Emotion 2008, Vol. 8, No. 3, 352–363
Copyright 2008 by the American Psychological Association 1528-3542/08/$12.00 DOI: 10.1037/1528-35126.96.36.1992
Fuzzy Picture Processing: Effects of Size Reduction and Blurring on Emotional Processing Andrea De Cesarei and Maurizio Codispoti University of Bologna Previous studies have suggested that picture size reduction affects emotional reactions, possibly because scenes subtending a small visual angle are perceived as being more distant and less relevant compared to larger stimuli. However, pictures that subtend a small visual angle also contain few fine-grained details, which may determine less vivid representations and responses compared to larger and more detailed images. Critically, the present study compared two different types of manipulations, namely size reduction and low-pass spatial filtering, which determined similar detail loss but affected visual angles differently. Affective modulation was assessed using an evaluative task and a behavioral interference task. Results showed that the availability of fine-grained details, independently of visual angle, modulated emotional evaluation. Moreover, interference in an unrelated task was unaffected by either size reduction or low-pass spatial filtering. These findings suggest that high spatial frequencies affect subjective emotional response whereas attentional capture by affective stimuli seems to rely on information that is sufficient to allow a categorization of picture content. Keywords: emotion, spatial frequencies, arousal, stimulus size, attention
cesses, which facilitate perceptual encoding and recognition in sensory systems (Beaver, Mogg, & Bradley, 2005; Derryberry & Tucker, 1994; Fox et al., 2000; Fox, Russo, & Dutton, 2002; Lang, Bradley, & Cuthbert, 1990, 1997; Mogg & Bradley, 1999; Phelps, Ling, & Carrasco, 2006; Vuilleumier & Driver, 2007). A number of physiological and behavioral measures are consistent in suggesting greater attention allocation to emotional pictures compared with neutral ones. For example, in a series of studies Margaret Bradley and collaborators (Bradley, Cuthbert, & Lang, 1996a, 1999) explored attentional processes during picture perception by delivering a tone probe at various temporal intervals after picture onset and asking participants to ignore the picture and press a button as fast as possible when the tone probe was presented. Participants showed longer reaction times early on in the picture interval, and even longer reaction times for emotionally arousing pictures, compared with neutral ones. Interference effects caused by task-unrelated emotional pictures have been observed during a variety of visual and acoustic tasks, suggesting that motivationally relevant stimuli draw more on attentional resources, leaving them less available for task processing (Bradley et al., 1996, 1999; Calvo & Nummenmaa, 2007; OkonSinger, Tzelgov, & Henik, 2007; Pereira et al., 2006). Although some studies have suggested that unpleasant stimuli capture greater attention than pleasant stimuli do (Hartikainen, Ogawa, & ¨ hman, Lundqvist, & Esteves, 2001; Pratto & John, Knight, 2000; O 1991), when pleasant and unpleasant stimuli were equated in terms of arousal ratings, no difference as a function of valence was found, and highly arousing picture contents (pleasant and unpleasant) captured greater attentional resources than low arousing stimuli (Blair et al., 2007; Bradley et al., 1999; Bradley, Drobes, & Lang, 1996; Nummenmaa, Hyo¨na¨, & Calvo, 2006; Schimmack, 2005; Verbruggen & De Houwer, 2007). A large amount of literature ascribed the effects of emotional stimuli on subjective ratings and cognition to their intrinsic rele-
Emotional processing has been widely investigated via the presentation of affective pictures that are effective cues in evoking a broad range of emotional reactions, varying in intensity, and involving both pleasant and unpleasant affects. Following a dimensional approach, emotional behavior can be defined by a first dimension of affective valence, which controls the direction of emotional engagement (approach or withdrawal), and a second dimension of arousal, which dictates the intensity of affective response (Dickinson & Dearing, 1979; Konorski, 1967; Lang, Bradley, & Cuthbert, 1990; Russell & Barrett, 1999; Schneirla, 1959; Wundt, 1896). Several studies have shown that valence and arousal are the most important dimensions of connotative meaning and feeling states, and that they control most of the variance of subjective reports (Bradley, Codispoti, Cuthbert, & Lang, 2001; Osgood, 1969; Russell & Barrett, 1999). The affective space defined by pleasure and arousal ratings is very similar across sensory modalities, in fact similar findings have been obtained using pictures, sounds, odorants, or words (Bensafi et al., 2002; Bradley & Lang, 2007). Moreover, subjective ratings of valence and arousal covary systematically with the biological reflexes that are associated with activation of appetitive and defensive motive systems (Bradley et al., 2001; Bradley, Codispoti, & Lang, 2006; Bradley & Lang, 2007). Motivationally relevant stimuli (i.e., those that strongly activate either the appetitive or defensive system) engage attentional pro-
Andrea De Cesarei and Maurizio Codispoti, Department of Psychology, University of Bologna. We thank Geoff Loftus for helpful discussions. We also thank Giulia Viconi for her help in data acquisition, and all participants for having taken part in the studies. Correspondence concerning this article should be addressed to Andrea De Cesarei, Department of Psychology, University of Bologna, Viale Berti Pichat, 5– 40127 Bologna, Italy. E-mail: [email protected]
EMOTIONAL PROCESSING AND STIMULUS DEGRADATION
vance, related to the evolutionary significance of appetitive or threatening stimuli. However, stimulus relevance varies with both stimulus content and contextual factors such as distance from the observer. Distance may either emphasize or reduce stimulus relevance (Lang et al., 1997), and modulate emotional response (Teghtsoonian & Frost, 1982). As physical distance is closely related to retinal size (Graham, 1965; Loftus & Harley, 2005), recent studies have investigated the effects of picture size on emotional responses, showing that subjective ratings of valence and arousal vary with picture size (Detenber & Reeves, 1996), with less affective modulation for smaller compared to larger pictures (Codispoti & De Cesarei, 2007). Picture size reduction has been shown to modulate emotional processing, possibly because of its direct relation to distance. However, a number of visual features vary with picture size, and size reduction may be considered just one out of many possible ways to degrade an image. In particular, when an object is reduced in size, fine-grained details fall below the threshold of discriminability and are not perceived. Similarly, under a variety of suboptimal visual conditions such as dark, fog, or when stimuli are in parafoveal or peripheral vision, small details in a scene are not perceived, reducing the feeling of presence (Lombard, Reich, Grabe, Bracken, & Ditton, 2000). Thus, it is possible that the availability of fine-grained perceptual details, which correspond to the higher portion of the spatial frequency spectrum, modulate affective processing even in absence of changes in visual angle. By comparing the effects of size reduction to the effects of low-pass spatial filtering on affective response, the present study tried to disentangle this issue.
The Research Problem The present experiments were aimed at assessing the effects on emotional processing of two distinct perceptual manipulations, namely picture size reduction and low-pass spatial filtering. The rationale of this comparison was that, if the effects of size reduction on affective processing are due to the lower relevance of stimuli that subtend a small visual angle, and may therefore be either distant or small, then no effect of perceptual degradation should be observed for low-pass filtered, full-size stimuli. On the other hand, if generic vividness of the visual scene is associated with enhanced affective processing, then a change in emotional response should be observed following both size reduction and low-pass spatial filtering. Following this strategy, the first study compared intact pictures to filtered or size-reduced pictures with regards to subjective ratings of emotional experience, whereas the second study examined the effects of stimulus degradation on attention allocation to emotional scenes presented as distractors while performing an auditory discrimination task.
Experiment 1 In Experiment 1, we investigated the combined effects of perceptual degradation and stimulus content (positive, neutral, or negative) on subjective emotional response. Stimuli were presented in different sizes ranging from full screen (21.23° horizontal ⫻ 16.22° vertical degrees of visual angle) to very small (2.68° ⫻ 2.05°) sizes. Moreover, additional perceptual degradation con-
ditions were obtained by applying a low-pass spatial filter that progressively filtered out a larger amount of sharp details. Following the presentation of each picture, participants were asked to rate the pleasantness and arousal of the emotional state elicited by the image. The amount of fine-grained details (e.g., wrinkles) in an image is indexed by the power of high spatial frequencies in the frequency spectrum. Size reduction, as well as low-pass spatial filtering, makes fine-grained details more difficult to perceive and reduces high spatial frequency (HSF) power, expressed in cycles/ image (cpi). Using an appropriate low-pass spatial filtering function, we tried to equate the loss of details across size-reduced and filtered pictures. To check the efficacy of our manipulation, participants were asked to rate each picture according to the extent to which they were able to identify fine-grained details.
Method Participants. Forty students (21 men) from the University of Bologna took part in the present study. Age ranged from 18 to 39 years (M ⫽ 22.13, SD ⫽ 3.28). Before agreeing to participate in the study and signing an informed consent form, all participants were warned that they would see potentially arousing pictures. Stimuli and equipment. One hundred eighty grayscale images were selected from various sources, including document scanning, the Internet and the International Affective Picture System (IAPS; Lang, Bradley, & Cuthbert, 2001). Twenty images were chosen for each of eight different contents: erotic couples, opposite sex nudes (20 males and 20 females), babies, neutral animals, people in urban or natural contexts, animal attack, human attack, or mutilated bodies. All images were adjusted to the same overall value of brightness and contrast. Pictures were presented using E-Prime software (Schneider, Eschman, & Zuccolotto, 2002) on a 19“ monitor. The distance between the monitor and the participant was 1m, and a chinrest assured that the distance remained constant between and across participants. Manipulation parameters. Pictures were resized to one of four different sizes (100%, 50%, 25%, 12.5%). Visual angles subtended by stimuli in each size were 21.23° horizontal ⫻ 16.22° vertical (100%), 10.71° ⫻ 8.21° (50%), 5.36° ⫻ 4.11° (25%), and 2.68° ⫻ 2.05° (12.5%). For each level of size reduction, a low-pass spatial filter condition that ensured a similar loss of details was created. We used a three-octave wide cosine filter, passing all spatial frequencies below a roll-off frequency (F0), degrading with a parabolic function reaching zero at a cut-off frequency of F1 ⫽ 3 ⫻ F0, and completely removing spatial frequencies above this cut-off point. In the remainder of the paper, we will describe this filter by its cut-off frequency F1. Using this type of spatial filter, a previous study demonstrated that spatial frequencies of 30 cpi or higher do not influence a face categorization task for faces subtending 1° of visual angle (Loftus & Harley, 2005). Accordingly, spatial frequencies above about 480 cpi should not influence categorization of images subtending about 16° degrees of visual angle. Therefore, filter cut-offs used to match full-screen filtered images to progressively smaller pictures were determined as 486.6, 243.3, 121.65, and 60.83 cpi. Both resized and filtered images were pasted on a gray background of the same brightness as the
DE CESAREI AND CODISPOTI
average of the brightness of all the images (Codispoti & De Cesarei, 2007; De Cesarei & Codispoti, 2006). Procedure. At the beginning of the experiment, a practice sequence of eight trials was performed. Images in the practice block were not presented during the test block, to prevent distortion in ratings due to habituation to specific pictures (e.g., Codispoti, Ferrari, & Bradley, 2006, 2007). Data from the practice trials were not analyzed. During the test block, 80 pictures (10 for each stimulus category) were selected from the complete dataset and assigned to one of the 4 (amount of degradation) ⫻ 2 (type of degradation) experimental conditions. Each participant was allowed to see each image only once, and in only one degradation condition. Across all participants, all pictures were presented an equal number of times in all degradation conditions. During each trial, a fixation cross appeared for 1s. Then, the image was presented for 1s. After the image disappeared and a 1s delay elapsed, the rating screens were presented. After the last rating screen and a blank screen of 1s had been presented, the next trial began. Ratings. After each picture, participants were requested to rate their emotional state, and the vividness of the image presented. It was made clear to participants that they had to rate their experienced emotional state, and not give a semantic judgment on the picture. Emotional ratings were collected using the SelfAssessment Manikin (SAM; Lang, 1980), a visual rating scale (9 levels) that allows the rating of the pleasantness and intensity of the experienced emotional state. Moreover, participants were told that pictures would be degraded to various extents, and that they should rate the extent to which they were able to identify details in the picture. To rate vividness, participants used a 9-level visual scale ranging from very low (almost no details were recognizable in the picture) to very high (picture details were very well-defined) vividness. The vividness scale was created by presenting the intermediate item of the SAM valence scale with 9 progressive levels of contrast, ranging from very low to very high. Data analysis. For each participant, SAM and vividness ratings were averaged in all experimental conditions. Data were analyzed through repeated-measures analyses of variances (ANOVAs) with the factors Type of Manipulation (size or filter),
Amount of Manipulation (4 levels), and Category (8 levels). To deal with sphericity violations that increase the probability of Type I error, a Huynh–Feldt correction was applied to the degrees of freedom. Moreover, contrast analyses were conducted (Loftus, 1996) by fitting average values to the degradation coefficients 1/1, 1/2, 1/4, 1/8. The rating-degradation R2 value, indicating the goodness-of-fit between the observed values and the a priori coefficients, describes the strength of the relationship between the observed variable and perceptual degradation.
Results Vividness. Vividness ratings are reported in Table 1 and shown in Figure 1. Vividness decreased progressively across different levels of manipulation, similarly for filtered and resized pictures. No main effects or interactions involving Type of Manipulation were observed. A main effect of Amount of Manipulation was observed, F(3, 96) ⫽ 243.98, p ⬍ .001, 2 ⫽ .884, indicating lower vividness ratings for degraded (both smaller and more filtered) pictures, rating-degradation R2 ⫽ .69. A significant Category effect was observed, F(7, 224) ⫽ 10.3, p ⬍ .001, 2 ⫽ .24, which was further qualified by the higher order Amount of Manipulation ⫻ Category interaction, F(21, 672) ⫽ 1.93, p ⬍ .05, 2 ⫽ .057. Pairwise comparisons with ␣ ⫽ .05 highlighted significantly lower vividness for neutral people compared to all other categories. Pictures of opposite sex nudes and babies were rated as more vivid compared to animals, animal and human attack, and mutilated bodies. Finally, erotic couples were rated as less vivid compared to opposite sex nudes. Separate ANOVAs with the factor Amount of Manipulation were carried out for each category. For all categories, the main effect Amount of Manipulation was significant, indicating progressively lower vividness as images were degraded. To characterize this interaction, a contrast analysis was conducted (see Table 1). Similarly high R2 values were observed, with the exception of animal and human attack, which showed a slightly worse fit compared to all other categories. Valence. Valence ratings are reported in Table 1 and plotted in Figure 1. Categories were rated as different in valence, with
Table 1 Results of Experiment 1 Vividness 2
Couples Opposite sex nudes Babies Animals Neutral people Animal attack Human attack Mutilated bodies Total
.70 .68 .75 .75 .75 .62 .51 .72 .69
7.74 8.00 7.89 7.76 7.46 7.57 7.21 7.45 7.64
7.45 7.64 7.54 7.43 6.86 7.54 7.30 7.11 7.36
6.39 6.66 6.48 6.08 5.80 6.37 6.60 6.20 6.32
4.79 4.74 5.12 4.65 3.94 4.56 4.71 4.71 4.65
6.59 6.76 6.76 6.48 6.02 6.51 6.46 6.37 6.49
.66 .53 .61 .86 .28 .00 .77 .50 .43
6.81 6.96 7.16 6.16 4.89 4.27 3.25 1.95 5.18
6.77 6.97 7.25 5.98 5.09 4.67 3.61 2.00 5.29
6.42 6.67 6.60 5.66 4.81 4.57 3.73 2.06 5.07
5.91 5.86 5.97 5.39 4.59 4.17 3.59 2.56 4.76
6.48 6.62 6.75 5.80 4.85 4.42 3.55 2.14 5.07
.85 .78 .57 .77 .26 .93 .90 .64 .77
5.51 5.01 4.54 3.58 2.82 4.98 4.72 6.61 4.72
5.27 4.59 4.42 3.54 3.15 4.34 4.52 6.46 4.54
5.02 4.54 4.31 3.15 2.66 4.30 4.30 6.07 4.29
4.63 3.89 3.59 3.00 2.42 3.91 4.07 5.07 3.82
5.11 4.51 4.22 3.32 2.76 4.38 4.40 6.05 4.34
Note. The effects of stimulus category and level of degradation (expressed as the ratio with respect to the intact picture) on each measure. Separately for each category and measure, the R2 indexes of goodness-of-fit between the observed measure and the level of degradation are reported, together with the raw data (means).
EMOTIONAL PROCESSING AND STIMULUS DEGRADATION
Figure 1. The effects of picture degradation and stimulus content on ratings of arousal, pleasure and vividness. From left to right, stimuli are progressively degraded either by reducing size (white circles) or by applying a low-pass filter (black circles).
positive stimuli being rated as more pleasant compared to neutral and unpleasant stimuli. More interesting, this difference progressively decreased with picture degradation. No difference between filtered and resized stimuli was observed. A significant main effect of Amount of Manipulation was observed, F(3, 96) ⫽ 18.29, p ⬍ .001, 2 ⫽ .38, with pictures that were degraded to a greater extent (either smaller or more blurred) being rated as less pleasant compared to those degraded to a lesser extent, rating-degradation R2 ⫽ .43. Moreover, a significant Category effect was observed, F(7, 210) ⫽ 70.49, p ⬍ .001, 2 ⫽ .7, which was further characterized by a significant interaction between Amount of Manipulation and Category, F(21, 630) ⫽ 5.3, p ⬍ .001, 2 ⫽ .15. Investigating the linear
fit between pleasantness and degradation (see Table 1), moderate R2 values were observed for all categories, with the exception of pictures of neutral people and animal attack that showed low R2 values. Significant effects of Category were observed in all conditions, Fs(7, 266) ⬎ 19.73, ps ⬍ .001, 2s ⬎ .34. Pairwise comparisons highlighted significant differences between all categories, ps ⬍ .05, with the exception of pleasant contents (erotic couples, opposite sex nudes, and babies), which showed no difference between categories, and pictures depicting animal attack, which showed no difference when compared to those of neutral people. Arousal. Arousal ratings are reported in Table 1 and Figure 1. Picture degradation resulted in lower arousal, and this pattern appeared similarly pronounced for both size-reduced and low-pass
DE CESAREI AND CODISPOTI
filtered pictures. In particular, arousing categories showed a more marked decrease compared to neutral stimuli. Statistical analysis revealed a significant effect of Type of Manipulation, F(1, 32) ⫽ 23.8, p ⬍ .001, 2 ⫽ .43, indicating slightly higher arousal ratings for filtered compared to resized pictures. Moreover, a significant Amount of Manipulation effect was observed, F(3, 96) ⫽ 38.67, p ⬍ .001, 2 ⫽ .55, indicating lower arousal ratings for more impoverished pictures (either smaller or more filtered) compared to those which remained intact, rating-degradation R2 ⫽ .77. A significant Category effect was observed, F(7, 224) ⫽ 26.62, p ⬍ .001, 2 ⫽ .45, which was further characterized by an interaction between Amount of Manipulation and Category, F(21, 672) ⫽ 1.91, p ⬍ .05, 2 ⫽ .056. As reported in Table 1, moderate to high R2 values were observed for all categories, except for the neutral people category that showed the smallest change. A significant main effect of Category was observed in all conditions, Fs(7, 259) ⬎ 7.21, ps ⬍ .001, 2s ⬎ .16. Pairwise comparisons highlighted significant differences between all categories, ps ⬍ .05, with the exception of babies, opposite sex nudes, animal and human attack, which showed no difference when compared.
expected to be less powerful stimuli for attention capture compared to intact pictures. The aim of Experiment 2 was to investigate this possibility, by probing affective interference on a secondary task using variously degraded stimuli. In a dual task paradigm, participants were asked to perform an auditory discrimination task while viewing taskunrelated emotional and neutral pictures that could be either reduced in size or low-pass filtered. Building on the same strategy as in Experiment 1, pictures were either blurred or reduced in size so that detail loss would be similar across the two types of manipulation. In addition, we were interested in determining whether changes in stimulus size, even in the absence of meaningful picture content, would modulate attentional capture (Castiello & Umilta`, 1990). To this aim, an additional condition was created by first scrambling each picture and then reducing it in size. Moreover, using stimulus categories that ranged from positive to negative, and from low to highly arousing, Experiment 2 tried to assess which specific contents are associated with disruptive effects on a secondary task. As in Experiment 1, to prevent systematic category-color associations that may confound results, pictures were presented in grayscale.
Consistent with previous studies (Bradley et al., 2001), the results of Experiment 1 showed that viewing grayscale pictures of natural scenes affected subjective ratings of arousal and pleasure. Stimuli were manipulated by either size reduction or low-pass filtering, so that the loss of sharp details would not differ between the two types of degradation. Confirming the efficacy of this manipulation, ratings of vividness did not differ between the two conditions. Ratings of affective state were clearly affected by picture degradation. More specifically, size reduction and low-pass spatial filtering influenced affective ratings in a similar manner, with less-pronounced differences between neutral and emotional stimuli for degraded compared to intact pictures. Impoverished pictures were rated as less vivid, less arousing and less pleasant compared to intact pictures. Moreover, as reflected by the RatingDegradation R2, the effects of stimulus degradation on arousal appeared more pronounced than those regarding valence. Experiment 1 clearly demonstrated that two different types of perceptual degradation similarly modulated the emotional state elicited by natural scenes. However, subjective ratings are prone to being biased by beliefs and expectations, and capture only part of the whole range of emotional responses. In Experiment 2, we tried to extend the present results to an index of emotional response that is not under explicit control, namely interference through affective stimuli during an unrelated task.
Participants. Twenty-nine students (14 men) from the University of Bologna participated in the present study. Age ranged from 19 to 33 years (M ⫽ 24.52, SD ⫽ 3.79). Before agreeing to participate in the study and signing an informed consent form, all participants were warned that they would see potentially arousing pictures. Stimuli and equipment. Two hundred forty-three grayscale images (27 for each content) were selected from the same sources as in Experiment 1. Several stimulus categories were chosen, including erotic couples, opposite sex nudes (27 males and 27 females), babies, neutral animals, people in urban or natural contexts, animal attack, human attack, and mutilated bodies. Nine versions of each of the 243 images were created by either reducing size, low-pass filtering, or scrambling the original picture in one of three different levels. In the size reduction condition, pictures were resized to one of three different sizes, respectively 100%, 33% or 12.5%. Visual angles subtended in each condition were 20.96 horizontal ⫻ 15.66 vertical (100%), 6.99 ⫻ 5.22 (33%), and 2.62 ⫻ 1.96 (12.5%). In the low-pass spatial filter condition, a filter was applied to the original images to simulate the high frequency loss of the resized images. The low-pass thresholds used were determined as in Experiment 1, and corresponded to 469.8, 156.6, and 58.73 cpi. Finally, scrambled pictures were created by dividing each image into 10-pixel squares and rearranging them 1,000 times, until the original content was unrecognizable. These patterns were resized to 100%, 33%, or 12.5% of the original size. Resized, filtered, and scrambled images were pasted on a gray background of the same brightness as the average of all images. Two acoustic tones were created, at a frequency of 1,000 and 1,050 Hz. Each tone lasted 200 ms, with a rise and fall time of 10 ms. Volumes were adjusted to 75 dB. The tones were presented using a set of binaural headphones. Equipment and laboratory settings were otherwise identical to Experiment 1.
Experiment 2 Experiment 1 demonstrated that two different types of perceptual manipulation affect stimulus relevance similarly, with lower relevance for degraded compared to intact stimuli. Interference in ongoing tasks caused by the presence of emotionally charged distractors has been repeatedly observed and related to the intrinsic relevance value of affective stimuli. Accordingly, if picture degradation affects stimulus relevance, degraded images should be
.93 (.26) .95 (.22) .92 (.27) .91 (.28) .95 (.22) .93 (.26) .93 (.25) .95 (.23) .93 (.25) .97 (.18) .94 (.23) .92 (.27) .89 (.32) .94 (.23) .91 (.29) .94 (.23) .95 (.21) .93 (.25) Note.
cpi ⫽ cycles per image. Sizes are relative to the fullscreen size, measuring 20.96° horizontal ⫻ 15.66° vertical degrees of visual angle.
.93 (.25) .95 (.21) .94 (.23) .93 (.25) .95 (.21) .95 (.21) .91 (.29) .95 (.21) .94 (.24) .89 (.32) .95 (.21) .91 (.29) .92 (.27) .95 (.21) .92 (.27) .94 (.23) .93 (.25) .93 (.26) .93 (.26) .95 (.21) .95 (.23) .93 (.25) .95 (.23) .95 (.23) .95 (.21) .94 (.24) .94 (.23) .91 (.29) .98 (.15) .95 (.21) .92 (.27) .93 (.25) .94 (.23) .95 (.21) .99 (.11) .95 (.22) .93 (.25) .97 (.18) .92 (.27) .97 (.18) .95 (.21) .89 (.32) .98 (.15) .92 (.27) .94 (.24) Erotic couples Opposite sex nudes Babies Animals Neutral people Animal attack Human attack Mutilation Total
.90 (.31) .92 (.27) .94 (.23) .95 (.21) .94 (.23) .95 (.21) .95 (.21) .92 (.27) .94 (.25)
.92 (.27) .98 (.15) .97 (.18) .94 (.23) .94 (.23) .95 (.21) .92 (.27) .97 (.18) .95 (.22)
.92 (.28) .95 (.21) .94 (.23) .95 (.21) .95 (.23) .93 (.25) .95 (.22) .93 (.25) .94 (.24)
.95 (.21) .92 (.27) .95 (.21) .95 (.21) .97 (.18) .95 (.21) .94 (.23) .90 (.31) .94 (.23)
.92 (.27) .97 (.18) .93 (.25) .93 (.25) .94 (.23) .94 (.23) .97 (.18) .93 (.25) .94 (.24)
12.5% 33% 33% 58.73 cpi 156.6 cpi 469.8 cpi Category
Table 2 Accuracy in the Secondary Discrimination Task (M and SD)
EMOTIONAL PROCESSING AND STIMULUS DEGRADATION
Procedure. At the beginning of the experiment, a practice sequence of 48 trials was performed. Images in the practice block were not repeated during test blocks, to prevent habituation to specific pictures. During the test block, 216 pictures were selected from the original picture set and presented. Each picture was assigned to one of the 3 (type of manipulation: resizing, low-pass filtering, or scrambling) ⫻ 3 (amounts of manipulation) degradation conditions. To avoid repetition effects, each picture was seen only once by each participant, and in only one of the nine degradation conditions. Across all participants, all pictures were presented in all degradation conditions. During each trial, a fixation cross appeared for 1s. Then the image was presented and remained on the screen for 3s. After 250ms from picture onset, a tone was presented binaurally. After picture offset, a blank screen lasting between 2,800 and 3,800 ms was presented (intertrial interval, ITI). During the intertrial interval, a fixation cross was presented in the center of the screen. Participants were instructed to respond as to whether they heard a 1,000 or 1,050 Hz tone by pressing one of two digits on the keyboard (“z” or “m”) as fast and accurately as possible. For half of the participants, the tone-digit mapping was reversed. Data analysis. Only response times of accurate trials were analyzed. Two participants showing an exaggerated percentage of errors and slow responses (52% and 33% compared to an average of 5.9%) were discarded from analysis. To deal with abnormally slow or fast responses, we aggregated data within each participant and conditions using the median as an index of central tendency.1 Reaction time and accuracy were aggregated in all experimental conditions. Data were analyzed using a repeated-measures ANOVA with the factors Type of Manipulation (3 levels: size, filter, or scrambled), Amount of Manipulation (3 levels), and Category (8 levels).
Results Accuracy. Accuracy data are reported in Table 2. More accurate responses for degraded compared to intact stimuli were observed, Amount of Manipulation F(2, 52) ⫽ 3.67, p ⬍ .05, 2 ⫽ .12. Pairwise comparisons highlighted a significant difference between intact and slightly degraded stimuli, p ⬍ .05, and no difference between slightly degraded and the most degraded stimuli. No other significant main effects or interactions involving either Type of Manipulation or Category were observed in this analysis. Reaction times. Reaction times to stimuli varying in content and perceptual manipulation are reported in Figure 2. Reaction times appeared to be modulated by stimulus category, but not by the amount of degradation. Effects of stimulus category were observed for both filtered and resized pictures but not for scrambled images, confirming the efficacy of the scrambling manipulation. 1 Using the median as an index of central tendency assures that no data are discarded and minimizes the influence of extreme values. However, in a separate analysis, we analyzed response-time data with the same ANOVA design as reported, using the mean as an index of central tendency and excluding trials that exceeded a ⫾ 2 SD from the mean threshold. No differences in terms of significant effects were observed with respect to the results reported.
DE CESAREI AND CODISPOTI
Figure 2. Response times to size-reduced (gray bars), low-pass filtered (triangles and lines) and scrambled (dots) stimuli. From left to right, panels represent reaction times to pictures with progressively higher levels of degradation.
The repeated-measures ANOVA revealed significant effects of Category, F(7, 182) ⫽ 12.96, p ⬍ .001, 2 ⫽ .33, and Type of Manipulation, F(2, 52) ⫽ 10.3, p ⬍ .01, 2 ⫽ .28. No effect of Amount of Manipulation was observed, F ⬍ 1. A significant interaction between Category and Type of Manipulation was observed, F(14, 364) ⫽ 4.11, p ⬍ .001, 2 ⫽ .14, which was due to significant effects of Category in the filtered and size-reduced condition, Fs(7, 182) ⬎ 8.32, ps ⬍ .001, 2s ⬎ .24, but not in the scrambled condition, F(7, 182) ⫽ 1.16, ns. In the filtered and size-reduced conditions, pairwise comparisons revealed significantly slower responses for pictures of erotic couples, opposite sex nudes and mutilation compared to all other contents, and faster responses for pictures of neutral animals or people compared to those depicting animal or human attack, ps ⬍ .05. A separate repeated-measures ANOVA with the factor Type of Manipulation was carried out for each stimulus category. A significant effect of Type of Manipulation was only observed for pictures depicting erotic couples, opposite sex nudes, and mutilations, Fs(2, 52) ⬎ 5.55, ps ⬍ .01, 2s ⬎ .18, indicating slower responses for both filtered and size-reduced compared to scrambled stimuli. Finally, as reported in Figure 3, a state-trace analysis (Bamber, 1979; Loftus, 2002) was conducted to visually represent the interaction between Type of Manipulation, Amount of Manipulation, and Category. In a series of scatterplots, reaction times
to stimuli presented in two different conditions (e.g., sizereduced vs. filtered) were compared. The strength of the relationship between the two conditions, expressed by the R2, suggests that the factors influencing responses under both conditions are the same, that is, the stimuli that prompt slower responses under the size-reduction condition also impair performance following low-pass spatial filtering. Although a strong relationship was observed between size-reduced and low-pass filtered stimuli, R2 ⫽ 0.63, almost no relationship between scrambled and either size-reduced or low-pass filtered stimuli was observed, R2 ⬍ 0.007.
Discussion Experiment 2 investigated the combined effects of stimulus content and perceptual degradation on interference during an unrelated auditory task. Using the same parameters as Experiment 1, a similar loss of details was achieved across size reduction and low-pass spatial filtering. Interference caused by highly arousing contents such as erotic couples, opposite sex nudes, and mutilated bodies was observed, and no difference between similarly highly arousing stimuli was observed (Bradley, Cuthbert, et al., 1996; Bradley et al.,1999; Bradley, Drobes, et al., 1996; Schimmack, 2005).
EMOTIONAL PROCESSING AND STIMULUS DEGRADATION
Contrary to the hypothesis that lower stimulus relevance achieved through perceptual degradation would have reduced attentional capture by task-unrelated emotional stimuli, similar affective interference was observed when using both intact and degraded stimuli. In contrast, no effects of picture category were observed for scrambled pictures, confirming that picture content was unrecognizable to a level that eliminated interference with the present task. Moreover, no differences in attentional capture were observed for neutral pictures compared to scrambled patterns, suggesting that simply discriminating a meaningful image from a random pattern is not enough to divert attention from an ongoing attentional task. The interference effect that was observed for highly arousing stimuli might depend on processes that are separate from scene recognition and are only triggered by highly arousing affective stimuli. The slowing of reaction times due to the interference of highly arousing task-unrelated pictures did not depend on the amount of perceptual detail of the image, but rather appeared to be an “all or nothing” phenomenon that occurred whenever a highly arousing stimulus was recognized.
General Discussion In the two separate experiments it was observed that, although changes in perceptual detailedness decreased the affective value of stimuli as reflected by subjective ratings (arousal and valence, Experiment 1), even the most degraded pictures showed the same capacity as did intact pictures for attracting attentional resources and interfering with an unrelated discrimination task (Experiment 2). Moreover, no differences were observed between the effects of size reduction and low-pass filtering on affective modulation of subjective ratings, suggesting a role for high spatial frequencies in the modulation of the emotional state. A similar loss of fine-grained details, defined as HSF power, was obtained using two different perceptual manipulations. Accordingly, similar changes in affective modulations of emotional state were observed following size reduction and low-pass spatial filtering, suggesting that HSFs may modulate subjective affective response. One possible interpretation is that a vivid picture may be more effective compared to a degraded stimulus in activating an internal representation of a relevant scene. Dealing with stimuli that have an emotional connotation activates associative networks which connect sensory, semantic and procedural information (Lang, 1979, 1995). When an emotional event, or stimulus, is imagined or perceived some nodes are activated, and their activation propagates to other nodes, prompting physiological changes (e.g., increase in the activity of sweat glands), explicit behavioral actions (e.g., trying to escape from an uncomfortable situation), and changes in subjective emotional state. More important, not all kinds of stimuli are similarly effective in activating a semantic network. The largest activation follows real world stimuli that
Figure 3. Scatterplots report state-trace analysis of the interaction between Type of Manipulation, Amount of Manipulation and Category. In the upper panel a strong similarity between factors affecting responses in the size-reduced and low-pass filtered condition is highlighted. On the other hand, such a strong relationship was absent for scrambled and either size-reduction or low-pass filtering (middle and bottom panels).
DE CESAREI AND CODISPOTI
contain the largest number of descriptive features, for example, feeling the contact with a real poisonous snake. In comparison, progressively impoverished versions of the same stimulus that only preserve some of the original features (e.g., a picture or a drawing depicting a snake) would be less effective in activating nodes of associative networks and, therefore, would result in less-pronounced emotional responses (Simons, Detenber, Roedema, & Reiss, 1999). The present study suggests that stimulus degradation, obtained either through size reduction or low-pass spatial filtering, is effective in reducing activation of sensory nodes and, consequently, modulating affective responses. Presence of fine-grained HSF information might enhance the degree of activation of detailed representations, resulting in a more complete representation of relevant events and modulating the subjective emotional state.2 Highly arousing stimuli, which potentially capture attentional resources because of their intrinsic relevance, have been shown in previous studies to interfere with an ongoing unrelated task (Bradley, Cuthbert, et al., 1996; Bradley et al., 1999; Bradley, Drobes, et al, 1996). Accordingly, it was hypothesized that degraded pictures, which are rated as less arousing compared to intact pictures, would also result in lower interference on an unrelated task. Clearly, this hypothesis was disconfirmed, and affective interference was observed following both filtered and size-reduced stimuli, with no difference related to the amount of manipulation. Attentional resources devoted to the processing of arousing stimuli are reflected both in the interference during an unrelated task, and in the modulation of psychophysiological responses such as heart rate (HR) and cortical activity. In particular, it has been shown in previous studies that amplitude of the late positive potential (LPP) component of the event-related potentials (ERPs) is larger for arousing compared to neutral stimuli, presumably reflecting enhanced attention allocation to relevant events (Codispoti, Ferrari, De Cesarei, & Cardinale, 2006; Johnston, Miller, & Burleson, 1986; Radilova, 1982; Schupp et al., 2004). More interesting, in previous studies no effects of size reduction were observed on any of these measures, suggesting that attention allocation to stimuli depicting arousing contents is affected to a relatively small extent by changes in a low-level perceptual characteristic such as stimulus size (Codispoti & De Cesarei, 2007; De Cesarei & Codispoti, 2006; Sa´nchez-Navarro, Martı´nez-Selva, Roma´n, & Torrente, 2006). On the other hand, emotional reactions to affective stimuli as reflected by subjective ratings and measures of autonomic arousal were shown to be modulated by stimulus size. (Codispoti & De Cesarei, 2007). In particular, subjective ratings of pleasure and arousal, as shown in the present and a previous study, are modulated by both stimulus content and size, and affective modulation is more pronounced for larger compared to smaller stimuli (Codispoti & De Cesarei, 2007). Skin conductance change, which has been proposed to be an indicator of sympathetic activation related to action preparation (Lang et al., 1997; Roth, 1983), is only modulated by stimulus arousal when large stimuli are presented, and no modulation was observed for small stimuli, possibly reflecting the absence of action preparation when stimulus imminence or content arousal does not exceed a minimum threshold (Codispoti & De Cesarei, 2007; Ravaja, 2004a, 2004b). Arousal represents the intensity of activation of either the appetitive or the defensive system in response to significant stimuli and is a fundamental factor of the orienting response to affective cues. In the present and previous studies the manipulation of a
perceptual factor such as picture size allowed us to highlight dissociations in subjective, behavioral, and psychophysiological changes suggesting that they reflect different processes. Similarly, dissociations in affective modulation of separate responses were also observed in studies that manipulated stimulus relevance by either varying picture exposure time (Codispoti, Bradley, & Lang, 2001; Codispoti, Mazzetti, & Bradley, 2002) or repeatedly presenting the same stimulus (Bradley, Lang, & Cuthbert, 1993; Codispoti et al., 2006, 2007). The present results support the possibility that low spatial frequency (LSF) information, which is analyzed during the first stages of visual processing, may modulate subsequent processing and, in particular, divert attentional resources from the current task whenever a relevant stimulus is detected (Bar, 2003; Vuilleumier, Armony, Driver, & Dolan, 2003). LSFs convey information regarding scene configuration and are analyzed at a faster rate compared to high spatial frequencies (Loftus & Harley, 2004; Schyns & Oliva, 1994), providing convenient information with which to quickly react to relevant environmental stimuli.3 However, the present results do not allow for the unequivocal acceptance of this possibility, and alternative accounts are possible. Despite stimuli being degraded to 12.5% of their original size or frequency spectrum, in this and previous studies participants were nonetheless able to efficiently categorize picture content (De Cesarei & Codispoti, 2006; Roring, Hines, & Charness, 2006). Accordingly, it might be that attentional capture through affective stimuli is observed whenever a scene is successfully categorized, regardless of the type of degradation applied. Critically, these two scenarios predict opposite results for high-pass filtered stimuli. If low spatial frequencies (LSFs) play a special role in emotional perception, as suggested by some studies probing emotional modulation of blood oxygen level-dependent (BOLD) response in subcortical structures (e.g., Vuilleumier et al., 2003), then a reduced attentional capture should be observed following stimuli only containing HSFs. On the other hand, if any understandable 2 It could be that, as image size is reduced, participants fail to understand the content of some pictures, and consequently rate them as low arousing not because of their content, but because they are treated as “nonimages.” However, in a previous study presenting stimuli of comparable size, categorization accuracy exceeded 85% for the smallest pictures (De Cesarei & Codispoti, 2006). Moreover, a recent study (Roring, Hines, & Charness, 2006) found that the ability to correctly categorize a wide range of facial expressions, which presumably requires identification of finegrained details, is not affected by changes in stimulus size ranging from about 14° to 4° degrees of vertical visual angle. 3 Based on the possibility that different spatial frequencies are analyzed at different times (Bar, 2003; Loftus & Harley, 2004; Morrison & Schyns, 2001; Sanocki, 2001; Schyns & Oliva, 1994), it might be suggested that the 250ms temporal distance between the picture and the tone (stimulus-onset asynchrony, SOA) used in the present study probed attentional capture at a stage when HSF information was not yet being processed, preventing the appearance of effects related to the reduction in HSF power. However, the results of several studies make this possibility unlikely. For instance, starting from about 150 to 180ms after stimulus onset, people can recognize stimulus identity based on high spatial frequency information (Loftus & Harley, 2004; Schyns & Oliva, 1994). Moreover, SOA only constrains the delay between the onset of the processing of the two stimuli. Afterward, these processes proceed in a parallel fashion, and potentially interfere with each other, for several hundredths of milliseconds, until the execution of the response or later.
EMOTIONAL PROCESSING AND STIMULUS DEGRADATION
emotional stimulus can divert attentional resources from the ongoing task, then similar attentional capture by affective stimuli should be observed following stimuli containing either high or low spatial frequencies. The present results demonstrate that, although some scenes may appear with little or no color information associated, nevertheless they may capture attentional resources if their content is relevant. The present results were comparable to previous studies investigating behavioral interference caused by emotional pictures presented in color (Bradley, Cuthbert, et al., 1996; Bradley et al., 1999; Bradley, Drobes, et al., 1996), ruling out the possibility that systematic changes in the color of specific categories (e.g., red associated with mutilated bodies) might have confounded the results. Consistently, it has been shown that peripheral and central indexes of emotional processing are similarly affected by picture content whether pictures are presented in color or in grayscale (Bradley et al., 2001; Bradley et al., 2003; Sabatinelli, Bradley, Lang, Costa, & Versace, 2007).
Limitations and Future Directions Although the present study successfully identified conditions that reduced the subjective rating of emotional state, and established an important link between the effects of size reduction and low-pass spatial filtering on attentional capture, some limitations must be acknowledged. In the present study, the amount of filtering was chosen to mimic the detail loss that occurred when pictures were presented in progressively smaller sizes. However, even in the most degraded condition pictures appeared to be identifiable, and elicited subjective and attentional emotional responses. Therefore, it is not possible to relate the observed effects to emotional processes that rely on LSF information, or to the generic fact that participants were able to discriminate picture contents. Further studies might reduce the low-pass cutoff frequency to reduce picture identifiability, and determine the cut-off frequency that is necessary to eliminate affective modulation on an evaluative and attentional level. If subjective rating and attentional capture rely on different types of information, they are also expected to differ in the cut-off spatial frequency below which no affective modulation is observed. Finally, previous studies demonstrated that the effects of size on emotional response differ between different systems (Codispoti & De Cesarei, 2007; De Cesarei & Codispoti, 2006; Reeves, Lang, Kim, & Tatar, 1999). In particular, affective modulation of skin conductance appeared to be very sensitive to size reduction, with more pronounced responses to arousing compared to neutral pictures in large (21° ⫻ 16°) but not small (3° ⫻ 2°) sizes. Therefore, it would be interesting to investigate the contribution of different spatial frequencies to the affective modulation of autonomic changes.
References Bamber, D. (1979). State trace analysis: A method of testing simple theories of causation. Journal of Mathematical Psychology, 19, 137– 181. Bar, M. (2003). A cortical mechanism for triggering top-down facilitation in visual object recognition. Journal of Cognitive Neuroscience, 15, 600 – 609. Beaver, J. D., Mogg, K., & Bradley, B. P. (2005). Emotional conditioning
to masked stimuli and modulation of visuospatial attention. Emotion, 5, 67–79. Bensafi, M., Rouby, C., Farget, V., Bertrand, B., Vigouroux, M., & Holley, A. (2002). Autonomic nervous system responses to odours: The role of pleasantness and arousal. Chemical Senses, 27, 703–709. Blair, K. S., Smith, B. W., Mitchell, D. G., Morton, J., Vythilingam, M., Pessoa, L., et al. (2007). Modulation of emotion by cognition and cognition by emotion. Neuroimage, 35, 430 – 440. Bradley, M. M., Codispoti, M., Cuthbert, B. N., & Lang, P. J. (2001). Emotion and Motivation I: Defensive and appetitive reactions in picture processing. Emotion, 1, 276 –298. Bradley, M. M., Codispoti, M., & Lang, P. J. (2006). A multi-process account of startle modulation during affective perception. Psychophysiology, 43, 486 – 497. Bradley, M. M., Cuthbert, B. N., & Lang, P. J. (1996). Picture media and emotion: Effects of a sustained affective context. Psychophysiology, 33, 662– 670. Bradley, M. M., Cuthbert, B. N., & Lang, P. J. (1999). Affect and the startle reflex. In M. E. Dawson, A. M. Schell, & A. H. Boehmelt (Eds.), Startle modification: Implications for neuroscience, cognitive science, and clinical science (pp. 157–183). New York: Cambridge University Press. Bradley, M. M., Drobes, D., & Lang, P. J. (1996). A probe for all reasons: Reflex and RT measures in perception [Abstract]. Psychophysiology, 33(Suppl.), S25. Bradley, M. M., & Lang, P. J. (2007). Emotion and motivation. In J. T. Cacioppo, L. G. Tassinary, & G. Berntson (Eds.), Handbook of psychophysiology (3rd ed., pp. 581– 607). New York: Cambridge University Press. Bradley, M. M., Lang, P. J., & Cuthbert, B. N. (1993). Emotion, novelty, and the startle reflex: Habituation in humans. Behavioral Neuroscience, 107, 970 –980. Bradley, M. M., Sabatinelli, D., Lang, P. J., Fitzsimmons, J. R., King, W., & Desai, P. (2003). Activation of the visual cortex in motivated attention. Behavioral Neuroscience, 117, 369 –380. Calvo, M. G., & Nummenmaa, L. (2007). Processing of unattended emotional visual scenes. Journal of Experimental Psychology General, 136, 347–369. Castiello, U., & Umilta`, C. (1990). Size of the attentional focus and efficiency of processing. Acta Psychologica, 73, 195–209. Codispoti, M., Bradley, M. M., & Lang, P. J. (2001). Affective reactions to briefly presented pictures. Psychophysiology, 38, 474 – 478. Codispoti, M., & De Cesarei, A. (2007). Arousal and attention: Picture size and emotional reactions. Psychophysiology, 44, 680 – 686. Codispoti, M., Ferrari, V., & Bradley, M. M. (2006). Repetitive picture processing: Autonomic and cortical correlates. Brain Research, 1068, 213–222. Codispoti, M., Ferrari, V., & Bradley, M. M. (2007). Repetition and ERPs: Distinguishing between early and late processes in affective picture perception. Journal of Cognitive Neuroscience, 19, 577–586. Codispoti, M., Ferrari, V., De Cesarei, A., & Cardinale, R. (2006). Implicit and explicit categorization of natural scenes. Progress in Brain Research, 156, 53– 65. Codispoti, M., Mazzetti, M., & Bradley, M. M. (2002). Exposure time and affective modulation in picture perception [Abstract]. Psychophysiology, 39, S62. De Cesarei, A., & Codispoti, M. (2006). When does size not matter? Effects of stimulus size on affective modulation. Psychophysiology, 43, 207–215. Derryberry, D., & Tucker, D. M. (1994). Motivating the focus of attention. In P. M. Niedenthal & S. Kitayama (Eds.), The heart’s eye: Emotional influences in perception and attention (pp. 167–196). San Diego, CA: Academic. Detenber, B. H., & Reeves, B. (1996). A bio-informational theory of
DE CESAREI AND CODISPOTI
emotion: Motion and image size effects on viewers. Journal of Communication, 46(3), 66 – 84. Dickinson, A., & Dearing, M. F. (1979). Appetitive-aversive interactions and inhibitory processes. In A. Dickinson & R. A. Boakes (Eds.), Mechanisms of learning and motivation. A memorial volume to Jerzy Konorski (pp. 203–231). Hillsdale, NJ: Erlbaum. Fox, E., Lester, V., Russo, R., Bowles, R., Pichler, A., & Dutton, K. (2000). Facial expressions of emotion: Are angry faces detected more efficiently? Cognition and Emotion, 14, 61–92. Fox, E., Russo, R., & Dutton, K. (2002). Attentional bias for threat: Evidence for delayed disengagement from emotional faces. Cognition and Emotion, 16, 355–379. Graham C. H. (1965). Visual space perception. In C. H. Graham (Ed.), Vision and visual perception (pp. 504 –547). New York: Wiley. Hartikainen, K. M., Ogawa, K. H., & Knight, R. T. (2000). Transient interference of right hemispheric function due to automatic emotional processing. Neuropsychologia, 38, 1576 –1580. Johnston, V. S., Miller, D. R., & Burleson, M. H. (1986). Multiple P3s to emotional stimuli and their theoretical significance. Psychophysiology, 23, 684 – 694. Konorski, J. (1967). Integrative activity of the brain: An interdisciplinary approach. Chicago: University of Chicago Press. Lang, P. J. (1979). A bio-informational theory of emotional imagery. Psychophysiology, 16, 495–512. Lang, P. J. (1980). Behavioral treatment and bio-behavioral treatment: Computer applications. In J. B. Sidowsky, J. H. Johnson, & T. A. Williams (Eds.), Technology in mental health care delivery systems (pp. 119 –137). Norwood, NJ: Ablex. Lang, P. J. (1995). The network model of emotion: Motivational connections. In R. S. Wyer & T. K. Srull (Eds.), Advances in social cognition (Vol. 6, pp. 109 –133). Mahwah, NJ: Erlbaum. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1990). Emotion, attention, and the startle reflex. Psychological Review, 97, 377–395. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1997). Motivated attention: Affect, activation, and action. In P. J. Lang, R. F. Simons, & M. Balaban (Eds.), Attention and orienting (pp. 97–135). Mahwah, NJ: Erlbaum. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (2001). International affective picture system (IAPS): Instruction manual and affective ratings (Tech. Rep. No. A–5). Gainesville: University of Florida, Center for Research in Psychophysiology. Loftus, G. R. (1996). Psychology will be a much better science when we change the way we analyze data. Current Directions in Psychological Science, 5, 161–171. Loftus, G. R. (2002). Analysis, interpretation, and visual presentation of data. In H. Pashler, S. Yantis, D. Medin, R. Gallistel, & J. Wixted (Eds.), Stevens’ handbook of experimental psychology (Vol. 4, 3rd ed., pp. 339 –390). New York: Wiley. Loftus, G. R., & Harley, E. M. (2004). How different spatial-frequency components contribute to visual information acquisition. Journal of Experimental Psychology: Human Perception and Performance, 30, 104 –118. Loftus, G. R., & Harley, E. M. (2005). Why is it easier to identify someone close than far away? Psychonomic Bulletin & Review, 12, 43– 65. Lombard, M., Reich, R. D., Grabe, M. E., Bracken, C. C., & Ditton, T. B. (2000). Presence and television. The role of screen size. Human Communication Research, 26, 75–98. Mogg, K., & Bradley, B. P. (1999). Orienting of attention to threatening facial expression presented under conditions of restricted awareness. Cognition and Emotion, 13, 713–740. Morrison, D. J., & Schyns, P. G. (2001). Usage of spatial scales for the categorization of faces, objects, and scenes. Psychonomic Bulletin and Review, 8, 454 – 469.
Nummenmaa, L., Hyo¨na¨, J., & Calvo, M. G. (2006). Eye movement assessment of selective attentional capture by emotional pictures. Emotion, 6, 257–268. ¨ hman, A., Lundqvist, D., & Esteves, F. (2001). The face in the crowd O revisited: A threat advantage with schematic stimuli. Journal of Personality and Social Psychology, 80, 381–396. Okon-Singer, H., Tzelgov, J., & Henik, A. (2007). Distinguishing between automaticity and attention in the processing of emotionally significant stimuli. Emotion, 7, 147–157. Osgood, C. E. (1969). On the whys and wherefores of E, P, and A. Journal of Personality and Social Psychology, 12, 194 –199. Pereira, M. G., Volchan, E., De Souza, G. G., Oliveira, L., Campagnoli, R. R., Pinheiro, W. M., et al. (2006). Sustained and transient modulation of performance induced by emotional picture viewing. Emotion, 6, 622– 634. Phelps, E. A., Ling, S., & Carrasco, M. (2006). Emotion facilitates perception and potentiates the perceptual benefits of attention. Psychological Science, 17, 292–299. Pratto, F., & John, O. P. (1991). Automatic vigilance: The attentiongrabbing power of negative social information. Journal of Personality and Social Psychology, 61, 380 –391. Radilova, J. (1982). The late positive component of visual evoked response sensitive to emotional factors. Activitas Nervosa Superior, Suppl. 3(Pt. 2), 334 –337. Ravaja, N. (2004a). Effects of image motion on a small screen on emotion, attention, and memory: Moving-face versus static-face newscaster. Journal of Broadcasting & Electronic Media, 48, 108 –133. Ravaja, N. (2004b). Effects of a small talking facial image on autonomic activity: The moderating influence of dispositional BIS and BAS sensitivities and emotions. Biological Psychology, 65, 163–183. Reeves, B., Lang, A., Kim, E. Y., & Tatar, D. (1999). The effects of screen size and message on attention and arousal. Media Psychology, 1, 49 – 67. Roring, R. W., Hines, F. G., & Charness, N., C. (2006). Age-related identification of emotions at different image size. Human Factors, 48, 675– 681. Roth, W. T. (1983). A comparison of P300 and skin conductance response. In A. W. K. Gaillard & W. Ritter (Eds.), Tutorials in event related potentials: Endogenous components (pp. 177–199). Amsterdam: NorthHolland. Russell, J. A., & Barrett, L. F. (1999). Affect, prototypical emotional episodes, and other things called emotion: Dissecting the elephant. Journal of Personality and Social Psychology, 76, 805– 819. Sabatinelli, D., Bradley, M. M., Lang, P. J., Costa, V. D., Versace, F. (2007). Pleasure rather than salience activates human nucleus accumbens and medial prefrontal cortex. Journal of Neurophysiology, 98, 1374 –1379. Sa´nchez-Navarro, J. P., Martı´nez-Selva, J. M., Roma´n, F., & Torrente, G. (2006). The effect of content and physical properties of affective pictures on emotional responses. Spanish Journal of Psychology, 9, 145–153. Sanocki, T. (2001). Interaction of scale and time during object identification. Journal of experimental psychology: Human perception and performance, 27, 290 –302. Schimmack, U. (2005). Attentional interference effects of emotional pictures: Threat, negativity, or arousal? Emotion, 5, 55– 66. Schneider, W., Eschman, A., & Zuccolotto, A. (2002). E-prime user’s guide. Pittsburgh, PA: Psychology Software Tools. Schneirla, T. (1959). An evolutionary and developmental theory of biphasic processes underlying approach and withdrawal. In M. Jones (Ed.), Nebraska symposium on motivation (pp. 27–58). Lincoln: University of Nebraska Press. Schupp, H. T., Cuthbert, B. N., Bradley, M. M., Hillman, C. H., Hamm, A. O., & Lang, P. J. (2004). Brain processes in emotional perception: Motivated attention. Cognition and Emotion, 18, 593– 611.
EMOTIONAL PROCESSING AND STIMULUS DEGRADATION Schyns, P. G., & Oliva, A. (1994). From blobs to boundary edges: Evidence for time and spatial scale dependent scene recognition. Psychological Science, 5, 195–200. Simons, R. F., Detenber, B. H., Roedema, T. M., & Reiss, J. E. (1999). Emotion processing in three systems: The medium and the message. Psychophysiology, 36, 619 – 627. Teghtsoonian, R., & Frost, R. O. (1982). The effect of viewing distance on fear of snakes. Journal of Behavior Therapy and Experimental Psychiatry, 13, 181–190. Verbruggen, F., & De Houwer, J. (2007). Do emotional stimuli interfere with response inhibition? Evidence from the stop signal paradigm. Cognition and Emotion, 21, 391– 403.
Vuilleumier, P., Armony, J. L., Driver, J., & Dolan, R. J. (2003). Distinct spatial frequency sensitivities for processing faces and emotional expressions. Nature Neuroscience, 6, 624 – 631. Vuilleumier, P., & Driver, J. (2007). Modulation of visual processing by attention and emotion: Windows on causal interactions between human brain regions. Philosophical Transactions of the Royal Society of London Biological Sciences, 362, 837– 855. Wundt, W. (1896). Grundriss der Psychologie [Outlines of Psychology] (6th ed.). Leipzig, Germany: Engelman.
Received July 17, 2007 Revision received January 20, 2008 Accepted February 1, 2008 䡲
E-Mail Notification of Your Latest Issue Online! Would you like to know when the next issue of your favorite APA journal will be available online? This service is now available to you. Sign up at http://notify.apa.org/ and you will be notified by e-mail when issues of interest to you become available!