- Email: [email protected]
Is power–space a continuum? Distance effect during power judgments
Consciousness and Cognition 37 (2015) 8–15
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Is power–space a continuum? Distance effect during power judgments Tianjiao Jiang, Lei Zhu ⇑ Department of Psychology, Fudan University, Handan Road 220, Shanghai SH 200433, China
a r t i c l e
i n f o
Article history: Received 25 March 2015 Revised 23 July 2015 Accepted 5 August 2015
Keywords: Power judgment Distance effect Dichotomic Continuous
a b s t r a c t Despite the increasing evidence suggesting that power processing can activate vertical space schema, it still remains unclear whether this power–space is dichotomic or continuous. Here we tested the nature of the power–space by the distance effect, a continuous property of space cognition. In two experiments, participants were required to judge the power of one single word (Experiment 1) or compare the power of two words presented in pairs (Experiment 2). The power distance was indexed by the absolute difference of power ratings. Results demonstrated that reaction time decreased with the power distance, whereas accuracy increased with the power distance. The findings indicated that different levels of power were presented as different vertical heights, implying that there was a common mechanism underlying space and power cognition. Ó 2015 Elsevier Inc. All rights reserved.
How abstract concepts such as power are mentally represented is an essential question that has received much attention within the domain of cognitive psychology. In the psychological literature, power has been defined as the ability or capacity to influence others through the control of resources (Galinsky, Gruenfeld, & Magee, 2003; Keltner, Gruenfeld, & Anderson, 2003). In our daily life, when we talk about power, we often use vertical information in our language. For example, leaders who supervise their employees have ‘‘high” status, or are ‘‘up” in the hierarchy, whereas the employees are at the ‘‘lower” levels of the hierarchy. Simply put, power is metaphorically understood as vertical height in physical space: ‘‘control is up, lack of control is down” (Lakoff, 1987; Lakoff & Johnson, 1980). This idea is broadly in line with the grounded cognition framework (e.g., Barsalou, 1999, 2008; Glenberg, 1997), which argues that conceptual thinking involves perceptual simulation. Representing abstract concepts reactivates previously stored information from sensory-motor experience to form a simulation of this sensory-motor experience. Supporting this analogue, several lines of evidence suggest the interactions between sensory-motor experience and power (Chiao, 2010; Chiao et al., 2009; Giessner & Schubert, 2007; Mason, Magee, & Fiske, 2014; Schubert, 2005; Zanolie et al., 2012). First, power judgments are affected by spatial information in the vertical dimensions provided by vision. In one of their experiments, Schubert (2005) presented participants with a series of pairs of group labels (e.g., employer–employee, master– servant), one at the top and the other at the bottom of the screen, and required them to judge which label was powerful. Participants reacted faster when powerful group labels appeared at top and powerless group labels appeared at the bottom. In the other experiments, single words referring to powerful or powerless groups were presented. Participants decided whether the word represented a powerful or powerless group. The stimulus position (either at the top or at the bottom of the computer screen) or response key (up or down cursor keys) was manipulated. Interactions between stimulus position ⇑ Corresponding author. E-mail address: [email protected] (L. Zhu). http://dx.doi.org/10.1016/j.concog.2015.08.002 1053-8100/Ó 2015 Elsevier Inc. All rights reserved.
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
9
or response key and power were found, i.e. participants responded faster to powerful groups when they appeared at the top of the screen and to powerless groups when they appeared at the bottom of the screen, and they responded faster to powerful groups with the up cursor key and to powerless groups with down cursor key. Our study further found that such interactions also appeared during tasks without explicit power evaluation, i.e., requiring participants to report whether the words represented people or animals (Jiang, Sun, & Zhu, 2015). Second, neuroimaging studies on a related abstract concept (i.e. social status) demonstrated that status judgment was mediated by parietal lobe regions that process up and down in allocentric space (Chiao, 2010; Chiao et al., 2009; Mason et al., 2014); these findings also point to partially overlapping mental representations of power and space. Given that people who have high status often hold power, status judgments likely also include power appraisal. Results from these neuroimaging studies thus implicate the parietal lobes as a locus of interactions between space and power. A third stream of research has shown that processing the concept of power produced an implicit reorientation of visuospatial attention, whereby powerful words bias attention upwards and powerless words bias attention downwards. In a recent study, target letters preceded by power judgment of words were identified faster when their spatial position was congruent with the perceived power of the preceding word than when it was incongruent (Zanolie et al., 2012). Higher P1 amplitude was also recorded for congruent trials (Zanolie et al., 2012). These effects are thought to result from an up-down image schema automatically activated during power processing. Although all the previous studies suggested that power processing activated vertical space schema, it still remains unclear whether this power–space is dichotomic or continuous. One conjecture is that different levels of power are presented as different vertical heights. However, the dichotomic manipulation of power (powerful vs. powerless) and vertical space (up vs. down) in the previous studies cannot rule out another possibility. That is, participants might just relate powerful to up and powerless to down. If the concept power and spatial schema were both activated, participants would easily notice the dichotomic relation between power and space. Distance effect could be used to test the nature of the power–space. Distance effect has been found in the judgment of many abstract concepts which were represented as a continuous space, such as numbers or status. It denotes the phenomenon that the amount of time it takes to compare two items of a concept is an inverse function of how much numerical distance separates those items. For example, people responded more slowly when they compare the numbers that are closer in quantity (e.g., 98 vs. 99) relative to those farther in quantity (e.g., 11 vs. 99, Dehaene, Piazza, Pinel, & Cohen, 2003; Moyer & Landauer, 1967). The distance effect has also been demonstrated in the social domains such as status (Chiao, Bordeaux, & Ambady, 2004; Chiao et al., 2009). Chiao et al. (2004) used ‘‘assistant professor” as an anchor and asked participants to decide whether the presented occupations (e.g., assistant professor, secretary) were higher, lower than or equal to the status of the anchor. They found that participants responded faster and faster with the increased status distance of each occupation from the anchor (i.e., long RT for assistant professor than secretary). Von Hecker, Klauer, and Sankaran (2013) found similar results by asking participants to compare the status of novel persons based on their interactions the participants previously learned from a story. Given that the distance effect could not be applied to dichotomic representations, if power–space is a continuum, distance effect could also apply to power judgment, like number or status. However, unlike status, the distance on power cannot be directly identified.1 It is difficult to judge which one is close to employee on the power rank, employer or manager. Compared with status, the concept of power is more likely to be presented continuously in the brain. Thus, for power, it is hard to select an explicit anchor and different items with various distances from the anchor. Therefore, we adopted a power judgment task without anchor in two experiments. In Experiment 1, participants explicitly judged whether the words represented a powerful or powerless group. We argued that participants also needed an intrinsic anchor which was supposed to be around the middle of the power rank when completing the task (Mussweiler, 2003). They judged the word as powerful, if its power was greater than this intrinsic anchor, and powerless if its power was lower than it. Then, the power distance could be identified referring to the power ratings of the words. For powerful words, the higher the power rating of a word is, the farther the word departs from the middle anchor. On the other hand, for powerless words, the lower the power rating of a word is, the farther the word departs from the anchor. Thus, 32 words (16 powerful words) denoting people taken from our previous study (Jiang et al., 2015, see Appendix A) were used as materials. The power of these words was already rated. Given the high rater-consistency of power (Kendall’s W = .76, cf. methods), the power ratings in Jiang et al.’s study can be used to assess the distance. The mean rating of all the words (3.76) was used as anchor to calculate power distance. In Experiment 2, participants were presented two-word pairs and asked to judge which one represents a more powerful group. The power distance could be identified referring to each participant’s power ratings of the words. For example, if a participant rated ‘‘king” as 9 and ‘‘soldier” as 4, then the distance was 5. Similarly, 12 words denoting people taken from Jiang et al. (2015, see Appendix A) were used as materials. It was predicted that reaction time would decrease with the power distance. 1. Experiment 1 In the experiment, participants viewed a series of eye gazes change (upward vs. downward), while hearing the 32 words described before, one at a time. They were required to judge whether the words represented powerful or powerless groups. It is predicted that reaction time would decrease with the power distance of each word.
1
Besides, power and status are conceptually different. Power is a kind of resources and status is a kind of respect (Hays, 2013).
10
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
Fig. 1. (A) Gaze stimuli. (B) Average log RT for each word decreased with the distance, whereas average accuracy for each word increased with the distance.
1.1. Method 1.1.1. Participants Twenty four right-handed volunteers from the university community with normal or corrected-to-normal vision (fifteen males, 24.33 years old (SD = 3.31)) participated in this experiment. Before the experiment, all the participants were asked which was their dominant hand, i.e., which hand they used to write or hold chopsticks. All the participants reported that they were right-handed. 1.1.2. Materials Thirty-two words denoting people taken from our previous study (Jiang et al., 2015, see Appendix A) were used as materials. Half of them denote powerful groups (e.g., king) and the other half denote powerless groups (e.g., servant). Another 5 words were also adopted as practice items appeared only once in the beginning. The power distance was indexed by the absolute value of difference between each word’s power rating and the mean. The power ratings correlated well across participants (Kendall’s W = .76, p = .000). The powerful words were rated to be more powerful than the powerless words (t(30) = 7.32, p < .001). The valence was counterbalanced between two kinds of words (t(30) = 0.03, p = .97). In the experiment, all the words were spoken by a female voice. The time durations of the words were 530–590 ms. The time durations of the powerful words were not significantly different from those of the powerless words (t(30) = 1.00, p > .05). The gaze change stimuli were created by eye pictures of a human face generated with the FaceGen (http://www.facegen. com). The eye pictures were 7.7 cm long and 1.2 cm high (Fig. 1A). A picture showing a direct gaze was followed by one of the two gaze changes: either upward or downward. The red2-color change pictures were also adopted as catch trials. All pictures were presented in the center of the screen on a black background. 1.1.3. Procedures The experiment consisted of 5 practice trials and 96 formal trials, 32 for each change (upward, downward and color). All the words appeared three times, once following each change. The presentation order of the 96 formal trails was randomized. The participants wore an earphone throughout the experiment. Each trial started with a 500 ms white cross presented in the center of the screen on a black background, followed by a direct gaze eye picture. Referring to Grade, Lefèvre, and Pesenti (2013), the direct gazes were presented for random durations of 500, 700, or 900 ms to prevent response anticipation. Then, the gaze changed to upward, downward or red for 2 s. At the same time, a word spoken by a female voice was heard from the earphone. To ensure the processing of gaze change, 32 catch trials with the picture changing to red color were added. Participants were required to judge the power of the word by pressing two horizontally aligned keys ‘‘j” and ‘‘k” on the keyboard as quickly and as accurately as possible when gazes changed to upward or downward, and to do nothing when the pictures turned to red. The response mappings (‘‘j” or ‘‘k” for powerful words) were counterbalanced across participants. 1.2. Results and discussion 1.2.1. Reaction times Participants responded in 2.86% (SD = 4.69%) trials during color changes. For each word in two gaze change conditions, more than 88% participants classified them as powerful or powerless correctly. Then, according to the general procedure in the literature (Chiao et al., 2004), correct reaction times (RTs) underwent logarithmic transformation to reduce skewness of latency distributions. Referring to Judd, Westfall, and Kenny (2012), we
2
For interpretation of color in Fig. 1, the reader is referred to the web version of this article.
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
11
carried out a mixed model analysis on log RT with participants and targets nested within power as random factors, power (powerful vs. powerless) and gaze locations as a fixed factors and distance as a fixed covariate. It was revealed that the effect of distance was significant (b = 0.016, t(1443.50) = 4.39, p < .001), indicating a negative correlation between distance and log RT (Fig. 1B). Consistent with previous studies (Jiang et al., 2015; Schubert, 2005), powerful words were judged faster than powerless words (b = 0.033, t(1443.73) = 5.55, p < .001) and the interaction between power and vertical gaze location was significant (b = 0.033, t(745.50) = 4.14, p < .001). Participants responded faster in downward-gaze trials than in upward-gaze trials (b = 0.027, t(753.99) = 3.21, p < .01). None of the other effects were significant. 1.2.2. Accuracy Then, we carried out a mixed model analysis on accuracy with participants and targets nested within power as random factors, power (powerful vs. powerless) and gaze locations as a fixed factors and distance as a fixed covariate. It was demonstrated that accuracy improved with the increase of the distance of each word (b = 0.021, t(1491.05) = 3.00, p < 0.01, Fig. 1B). None of the other effects were significant. 2. Experiment 2 The findings of Experiment 1 indicated that like other social concepts, distance effect could also be applied to power. However, the findings might be confounded with typicality effects. The observed data pattern could also be interpreted as that extreme exemplars of a power category are more typical and thus easier to judge. On the other hand, it is uncertain whether single-item judgment really involved comparisons to an implicit anchor. Thus, in Experiment 2, we used two-word comparisons. Participants viewed a series of word pairs and were required to judge which word represented a more powerful group. Then, they rated the power of each word. In order to rule out the influence of typicality, we classified word pairs to typical and atypical ones and analyzed them separately. 2.1. Method 2.1.1. Participants Twenty right-handed volunteers from the university community with normal or corrected-to-normal vision (eight males, M = 26.55 years old (SD = 3.41)) participated in this experiment. 2.1.2. Materials Twelve words denoting people taken from our previous study (Jiang et al., 2015, see Appendix A) were used as materials. A word pair was composed of any two words from the word list. Thus, there were sixty-six word pairs in all. All the word pairs were presented in the vertical center of the screen on a white background, one on the left and the other on the right. The location of each word in one pair (i.e., which one on the left and which one on the right) was randomized across participants. For example, for the word pair ‘‘king-soldier”, king appeared on the left for some participants and on the right for the other participants. The probability that each word appeared on the left was not significantly different from the probability that they appeared on the right (t(11) = 0.00, p = 1.00). All the words were presented in black, 48-point HEITI (a kind of Chinese font). Another 6 words composing 6 word pairs were also adopted as practice items appeared in the beginning. 2.1.3. Procedures The experiment consisted of 6 practice trials and 66 formal trials, each for one pair of words. The presentation order of the 66 formal trails was randomized. Each trial started with a 500 ms black cross presented in the center of the screen on a white background, followed by a word pair presented for 3 s. Participants were required to judge which word represented a more powerful group by pressing two horizontally aligned keys on the keyboard as quickly and accurately as possible. If the left word was more powerful, they pressed ‘‘F” with their left index finger. If the right word was more powerful, they pressed ‘‘J” with their right index finger. After the experiment, participants were presented with the 12 words in a random order and asked to rate the power of each word on a 9-point Likert scale, 1 indicating extremely powerless and 9 indicating extremely powerful. 2.2. Results and discussion 2.2.1. Reaction times For each participant, the power distance of a word pair was indexed by the absolute value of the difference between two power ratings. It could be an integer varying from 0 to 8. Then, the correct answer of each pair was also identified by participants’ own power ratings. If the power rating of the left word was higher than that of the right word, the correct answer was ‘‘F”. On the contrary, if the power rating of the left word was lower than that of the right word, the correct answer was ‘‘J”. If the power ratings of two words were equal, either ‘‘F” or ‘‘J” was considered as correct. Then, according to the general
12
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
Fig. 2. (A) Average log RT for each pair decreased with the distance. (B) Average log RT for atypical pair decreased with the distance. (C) Average log RT for typical pair decreased with the distance.
procedure in the literature (Chiao et al., 2004), correct reaction times (RTs) underwent logarithmic transformation to reduce skewness of latency distributions. Referring to Judd et al. (2012), we carried out a mixed model analysis on log RT with participants and targets as random factors and distance as a fixed covariate. It was revealed that the effect of distance was significant (b = 0.025, t (1187.76) = 14.50, p < .001), indicating a negative correlation between distance and log RT (Fig. 2A). However, some pairs were more typical than others as exemplars of power. In our daily life, the scene that a general controls a soldier appears more often than the scene that a general controls a baby. This might confound with the distance effect, given that responses to typical pairs might be faster than response to atypical pairs. That is, the distance effect could also emerge when typical pairs had long distance between each other. Thus, we classified the word pairs Nurserymaid– Baby, Nurserymaid–Child, Nurserymaid–Sick person, Nurserymaid–Host, Attendant–Host, Attendant–King, Attendant–General, Attendant–Overlord, Soldier–King, Soldier–General, General–King as typical pairs and the others as atypical pairs. It is found that the distance of typical pairs was not significantly different from that of atypical pairs (t(64) = 0.36, p > .05) and the difference of log RT between typical and atypical pairs was also not significant (t(19) = 0.55, p > .05). Then, two mixed models were created to analyze typical and atypical pairs separately. The results showed that distance effect could be applied to both atypical and typical pairs (atypical: b = 0.026, t(984.14) = 13.56, p < .001; typical: b = 0.022, t (188.74) = 4.77, p < .001, Fig. 2B and C). In addition, responses given by the right hand (‘‘J”) might be faster than the responses given by the left hand (‘‘F”). However, the ‘‘F” and ‘‘J” response frequency could not be counterbalanced before the experiment. This might confound with the distance effect. That is, the distance effect could also emerge when there were more ‘‘F” responses for the near distance and more ‘‘J” responses for the far distance. Thus, we examined whether the frequency of ‘‘F” and ‘‘J” responses was balanced for each distance. It was revealed that the frequency of ‘‘F” responses was not significantly different from that of ‘‘J” responses for all the distances (ps > .05). Besides, the log RTs of the ‘‘F” responses were not significantly different from those of the ‘‘J” responses (t(19) = 2.07, p > .05). 2.2.2. Accuracy First, the accuracy of the ‘‘F” responses was not significantly different from that of the ‘‘J” responses (t(19) = 0.21, p > .05) and the difference of accuracy between typical and atypical pairs was also not significant (t(19) = 1.06, p > .05). Then, we carried out three mixed model analyses for all the pairs, typical pairs and atypical pairs separately. It was demonstrated that accuracy improved with the increase of the distance between two words (b = 0.029, t(1318) = 8.14, p < 0.001, Fig. 3). And, this trend was significant for both atypical pairs (b = 0.028, t(1098) = 7.20, p < 0.001) and typical pairs (b = 0.034, t(218) = 3.86, p < 0.001). 3. General discussion The present study addressed the question of whether the power space is dichotomic or continuous. To answer this question, we tested the distance effect of power judgment. Distance effect has been found in the judgments of many abstract concepts represented as a continuous space, such as numbers or status. It denotes the phenomenon that the amount of time it takes to compare two items of a concept is an inverse function of how much numerical distance separates those items. For example, people responded more slowly when comparing numbers and social status that were closer relative to those farther (Chiao et al., 2004; Dehaene et al., 2003; Moyer & Landauer, 1967; Von Hecker et al., 2013).
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
13
Fig. 3. (A) Average accuracy for each pair increased with the distance. (B) Average accuracy for atypical pair increased with the distance. (C) Average accuracy for typical pair increased with the distance.
Chiao et al. (2009) further revealed that this distance effect was mediated by the intraparietal sulcus (IPS) implicated in processing continuous spatial information. In the present study, reaction times were also associated with distance such that those word or word pairs with long distance were judged more quickly and accurately. Given that the distance effect did not apply to dichotomic representations, this finding implied that the different levels of power were presented as different vertical heights. In addition, in the studies on distance effect of status (Chiao et al., 2004, 2009; Von Hecker et al., 2013), the distance was identified by researchers or other participants in a pilot study. In Experiment 2, however, the power distance was identified referring to each participant’s own power ratings, which increased the reliability of the results. The findings of the present study completely ruled out the possibility that power was related to space in a dichotomic manner (i.e., powerful = up and powerless = down) and people were inclined to notice this relation when both power and vertical space were manipulated with only two levels. Zanolie et al. (2012) argued that participants might have a response bias during power judgment: responding ‘‘powerful” to stimuli at the top of the screen and ‘‘powerless” to stimuli at the bottom of the screen. Power judgments did not trigger a space schema. Rather, people just noticed the dichotomic power–space relation when concepts of power and space were both activated. Thus, Zanolie et al. (2012) did not manipulate the vertical positions of the words. Instead, Zanolie et al. used an attentional shift paradigm to remove such response bias. In their experiment, participants were required to identify the target letters preceded by power judgment of words. The space locations of the words were manipulated by two levels: up and down. It is found that target letters were identified faster when their spatial positions were congruent with the perceived power of the preceding word than when they were incongruent. However, as Zanolie et al. claimed, the concepts of power and space were both activated. And, in such situation, it was easier for participants to notice the dichotomic power–space relation. On the contrary, the distance effect demonstrated in our study directly reflected the spatial properties of power cognition. More broadly, our results support the grounded cognition framework (e.g., Barsalou, 1999, 2008; Glenberg, 1997). The basic tenet of this framework is that human cognition is body based. That is, conceptual thinking often relies on simulations of perceptual symbols, which are schematized perceptual experiences including senses, proprioception, introspection, and motor programs. There is now increasing evidence that abstract concepts are also grounded in sensory-motor representations through metaphors (Conceptual Metaphor Theory, Gibbs, 1994; Lakoff & Johnson, 1980, 1999). For example, power mapped to verticality (Giessner & Schubert, 2007; Schubert, 2005; our study), physical force (Schubert, 2004) and size (Schubert, Waldzus, & Giessner, 2009), social proximity mapped to temperature (Ijzerman & Semin, 2009), and social distance mapped to spatial distance (Parkinson & Wheatley, 2013). Acknowledgments This research was supported by National Natural Science Foundation of China (31100728), Projects Planning in Shanghai Philosophy and Social Sciences Research (2012JJY001) and the research fund of the School of Social Development and Public Policy at Fudan University. Appendix A See Tables 1 and 2.
14
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
Table 1 Stimuli used in Experiment 1. Powerless
Powerful
Word
English
Power
Valence
Duration (ms)
Word
English
Power
Valence
Duration (ms)
婴儿 孤儿 病人 奴隶 孩子 丫头 仆人 随从 实习生 保姆 学徒 学生 农夫 农民 水手 士兵
Baby Orphan Sick person Slave Child Bonne Servant Attendant Intern Nurserymaid Apprentice Student Plowman Farmer Sailor Soldier
1.58 1.67 1.75 1.75 1.92 2.00 2.08 2.42 2.50 2.58 2.58 3.00 3.33 3.33 3.83 4.08
6.17 3.25 2.83 2.17 5.92 4.83 3.25 3.08 4.08 3.67 3.67 5.00 4.58 4.33 5.00 4.50
540 570 550 540 560 550 550 540 560 560 550 530 560 530 540 540
国王 总统 君主 统治者 霸主 政治家 官员 老板 将军 总理 首领 教练 主任 经理 校长 主人
King President Sovereign Governor Overlord Politician Officer Boss General Premier Leader Coach Director Manager Principal Host
5.83 6.17 6.33 6.50 6.75 4.33 4.50 4.58 5.42 5.42 5.50 3.17 3.33 3.75 4.17 4.25
4.50 4.92 5.42 3.42 4.75 2.83 2.92 3.67 5.17 4.75 4.92 4.08 3.25 3.92 4.33 3.50
550 550 550 570 540 590 550 540 550 550 560 540 540 550 560 550
Table 2 Stimuli used in Experiment 2. Word
In English
Power
霸主 保姆 病人 国王 孩子 将军 经理 士兵 随从 学生 婴儿 主人
Overlord Nurserymaid Sick person King Child General Manager Soldier Attendant Student Baby Host
8.23 2.87 2.42 8.60 3.09 7.63 6.70 3.84 3.24 3.83 2.28 6.65
Note. Power indicates the average power ratings for each word across all the participants.
References Barsalou, L. W. (1999). Perceptual symbol systems. Behavioral and Brain Sciences, 22, 577–609. Barsalou, L. W. (2008). Cognitive and neural contributions to understanding the conceptual system. Current Directions in Psychological Science, 17, 91–95. Chiao, J. Y. (2010). Neural basis of social status hierarchy across species. Current Opinion in Neurobiology, 20, 803–809. Chiao, J. Y., Bordeaux, A. R., & Ambady, N. (2004). Mental representations of social status. Cognition, 93, B49–B57. Chiao, J. Y., Harada, T., Oby, E. R., Li, Z., Parrish, T., & Bridge, D. J. (2009). Neural representations of social status hierarchy in human inferior parietal cortex. Neuropsychologia, 47, 354–363. Dehaene, S., Piazza, M., Pinel, P., & Cohen, L. (2003). Three parietal circuits for number processing. Cognitive Neuropsychology, 20, 487–506. Galinsky, A. D., Gruenfeld, D., & Magee, J. C. (2003). From power to action. Journal of Personality and Social Psychology, 85, 453–466. Gibbs, R. W. (1994). The poetics of mind: Figurative thought, language, and understanding. New York, NY: Cambridge University Press. Giessner, S. R., & Schubert, T. W. (2007). High in the hierarchy: How vertical location and judgments of leaders’ power are interrelated. Organizational Behavior and Human Decision Processes, 104, 30–44. Glenberg, A. M. (1997). What memory is for? Behavioral and Brain Sciences, 20, 1–55. Grade, S., Lefèvre, N., & Pesenti, M. (2013). Influence of gaze observation on random number generation. Experimental Psychology, 60, 122–130. Hays, N. A. (2013). Fear and loving in social hierarchy: Sex differences in preferences for power versus status. Journal of Experimental Social Psychology, 49, 1130–1136. Ijzerman, H., & Semin, G. R. (2009). The thermometer of social relations. Mapping social proximity on temperature. Psychological Science, 20, 1214–1220. Jiang, T., Sun, L., & Zhu, L. (2015). The influence of vertical motor responses on explicit and incidental processing of power words. Consciousness and Cognition, 34, 33–42. Judd, C. M., Westfall, J., & Kenny, D. A. (2012). Treating stimuli as a random factor in social psychology: A new and comprehensive solution to a pervasive but largely ignored problem. Journal of Personality and Social Psychology, 103, 54–69. Keltner, D., Gruenfeld, D., & Anderson, C. P. (2003). Power, approach, and inhibition. Psychological Review, 110, 265–284. Lakoff, G. (1987). Women, fire, and dangerous things. Chicago: University of Chicago Press. Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: University of Chicago Press. Lakoff, G., & Johnson, M. (1999). Philosophy in the flesh. New York, NY: Basic Books. Mason, M., Magee, J. C., & Fiske, S. T. (2014). Neural substrates of social status inference: Roles of medial prefrontal cortex and superior temporal sulcus. Journal of Cognitive Neuroscience, 26, 1131–1140. Moyer, R. S., & Landauer, T. K. (1967). Time required for judgments of numerical inequality. Nature, 215, 1519–1520. Mussweiler, T. (2003). Comparison processes in social judgment: Mechanism and consequences. Psychological Review, 110, 472–489.
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
15
Parkinson, C., & Wheatley, T. (2013). Old cortex, new contexts: Re-purposing spatial perception for social cognition. Frontiers in Human Neuroscience, 7, 645. Schubert, T. W. (2004). The power in your hand: Gender differences in bodily feedback from making a fist. Personality and Social Psychology Bulletin, 30, 757–769. Schubert, T. W. (2005). Your highness: Vertical positions as perceptual symbols of power. Journal of Personality and Social Psychology, 89, 1–21. Schubert, T. W., Waldzus, S., & Giessner, S. R. (2009). Control over the association of power and size. Social Cognition, 27, 1–19. Von Hecker, U., Klauer, C. K., & Sankaran, S. (2013). Embodiment of social status: Verticality effects in multi-level rank-orders. Social Cognition, 31, 374–389. Zanolie, K., Dantzig, S., Boot, I., Wijnen, J., Schubert, T. W., Giessner, S. R., et al (2012). Mighty metaphors: Behavioral and ERP evidence that power shifts attention on a vertical dimension. Brain and Cognition, 78, 50–58.
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Is power–space a continuum? Distance effect during power judgments Tianjiao Jiang, Lei Zhu ⇑ Department of Psychology, Fudan University, Handan Road 220, Shanghai SH 200433, China
a r t i c l e
i n f o
Article history: Received 25 March 2015 Revised 23 July 2015 Accepted 5 August 2015
Keywords: Power judgment Distance effect Dichotomic Continuous
a b s t r a c t Despite the increasing evidence suggesting that power processing can activate vertical space schema, it still remains unclear whether this power–space is dichotomic or continuous. Here we tested the nature of the power–space by the distance effect, a continuous property of space cognition. In two experiments, participants were required to judge the power of one single word (Experiment 1) or compare the power of two words presented in pairs (Experiment 2). The power distance was indexed by the absolute difference of power ratings. Results demonstrated that reaction time decreased with the power distance, whereas accuracy increased with the power distance. The findings indicated that different levels of power were presented as different vertical heights, implying that there was a common mechanism underlying space and power cognition. Ó 2015 Elsevier Inc. All rights reserved.
How abstract concepts such as power are mentally represented is an essential question that has received much attention within the domain of cognitive psychology. In the psychological literature, power has been defined as the ability or capacity to influence others through the control of resources (Galinsky, Gruenfeld, & Magee, 2003; Keltner, Gruenfeld, & Anderson, 2003). In our daily life, when we talk about power, we often use vertical information in our language. For example, leaders who supervise their employees have ‘‘high” status, or are ‘‘up” in the hierarchy, whereas the employees are at the ‘‘lower” levels of the hierarchy. Simply put, power is metaphorically understood as vertical height in physical space: ‘‘control is up, lack of control is down” (Lakoff, 1987; Lakoff & Johnson, 1980). This idea is broadly in line with the grounded cognition framework (e.g., Barsalou, 1999, 2008; Glenberg, 1997), which argues that conceptual thinking involves perceptual simulation. Representing abstract concepts reactivates previously stored information from sensory-motor experience to form a simulation of this sensory-motor experience. Supporting this analogue, several lines of evidence suggest the interactions between sensory-motor experience and power (Chiao, 2010; Chiao et al., 2009; Giessner & Schubert, 2007; Mason, Magee, & Fiske, 2014; Schubert, 2005; Zanolie et al., 2012). First, power judgments are affected by spatial information in the vertical dimensions provided by vision. In one of their experiments, Schubert (2005) presented participants with a series of pairs of group labels (e.g., employer–employee, master– servant), one at the top and the other at the bottom of the screen, and required them to judge which label was powerful. Participants reacted faster when powerful group labels appeared at top and powerless group labels appeared at the bottom. In the other experiments, single words referring to powerful or powerless groups were presented. Participants decided whether the word represented a powerful or powerless group. The stimulus position (either at the top or at the bottom of the computer screen) or response key (up or down cursor keys) was manipulated. Interactions between stimulus position ⇑ Corresponding author. E-mail address: [email protected] (L. Zhu). http://dx.doi.org/10.1016/j.concog.2015.08.002 1053-8100/Ó 2015 Elsevier Inc. All rights reserved.
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
9
or response key and power were found, i.e. participants responded faster to powerful groups when they appeared at the top of the screen and to powerless groups when they appeared at the bottom of the screen, and they responded faster to powerful groups with the up cursor key and to powerless groups with down cursor key. Our study further found that such interactions also appeared during tasks without explicit power evaluation, i.e., requiring participants to report whether the words represented people or animals (Jiang, Sun, & Zhu, 2015). Second, neuroimaging studies on a related abstract concept (i.e. social status) demonstrated that status judgment was mediated by parietal lobe regions that process up and down in allocentric space (Chiao, 2010; Chiao et al., 2009; Mason et al., 2014); these findings also point to partially overlapping mental representations of power and space. Given that people who have high status often hold power, status judgments likely also include power appraisal. Results from these neuroimaging studies thus implicate the parietal lobes as a locus of interactions between space and power. A third stream of research has shown that processing the concept of power produced an implicit reorientation of visuospatial attention, whereby powerful words bias attention upwards and powerless words bias attention downwards. In a recent study, target letters preceded by power judgment of words were identified faster when their spatial position was congruent with the perceived power of the preceding word than when it was incongruent (Zanolie et al., 2012). Higher P1 amplitude was also recorded for congruent trials (Zanolie et al., 2012). These effects are thought to result from an up-down image schema automatically activated during power processing. Although all the previous studies suggested that power processing activated vertical space schema, it still remains unclear whether this power–space is dichotomic or continuous. One conjecture is that different levels of power are presented as different vertical heights. However, the dichotomic manipulation of power (powerful vs. powerless) and vertical space (up vs. down) in the previous studies cannot rule out another possibility. That is, participants might just relate powerful to up and powerless to down. If the concept power and spatial schema were both activated, participants would easily notice the dichotomic relation between power and space. Distance effect could be used to test the nature of the power–space. Distance effect has been found in the judgment of many abstract concepts which were represented as a continuous space, such as numbers or status. It denotes the phenomenon that the amount of time it takes to compare two items of a concept is an inverse function of how much numerical distance separates those items. For example, people responded more slowly when they compare the numbers that are closer in quantity (e.g., 98 vs. 99) relative to those farther in quantity (e.g., 11 vs. 99, Dehaene, Piazza, Pinel, & Cohen, 2003; Moyer & Landauer, 1967). The distance effect has also been demonstrated in the social domains such as status (Chiao, Bordeaux, & Ambady, 2004; Chiao et al., 2009). Chiao et al. (2004) used ‘‘assistant professor” as an anchor and asked participants to decide whether the presented occupations (e.g., assistant professor, secretary) were higher, lower than or equal to the status of the anchor. They found that participants responded faster and faster with the increased status distance of each occupation from the anchor (i.e., long RT for assistant professor than secretary). Von Hecker, Klauer, and Sankaran (2013) found similar results by asking participants to compare the status of novel persons based on their interactions the participants previously learned from a story. Given that the distance effect could not be applied to dichotomic representations, if power–space is a continuum, distance effect could also apply to power judgment, like number or status. However, unlike status, the distance on power cannot be directly identified.1 It is difficult to judge which one is close to employee on the power rank, employer or manager. Compared with status, the concept of power is more likely to be presented continuously in the brain. Thus, for power, it is hard to select an explicit anchor and different items with various distances from the anchor. Therefore, we adopted a power judgment task without anchor in two experiments. In Experiment 1, participants explicitly judged whether the words represented a powerful or powerless group. We argued that participants also needed an intrinsic anchor which was supposed to be around the middle of the power rank when completing the task (Mussweiler, 2003). They judged the word as powerful, if its power was greater than this intrinsic anchor, and powerless if its power was lower than it. Then, the power distance could be identified referring to the power ratings of the words. For powerful words, the higher the power rating of a word is, the farther the word departs from the middle anchor. On the other hand, for powerless words, the lower the power rating of a word is, the farther the word departs from the anchor. Thus, 32 words (16 powerful words) denoting people taken from our previous study (Jiang et al., 2015, see Appendix A) were used as materials. The power of these words was already rated. Given the high rater-consistency of power (Kendall’s W = .76, cf. methods), the power ratings in Jiang et al.’s study can be used to assess the distance. The mean rating of all the words (3.76) was used as anchor to calculate power distance. In Experiment 2, participants were presented two-word pairs and asked to judge which one represents a more powerful group. The power distance could be identified referring to each participant’s power ratings of the words. For example, if a participant rated ‘‘king” as 9 and ‘‘soldier” as 4, then the distance was 5. Similarly, 12 words denoting people taken from Jiang et al. (2015, see Appendix A) were used as materials. It was predicted that reaction time would decrease with the power distance. 1. Experiment 1 In the experiment, participants viewed a series of eye gazes change (upward vs. downward), while hearing the 32 words described before, one at a time. They were required to judge whether the words represented powerful or powerless groups. It is predicted that reaction time would decrease with the power distance of each word.
1
Besides, power and status are conceptually different. Power is a kind of resources and status is a kind of respect (Hays, 2013).
10
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
Fig. 1. (A) Gaze stimuli. (B) Average log RT for each word decreased with the distance, whereas average accuracy for each word increased with the distance.
1.1. Method 1.1.1. Participants Twenty four right-handed volunteers from the university community with normal or corrected-to-normal vision (fifteen males, 24.33 years old (SD = 3.31)) participated in this experiment. Before the experiment, all the participants were asked which was their dominant hand, i.e., which hand they used to write or hold chopsticks. All the participants reported that they were right-handed. 1.1.2. Materials Thirty-two words denoting people taken from our previous study (Jiang et al., 2015, see Appendix A) were used as materials. Half of them denote powerful groups (e.g., king) and the other half denote powerless groups (e.g., servant). Another 5 words were also adopted as practice items appeared only once in the beginning. The power distance was indexed by the absolute value of difference between each word’s power rating and the mean. The power ratings correlated well across participants (Kendall’s W = .76, p = .000). The powerful words were rated to be more powerful than the powerless words (t(30) = 7.32, p < .001). The valence was counterbalanced between two kinds of words (t(30) = 0.03, p = .97). In the experiment, all the words were spoken by a female voice. The time durations of the words were 530–590 ms. The time durations of the powerful words were not significantly different from those of the powerless words (t(30) = 1.00, p > .05). The gaze change stimuli were created by eye pictures of a human face generated with the FaceGen (http://www.facegen. com). The eye pictures were 7.7 cm long and 1.2 cm high (Fig. 1A). A picture showing a direct gaze was followed by one of the two gaze changes: either upward or downward. The red2-color change pictures were also adopted as catch trials. All pictures were presented in the center of the screen on a black background. 1.1.3. Procedures The experiment consisted of 5 practice trials and 96 formal trials, 32 for each change (upward, downward and color). All the words appeared three times, once following each change. The presentation order of the 96 formal trails was randomized. The participants wore an earphone throughout the experiment. Each trial started with a 500 ms white cross presented in the center of the screen on a black background, followed by a direct gaze eye picture. Referring to Grade, Lefèvre, and Pesenti (2013), the direct gazes were presented for random durations of 500, 700, or 900 ms to prevent response anticipation. Then, the gaze changed to upward, downward or red for 2 s. At the same time, a word spoken by a female voice was heard from the earphone. To ensure the processing of gaze change, 32 catch trials with the picture changing to red color were added. Participants were required to judge the power of the word by pressing two horizontally aligned keys ‘‘j” and ‘‘k” on the keyboard as quickly and as accurately as possible when gazes changed to upward or downward, and to do nothing when the pictures turned to red. The response mappings (‘‘j” or ‘‘k” for powerful words) were counterbalanced across participants. 1.2. Results and discussion 1.2.1. Reaction times Participants responded in 2.86% (SD = 4.69%) trials during color changes. For each word in two gaze change conditions, more than 88% participants classified them as powerful or powerless correctly. Then, according to the general procedure in the literature (Chiao et al., 2004), correct reaction times (RTs) underwent logarithmic transformation to reduce skewness of latency distributions. Referring to Judd, Westfall, and Kenny (2012), we
2
For interpretation of color in Fig. 1, the reader is referred to the web version of this article.
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
11
carried out a mixed model analysis on log RT with participants and targets nested within power as random factors, power (powerful vs. powerless) and gaze locations as a fixed factors and distance as a fixed covariate. It was revealed that the effect of distance was significant (b = 0.016, t(1443.50) = 4.39, p < .001), indicating a negative correlation between distance and log RT (Fig. 1B). Consistent with previous studies (Jiang et al., 2015; Schubert, 2005), powerful words were judged faster than powerless words (b = 0.033, t(1443.73) = 5.55, p < .001) and the interaction between power and vertical gaze location was significant (b = 0.033, t(745.50) = 4.14, p < .001). Participants responded faster in downward-gaze trials than in upward-gaze trials (b = 0.027, t(753.99) = 3.21, p < .01). None of the other effects were significant. 1.2.2. Accuracy Then, we carried out a mixed model analysis on accuracy with participants and targets nested within power as random factors, power (powerful vs. powerless) and gaze locations as a fixed factors and distance as a fixed covariate. It was demonstrated that accuracy improved with the increase of the distance of each word (b = 0.021, t(1491.05) = 3.00, p < 0.01, Fig. 1B). None of the other effects were significant. 2. Experiment 2 The findings of Experiment 1 indicated that like other social concepts, distance effect could also be applied to power. However, the findings might be confounded with typicality effects. The observed data pattern could also be interpreted as that extreme exemplars of a power category are more typical and thus easier to judge. On the other hand, it is uncertain whether single-item judgment really involved comparisons to an implicit anchor. Thus, in Experiment 2, we used two-word comparisons. Participants viewed a series of word pairs and were required to judge which word represented a more powerful group. Then, they rated the power of each word. In order to rule out the influence of typicality, we classified word pairs to typical and atypical ones and analyzed them separately. 2.1. Method 2.1.1. Participants Twenty right-handed volunteers from the university community with normal or corrected-to-normal vision (eight males, M = 26.55 years old (SD = 3.41)) participated in this experiment. 2.1.2. Materials Twelve words denoting people taken from our previous study (Jiang et al., 2015, see Appendix A) were used as materials. A word pair was composed of any two words from the word list. Thus, there were sixty-six word pairs in all. All the word pairs were presented in the vertical center of the screen on a white background, one on the left and the other on the right. The location of each word in one pair (i.e., which one on the left and which one on the right) was randomized across participants. For example, for the word pair ‘‘king-soldier”, king appeared on the left for some participants and on the right for the other participants. The probability that each word appeared on the left was not significantly different from the probability that they appeared on the right (t(11) = 0.00, p = 1.00). All the words were presented in black, 48-point HEITI (a kind of Chinese font). Another 6 words composing 6 word pairs were also adopted as practice items appeared in the beginning. 2.1.3. Procedures The experiment consisted of 6 practice trials and 66 formal trials, each for one pair of words. The presentation order of the 66 formal trails was randomized. Each trial started with a 500 ms black cross presented in the center of the screen on a white background, followed by a word pair presented for 3 s. Participants were required to judge which word represented a more powerful group by pressing two horizontally aligned keys on the keyboard as quickly and accurately as possible. If the left word was more powerful, they pressed ‘‘F” with their left index finger. If the right word was more powerful, they pressed ‘‘J” with their right index finger. After the experiment, participants were presented with the 12 words in a random order and asked to rate the power of each word on a 9-point Likert scale, 1 indicating extremely powerless and 9 indicating extremely powerful. 2.2. Results and discussion 2.2.1. Reaction times For each participant, the power distance of a word pair was indexed by the absolute value of the difference between two power ratings. It could be an integer varying from 0 to 8. Then, the correct answer of each pair was also identified by participants’ own power ratings. If the power rating of the left word was higher than that of the right word, the correct answer was ‘‘F”. On the contrary, if the power rating of the left word was lower than that of the right word, the correct answer was ‘‘J”. If the power ratings of two words were equal, either ‘‘F” or ‘‘J” was considered as correct. Then, according to the general
12
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
Fig. 2. (A) Average log RT for each pair decreased with the distance. (B) Average log RT for atypical pair decreased with the distance. (C) Average log RT for typical pair decreased with the distance.
procedure in the literature (Chiao et al., 2004), correct reaction times (RTs) underwent logarithmic transformation to reduce skewness of latency distributions. Referring to Judd et al. (2012), we carried out a mixed model analysis on log RT with participants and targets as random factors and distance as a fixed covariate. It was revealed that the effect of distance was significant (b = 0.025, t (1187.76) = 14.50, p < .001), indicating a negative correlation between distance and log RT (Fig. 2A). However, some pairs were more typical than others as exemplars of power. In our daily life, the scene that a general controls a soldier appears more often than the scene that a general controls a baby. This might confound with the distance effect, given that responses to typical pairs might be faster than response to atypical pairs. That is, the distance effect could also emerge when typical pairs had long distance between each other. Thus, we classified the word pairs Nurserymaid– Baby, Nurserymaid–Child, Nurserymaid–Sick person, Nurserymaid–Host, Attendant–Host, Attendant–King, Attendant–General, Attendant–Overlord, Soldier–King, Soldier–General, General–King as typical pairs and the others as atypical pairs. It is found that the distance of typical pairs was not significantly different from that of atypical pairs (t(64) = 0.36, p > .05) and the difference of log RT between typical and atypical pairs was also not significant (t(19) = 0.55, p > .05). Then, two mixed models were created to analyze typical and atypical pairs separately. The results showed that distance effect could be applied to both atypical and typical pairs (atypical: b = 0.026, t(984.14) = 13.56, p < .001; typical: b = 0.022, t (188.74) = 4.77, p < .001, Fig. 2B and C). In addition, responses given by the right hand (‘‘J”) might be faster than the responses given by the left hand (‘‘F”). However, the ‘‘F” and ‘‘J” response frequency could not be counterbalanced before the experiment. This might confound with the distance effect. That is, the distance effect could also emerge when there were more ‘‘F” responses for the near distance and more ‘‘J” responses for the far distance. Thus, we examined whether the frequency of ‘‘F” and ‘‘J” responses was balanced for each distance. It was revealed that the frequency of ‘‘F” responses was not significantly different from that of ‘‘J” responses for all the distances (ps > .05). Besides, the log RTs of the ‘‘F” responses were not significantly different from those of the ‘‘J” responses (t(19) = 2.07, p > .05). 2.2.2. Accuracy First, the accuracy of the ‘‘F” responses was not significantly different from that of the ‘‘J” responses (t(19) = 0.21, p > .05) and the difference of accuracy between typical and atypical pairs was also not significant (t(19) = 1.06, p > .05). Then, we carried out three mixed model analyses for all the pairs, typical pairs and atypical pairs separately. It was demonstrated that accuracy improved with the increase of the distance between two words (b = 0.029, t(1318) = 8.14, p < 0.001, Fig. 3). And, this trend was significant for both atypical pairs (b = 0.028, t(1098) = 7.20, p < 0.001) and typical pairs (b = 0.034, t(218) = 3.86, p < 0.001). 3. General discussion The present study addressed the question of whether the power space is dichotomic or continuous. To answer this question, we tested the distance effect of power judgment. Distance effect has been found in the judgments of many abstract concepts represented as a continuous space, such as numbers or status. It denotes the phenomenon that the amount of time it takes to compare two items of a concept is an inverse function of how much numerical distance separates those items. For example, people responded more slowly when comparing numbers and social status that were closer relative to those farther (Chiao et al., 2004; Dehaene et al., 2003; Moyer & Landauer, 1967; Von Hecker et al., 2013).
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
13
Fig. 3. (A) Average accuracy for each pair increased with the distance. (B) Average accuracy for atypical pair increased with the distance. (C) Average accuracy for typical pair increased with the distance.
Chiao et al. (2009) further revealed that this distance effect was mediated by the intraparietal sulcus (IPS) implicated in processing continuous spatial information. In the present study, reaction times were also associated with distance such that those word or word pairs with long distance were judged more quickly and accurately. Given that the distance effect did not apply to dichotomic representations, this finding implied that the different levels of power were presented as different vertical heights. In addition, in the studies on distance effect of status (Chiao et al., 2004, 2009; Von Hecker et al., 2013), the distance was identified by researchers or other participants in a pilot study. In Experiment 2, however, the power distance was identified referring to each participant’s own power ratings, which increased the reliability of the results. The findings of the present study completely ruled out the possibility that power was related to space in a dichotomic manner (i.e., powerful = up and powerless = down) and people were inclined to notice this relation when both power and vertical space were manipulated with only two levels. Zanolie et al. (2012) argued that participants might have a response bias during power judgment: responding ‘‘powerful” to stimuli at the top of the screen and ‘‘powerless” to stimuli at the bottom of the screen. Power judgments did not trigger a space schema. Rather, people just noticed the dichotomic power–space relation when concepts of power and space were both activated. Thus, Zanolie et al. (2012) did not manipulate the vertical positions of the words. Instead, Zanolie et al. used an attentional shift paradigm to remove such response bias. In their experiment, participants were required to identify the target letters preceded by power judgment of words. The space locations of the words were manipulated by two levels: up and down. It is found that target letters were identified faster when their spatial positions were congruent with the perceived power of the preceding word than when they were incongruent. However, as Zanolie et al. claimed, the concepts of power and space were both activated. And, in such situation, it was easier for participants to notice the dichotomic power–space relation. On the contrary, the distance effect demonstrated in our study directly reflected the spatial properties of power cognition. More broadly, our results support the grounded cognition framework (e.g., Barsalou, 1999, 2008; Glenberg, 1997). The basic tenet of this framework is that human cognition is body based. That is, conceptual thinking often relies on simulations of perceptual symbols, which are schematized perceptual experiences including senses, proprioception, introspection, and motor programs. There is now increasing evidence that abstract concepts are also grounded in sensory-motor representations through metaphors (Conceptual Metaphor Theory, Gibbs, 1994; Lakoff & Johnson, 1980, 1999). For example, power mapped to verticality (Giessner & Schubert, 2007; Schubert, 2005; our study), physical force (Schubert, 2004) and size (Schubert, Waldzus, & Giessner, 2009), social proximity mapped to temperature (Ijzerman & Semin, 2009), and social distance mapped to spatial distance (Parkinson & Wheatley, 2013). Acknowledgments This research was supported by National Natural Science Foundation of China (31100728), Projects Planning in Shanghai Philosophy and Social Sciences Research (2012JJY001) and the research fund of the School of Social Development and Public Policy at Fudan University. Appendix A See Tables 1 and 2.
14
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
Table 1 Stimuli used in Experiment 1. Powerless
Powerful
Word
English
Power
Valence
Duration (ms)
Word
English
Power
Valence
Duration (ms)
婴儿 孤儿 病人 奴隶 孩子 丫头 仆人 随从 实习生 保姆 学徒 学生 农夫 农民 水手 士兵
Baby Orphan Sick person Slave Child Bonne Servant Attendant Intern Nurserymaid Apprentice Student Plowman Farmer Sailor Soldier
1.58 1.67 1.75 1.75 1.92 2.00 2.08 2.42 2.50 2.58 2.58 3.00 3.33 3.33 3.83 4.08
6.17 3.25 2.83 2.17 5.92 4.83 3.25 3.08 4.08 3.67 3.67 5.00 4.58 4.33 5.00 4.50
540 570 550 540 560 550 550 540 560 560 550 530 560 530 540 540
国王 总统 君主 统治者 霸主 政治家 官员 老板 将军 总理 首领 教练 主任 经理 校长 主人
King President Sovereign Governor Overlord Politician Officer Boss General Premier Leader Coach Director Manager Principal Host
5.83 6.17 6.33 6.50 6.75 4.33 4.50 4.58 5.42 5.42 5.50 3.17 3.33 3.75 4.17 4.25
4.50 4.92 5.42 3.42 4.75 2.83 2.92 3.67 5.17 4.75 4.92 4.08 3.25 3.92 4.33 3.50
550 550 550 570 540 590 550 540 550 550 560 540 540 550 560 550
Table 2 Stimuli used in Experiment 2. Word
In English
Power
霸主 保姆 病人 国王 孩子 将军 经理 士兵 随从 学生 婴儿 主人
Overlord Nurserymaid Sick person King Child General Manager Soldier Attendant Student Baby Host
8.23 2.87 2.42 8.60 3.09 7.63 6.70 3.84 3.24 3.83 2.28 6.65
Note. Power indicates the average power ratings for each word across all the participants.
References Barsalou, L. W. (1999). Perceptual symbol systems. Behavioral and Brain Sciences, 22, 577–609. Barsalou, L. W. (2008). Cognitive and neural contributions to understanding the conceptual system. Current Directions in Psychological Science, 17, 91–95. Chiao, J. Y. (2010). Neural basis of social status hierarchy across species. Current Opinion in Neurobiology, 20, 803–809. Chiao, J. Y., Bordeaux, A. R., & Ambady, N. (2004). Mental representations of social status. Cognition, 93, B49–B57. Chiao, J. Y., Harada, T., Oby, E. R., Li, Z., Parrish, T., & Bridge, D. J. (2009). Neural representations of social status hierarchy in human inferior parietal cortex. Neuropsychologia, 47, 354–363. Dehaene, S., Piazza, M., Pinel, P., & Cohen, L. (2003). Three parietal circuits for number processing. Cognitive Neuropsychology, 20, 487–506. Galinsky, A. D., Gruenfeld, D., & Magee, J. C. (2003). From power to action. Journal of Personality and Social Psychology, 85, 453–466. Gibbs, R. W. (1994). The poetics of mind: Figurative thought, language, and understanding. New York, NY: Cambridge University Press. Giessner, S. R., & Schubert, T. W. (2007). High in the hierarchy: How vertical location and judgments of leaders’ power are interrelated. Organizational Behavior and Human Decision Processes, 104, 30–44. Glenberg, A. M. (1997). What memory is for? Behavioral and Brain Sciences, 20, 1–55. Grade, S., Lefèvre, N., & Pesenti, M. (2013). Influence of gaze observation on random number generation. Experimental Psychology, 60, 122–130. Hays, N. A. (2013). Fear and loving in social hierarchy: Sex differences in preferences for power versus status. Journal of Experimental Social Psychology, 49, 1130–1136. Ijzerman, H., & Semin, G. R. (2009). The thermometer of social relations. Mapping social proximity on temperature. Psychological Science, 20, 1214–1220. Jiang, T., Sun, L., & Zhu, L. (2015). The influence of vertical motor responses on explicit and incidental processing of power words. Consciousness and Cognition, 34, 33–42. Judd, C. M., Westfall, J., & Kenny, D. A. (2012). Treating stimuli as a random factor in social psychology: A new and comprehensive solution to a pervasive but largely ignored problem. Journal of Personality and Social Psychology, 103, 54–69. Keltner, D., Gruenfeld, D., & Anderson, C. P. (2003). Power, approach, and inhibition. Psychological Review, 110, 265–284. Lakoff, G. (1987). Women, fire, and dangerous things. Chicago: University of Chicago Press. Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: University of Chicago Press. Lakoff, G., & Johnson, M. (1999). Philosophy in the flesh. New York, NY: Basic Books. Mason, M., Magee, J. C., & Fiske, S. T. (2014). Neural substrates of social status inference: Roles of medial prefrontal cortex and superior temporal sulcus. Journal of Cognitive Neuroscience, 26, 1131–1140. Moyer, R. S., & Landauer, T. K. (1967). Time required for judgments of numerical inequality. Nature, 215, 1519–1520. Mussweiler, T. (2003). Comparison processes in social judgment: Mechanism and consequences. Psychological Review, 110, 472–489.
T. Jiang, L. Zhu / Consciousness and Cognition 37 (2015) 8–15
15
Parkinson, C., & Wheatley, T. (2013). Old cortex, new contexts: Re-purposing spatial perception for social cognition. Frontiers in Human Neuroscience, 7, 645. Schubert, T. W. (2004). The power in your hand: Gender differences in bodily feedback from making a fist. Personality and Social Psychology Bulletin, 30, 757–769. Schubert, T. W. (2005). Your highness: Vertical positions as perceptual symbols of power. Journal of Personality and Social Psychology, 89, 1–21. Schubert, T. W., Waldzus, S., & Giessner, S. R. (2009). Control over the association of power and size. Social Cognition, 27, 1–19. Von Hecker, U., Klauer, C. K., & Sankaran, S. (2013). Embodiment of social status: Verticality effects in multi-level rank-orders. Social Cognition, 31, 374–389. Zanolie, K., Dantzig, S., Boot, I., Wijnen, J., Schubert, T. W., Giessner, S. R., et al (2012). Mighty metaphors: Behavioral and ERP evidence that power shifts attention on a vertical dimension. Brain and Cognition, 78, 50–58.