- Email: [email protected]
Subliminal Gestalt grouping: Evidence of perceptual grouping by proximity and similarity in absence of conscious perception
Consciousness and Cognition 25 (2014) 1–8
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Subliminal Gestalt grouping: Evidence of perceptual grouping by proximity and similarity in absence of conscious perception Pedro R. Montoro a,⇑, Dolores Luna a, Juan J. Ortells b a b
Departamento de Psicología Básica 1, UNED, Spain Departamento de Psicología, Universidad de Almeria, Spain
a r t i c l e
i n f o
Article history: Received 30 July 2013 Available online 8 February 2014 Keywords: Perceptual grouping Visual masking Subliminal priming Proximity Similarity
a b s t r a c t Previous studies making use of indirect processing measures have shown that perceptual grouping can occur outside the focus of attention. However, no previous study has examined the possibility of subliminal processing of perceptual grouping. The present work steps forward in the study of perceptual organization, reporting direct evidence of subliminal processing of Gestalt patterns. In two masked priming experiments, Gestalt patterns grouped by proximity or similarity that induced either a horizontal or vertical global orientation of the stimuli were presented as masked primes and followed by visible targets that could be congruent or incongruent with the orientation of the primes. The results showed a reliable priming effect in the complete absence of prime awareness for both proximity and similarity grouping principles. These findings suggest that a phenomenal report of the Gestalt pattern is not mandatory to observe an effect on the response based on the global properties of Gestalt stimuli. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction The world we perceive consists of objects and their interrelations coherently arranged in scenes. The processes of perceptual organization are responsible for structuring the retinal mosaic into the global stimuli of perceived objects (Palmer, 1999). One century ago, the Gestalt school of psychology was the first to deal with the problem of perceptual organization (see Wagemans et al., 2012, for a review). According to Max Wertheimer, organization is basically composed of grouping and segregation processes. The well-know principles of grouping describe the stimulus factors that determine the visual grouping of discrete elements, including proximity, similarity, common fate, good continuation, closure (Wertheimer, 1923), and, more recently, other new principles such as common region, connectedness or synchrony have been proposed (Palmer, 1992; Palmer & Rock, 1994). Classical cognitive theories of visual processing have assumed that perceptual grouping occurs preattentively in absence of attention, producing segmented perceptual units or ‘proto-objects’ for analysis by subsequent processes (Julesz, 1981; Kahneman & Henik, 1981; Neisser, 1967; Treisman, 1982, 1985, 1988). This assumption was challenged by the results of studies showing that perceptual grouping is not perceived under strict conditions of inattention (Mack & Rock, 1998; Mack, Tang, Tuma, Kahn, & Rock, 1992; Rock, Linnett, Grant, & Mack, 1992). However, Moore and Egeth (1997), using inattentional conditions similar to that of Mack and colleagues but crucially introducing on-line measures of unattended processing, showed that grouping could occur outside the focus of attention. Moore and Egeth suggested that grouped patterns could ⇑ Corresponding author. Address: Departamento de Psicología Básica 1, Facultad de Psicología, UNED, C/ Juan del Rosal 10, 28040 Madrid, Spain. E-mail address: [email protected] (P.R. Montoro). http://dx.doi.org/10.1016/j.concog.2014.01.004 1053-8100/Ó 2014 Elsevier Inc. All rights reserved.
2
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
have been perceived without attention but not encoded in memory. Later studies using similar methodological strategies have provided further evidence that perceptual grouping does not require attention (Chan & Chua, 2003; Driver, Davis, Russell, Turatto, & Freeman, 2001; Kimchi & Razpurker-Apfeld, 2004; Lamy, Segal, & Ruderman, 2006; Mitroff & Scholl, 2005; Russell & Driver, 2005). Interestingly, in a similar way, two recent studies have reported evidence for subliminal processing of other types of perceptual organization such as contour integration (Rosenthal & Humphreys, 2010) and illusory contour completion (Wang, Weng, & He, 2012). Rosenthal and Humphreys found a significant subliminal learning of global contours generated by integration of Gabor elements, despite being irrelevant to the main task. Using a suppression time paradigm, Wang et al. (2012) have recently observed that Kanizsa-like stimuli can be processed without awareness. These results are impressive examples of unconscious organization of the discrete elements to form emergent patterns that are qualitatively different from the sum of their parts. In addition, previous works provide a theoretical basis for a grouping effect occurring at an unconscious level. Several models for contour processing have analyzed the possible mechanism underlying grouping effects and have suggested that grouping could occur at an unconscious level. For example, Grossberg and Mingolla (1985a, 1985b) proposed a model of recurrent networks that considers unconscious computation of boundary contours. The model distinguishes between a Boundary Contour System (BCS) and a Feature Contour System (FCS) that extract different information concerning contour at an early processing stage. Information on edges and boundaries (i.e. the form outline of an object) is processed in the BCS whereas information on surface features (i.e. color or brightness) that fills in the area delimited by the BCS is processed in the FCS. The model assumes that the processing of form in BCS, which extracts edge information, occurs at non-conscious levels. Neumann and Sepp (1999) developed a neuropsychological model for contour processing based on interactions between V1 and V2. The model suggests an early occurrence for grouping in the stream of these interactions. Further studies in the field of visual neuroscience (see Breitmeyer & Tapia, 2011, for a review) assuming separate pathways for the processing of form and surface and using a metacontrast masking procedure showed that the conscious processing of form requires previous fill in of surface properties. The results also suggest that form processing proceeds faster than surface processing at unconscious levels whereas at conscious level it proceeds slower than surface processing. However, to our knowledge, no previous study has examined the organization of visual elements into global patterns by means of Gestalt principles of grouping in absence of conscious perception. The present work steps forward in the study of perceptual grouping processes examining the possibility of a subliminal processing of Gestalt patterns generated by the action of grouping principles. This issue is relevant, as recent theoretical contributions have differentiated between consciousness and attention, and, by extension, between unconscious and inattentional processes (Dehaene & Changeux, 2011; Dehaene, Changeux, Naccache, Sackur, & Sergent, 2006; Koch & Tsuchiya, 2007; Lamme, 2003). In this line, Dehaene and collaborators have proposed a tripartite distinction of subliminal, preconscious and conscious processing (see Dehaene & Changeux, 2011, for a review). According to this taxonomy, a subliminal stimulus is an invisible, undetectable one, even with focused attention. In contrast, a preconscious stimulus is potentially visible but is not consciously perceived due to distraction or inattention. From this view, the results of studies supporting perceptual grouping without attention are samples of preconscious processing because their procedures involve supraliminal patterns that could be reported if they were attended (e.g., see Full Attention condition in Moore and Egeth (1997)’s experiments). This implies that the findings of grouping without attention cannot be directly generalized to a hypothetical subliminal processing of grouping. Consequently, an investigation on the possibility of Gestalt grouping under exhaustive subliminal conditions is necessary to extend our knowledge of perceptual organization operations in the human visual system. Our aim is to determine whether a phenomenal report of the patterns is mandatory to observe an effect on the response based on the global properties of grouped stimuli. This goal is especially relevant considering that Gestalt psychologists focused their attention on phenomenology and, coherently, proposed their classical principles of grouping based exclusively on subjective measures obtained by phenomenal demonstrations (Lamy et al., 2006; Spillmann, 2009). In order to achieve our purpose, we had to make two important decisions relative to (1) the subliminal method to be implemented as well as (2) the representative principles of perceptual grouping to be included in the study. Regarding the subliminal method, we used a masked priming task, the most frequently used technique for studying the extent of processing during subliminal perception (see Kim & Blake, 2005; Kouider & Dehaene, 2007, for reviews). Specifically, a double masking procedure combining forward and backward masks was applied to the grouped pattern. As principles of grouping, we selected proximity and luminance similarity (see Fig. 1), two classical Gestalt laws already described by Wertheimer (1923) and widely studied in numerous previous works (e.g., Ben-Av & Sagi, 1995; Kubovy, Holcombe, & Wagemans, 1998; Luna & Montoro, 2011; Quinlan & Wilton, 1998). Note that these principles base their capacity for grouping on the analysis of different properties of the elements, i.e., spatial relationships between single units in the case of proximity and relative levels of luminance concerning similarity. Due to the temporal and computational differences between grouping principles showed by previous studies (Han, Ding, & Song, 2002; Han, Song, Ding, Yund, & Woods, 2001; Kurylo, 1997; Luna & Montoro, 2008), this selection might provide us with a more extensive exploration of the boundary conditions for observing a subliminal processing of Gestalt patterns. In two different experiments, the grouping principles were applied to disk lattices to form horizontally or vertically oriented global patterns, which could be congruent or incongruent with the orientation of the target stimuli displayed afterwards. Thus, the priming effect was based on the global orientation of the stimuli and not on any local characteristic. Several kinds of targets with different physical appearances were displayed in order to force participants to base their response exclusively on the orientation of the stimuli.
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
3
Fig. 1. Stimuli and sequence of events used in the experiments. (a) Grouped patterns displayed as primes in Experiment 1. (b) Grouped patterns used in Experiment 2. (c) Uniform patterns used as primes in half of the trials in both forced-choice prime detection tasks.
A crucial control parameter included in our priming procedure was to avoid displaying the Gestalt patterns as visible targets, and we used them exclusively as masked primes. Previous studies have suggested that subliminal priming effects could be due to the fact that the masked primes had been also displayed as conscious targets in other phases of the experiment (Abrams & Greenwald, 2000; Damian, 2001; Eimer & Schlaghecken, 1998; Neumann & Klotz, 1994). However, other studies have disconfirmed this hypothesis by showing subliminal priming effects for primes that were never presented as visible targets (Abrams, Klinger, & Greenwald, 2002; Kiesel, Kunde, Pohl, & Hoffmann, 2006; Naccache & Dehaene, 2001; Ortells, Marí-Beffa, & Plaza-Ayllón, 2013; Van den Bussche, Notebaert, & Reynvoet, 2009). Therefore, we considered that it was necessary to control this experimental confound in order to determine whether genuine subliminal Gestalt grouping is possible. Thus, the Gestalt patterns were exclusively displayed as primes under masked conditions of presentation. 2. Experiment 1: Grouping by proximity 2.1. Method 2.1.1. Participants Thirty-eight undergraduate students (five men; age range = 19–59 years, M = 32.02, SD = 10.86) from the UNED participated in the main experiment consisting of two consecutive tasks: (1) a masked priming task and (2) a prime visibility discrimination task. All of them had normal or corrected-to-normal vision and received course credits for their participation. A different sample of twenty-three undergraduate students (ten men; age range = 20–64 years, M = 35.5, SD = 12.1) from the UNED were recruited for an additional control experiment consisting of a forced-choice prime detection task. All of them had normal or corrected-to-normal vision and received course credits for their participation. The experimental procedures were approved by the Local Ethics Committee and conform to the Declaration of Helsinki.
4
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
2.1.2. Stimuli and apparatus The stimuli were displayed on a 19-in. LCD–LED Samsung 943N color monitor with a 75-Hz refresh rate, a 5:4 aspect ratio, a light gray background (RGB: 198; 36 cd/m2) and a resolution of 1024 768 controlled by a computer running E-Prime 1.2 software (Psychology Software Tools, 1996–2002). Viewing distance was approximately 57 cm. All the stimuli were displayed on a central dark gray square background (RGB: 95; 6 cd/m2) subtending a visual angle of 5.3° 5.3°. The Gestalt patterns presented as primes consisted of disk lattices of light gray (RGB: 191; 33 cd/m2) or black (RGB: 0; 0 cd/m2) elements. Each local element subtended a visual angle of 0.40° 0.40°. The proximity-grouping patterns consisted of 8 6 elements, or vice versa, which were arranged in rows or columns so that the distances between near or remote rows or columns were 0.40° and 1.0°, respectively. The masking patterns were slightly larger random light gray and black disk lattices arranged in a 13 13-matrix. There was the same proportion of black and light gray disks in the masks. The target stimuli consisted of five continuous zigzag or straight lines disposed in a vertical or horizontal orientation. There were six different target stimuli in order to force participants to base their response on the global orientation of the stimuli regardless of the physical appearance of the stimulus (see Fig. 1). 2.2. Procedure and design Participants in the main experiment were tested individually in a dimly lit room in two consecutive sessions, one for the masked priming task, and another one for the prime visibility discrimination task. In the masked priming task, participants carried out a forced-choice reaction time (RT) task. Subjects were told that they would see target lines displayed on the screen, and that they would have to indicate the vertical or horizontal orientation by pressing one of two response buttons (number 1 or 2, respectively) with their middle and index fingers of the dominant hand as fast as possible but avoid making mistakes. Crucially, subjects were not told about the masked primes, but were merely informed that each trial would begin with the presentation of a ‘flash signal’ to warn them that the target stimulus was going to appear afterwards. The sequence of events in a trial for each grouping condition is depicted in Fig. 1. Each trial started with a mask pattern; 147 ms later, a Gestalt grouped pattern was presented for 53 ms; then, a second mask pattern, different from the first one, was displayed during 80 ms; and, finally, the target stimulus appeared on the screen and remained until response. After the end of the trial, a pause of 800 ms ensued before the start of the next trial. There were a practise block with 24 trials and seven experimental blocks consisting of 96 trials each, for a total of 672 experimental trials (one half congruent trials and one half incongruent ones). Feedback was provided only for the practise trials. Immediately after the end of the priming task, participants were systematically asked several questions concerning the ‘flash signal’, namely: Have you noticed any strange or curious thing during the experiment? Have you seen any strange or curious thing related to the ‘flash signal’? Have you seen any figure, configuration or recognizable stimulus during the ‘flash signal’? After that, they were fully informed of the nature of the ‘flash signal’ and were asked to perform a forced-choice prime discrimination task designed to obtain an objective index (i.e., d0 ) of prime visibility. In this task, consisting of 12 practise trials followed by 120 experimental trials, participants were instructed to pay attention to the prime stimulus that was displayed between the two masks, and to perform a forced-choice discrimination task indicating the horizontal or vertical orientation of that stimulus. For this task, the sequence of events was identical to that of the priming task, with the following exceptions: (1) the target stimuli were displayed during a fixed time of 507 ms, (2) after the target offset, a horizontal and a vertical rectangle with the corresponding labels ‘HORIZONTAL’ and ‘VERTICAL’ inside were displayed on the screen so that the participants provided their response by clicking the mouse on the rectangle containing the chosen orientation without response time demand; and (3) the trials were self-administered in order to ensure that participants were as ready as possible to discriminate the masked prime. Participants were instructed to try to be as accurate as possible and to guess on trials in which they could not identify the primes. The additional sample of participants performing the forced-choice prime detection task also received exhaustive information about the primes and the masking procedure. The procedure in this task was identical to that of the forced-choice discrimination task with the following exceptions: (1) two new uniform patterns were used (see Fig. 1) consisting of alternating gray and black disks distributed evenly across the lattice; (2) half of the trials presented the uniform patterns as primes, and the other half was composed of the grouped patterns; and (3) participants were instructed to discriminate between grouped and uniform patterns by clicking the mouse on the rectangle containing the selected choice without response time demand. The design included a within-subjects factor comparing congruent versus incongruent trials as a function of the horizontal–vertical congruency between prime and target. There was the same proportion of congruent as incongruent trials in each single block of the experiment. 2.3. Results 2.3.1. Priming task Mean RTs of correct responses and mean accuracy were submitted to separate analyses of variance (ANOVAs) with Prime–Target Congruency as within-subjects factor (2 levels: Congruent or Incongruent). Mean RTs and mean accuracy (hit rates) as a function of this factor are reported in Table 1.
5
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8 Table 1 Mean (SD) RTs (in ms), and mean accuracy (hit rate) for the congruent and incongruent trials, and the amount of priming (incongruent–congruent) as a function of Experiment (grouping by proximity or luminance similarity).
*
Incongruent
Congruent
Priming effect
Experiment 1: Proximity 543 (56.9) .97 (0.01)
532 (58.1) .98 (0.01)
+11*** 1
Experiment 2: Similarity 533 (53.7) .97 (0.02)
524 (54.8) .98 (0.02)
+9*** 1*
p < .05. p < .001.
***
2.3.1.1. RT analysis. Inaccurate responses (on average 2.4%) and RTs greater than 1500 or less than 200 ms (2.6% of trials) were excluded from the RT analysis. There was a significant main effect of priming, F(1, 37) = 74.69, MSe = 28.80, p < .001, g2p ¼ :67, indicating that RTs in congruent trials were reliably shorter (532 ms) than those in incongruent trials (543 ms). 2.3.1.2. Accuracy analysis. Hit rates oscillated between 94.5% and 99.5%. The same ANOVA was conducted on accuracy rates, revealing a marginally significant effect of priming, F(1, 37) = 3.83, MSe < .001, p = .058, g2p ¼ :09. These results mirrored those of RTs (Congruent: 97.7%; Incongruent: 97.3%). 2.3.2. Prime visibility discrimination task To measure subjective visibility of the masked primes, participants were asked to report what patterns they had seen on the screen before the presentation of the target during the experiment. None of the participants reported having seen any horizontal- or vertical-oriented patterns before the presentation of the target. This finding suggests that participants had no subjective awareness of the primes. Mean accuracy in the prime visibility discrimination task across vertical and horizontal primes was .51 (SD = .03; individual rates from .42 to .58), which suggests that primes could not be classified above chance. A direct measure of prime visibility (d0 ) was calculated for each participant. The measures were obtained by treating one level (i.e., horizontal) as signal and the other level (i.e., vertical) as noise. The individual d0 values ranged between .42 and .54, and the overall mean was .02 (SD = .19). A single sample t-test showed that this d’ was not significantly different from zero, t (37) = .287, p = .78, thus suggesting that participants were not aware of the prime orientation in the main task. A linear regression analysis was further conducted to examine the relationship between individual priming effect and prime visibility across participants. For each participant, a priming effect (D = RTincongruent RTcongruent) was calculated and correlated with the corresponding value of the d0 in the forced-choice prime discrimination task. There was no significant linear relationship between the priming effect and prime visibility (R2 = .157, p = .35). In sum, the results revealed that the primes were largely invisible. However, it could be possible that masked stimuli were unidentifiable yet detectable (Kim & Blake, 2005). In our study, participants could have been aware that a grouped pattern appeared without being aware of its concrete orientation. To examine that possibility, in this and the next study, an additional control experiment consisting of a forced-choice detection task was conducted on a different sample of participants. 2.3.3. Forced-choice detection task Individual hit rates oscillated between .44 and .59 (M = .51, SD = .04). Individual d0 indexes were computed similarly to the forced-choice discrimination task, ranging between .49 and .39 (M = .01, SD = .24). A single sample t-test, t (22) = .524, p = .61, showed that mean performance was not significantly different from zero, suggesting that participants were not capable of detecting the presence of a grouped pattern under the masking conditions displayed. 3. Experiment 2: Grouping by similarity 3.1. Method 3.1.1. Participants Thirty-eight undergraduate students (three men; age range = 19–51 years, M = 29.24, SD = 10.5) from the UNED participated in the main experiment consisting of two consecutive tasks: (1) a masked priming task and (2) a prime visibility discrimination task. All of them had normal or corrected-to-normal vision and received course credits for their participation. A different sample of twenty-three undergraduate students (ten men; age range = 20–64 years, M = 35.5, SD = 12.1) from the UNED were recruited for an additional control experiment consisting of a forced-choice prime detection task. All of them had normal or corrected-to-normal vision and received course credits for their participation. The experimental procedures were approved by the Local Ethics Committee and conform to the Declaration of Helsinki.
6
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
3.1.2. Stimuli and apparatus The stimuli and apparatus were identical to those of Experiment 1, with the sole exception that the Gestalt patterns consisted of a 6 6 array, forming rows or columns with elements of identical luminance. The horizontal and vertical distance between adjacent local elements was 0.50°. 3.1.3. Procedure and design The procedure and design were identical to those of Experiment 1. 3.2. Results 3.2.1. Priming task Mean RTs of correct responses and mean accuracy were submitted to separate ANOVA with Prime–Target Congruency as within-subjects factor (2 levels: Congruent or Incongruent). Mean RTs and mean accuracy (hit rates) as a function of this factor are reported in Table 1. 2.3.1.3. RT analysis. Inaccurate responses (on average 2.6%), and trials with RTs greater than 1500 or less than 200 ms (2.6% of trials) were excluded from the RT analysis. There was a significant main effect of priming, F(1, 37) = 41.86, MSe = 36.42, p < .001, g2p ¼ :53, indicating that RTs in congruent trials were reliably shorter (533 ms) than those in incongruent trials (542 ms). 2.3.1.4. Accuracy analysis. Hit rates oscillated between 92% and 99.5%. The same ANOVA was conducted on accuracy rates, revealing a significant effect of priming, F(1, 37) = 4.18, MSe < .001, p = .048, g2p ¼ :10. These results mirrored those of RTs (Congruent: 97.5%; Incongruent: 97.1%). 3.2.2. Prime visibility discrimination task As in Experiment 1, none of the participants reported having seen any horizontal- or vertical-oriented patterns before the presentation of the target. Mean accuracy in the prime visibility task across vertical and horizontal primes was .50 (SD = .04; individual rates ranged from .40 to .57). The individual d0 values oscillated between .53 and .38, and the overall mean was .01 (SD = .22). A single sample t-test showed that this d0 was not significantly different from zero, suggesting that participants were not aware of the prime orientation in the main task, t (37) = .287, p = .78. Furthermore, there was no significant linear relationship between priming effect and prime visibility (R2 = .121, p > .45). 3.2.3. Forced-choice detection task The procedure and the analysis were identical to those of Experiment 1. Individual hit rates oscillated between .44 and .59 (M = .51, SD = .04). Individual d0 indexes ranged between .49 and .39 (M = .01, SD = .24). A single sample t-test, t (22) = .524, p = .61, showed that mean performance was not significantly different from zero, supporting that participants could not detect the presence of a grouped pattern under the masking conditions displayed. 4. Combined analysis of Experiments 1 and 2 Finally, a combined analysis of Experiment 1 and Experiment 2 was conducted in order to compare the effect of each grouping principle. Previous studies have suggested temporal and computational differences between grouping principles (Han et al., 2001, 2002; Kurylo, 1997; Luna & Montoro, 2008). Thus, we examined whether grouping by proximity versus by similarity also differ when they are processed under subliminal conditions. To test this possibility, mean RTs and mean accuracy were submitted to two separate mixed ANOVAs with Perceptual Grouping Condition as a between-subjects factor (2 levels: Proximity or Similarity), and Prime–Target Congruency as within-subjects factor (2 levels: Congruent or Incongruent). 2.4. RT analysis There was a significant main effect of priming, F(1, 74) = 111.9, MSe = 26.93, p < .001, g2p ¼ :60, indicating that RTs in congruent trials were shorter (528 ms) than those in incongruent trials (538 ms). Neither the between-subjects factor nor the interaction was significant (F < 1). 2.5. Accuracy analysis The same ANOVA was conducted on accuracy rates, revealing a significant effect of priming, F(1, 74) = 8, MSe < .001, p < .01, g2p ¼ :10, showing a higher hit rate in congruent trials (97.6%) than in incongruent trials (97.2%). Neither the between-subjects factor nor the interaction was significant (F < 1). These results mirrored those of RTs.
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
7
5. Discussion Can perceptual grouping occur unconsciously in absence of a phenomenal report of the Gestalt pattern displayed? Previous studies making use of indirect processing measures have shown that perceptual grouping can occur even under the strictest conditions of inattention proposed by Mack and Rock’s (1998) paradigm (Chan & Chua, 2003; Kimchi & Razpurker-Apfeld, 2004; Lamy et al., 2006; Mitroff & Scholl, 2005; Russell & Driver, 2005). However, those results do not allow us to determine whether perceptual grouping occurs under subliminal conditions (Dehaene & Changeux, 2011). For this reason, the aim of the present study was to examine whether perceptual grouping requires consciousness. According to a dissociation paradigm (Merikle, Smilek, & Eastwood, 2001; Reingold & Merikle, 1988, 1993), we had to verify two empirical aspects in order to answer this question. First, we ruled out that grouping was processed by conscious vision, as masked grouped patterns were largely invisible, as suggested both by subjective and objective (prime discrimination and detection tasks) measures. Second, we found that masked visual groups produced significant effects on an indirect measure of processing, that is, the orientation-based priming effect. Thus, our results provide the first evidence to support that grouping principles can be applied to discrete visual elements to form Gestalt groups under subliminal conditions, at least for the cases of proximity and luminance similarity examined in the current work. This implies that a phenomenal report of the pattern is not mandatory to observe an effect on the response based on the global properties of grouped stimuli. In addition, similar effects for proximity and similarity grouped patterns were observed in the present study, showing that both grouped primes similarly facilitated the processing of subsequent target orientations. Although the magnitude of the priming effects found in the present research was relatively small (D 10 ms), the effects were consistent both in Experiment 1 and Experiment 2 for RTs as well as for accuracy. In addition, the effect size estimated by means of the partial eta squared index reached relatively large values (g2p ¼ :67 for proximity and g2p ¼ :53 for similarity), thus suggesting that an important proportion of the variance in the dependent variable is explained by the experimental factor (Richardson, 2011). One of the possible causes for such a reduced magnitude of the priming effect might be related to the decision of only displaying the Gestalt patterns as primes. As previous studies have shown, masked priming effects produced by primes never presented as visible targets are significantly smaller in magnitude than primes previously displayed in a visible form (for reviews, Forster, 1998; Van den Bussche, Van den Noortgate, & Reynvoet, 2009). For instance, the results of a picture prime study by Van den Bussche, Notebaert, et al. (2009), which made use of masked line drawings never presented as targets, showed effects size of about 8–9 ms, in a similar manner to our data. In any event, this control allows discarding alternative explanations of the priming effects and obtaining a more conservative measure of a genuine subliminal grouping effect. Our findings are compatible with recent previous studies reporting Gestalt effects in absence of awareness with other demonstrations of perceptual organization such as contour integration or illusory contours (Rosenthal & Humphreys, 2010; Wang et al., 2012). The present results also are compatible with the proposals of models for contour integration that have suggested that grouping is computed in early stages of the visual processing thorough V1–V2 interactions (Neumann & Sepp, 1999) or assumed an unconscious computation of object boundaries by means of a Boundary Contour System that extracts information about edges and boundaries of the objects at an early processing stage that Grossberg and Mingolla’s model (1985a, 1985b). From a more general view, the present contribution is coherent with studies supporting perceptual grouping without attention (Chan & Chua, 2003; Driver et al., 2001; Kimchi & Razpurker-Apfeld, 2004; Lamy et al., 2006; Mitroff & Scholl, 2005; Russell & Driver, 2005). Taken together, all these results showed the automatic and flexible nature of perceptual organization operations in the human visual system and suggest that, even when visual processing is disrupted or hindered, organizational processes can proceed efficiently to structure the visual input (Mitroff & Scholl, 2005; Shomstein, Kimchi, Hammer, & Behrmann, 2010). Acknowledgment This work was supported by Grant 2012V/PUNED/0009 from the UNED to PRM and DL. References Abrams, R. L., & Greenwald, A. G. (2000). Parts outweigh the whole (word) in unconscious analysis of meaning. Psychological Science, 11, 118–124. Abrams, R. L., Klinger, M. R., & Greenwald, A. G. (2002). Subliminal words activate semantic categories (not automated motor responses). Psychonomic Bulletin & Review, 9, 100–106. Ben-Av, M. B., & Sagi, D. (1995). Perceptual grouping by similarity and proximity: Experimental results can be predicted by intensity autocorrelations. Vision Research, 35, 853–866. Breitmeyer, B. G., & Tapia, E. (2011). Roles of contour and surface processing in microgenesis of object perception and visual consciousness. Advances in Cognitive Psychology, 7, 61–74. Chan, W. Y., & Chua, F. K. (2003). Grouping with and without attention. Psychonomic Bulletin & Review, 10, 932–938. Damian, M. F. (2001). Congruity effects evoked by subliminally presented primes: Automaticity rather than semantic processing. Journal of Experimental Psychology: Human Perception and Performance, 27, 154–165. Dehaene, S., & Changeux, J. P. (2011). Experimental and theoretical approaches to conscious processing. Neuron, 70, 200–227. Dehaene, S., Changeux, J. P., Naccache, L., Sackur, J., & Sergent, C. (2006). Conscious, preconscious, and subliminal processing: A testable taxonomy. Trends in Cognitive Sciences, 10, 204–211. Driver, J., Davis, G., Russell, C., Turatto, M., & Freeman, E. (2001). Segmentation, attention and phenomenal visual objects. Cognition, 80, 61–95.
8
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
Eimer, M., & Schlaghecken, F. (1998). Effects of masked stimuli on motor activation: Behavioral and electrophysiological evidence. Journal of Experimental Psychology: Human Perception and Performance, 24, 1737–1747. Forster, K. I. (1998). The pros and cons of masked priming. Journal of Psycholinguistic Research, 27, 203–233. Grossberg, S., & Mingolla, E. (1985a). Neural dynamics of perceptual grouping: Textures, boundaries, and emergent segmentations. Perception & Psychophysics, 38, 141–171. Grossberg, S., & Mingolla, E. (1985b). Neural dynamics of form perception: Boundary completion, illusory figures, and neon color spreading. Psychological Review, 92, 173–211. Han, S., Ding, Y., & Song, Y. (2002). Neural mechanisms of perceptual grouping in humans as revealed by high density event related potentials. Neuroscience Letters, 319, 29–32. Han, S., Song, Y., Ding, Y., Yund, E. W., & Woods, D. L. (2001). Neural substrates for visual perceptual grouping in humans. Psychophysiology, 38, 926–935. Julesz, B. (1981). Textons, the elements of texture perception, and their interactions. Nature, 290, 91–97. Kahneman, D., & Henik, A. (1981). Perceptual organization and attention. In M. Kubovy & J. R. Pomerantz (Eds.), Perceptual organization (pp. 181–221). Hillsdale, NJ: Erlbaum. Kiesel, A., Kunde, W., Pohl, C., & Hoffmann, J. (2006). Priming from novel masked stimuli depends on target set size. Advances in Cognitive Psychology, 2, 37–45. Kim, C. Y., & Blake, R. (2005). Psychophysical strategies for rendering the normally visible ‘‘invisible’’. Trends in Cognitive Sciences, 9, 381–388. Kimchi, R., & Razpurker-Apfeld, I. (2004). Perceptual grouping and attention: Not all groupings are equal. Psychonomic Bulletin & Review, 11, 687–696. Koch, C., & Tsuchiya, N. (2007). Attention and consciousness: Two distinct brain processes. Trends in Cognitive Sciences, 11, 16–22. Kouider, S., & Dehaene, S. (2007). Levels of processing during non-conscious perception: A critical review. Philosophical Transactions of the Royal Society of London B, 362, 857–875. Kubovy, M., Holcombe, A. O., & Wagemans, J. (1998). On the lawfulness of grouping by proximity. Cognitive Psychology, 35, 71–98. Kurylo, D. D. (1997). Time course of perceptual grouping. Perception & Psychophysics, 59, 142–147. Lamme, V. A. F. (2003). Why visual attention and awareness are different. Trends in Cognitive Sciences, 7, 12–18. Lamy, D., Segal, H., & Ruderman, L. (2006). Grouping does not require attention. Perception & Psychophysics, 68, 17–31. Luna, D., & Montoro, P. R. (2008). The distortion of spatial relationships between local elements in hierarchical patterns decreases the global advantage effect. Psychological Research, 72, 168–175. Luna, D., & Montoro, P. R. (2011). Interactions between intrinsic principles of similarity and proximity and extrinsic principle of common region in visual perception. Perception, 40, 1467–1477. Mack, A., & Rock, I. (1998). Inattentional blindness. Cambridge, MA: MIT Press. Mack, A., Tang, B., Tuma, R., Kahn, S., & Rock, I. (1992). Perceptual organization and attention. Cognitive Psychology, 24, 475–501. Merikle, P. M., Smilek, D., & Eastwood, J. D. (2001). Perception without awareness: Perspectives from cognitive psychology. Cognition, 79, 115–134. Mitroff, S. R., & Scholl, B. J. (2005). Forming and updating object representations without awareness: Evidence from motion-induced blindness. Vision Research, 45, 961–967. Moore, C. M., & Egeth, H. (1997). Perception without attention: Evidence of grouping under conditions of inattention. Journal of Experimental Psychology: Human Perception and Performance, 23, 339–352. Naccache, L., & Dehaene, S. (2001). Unconscious semantic priming extends to novel unseen stimuli. Cognition, 80, 215–229. Neisser, U. (1967). Cognitive psychology. East Norwalk, CT: Appleton-Century-Crofts. Neumann, O., & Klotz, W. (1994). Motor responses to non-reportable masked stimuli: Where is the limit of direct motor specification? In C. Umilta & M. Moscovitch (Eds.), Attention and performance XV (pp. 123–150). Cambridge, MA: MIT Press. Neumann, H., & Sepp, W. (1999). Recurrent V1–V2 interaction in early visual boundary processing. Biological Cybernetics, 81, 425–444. Ortells, J. J., Marí-Beffa, P., & Plaza-Ayllón, V. (2013). Unconscious congruency priming from unpractised words is modulated by prime–target semantic relatedness. Journal of Experimental Psychology: Learning, Memory and Cognition, 39, 394–413. Palmer, S. E. (1992). Common region: A new principle of perceptual grouping. Cognitive Psychology, 24, 433–447. Palmer, S. E. (1999). Vision science. Photons to phenomenology. Cambridge, MA: MIT Press. Palmer, S. E., & Rock, I. (1994). Rethinking perceptual organization: The role of uniform connectedness. Psychonomic Bulletin & Review, 1, 29–55. Quinlan, P. T., & Wilton, R. L. (1998). Grouping by proximity or similarity? Competition between the Gestalt principles in vision. Perception, 27, 417–430. Reingold, E. M., & Merikle, P. M. (1988). Using direct and indirect measures to study perception without awareness. Perception & Psychophysics, 44, 563–575. Reingold, E. M., & Merikle, P. M. (1993). Theory and measurement in the study of unconscious processes. In M. Davies & G. Humphreys (Eds.), Consciousness (pp. 40–57). Oxford: Basil Blackwell Ltd.. Richardson, J. T. E. (2011). Eta squared and partial eta squared as measures of effect size in educational research. Educational Research Review, 6, 135–147. Rock, I., Linnett, C. M., Grant, P., & Mack, A. (1992). Perception without attention: Results of a new method. Cognitive Psychology, 24, 502–534. Rosenthal, O., & Humphreys, G. (2010). Perceptual organization without perception: The subliminal learning of global. Psychological Science, 21, 1751–1758. Russell, C., & Driver, J. (2005). New indirect measures of ‘inattentive’ visual grouping in a change-detection task. Perception & Psychophysics, 67, 606–623. Shomstein, S., Kimchi, R., Hammer, M., & Behrmann, M. (2010). Perceptual grouping operates independently of attentional selection: Evidence from hemispatial neglect. Attention, Perception, & Psychophysics, 72, 607–618. Spillmann, L. (2009). Phenomenology and neurophysiological correlations: Two approaches to perception research. Vision Research, 49, 1507–1521. Treisman, A. (1982). Perceptual grouping and attention in visual search for features and for objects. Journal of Experimental Psychology: Human Perception and Performance, 8, 194–214. Treisman, A. (1985). Preattentive processing in vision. Computer Vision, Graphics, and Image Processing, 31, 156–177. Treisman, A. (1988). Features and objects: The fourteenth Bartlett Memorial Lecture. Quarterly Journal of Experimental Psychology, 40A, 201–237. Van den Bussche, E., Notebaert, K., & Reynvoet, B. (2009). Masked primes can be genuinely semantically processed: A picture prime study. Experimental Psychology, 56, 295–300. Van den Bussche, E., Van den Noortgate, W., & Reynvoet, B. (2009). Mechanisms of masked priming: A meta-analysis. Psychological Bulletin, 135, 452–477. Wagemans, J., Elder, J. H., Kubovy, M., Palmer, S. E., Peterson, M. A., Singh, M., et al (2012). A century of Gestalt psychology in visual perception: I. Perceptual grouping and figure–ground organization. Psychological Bulletin, 138, 1172–1217. Wang, L., Weng, X., & He, S. (2012). Perceptual grouping without awareness: Superiority of Kanizsa triangle in breaking interocular suppression. PLoS One, 7(6), e40106. http://dx.doi.org/10.1371/journal.pone.0040106. Wertheimer, M. (1923). Untersuchungen zur Lehre von der Gestalt II. Psycholosgische Forschung, 4, 301–350.
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Subliminal Gestalt grouping: Evidence of perceptual grouping by proximity and similarity in absence of conscious perception Pedro R. Montoro a,⇑, Dolores Luna a, Juan J. Ortells b a b
Departamento de Psicología Básica 1, UNED, Spain Departamento de Psicología, Universidad de Almeria, Spain
a r t i c l e
i n f o
Article history: Received 30 July 2013 Available online 8 February 2014 Keywords: Perceptual grouping Visual masking Subliminal priming Proximity Similarity
a b s t r a c t Previous studies making use of indirect processing measures have shown that perceptual grouping can occur outside the focus of attention. However, no previous study has examined the possibility of subliminal processing of perceptual grouping. The present work steps forward in the study of perceptual organization, reporting direct evidence of subliminal processing of Gestalt patterns. In two masked priming experiments, Gestalt patterns grouped by proximity or similarity that induced either a horizontal or vertical global orientation of the stimuli were presented as masked primes and followed by visible targets that could be congruent or incongruent with the orientation of the primes. The results showed a reliable priming effect in the complete absence of prime awareness for both proximity and similarity grouping principles. These findings suggest that a phenomenal report of the Gestalt pattern is not mandatory to observe an effect on the response based on the global properties of Gestalt stimuli. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction The world we perceive consists of objects and their interrelations coherently arranged in scenes. The processes of perceptual organization are responsible for structuring the retinal mosaic into the global stimuli of perceived objects (Palmer, 1999). One century ago, the Gestalt school of psychology was the first to deal with the problem of perceptual organization (see Wagemans et al., 2012, for a review). According to Max Wertheimer, organization is basically composed of grouping and segregation processes. The well-know principles of grouping describe the stimulus factors that determine the visual grouping of discrete elements, including proximity, similarity, common fate, good continuation, closure (Wertheimer, 1923), and, more recently, other new principles such as common region, connectedness or synchrony have been proposed (Palmer, 1992; Palmer & Rock, 1994). Classical cognitive theories of visual processing have assumed that perceptual grouping occurs preattentively in absence of attention, producing segmented perceptual units or ‘proto-objects’ for analysis by subsequent processes (Julesz, 1981; Kahneman & Henik, 1981; Neisser, 1967; Treisman, 1982, 1985, 1988). This assumption was challenged by the results of studies showing that perceptual grouping is not perceived under strict conditions of inattention (Mack & Rock, 1998; Mack, Tang, Tuma, Kahn, & Rock, 1992; Rock, Linnett, Grant, & Mack, 1992). However, Moore and Egeth (1997), using inattentional conditions similar to that of Mack and colleagues but crucially introducing on-line measures of unattended processing, showed that grouping could occur outside the focus of attention. Moore and Egeth suggested that grouped patterns could ⇑ Corresponding author. Address: Departamento de Psicología Básica 1, Facultad de Psicología, UNED, C/ Juan del Rosal 10, 28040 Madrid, Spain. E-mail address: [email protected] (P.R. Montoro). http://dx.doi.org/10.1016/j.concog.2014.01.004 1053-8100/Ó 2014 Elsevier Inc. All rights reserved.
2
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
have been perceived without attention but not encoded in memory. Later studies using similar methodological strategies have provided further evidence that perceptual grouping does not require attention (Chan & Chua, 2003; Driver, Davis, Russell, Turatto, & Freeman, 2001; Kimchi & Razpurker-Apfeld, 2004; Lamy, Segal, & Ruderman, 2006; Mitroff & Scholl, 2005; Russell & Driver, 2005). Interestingly, in a similar way, two recent studies have reported evidence for subliminal processing of other types of perceptual organization such as contour integration (Rosenthal & Humphreys, 2010) and illusory contour completion (Wang, Weng, & He, 2012). Rosenthal and Humphreys found a significant subliminal learning of global contours generated by integration of Gabor elements, despite being irrelevant to the main task. Using a suppression time paradigm, Wang et al. (2012) have recently observed that Kanizsa-like stimuli can be processed without awareness. These results are impressive examples of unconscious organization of the discrete elements to form emergent patterns that are qualitatively different from the sum of their parts. In addition, previous works provide a theoretical basis for a grouping effect occurring at an unconscious level. Several models for contour processing have analyzed the possible mechanism underlying grouping effects and have suggested that grouping could occur at an unconscious level. For example, Grossberg and Mingolla (1985a, 1985b) proposed a model of recurrent networks that considers unconscious computation of boundary contours. The model distinguishes between a Boundary Contour System (BCS) and a Feature Contour System (FCS) that extract different information concerning contour at an early processing stage. Information on edges and boundaries (i.e. the form outline of an object) is processed in the BCS whereas information on surface features (i.e. color or brightness) that fills in the area delimited by the BCS is processed in the FCS. The model assumes that the processing of form in BCS, which extracts edge information, occurs at non-conscious levels. Neumann and Sepp (1999) developed a neuropsychological model for contour processing based on interactions between V1 and V2. The model suggests an early occurrence for grouping in the stream of these interactions. Further studies in the field of visual neuroscience (see Breitmeyer & Tapia, 2011, for a review) assuming separate pathways for the processing of form and surface and using a metacontrast masking procedure showed that the conscious processing of form requires previous fill in of surface properties. The results also suggest that form processing proceeds faster than surface processing at unconscious levels whereas at conscious level it proceeds slower than surface processing. However, to our knowledge, no previous study has examined the organization of visual elements into global patterns by means of Gestalt principles of grouping in absence of conscious perception. The present work steps forward in the study of perceptual grouping processes examining the possibility of a subliminal processing of Gestalt patterns generated by the action of grouping principles. This issue is relevant, as recent theoretical contributions have differentiated between consciousness and attention, and, by extension, between unconscious and inattentional processes (Dehaene & Changeux, 2011; Dehaene, Changeux, Naccache, Sackur, & Sergent, 2006; Koch & Tsuchiya, 2007; Lamme, 2003). In this line, Dehaene and collaborators have proposed a tripartite distinction of subliminal, preconscious and conscious processing (see Dehaene & Changeux, 2011, for a review). According to this taxonomy, a subliminal stimulus is an invisible, undetectable one, even with focused attention. In contrast, a preconscious stimulus is potentially visible but is not consciously perceived due to distraction or inattention. From this view, the results of studies supporting perceptual grouping without attention are samples of preconscious processing because their procedures involve supraliminal patterns that could be reported if they were attended (e.g., see Full Attention condition in Moore and Egeth (1997)’s experiments). This implies that the findings of grouping without attention cannot be directly generalized to a hypothetical subliminal processing of grouping. Consequently, an investigation on the possibility of Gestalt grouping under exhaustive subliminal conditions is necessary to extend our knowledge of perceptual organization operations in the human visual system. Our aim is to determine whether a phenomenal report of the patterns is mandatory to observe an effect on the response based on the global properties of grouped stimuli. This goal is especially relevant considering that Gestalt psychologists focused their attention on phenomenology and, coherently, proposed their classical principles of grouping based exclusively on subjective measures obtained by phenomenal demonstrations (Lamy et al., 2006; Spillmann, 2009). In order to achieve our purpose, we had to make two important decisions relative to (1) the subliminal method to be implemented as well as (2) the representative principles of perceptual grouping to be included in the study. Regarding the subliminal method, we used a masked priming task, the most frequently used technique for studying the extent of processing during subliminal perception (see Kim & Blake, 2005; Kouider & Dehaene, 2007, for reviews). Specifically, a double masking procedure combining forward and backward masks was applied to the grouped pattern. As principles of grouping, we selected proximity and luminance similarity (see Fig. 1), two classical Gestalt laws already described by Wertheimer (1923) and widely studied in numerous previous works (e.g., Ben-Av & Sagi, 1995; Kubovy, Holcombe, & Wagemans, 1998; Luna & Montoro, 2011; Quinlan & Wilton, 1998). Note that these principles base their capacity for grouping on the analysis of different properties of the elements, i.e., spatial relationships between single units in the case of proximity and relative levels of luminance concerning similarity. Due to the temporal and computational differences between grouping principles showed by previous studies (Han, Ding, & Song, 2002; Han, Song, Ding, Yund, & Woods, 2001; Kurylo, 1997; Luna & Montoro, 2008), this selection might provide us with a more extensive exploration of the boundary conditions for observing a subliminal processing of Gestalt patterns. In two different experiments, the grouping principles were applied to disk lattices to form horizontally or vertically oriented global patterns, which could be congruent or incongruent with the orientation of the target stimuli displayed afterwards. Thus, the priming effect was based on the global orientation of the stimuli and not on any local characteristic. Several kinds of targets with different physical appearances were displayed in order to force participants to base their response exclusively on the orientation of the stimuli.
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
3
Fig. 1. Stimuli and sequence of events used in the experiments. (a) Grouped patterns displayed as primes in Experiment 1. (b) Grouped patterns used in Experiment 2. (c) Uniform patterns used as primes in half of the trials in both forced-choice prime detection tasks.
A crucial control parameter included in our priming procedure was to avoid displaying the Gestalt patterns as visible targets, and we used them exclusively as masked primes. Previous studies have suggested that subliminal priming effects could be due to the fact that the masked primes had been also displayed as conscious targets in other phases of the experiment (Abrams & Greenwald, 2000; Damian, 2001; Eimer & Schlaghecken, 1998; Neumann & Klotz, 1994). However, other studies have disconfirmed this hypothesis by showing subliminal priming effects for primes that were never presented as visible targets (Abrams, Klinger, & Greenwald, 2002; Kiesel, Kunde, Pohl, & Hoffmann, 2006; Naccache & Dehaene, 2001; Ortells, Marí-Beffa, & Plaza-Ayllón, 2013; Van den Bussche, Notebaert, & Reynvoet, 2009). Therefore, we considered that it was necessary to control this experimental confound in order to determine whether genuine subliminal Gestalt grouping is possible. Thus, the Gestalt patterns were exclusively displayed as primes under masked conditions of presentation. 2. Experiment 1: Grouping by proximity 2.1. Method 2.1.1. Participants Thirty-eight undergraduate students (five men; age range = 19–59 years, M = 32.02, SD = 10.86) from the UNED participated in the main experiment consisting of two consecutive tasks: (1) a masked priming task and (2) a prime visibility discrimination task. All of them had normal or corrected-to-normal vision and received course credits for their participation. A different sample of twenty-three undergraduate students (ten men; age range = 20–64 years, M = 35.5, SD = 12.1) from the UNED were recruited for an additional control experiment consisting of a forced-choice prime detection task. All of them had normal or corrected-to-normal vision and received course credits for their participation. The experimental procedures were approved by the Local Ethics Committee and conform to the Declaration of Helsinki.
4
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
2.1.2. Stimuli and apparatus The stimuli were displayed on a 19-in. LCD–LED Samsung 943N color monitor with a 75-Hz refresh rate, a 5:4 aspect ratio, a light gray background (RGB: 198; 36 cd/m2) and a resolution of 1024 768 controlled by a computer running E-Prime 1.2 software (Psychology Software Tools, 1996–2002). Viewing distance was approximately 57 cm. All the stimuli were displayed on a central dark gray square background (RGB: 95; 6 cd/m2) subtending a visual angle of 5.3° 5.3°. The Gestalt patterns presented as primes consisted of disk lattices of light gray (RGB: 191; 33 cd/m2) or black (RGB: 0; 0 cd/m2) elements. Each local element subtended a visual angle of 0.40° 0.40°. The proximity-grouping patterns consisted of 8 6 elements, or vice versa, which were arranged in rows or columns so that the distances between near or remote rows or columns were 0.40° and 1.0°, respectively. The masking patterns were slightly larger random light gray and black disk lattices arranged in a 13 13-matrix. There was the same proportion of black and light gray disks in the masks. The target stimuli consisted of five continuous zigzag or straight lines disposed in a vertical or horizontal orientation. There were six different target stimuli in order to force participants to base their response on the global orientation of the stimuli regardless of the physical appearance of the stimulus (see Fig. 1). 2.2. Procedure and design Participants in the main experiment were tested individually in a dimly lit room in two consecutive sessions, one for the masked priming task, and another one for the prime visibility discrimination task. In the masked priming task, participants carried out a forced-choice reaction time (RT) task. Subjects were told that they would see target lines displayed on the screen, and that they would have to indicate the vertical or horizontal orientation by pressing one of two response buttons (number 1 or 2, respectively) with their middle and index fingers of the dominant hand as fast as possible but avoid making mistakes. Crucially, subjects were not told about the masked primes, but were merely informed that each trial would begin with the presentation of a ‘flash signal’ to warn them that the target stimulus was going to appear afterwards. The sequence of events in a trial for each grouping condition is depicted in Fig. 1. Each trial started with a mask pattern; 147 ms later, a Gestalt grouped pattern was presented for 53 ms; then, a second mask pattern, different from the first one, was displayed during 80 ms; and, finally, the target stimulus appeared on the screen and remained until response. After the end of the trial, a pause of 800 ms ensued before the start of the next trial. There were a practise block with 24 trials and seven experimental blocks consisting of 96 trials each, for a total of 672 experimental trials (one half congruent trials and one half incongruent ones). Feedback was provided only for the practise trials. Immediately after the end of the priming task, participants were systematically asked several questions concerning the ‘flash signal’, namely: Have you noticed any strange or curious thing during the experiment? Have you seen any strange or curious thing related to the ‘flash signal’? Have you seen any figure, configuration or recognizable stimulus during the ‘flash signal’? After that, they were fully informed of the nature of the ‘flash signal’ and were asked to perform a forced-choice prime discrimination task designed to obtain an objective index (i.e., d0 ) of prime visibility. In this task, consisting of 12 practise trials followed by 120 experimental trials, participants were instructed to pay attention to the prime stimulus that was displayed between the two masks, and to perform a forced-choice discrimination task indicating the horizontal or vertical orientation of that stimulus. For this task, the sequence of events was identical to that of the priming task, with the following exceptions: (1) the target stimuli were displayed during a fixed time of 507 ms, (2) after the target offset, a horizontal and a vertical rectangle with the corresponding labels ‘HORIZONTAL’ and ‘VERTICAL’ inside were displayed on the screen so that the participants provided their response by clicking the mouse on the rectangle containing the chosen orientation without response time demand; and (3) the trials were self-administered in order to ensure that participants were as ready as possible to discriminate the masked prime. Participants were instructed to try to be as accurate as possible and to guess on trials in which they could not identify the primes. The additional sample of participants performing the forced-choice prime detection task also received exhaustive information about the primes and the masking procedure. The procedure in this task was identical to that of the forced-choice discrimination task with the following exceptions: (1) two new uniform patterns were used (see Fig. 1) consisting of alternating gray and black disks distributed evenly across the lattice; (2) half of the trials presented the uniform patterns as primes, and the other half was composed of the grouped patterns; and (3) participants were instructed to discriminate between grouped and uniform patterns by clicking the mouse on the rectangle containing the selected choice without response time demand. The design included a within-subjects factor comparing congruent versus incongruent trials as a function of the horizontal–vertical congruency between prime and target. There was the same proportion of congruent as incongruent trials in each single block of the experiment. 2.3. Results 2.3.1. Priming task Mean RTs of correct responses and mean accuracy were submitted to separate analyses of variance (ANOVAs) with Prime–Target Congruency as within-subjects factor (2 levels: Congruent or Incongruent). Mean RTs and mean accuracy (hit rates) as a function of this factor are reported in Table 1.
5
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8 Table 1 Mean (SD) RTs (in ms), and mean accuracy (hit rate) for the congruent and incongruent trials, and the amount of priming (incongruent–congruent) as a function of Experiment (grouping by proximity or luminance similarity).
*
Incongruent
Congruent
Priming effect
Experiment 1: Proximity 543 (56.9) .97 (0.01)
532 (58.1) .98 (0.01)
+11*** 1
Experiment 2: Similarity 533 (53.7) .97 (0.02)
524 (54.8) .98 (0.02)
+9*** 1*
p < .05. p < .001.
***
2.3.1.1. RT analysis. Inaccurate responses (on average 2.4%) and RTs greater than 1500 or less than 200 ms (2.6% of trials) were excluded from the RT analysis. There was a significant main effect of priming, F(1, 37) = 74.69, MSe = 28.80, p < .001, g2p ¼ :67, indicating that RTs in congruent trials were reliably shorter (532 ms) than those in incongruent trials (543 ms). 2.3.1.2. Accuracy analysis. Hit rates oscillated between 94.5% and 99.5%. The same ANOVA was conducted on accuracy rates, revealing a marginally significant effect of priming, F(1, 37) = 3.83, MSe < .001, p = .058, g2p ¼ :09. These results mirrored those of RTs (Congruent: 97.7%; Incongruent: 97.3%). 2.3.2. Prime visibility discrimination task To measure subjective visibility of the masked primes, participants were asked to report what patterns they had seen on the screen before the presentation of the target during the experiment. None of the participants reported having seen any horizontal- or vertical-oriented patterns before the presentation of the target. This finding suggests that participants had no subjective awareness of the primes. Mean accuracy in the prime visibility discrimination task across vertical and horizontal primes was .51 (SD = .03; individual rates from .42 to .58), which suggests that primes could not be classified above chance. A direct measure of prime visibility (d0 ) was calculated for each participant. The measures were obtained by treating one level (i.e., horizontal) as signal and the other level (i.e., vertical) as noise. The individual d0 values ranged between .42 and .54, and the overall mean was .02 (SD = .19). A single sample t-test showed that this d’ was not significantly different from zero, t (37) = .287, p = .78, thus suggesting that participants were not aware of the prime orientation in the main task. A linear regression analysis was further conducted to examine the relationship between individual priming effect and prime visibility across participants. For each participant, a priming effect (D = RTincongruent RTcongruent) was calculated and correlated with the corresponding value of the d0 in the forced-choice prime discrimination task. There was no significant linear relationship between the priming effect and prime visibility (R2 = .157, p = .35). In sum, the results revealed that the primes were largely invisible. However, it could be possible that masked stimuli were unidentifiable yet detectable (Kim & Blake, 2005). In our study, participants could have been aware that a grouped pattern appeared without being aware of its concrete orientation. To examine that possibility, in this and the next study, an additional control experiment consisting of a forced-choice detection task was conducted on a different sample of participants. 2.3.3. Forced-choice detection task Individual hit rates oscillated between .44 and .59 (M = .51, SD = .04). Individual d0 indexes were computed similarly to the forced-choice discrimination task, ranging between .49 and .39 (M = .01, SD = .24). A single sample t-test, t (22) = .524, p = .61, showed that mean performance was not significantly different from zero, suggesting that participants were not capable of detecting the presence of a grouped pattern under the masking conditions displayed. 3. Experiment 2: Grouping by similarity 3.1. Method 3.1.1. Participants Thirty-eight undergraduate students (three men; age range = 19–51 years, M = 29.24, SD = 10.5) from the UNED participated in the main experiment consisting of two consecutive tasks: (1) a masked priming task and (2) a prime visibility discrimination task. All of them had normal or corrected-to-normal vision and received course credits for their participation. A different sample of twenty-three undergraduate students (ten men; age range = 20–64 years, M = 35.5, SD = 12.1) from the UNED were recruited for an additional control experiment consisting of a forced-choice prime detection task. All of them had normal or corrected-to-normal vision and received course credits for their participation. The experimental procedures were approved by the Local Ethics Committee and conform to the Declaration of Helsinki.
6
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
3.1.2. Stimuli and apparatus The stimuli and apparatus were identical to those of Experiment 1, with the sole exception that the Gestalt patterns consisted of a 6 6 array, forming rows or columns with elements of identical luminance. The horizontal and vertical distance between adjacent local elements was 0.50°. 3.1.3. Procedure and design The procedure and design were identical to those of Experiment 1. 3.2. Results 3.2.1. Priming task Mean RTs of correct responses and mean accuracy were submitted to separate ANOVA with Prime–Target Congruency as within-subjects factor (2 levels: Congruent or Incongruent). Mean RTs and mean accuracy (hit rates) as a function of this factor are reported in Table 1. 2.3.1.3. RT analysis. Inaccurate responses (on average 2.6%), and trials with RTs greater than 1500 or less than 200 ms (2.6% of trials) were excluded from the RT analysis. There was a significant main effect of priming, F(1, 37) = 41.86, MSe = 36.42, p < .001, g2p ¼ :53, indicating that RTs in congruent trials were reliably shorter (533 ms) than those in incongruent trials (542 ms). 2.3.1.4. Accuracy analysis. Hit rates oscillated between 92% and 99.5%. The same ANOVA was conducted on accuracy rates, revealing a significant effect of priming, F(1, 37) = 4.18, MSe < .001, p = .048, g2p ¼ :10. These results mirrored those of RTs (Congruent: 97.5%; Incongruent: 97.1%). 3.2.2. Prime visibility discrimination task As in Experiment 1, none of the participants reported having seen any horizontal- or vertical-oriented patterns before the presentation of the target. Mean accuracy in the prime visibility task across vertical and horizontal primes was .50 (SD = .04; individual rates ranged from .40 to .57). The individual d0 values oscillated between .53 and .38, and the overall mean was .01 (SD = .22). A single sample t-test showed that this d0 was not significantly different from zero, suggesting that participants were not aware of the prime orientation in the main task, t (37) = .287, p = .78. Furthermore, there was no significant linear relationship between priming effect and prime visibility (R2 = .121, p > .45). 3.2.3. Forced-choice detection task The procedure and the analysis were identical to those of Experiment 1. Individual hit rates oscillated between .44 and .59 (M = .51, SD = .04). Individual d0 indexes ranged between .49 and .39 (M = .01, SD = .24). A single sample t-test, t (22) = .524, p = .61, showed that mean performance was not significantly different from zero, supporting that participants could not detect the presence of a grouped pattern under the masking conditions displayed. 4. Combined analysis of Experiments 1 and 2 Finally, a combined analysis of Experiment 1 and Experiment 2 was conducted in order to compare the effect of each grouping principle. Previous studies have suggested temporal and computational differences between grouping principles (Han et al., 2001, 2002; Kurylo, 1997; Luna & Montoro, 2008). Thus, we examined whether grouping by proximity versus by similarity also differ when they are processed under subliminal conditions. To test this possibility, mean RTs and mean accuracy were submitted to two separate mixed ANOVAs with Perceptual Grouping Condition as a between-subjects factor (2 levels: Proximity or Similarity), and Prime–Target Congruency as within-subjects factor (2 levels: Congruent or Incongruent). 2.4. RT analysis There was a significant main effect of priming, F(1, 74) = 111.9, MSe = 26.93, p < .001, g2p ¼ :60, indicating that RTs in congruent trials were shorter (528 ms) than those in incongruent trials (538 ms). Neither the between-subjects factor nor the interaction was significant (F < 1). 2.5. Accuracy analysis The same ANOVA was conducted on accuracy rates, revealing a significant effect of priming, F(1, 74) = 8, MSe < .001, p < .01, g2p ¼ :10, showing a higher hit rate in congruent trials (97.6%) than in incongruent trials (97.2%). Neither the between-subjects factor nor the interaction was significant (F < 1). These results mirrored those of RTs.
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
7
5. Discussion Can perceptual grouping occur unconsciously in absence of a phenomenal report of the Gestalt pattern displayed? Previous studies making use of indirect processing measures have shown that perceptual grouping can occur even under the strictest conditions of inattention proposed by Mack and Rock’s (1998) paradigm (Chan & Chua, 2003; Kimchi & Razpurker-Apfeld, 2004; Lamy et al., 2006; Mitroff & Scholl, 2005; Russell & Driver, 2005). However, those results do not allow us to determine whether perceptual grouping occurs under subliminal conditions (Dehaene & Changeux, 2011). For this reason, the aim of the present study was to examine whether perceptual grouping requires consciousness. According to a dissociation paradigm (Merikle, Smilek, & Eastwood, 2001; Reingold & Merikle, 1988, 1993), we had to verify two empirical aspects in order to answer this question. First, we ruled out that grouping was processed by conscious vision, as masked grouped patterns were largely invisible, as suggested both by subjective and objective (prime discrimination and detection tasks) measures. Second, we found that masked visual groups produced significant effects on an indirect measure of processing, that is, the orientation-based priming effect. Thus, our results provide the first evidence to support that grouping principles can be applied to discrete visual elements to form Gestalt groups under subliminal conditions, at least for the cases of proximity and luminance similarity examined in the current work. This implies that a phenomenal report of the pattern is not mandatory to observe an effect on the response based on the global properties of grouped stimuli. In addition, similar effects for proximity and similarity grouped patterns were observed in the present study, showing that both grouped primes similarly facilitated the processing of subsequent target orientations. Although the magnitude of the priming effects found in the present research was relatively small (D 10 ms), the effects were consistent both in Experiment 1 and Experiment 2 for RTs as well as for accuracy. In addition, the effect size estimated by means of the partial eta squared index reached relatively large values (g2p ¼ :67 for proximity and g2p ¼ :53 for similarity), thus suggesting that an important proportion of the variance in the dependent variable is explained by the experimental factor (Richardson, 2011). One of the possible causes for such a reduced magnitude of the priming effect might be related to the decision of only displaying the Gestalt patterns as primes. As previous studies have shown, masked priming effects produced by primes never presented as visible targets are significantly smaller in magnitude than primes previously displayed in a visible form (for reviews, Forster, 1998; Van den Bussche, Van den Noortgate, & Reynvoet, 2009). For instance, the results of a picture prime study by Van den Bussche, Notebaert, et al. (2009), which made use of masked line drawings never presented as targets, showed effects size of about 8–9 ms, in a similar manner to our data. In any event, this control allows discarding alternative explanations of the priming effects and obtaining a more conservative measure of a genuine subliminal grouping effect. Our findings are compatible with recent previous studies reporting Gestalt effects in absence of awareness with other demonstrations of perceptual organization such as contour integration or illusory contours (Rosenthal & Humphreys, 2010; Wang et al., 2012). The present results also are compatible with the proposals of models for contour integration that have suggested that grouping is computed in early stages of the visual processing thorough V1–V2 interactions (Neumann & Sepp, 1999) or assumed an unconscious computation of object boundaries by means of a Boundary Contour System that extracts information about edges and boundaries of the objects at an early processing stage that Grossberg and Mingolla’s model (1985a, 1985b). From a more general view, the present contribution is coherent with studies supporting perceptual grouping without attention (Chan & Chua, 2003; Driver et al., 2001; Kimchi & Razpurker-Apfeld, 2004; Lamy et al., 2006; Mitroff & Scholl, 2005; Russell & Driver, 2005). Taken together, all these results showed the automatic and flexible nature of perceptual organization operations in the human visual system and suggest that, even when visual processing is disrupted or hindered, organizational processes can proceed efficiently to structure the visual input (Mitroff & Scholl, 2005; Shomstein, Kimchi, Hammer, & Behrmann, 2010). Acknowledgment This work was supported by Grant 2012V/PUNED/0009 from the UNED to PRM and DL. References Abrams, R. L., & Greenwald, A. G. (2000). Parts outweigh the whole (word) in unconscious analysis of meaning. Psychological Science, 11, 118–124. Abrams, R. L., Klinger, M. R., & Greenwald, A. G. (2002). Subliminal words activate semantic categories (not automated motor responses). Psychonomic Bulletin & Review, 9, 100–106. Ben-Av, M. B., & Sagi, D. (1995). Perceptual grouping by similarity and proximity: Experimental results can be predicted by intensity autocorrelations. Vision Research, 35, 853–866. Breitmeyer, B. G., & Tapia, E. (2011). Roles of contour and surface processing in microgenesis of object perception and visual consciousness. Advances in Cognitive Psychology, 7, 61–74. Chan, W. Y., & Chua, F. K. (2003). Grouping with and without attention. Psychonomic Bulletin & Review, 10, 932–938. Damian, M. F. (2001). Congruity effects evoked by subliminally presented primes: Automaticity rather than semantic processing. Journal of Experimental Psychology: Human Perception and Performance, 27, 154–165. Dehaene, S., & Changeux, J. P. (2011). Experimental and theoretical approaches to conscious processing. Neuron, 70, 200–227. Dehaene, S., Changeux, J. P., Naccache, L., Sackur, J., & Sergent, C. (2006). Conscious, preconscious, and subliminal processing: A testable taxonomy. Trends in Cognitive Sciences, 10, 204–211. Driver, J., Davis, G., Russell, C., Turatto, M., & Freeman, E. (2001). Segmentation, attention and phenomenal visual objects. Cognition, 80, 61–95.
8
P.R. Montoro et al. / Consciousness and Cognition 25 (2014) 1–8
Eimer, M., & Schlaghecken, F. (1998). Effects of masked stimuli on motor activation: Behavioral and electrophysiological evidence. Journal of Experimental Psychology: Human Perception and Performance, 24, 1737–1747. Forster, K. I. (1998). The pros and cons of masked priming. Journal of Psycholinguistic Research, 27, 203–233. Grossberg, S., & Mingolla, E. (1985a). Neural dynamics of perceptual grouping: Textures, boundaries, and emergent segmentations. Perception & Psychophysics, 38, 141–171. Grossberg, S., & Mingolla, E. (1985b). Neural dynamics of form perception: Boundary completion, illusory figures, and neon color spreading. Psychological Review, 92, 173–211. Han, S., Ding, Y., & Song, Y. (2002). Neural mechanisms of perceptual grouping in humans as revealed by high density event related potentials. Neuroscience Letters, 319, 29–32. Han, S., Song, Y., Ding, Y., Yund, E. W., & Woods, D. L. (2001). Neural substrates for visual perceptual grouping in humans. Psychophysiology, 38, 926–935. Julesz, B. (1981). Textons, the elements of texture perception, and their interactions. Nature, 290, 91–97. Kahneman, D., & Henik, A. (1981). Perceptual organization and attention. In M. Kubovy & J. R. Pomerantz (Eds.), Perceptual organization (pp. 181–221). Hillsdale, NJ: Erlbaum. Kiesel, A., Kunde, W., Pohl, C., & Hoffmann, J. (2006). Priming from novel masked stimuli depends on target set size. Advances in Cognitive Psychology, 2, 37–45. Kim, C. Y., & Blake, R. (2005). Psychophysical strategies for rendering the normally visible ‘‘invisible’’. Trends in Cognitive Sciences, 9, 381–388. Kimchi, R., & Razpurker-Apfeld, I. (2004). Perceptual grouping and attention: Not all groupings are equal. Psychonomic Bulletin & Review, 11, 687–696. Koch, C., & Tsuchiya, N. (2007). Attention and consciousness: Two distinct brain processes. Trends in Cognitive Sciences, 11, 16–22. Kouider, S., & Dehaene, S. (2007). Levels of processing during non-conscious perception: A critical review. Philosophical Transactions of the Royal Society of London B, 362, 857–875. Kubovy, M., Holcombe, A. O., & Wagemans, J. (1998). On the lawfulness of grouping by proximity. Cognitive Psychology, 35, 71–98. Kurylo, D. D. (1997). Time course of perceptual grouping. Perception & Psychophysics, 59, 142–147. Lamme, V. A. F. (2003). Why visual attention and awareness are different. Trends in Cognitive Sciences, 7, 12–18. Lamy, D., Segal, H., & Ruderman, L. (2006). Grouping does not require attention. Perception & Psychophysics, 68, 17–31. Luna, D., & Montoro, P. R. (2008). The distortion of spatial relationships between local elements in hierarchical patterns decreases the global advantage effect. Psychological Research, 72, 168–175. Luna, D., & Montoro, P. R. (2011). Interactions between intrinsic principles of similarity and proximity and extrinsic principle of common region in visual perception. Perception, 40, 1467–1477. Mack, A., & Rock, I. (1998). Inattentional blindness. Cambridge, MA: MIT Press. Mack, A., Tang, B., Tuma, R., Kahn, S., & Rock, I. (1992). Perceptual organization and attention. Cognitive Psychology, 24, 475–501. Merikle, P. M., Smilek, D., & Eastwood, J. D. (2001). Perception without awareness: Perspectives from cognitive psychology. Cognition, 79, 115–134. Mitroff, S. R., & Scholl, B. J. (2005). Forming and updating object representations without awareness: Evidence from motion-induced blindness. Vision Research, 45, 961–967. Moore, C. M., & Egeth, H. (1997). Perception without attention: Evidence of grouping under conditions of inattention. Journal of Experimental Psychology: Human Perception and Performance, 23, 339–352. Naccache, L., & Dehaene, S. (2001). Unconscious semantic priming extends to novel unseen stimuli. Cognition, 80, 215–229. Neisser, U. (1967). Cognitive psychology. East Norwalk, CT: Appleton-Century-Crofts. Neumann, O., & Klotz, W. (1994). Motor responses to non-reportable masked stimuli: Where is the limit of direct motor specification? In C. Umilta & M. Moscovitch (Eds.), Attention and performance XV (pp. 123–150). Cambridge, MA: MIT Press. Neumann, H., & Sepp, W. (1999). Recurrent V1–V2 interaction in early visual boundary processing. Biological Cybernetics, 81, 425–444. Ortells, J. J., Marí-Beffa, P., & Plaza-Ayllón, V. (2013). Unconscious congruency priming from unpractised words is modulated by prime–target semantic relatedness. Journal of Experimental Psychology: Learning, Memory and Cognition, 39, 394–413. Palmer, S. E. (1992). Common region: A new principle of perceptual grouping. Cognitive Psychology, 24, 433–447. Palmer, S. E. (1999). Vision science. Photons to phenomenology. Cambridge, MA: MIT Press. Palmer, S. E., & Rock, I. (1994). Rethinking perceptual organization: The role of uniform connectedness. Psychonomic Bulletin & Review, 1, 29–55. Quinlan, P. T., & Wilton, R. L. (1998). Grouping by proximity or similarity? Competition between the Gestalt principles in vision. Perception, 27, 417–430. Reingold, E. M., & Merikle, P. M. (1988). Using direct and indirect measures to study perception without awareness. Perception & Psychophysics, 44, 563–575. Reingold, E. M., & Merikle, P. M. (1993). Theory and measurement in the study of unconscious processes. In M. Davies & G. Humphreys (Eds.), Consciousness (pp. 40–57). Oxford: Basil Blackwell Ltd.. Richardson, J. T. E. (2011). Eta squared and partial eta squared as measures of effect size in educational research. Educational Research Review, 6, 135–147. Rock, I., Linnett, C. M., Grant, P., & Mack, A. (1992). Perception without attention: Results of a new method. Cognitive Psychology, 24, 502–534. Rosenthal, O., & Humphreys, G. (2010). Perceptual organization without perception: The subliminal learning of global. Psychological Science, 21, 1751–1758. Russell, C., & Driver, J. (2005). New indirect measures of ‘inattentive’ visual grouping in a change-detection task. Perception & Psychophysics, 67, 606–623. Shomstein, S., Kimchi, R., Hammer, M., & Behrmann, M. (2010). Perceptual grouping operates independently of attentional selection: Evidence from hemispatial neglect. Attention, Perception, & Psychophysics, 72, 607–618. Spillmann, L. (2009). Phenomenology and neurophysiological correlations: Two approaches to perception research. Vision Research, 49, 1507–1521. Treisman, A. (1982). Perceptual grouping and attention in visual search for features and for objects. Journal of Experimental Psychology: Human Perception and Performance, 8, 194–214. Treisman, A. (1985). Preattentive processing in vision. Computer Vision, Graphics, and Image Processing, 31, 156–177. Treisman, A. (1988). Features and objects: The fourteenth Bartlett Memorial Lecture. Quarterly Journal of Experimental Psychology, 40A, 201–237. Van den Bussche, E., Notebaert, K., & Reynvoet, B. (2009). Masked primes can be genuinely semantically processed: A picture prime study. Experimental Psychology, 56, 295–300. Van den Bussche, E., Van den Noortgate, W., & Reynvoet, B. (2009). Mechanisms of masked priming: A meta-analysis. Psychological Bulletin, 135, 452–477. Wagemans, J., Elder, J. H., Kubovy, M., Palmer, S. E., Peterson, M. A., Singh, M., et al (2012). A century of Gestalt psychology in visual perception: I. Perceptual grouping and figure–ground organization. Psychological Bulletin, 138, 1172–1217. Wang, L., Weng, X., & He, S. (2012). Perceptual grouping without awareness: Superiority of Kanizsa triangle in breaking interocular suppression. PLoS One, 7(6), e40106. http://dx.doi.org/10.1371/journal.pone.0040106. Wertheimer, M. (1923). Untersuchungen zur Lehre von der Gestalt II. Psycholosgische Forschung, 4, 301–350.