision offered by feed-forward and recurrent processing
Trends in Neuroscience
(2000) - et al.
Feedback contributions to visual awareness in human occipital cortex
Current Biology
(2003) - et al.
A feature integration theory of attention
Cognitive Psychology
(1980) Visual masking. An integrative approach
(1984)- et al.
Implications of sustained and transient channels for theories of visual pattern masking, saccadic suppression, and information processing
Psychological Review
(1976) - et al.
On the role of stroboscopic motion in metacontrast
Bulletin of the Psychonomic Society
(1981) - et al.
Recent models and findings in visual backward masking: a comparison, review, and update
Perception & Psychophysics
(2000) - et al.
Stimulus-onset asynchrony is not necessary for motion perception or metacontrast masking
Psychological Science
(1993) - et al.
Object substitution: a new form of masking in unattended visual locations
Psychological Science
(1997) - et al.
Cortical dynamics of visual motion perception: short-range and long-range apparent motion
Psychological Review
(1992)
An onset-onset law for one case of apparent motion and metacontrast
Perception & Psychophysics
Aspects of apparent motion
Figural change in apparent motion
Journal of Experimental Psychology
Cited by (42)
Visual masking: Contributions from and comments on Bruce Bridgeman
2018, Consciousness and CognitionCitation Excerpt :Although Bridgeman accepted that processing of the perceptual results of masking can be influenced by attention, he kept the view that masking as such is not based on attention mechanism. This standpoint contradicts what some metacontrast researchers have argued for (Boyer & Ro, 2007; Ramachandran & Cobb, 1995). However, in later years some carefully conducted research supported the stance that attention and masking are autonomous phenomena (Agaoglu, Breitmeyer, & Öǧmen, 2016; Webb, Igelström, Schurger, & Graziano, 2016).
Perceptual learning effect on decision and confidence thresholds
2016, Consciousness and CognitionKeeping postdiction simple
2015, Consciousness and CognitionCitation Excerpt :The metacontrast masking effect is strongest with SOAs of around 50 ms (which corresponds to the timing of one local reentrant processing loop), while the object substitution masking effect occurs with SOAs between 100 and 300 ms. Moreover, in metacontrast masking, the processing of the target is disrupted at 110–140 ms (Fahrenfort, Scholte, & Lamme, 2007), whereas it continues for more than 200 ms in object substitution masking (Woodman & Luck, 2003). Finally, while attentional factors modulate the metacontrast masking but are not the main cause of it (e.g., Boyer & Ro, 2007; Ramachandran & Cobb, 1995; Shelley-Tremblay & Mack, 1999; Tata, 2002), object substitution masking is explained in terms of the target being replaced by the mask due to attentional factors (Brehaut, Enns, & Di Lollo, 1999; Giesbrecht & Di Lollo, 1998; Visser & Enns, 2001). Given these differences, it is more reasonable to think that the two phenomena are due to different processes rather than one single process.
Masking with faces in central visual field under a variety of temporal schedules
2015, Vision ResearchCitation Excerpt :A second way in which metacontrast masking can be compared to OSM (or, rather, to common onset four dot masking) is by how each is modulated by attention. There is evidence that metacontrast masking can be modulated by attention (Boyer & Ro, 2007; Ramachandran & Cobb, 1995). On the other hand, the role of attention in four dot masking is less clear.
- ☆
This manuscript was accepted under the editorship of Jacques Mehler.