Judging a book by its cover (and its background): effects of the metaphor intelligence is brightness on ratings of book images

Abstract

Visual metaphors have been the focus of experimental and corpus studies aiming to determine whether metaphors are conceptual or purely linguistic. However, in visual metaphor research, experimental and corpus approaches have each been directed at a distinct set of visual metaphors. Psychological experiments have focused on primary metaphors, whereas corpus studies have concentrated on non-primary metaphors. The current study suggests that when non-primary metaphors are examined experimentally, cognitive effects appear only when the metaphor is contextually relevant. intelligence is brightness, for example, is not relevant when the potential source of intelligence is inaccessible. In this study, images of open books with bright backgrounds are rated as more likely to represent works of genius than open books with darker backgrounds. When books are closed, their ‘genius’ is inaccessible and intelligence is brightness has no effect.

Keywords

complex mapping conceptual metaphor theory pictorial metaphor primary metaphor visual priming

1. Introduction

Visual metaphors have long been hailed as a means of assessing whether metaphors are conceptual or merely linguistic phenomena (Forceville, 1994); they have risen to prominence in psychological research and corpus studies, and have yielded promising results in both areas. One might expect that psychological and corpus-based approaches would complement and reinforce each other, as they have in areas such as discourse analysis (Waugh et al., 2007). However, in the study of visual metaphor, each approach has focused on a different type of metaphor. Psychological studies examine ‘primary metaphors’ (Grady, 1997) and corpus studies focus on ‘complex metaphors’ (Lakoff and Turner, 1989), a dichotomy examined in the next two sections. This dichotomy is perhaps unfortunate, as corpus and psychological methods can lead to different types of insights. For example, psychological studies examine the effect of visual metaphors on reasoning, whereas corpus studies consider contextual factors that are not easily examined in a laboratory context. Neither type of study reveals whether complex visual metaphors affect reasoning in the manner of primary visual metaphors.

The current psychological study of a complex metaphor, intelligence is brightness, suggests that the metaphor’s cognitive effects are more context dependent than those of previously examined primary metaphors. The study assesses the effects of visual stimuli related to brightness on judgements of intelligence, and suggests that intelligence is brightness is cognitively activated only when its component primary metaphor knowing is seeing is also accessible.

1.1 Primary metaphors in psychological studies

Psychological experiments on visual metaphor have focused almost exclusively on primary metaphors. Primary metaphors are those that are based on real-world co-occurrences or ‘primary scenes’ (Grady, 1997; Grady and Johnson, 1998). For instance, we learn to associate affection with warmth through physical contact that indicates affection, such as when a child is held by a parent. The metaphor affection is warmth, as in expressions such as warm welcome, is a primary metaphor because it is based on this co-occurrence. A metaphor may be extended from the initial primary correspondence and acquire complex mappings. For example, human beings are not literally capable of emitting cold when they dislike someone, only warmth when they embrace or touch. Therefore the mapping lack of affection is cold, as in chilly reception, is a complex mapping based on the primary metaphor affection is warmth. Many well-studied metaphors are complex, such as the mapping from buildings to theories (as in the foundation of an argument), which cannot be a primary metaphor because we do not tend to have life experiences in which theories and buildings co-occur (Grady, 1997).

Primary metaphors examined in psychological experiments include good is light (Frank and Gilovich, 1988; Meier et al., 2004; Peña et al., 2009; cf. Lakens et al., 2012; Nickels, 2007), similarity is closeness (Casasanto, 2008), control is up (Schubert, 2005; Valenzuela and Soriano, 2009), affection is warmth (Zhong and Leonardelli, 2008), importance is size (Schubert et al., 2009; Valenzuela and Soriano, 2008) the Moving Ego and Moving Time metaphors (Boroditsky, 2000; Casasanto et al., 2005), morality is cleanliness (Zhong and Liljenquist, 2006), and happy is up (Casasanto and Lozano, 2008; Meier and Robinson, 2004).

Only a few complex metaphors, such as anger is heated fluid in a container, have been examined psychologically (Gibbs and O’Brien, 1990; Gibbs et al., 1997; Valenzuela and Soriano, 2007). anger is heated fluid in a container is complex, in that angry behaviour and boiling containers do not literally co-occur in the world. However, studies of complex metaphors such as the above present the metaphors linguistically rather than visually. As such, they do not test whether the complex metaphors are active in cognition outside of language.

Recent evidence suggests that complex correspondences may have non-linguistic cognitive effects only if their component primary metaphors are activated. For example, good is light has been considered primary in metaphor theory, but Lakens et al. (2012) show that the metaphor is accessed only if dark is bad is also active. This finding suggests that the correspondence between dark and bad is a primary correspondence, based on the co-occurring experience of darkness and discomfort due to the diurnal nature of human beings (Williams and Morland, 1976). The association between good and light has then been added to this primary metaphor as a complex mapping. Lakens et al. (2012: 585) theorize that ‘white can become associated with positivity, but such a mapping is highly context specific, and it is hypothesized to become activated primarily in opposition to the negativity of black.’ In terms of primary and complex metaphor, this means that the ingrained primary association between dark and bad is extended to a complex association between good and light. Lakens et al. indicate that the complex association appears only in the presence of the primary association. Unfortunately, due to the focus on primary metaphor in experimental studies of visual metaphor, few studies indicate whether, and in what contexts, complex metaphors are visually activated.

1.2 Complex metaphors in corpus studies

Whereas experimental visual metaphor studies focus on primary metaphor, corpus studies of visual metaphors typically examine non-primary metaphors. For example, the Global Financial Crisis in Bounegru and Forceville’s (2011) survey of editorial cartoons is represented as a natural disaster, such as a tsunami or a fire, in 48 percent of the metaphors in the study. In the world, the financial crisis was not accompanied by tsunamis and fires, so the metaphor the financial crisis is a natural disaster is not based on this co-occurrence. Instead, the metaphor derives from a system of correspondences between the concept of the crisis and the concept of a natural disaster, as Bounegru and Forceville discuss, and so can be considered a complex metaphor.

The most frequent visual metaphors noted in El Refaie’s (2003) study of editorial cartoons are also complex, such as the metaphors mapping floods and invasions onto the influx of refugees into Italy (p. 83). These refugees do not bring with them floods or invasions; and even if their migration was motivated by these factors, they themselves are neither literally invaders nor floodwaters.

Primary metaphors are often ignored in such corpus studies. For example, many of the cartoons depicting the financial crisis in Bounegru and Forceville (2011) contain line graphs pointing downwards. These graphs involve two primary metaphors, more is up and the Moving Time metaphor. The metaphor more is up, as in expressions such as stocks fell, is based on the real-world correspondence between height of a pile of material or objects, and quantity of material or objects: if we put more objects in a pile the pile becomes higher (Grady, 1997). The Moving Time metaphor, as in Christmas is coming, reflects our experience of objects moving through space over time, such as when a ball flies towards us and we calculate how long it will take to arrive. Both metaphors are necessary to understand a stock-market graph, in which the x-axis represents time and the y-axis represents more (Lakoff and Núñez, 2000). Despite the ubiquity of this type of metaphor in visual metaphor corpora, primary visual metaphors are rarely discussed in studies of these corpora. The focus on non-primary visual metaphors in corpus studies is particularly striking considering that primary metaphors are the main focus of psychological studies of visual metaphor.

1.3. Intelligence is brightness and the current study

The current experiment employs visual stimuli to examine the non-primary mapping intelligence is brightness. This complex mapping relies on the primary metaphor knowing is seeing, which is based on the real-world correspondence between seeing and learning information. However, although intelligence facilitates knowing, and brightness facilitates seeing, the mapping intelligence is brightness is non-primary because intelligent people and ideas do not literally emit light.

Like other complex metaphors and mappings, intelligence is brightness is postulated on the basis of linguistic evidence. The mapping can form the basis for novel linguistic expressions (intellectual supernova, a guttering candle of an intellect) as well as quotidian ones (illuminate, shed light on, dim, bright, brilliant). The mapping is not universal, but does occur in a range of languages including Spanish (una persona brillante ‘a brilliant person’, tiene pocas luces ‘He or she doesn’t have many lights [i.e., is stupid]’). Given that complex metaphors and mappings such as intelligence is brightness lack real-world correspondences and are posited on the basis of linguistic data, it seems reasonable that they might not have the same cognitive status as primary metaphors, especially in visual manifestations where no language is involved. Psychological findings on visual primary metaphors therefore cannot be assumed to hold for visual complex metaphors, and psychological studies targeting visual complex metaphors are required.

2. Method

Overview

The current study evoked brightness through images of books displayed against either dark or light backgrounds, and assessed the effect of these backgrounds on the evaluation of the ‘genius’ of the depicted books. It was predicted that the light backgrounds would encourage impressions of genius and the dark backgrounds would lead to lower estimations of genius. Images of books were selected over images of people because books are presumably less likely to be judged on the basis of factors such as attractiveness or perceived personality, which can distort results for human images (Sullivan, 2013b). In addition, the use of books allowed the brightness of both the background and the image to be manipulated, since the same book could be presented with different cover colours, such as the white and grey variations employed in the current study. Variation in cover colour was intended to test whether the optical illusion of a brighter object, caused by a dark background, influenced participants’ judgements. For example, a grey book might appear brighter against a dark background than against a light background. However, a grey book would still not appear as light as a white book. Manipulating both background and cover colour therefore permits the identification of effects caused by optical illusion rather than directly by the background.

Bright backgrounds were predicted to lead to higher ratings of books’ ‘genius’. The white and grey covers were not predicted to affect ratings of ‘genius’, since cover colours do not represent or indicate light emitted from the book, and only emitted light is theorized to map to intelligence (Lakoff and Johnson, 1999; Sullivan, 2006). The two cover colours were intended only to control for the optical illusion noted above.

Participants

Participants were 70 undergraduate students at the University of Queensland in Australia, enrolled in an introductory phonology course. Participants included 61 females and 9 males, 48 native speakers of English and 22 with high English proficiency.

Materials

Participants viewed 30 stylized greyscale drawings of books with black outlines, consisting of 10 closed books with dark grey covers, 10 closed books with white covers, and 10 open books. No legible text appeared in any of the images. The sets of grey and white-cover closed book images were adapted from the same 10 drawings with the book covers coloured either grey or white. For each pair of otherwise identical grey and white books, one member of the pair was flipped, resulting in either a left-facing grey book and a right-facing white book, or a right-facing grey book and a left-facing white book (see Figure 1). Half of the open-book images were flipped (see Figure 2). Each book was centred against either a light grey background or a dark grey background.

Figure 1

Four versions of a closed book image. From left to right: white cover, dark background; white cover, light background; grey cover, dark background; grey cover, light background.

Figure 2.

Two versions of an open book image: dark background and light background.

Participants viewed the images in E-Prime 2 on a Macintosh with a Dell LCD monitor. Responses were recorded with an Empirisoft DirectIN button box.

Procedure

On-screen instructions advised participants that they would ‘see a series of icons of books’ and be asked ‘to judge how appropriate the icon is to represent a work of genius’ which may be ‘a scientific, literary or other type of genius’. These instructions were intended to dissuade religious interpretations of the books. Participants rated each image on a scale from 1–7, with ‘1’ indicating ‘not at all likely to be a work of genius’ and ‘7’ designating ‘very likely to be a work of genius’.

In each trial, a black fixation cross appeared for 200 msecs against a flat neutral grey background. The book image appeared for 1500 msecs on a light or dark background that filled the viewable area of the screen (see Figures 1 and 2). Then the following question appeared in black on flat neutral grey: ‘How likely is it that the book is a work of genius? Assign a rating between 1 (not at all likely) and 7 (very likely).’ The question remained on screen until the participant responded. The screen was then flat neutral grey for 2000 msecs between trials.

Design

Participants viewed a practice block of four trials, followed by 60 experimental trials. Each participant viewed all 30 experimental images twice, once with a dark grey background and once with a light grey background. Of the 60 experimental images, half were vertically flipped so that the book faced the opposite direction. Even-numbered participants viewed mirror images of all items (both those already flipped and those not previously flipped), so that any given item was right-facing for half the participants and left-facing for the other half. Images were flipped to lessen the impression of repeated items, since the closed-book images were repeated four times (with grey or with white covers, and with dark or with light backgrounds; see Figure 1) and the open-book images were repeated twice (with dark or light backgrounds; see Figure 2). Additionally, participants were advised in the on-screen instructions that ‘some images are extremely similar’ and instructed to react to each image without trying to remember how they previously rated similar images.

3. Results

In a factorial ANOVA with the factors image type (open book, closed white book, or closed grey book), first language of the participant (English or another language), and background (light or dark), a main effect of language was found, F(1, 69) = 4.41, p = .04. However, this factor did not interact with image type or background, so data from both native and non-native speakers of English were included in the study.

Participants judged the closed books, both with grey and white covers, as decisively more likely to be works of ‘genius’ than the open books. The mean rating for the closed books was 4.11, whereas the mean for open books was only 3.27. The size of this effect was moderate (d = .528). Open books may have been rated lower overall due to their smaller apparent size. Images were sized to balance the amount of white that was visible, rather than the apparent size of the book based on the size of its pages and cover, so the open books may have appeared smaller and less likely to contain ideas of genius. The smaller apparent size of the open books may have made them appear less important, via the well-documented primary metaphor importance is size (Schubert et al., 2009; Valenzuela and Soriano, 2008). Smaller books may appear to contain less important material. Additionally, a shorter or smaller book usually contains less text, which may decrease the perceived probability of its containing ideas of genius.

The effects of the different backgrounds and image types were assessed with factorial ANOVAs with the factors background (light or dark) and image type (open book, closed white book, or closed grey book). In all analyses by items, items were the original drawings. That is, the grey and white versions of a closed book were considered as the same item. The main effect of background was not significant in either the subjects or the items analysis, F₁(1, 69) = 1.64, p = .20, and F₂(1, 18) = 2.19, p = .16. The main effect of image type was significant in both analyses, F₁(1, 69) = 74.77, p = .00, and F₂(1, 18) = 5.69, p = .04. The interaction between background and image type approached significance in the participants analysis, F₁(1, 69) = 2.58, p = .08, but not in the items analysis, F₂(1, 18) = 1.30, p = .28.

When only the open books were considered, open books with light backgrounds received the mean rating 3.33 and open books with dark backgrounds received the mean rating 3.20, a main effect of background that was marginally significant by subjects, F₁(1, 69) = 3.65, p = .06 and highly significant by items, F₂ (1, 9) = 14.95, p = .00; d = .083. This effect was completely absent for the closed books, in which books with light backgrounds and books with dark backgrounds received the same mean rating of 4.11.

Grey-cover books were rated as more likely to be works of genius than white-cover books (4.16 as compared to 4.06 mean ratings, d = .06). In an ANOVA of the closed books with the factors background (light or dark) and cover colour (white or grey), the main effect of cover colour was significant both by subjects, F₁ (1, 69) = 6.60, p = .01 and by items, F₂ (1, 9) = 5.69, p = .04. The main effect of background was not significant, F₁ (1, 69) = .00, p = .95, F₂ (1, 9) = .00, p = .97. Interaction for background and cover colour was not significant, F₁ (1, 69) = 1.69, p = .20, F₂ (1, 9) = 1.30, p =.28. There was therefore no evidence that background colour influenced perceptions of book colour.

4. Discussion

The hypothesis was supported only for the open books. These books received a significantly higher rating of ‘genius’ if they appeared against a light background, whereas no effect of background was found for closed books. For the images of closed books against light backgrounds, the surrounding light did not appear to be associated with the contents of the books, presumably because the contents of the closed books were invisible and inaccessible. According to Conceptual Metaphor Theory, the complex mapping intelligence is brightness depends on the primary metaphor knowing is seeing (Lakoff and Johnson, 1999; Sullivan, 2006, 2013a). Without this more basic mapping, intelligence is brightness is not activated. That is, if something cannot be seen, it cannot be judged as bright; and if an idea cannot be known, its intelligence cannot be evaluated. This requirement is implicitly recognized in visual instances of intelligence is brightness in daily life, in which representations of books tend to be depicted as open if they are ‘emitting’ wisdom or genius, as in the wide-open books in the seals of several universities, including the University of California at Berkeley.

The other significant result of the experiment was the effect of cover colour on evaluations of closed books. Grey-cover books were rated as more likely to be works of genius than white-cover books. This result cannot be attributed to intelligence is brightness. However, light colours may appear to weigh less (Caivano, 1998; Déribéré, 1964; Pope, 1949). If the white books were perceived as physically lighter, they may have been interpreted as less likely to contain ideas of genius. In addition, dark covers may have been seen as an indication of seriousness and authority. Black uniforms lead to more positive impressions of police officers (Nickels, 2007). Dark covers on books may similarly encourage connotations of intellectual authority, seriousness, or other qualities associated with ‘genius’.

5. Conclusion

The current study suggests that the complex mapping intelligence is brightness is relevant in interpreting visual stimuli, but only when the mapping is contextually appropriate, such as when a book is open and emitting light. Visually presented complex metaphors appear to generate cognitive effects of the kind documented for visual primary metaphors. However, complex metaphors appear to generate these effects in a more limited range of contexts than is the case for primary metaphors.

One explanation for this more limited range is that complex metaphors may require activation of their component primary metaphors. Lakens et al. (2012) suggest that metaphors that are not based directly on real-world correspondences (such as complex metaphors) are activated only in conjunction with metaphors that are based on these correspondences. Primary metaphors, such as bad is dark, are activated automatically, and negative concepts are automatically associated with black. In contrast, associations that depend on bad is dark are activated only alongside bad is dark. The current study supports the notion that complex metaphors are only activated in the context of more primary ones. Without the basic connection between knowing and seeing offered by the open books in the study, the complex secondary connection between intelligence and brightness is not activated.

Considerations such as the open or closed state of the depicted books are predicted to recur in the study of complex metaphors. These metaphors are based on structural correspondences between concepts that may never co-occur in the world. As such, the visual juxtaposition of these concepts may not be interpreted as metaphoric unless other conditions are fulfilled. The fulfilment of these conditions may be explained in terms of the complex metaphors’ constituent primary metaphors. Additional studies with a range of visual complex metaphors are required to pinpoint exactly under what conditions these visual metaphors have cognitive effects, and whether these conditions can be explained partly or entirely in terms of the complex metaphors’ component primary metaphors.

Visual corpus studies may also play a role in identifying the contexts that activate complex metaphors. For example, a corpus study of university and institutional logos could be predicted to find that light-surrounded books of knowledge are generally depicted as open, not closed. This type of study would help identify the conditions under which intelligence is brightness is visually activated. Observations from visual corpora, in the future, may feed into psychological studies of visual metaphor and help to narrow down the contexts in which complex mappings are likely to produce cognitive effects. This would be one of many benefits of applying corpus-based and psychological methods to the same types of metaphor, instead of the two non-overlapping types that have previously been the focus of the two methods.

Footnotes

Biographical Note

KAREN SULLIVAN is a lecturer in Linguistics at the University of Queensland, Australia. Her work focuses on the linguistic and visual communication of metaphor. Sullivan’s article ‘The languages of art: How representational and abstract painters conceptualize their work in terms of language’ (Poetics Today 30(3), 2009) was awarded the 2010 Researching and Applying Metaphor (RaAM) Early Career Researcher Award. book Frames and Constructions in Metaphoric Language (John Benjamins) examines how the resources of language are employed to communicate conceptual metaphor.

Address: University of Queensland, School of Languages and Comparative Cultural Studies, Gordon Greenwood (32) level 3, Brisbane, QLD 4072, Australia. [email: ksull@uq.edu.au]

References

Boroditsky

(2000) Metaphoric structuring: Understanding time through spatial metaphors. Cognition 75(1): 1–28.

Bounegru

Forceville

(2011) Metaphors in editorial cartoons representing the global financial crisis. Visual Communication 10(2): 209–229.

Caivano

J.L

. (1998) Color and semiotics: A two-way street. COLOR Research & Application 23(6): 390–401.

Casasanto

. (2008) Similarity and proximity: When does close in space mean close in mind? Memory & Cognition 36(6): 1047–1056.

Casasanto

Lozano

(2008) The cognitive function of metaphorical gestures. In: Cienki

Müller

(eds) Metaphor and Gesture. Amsterdam: John Benjamins.

Casasanto

. (2005) How deep are the effects of language on thought? Time estimation in speakers of English, Indonesian, Greek, and Spanish. In: Forbus

. (eds) Proceedings of the 26th Annual Conference of the Cognitive Science Society. Hillsdale, NJ: Lawrence Erlbaum, 575–580.

Déribéré

(1964) La Couleur dans les activités humaines (El Color en las actividades humanas). Madrid: Tecnos.

El Refaie

(2003) Understanding visual metaphor: The example of newspaper cartoons. Visual Communication 2(1): 75–95.

Forceville

(1994) Pictorial metaphor in advertisements. Metaphor and Symbolic Activity 9(1): 1–29.

10.

Frank

M.G.

Gilovich

(1988) The dark side of self- and social perception: Black uniforms and aggression in professional sports. Journal of Personality and Social Psychology 54(1): 74–85.

11.

Gibbs

R.W

Jr O’Brien

(1990) Idioms and mental imagery: The metaphorical motivation for idiomatic meaning. Cognition 36: 35–68.

12.

Gibbs

R.W

Jr . (1997) Metaphor in idiom comprehension. Journal of Memory and Language 37: 141–154.

13.

Grady

J.E.

(1997) Foundations of meaning: Primary metaphors and primary scenes. PhD dissertation, University of California, Berkeley.

14.

Grady

Johnson

(1998) Converging evidence for the notions of subscene and primary scene. In: Proceedings of the 23rd Annual Meeting of the Berkeley Linguistics Society, Berkeley, CA, 123–136.

15.

Lakens

Semin

G.R.

Foroni

. (2012) But for the bad, there would not be good: Grounding valence in brightness through shared relational structures. Journal of Experimental Psychology: General 141(3): 584–594.

16.

Lakoff

Johnson

. (1999) Philosophy in the Flesh. New York: Basic Books.

17.

Lakoff

Núñez

R.E.

(2000) Where Mathematics Comes From: How the Embodied Mind Brings Mathematics into Being. New York: Basic Books.

18.

Lakoff

Turner

(1989) More Than Cool Reason. Chicago: University of Chicago Press.

19.

Meier

B.P.

Robinson

M.D.

(2004) Why the sunny side is up: Associations between affect and vertical position. Psychological Science 15(2): 243–247.

20.

Meier

B.P.

Robinson

Clore

. (2004) Why good guys wear white: Automatic inferences about stimulus valence based on brightness. Psychological Science 15(2): 82–87.

21.

Nickels

(2007) Good guys wear black: Uniform color and citizen impressions of police. Policing: An International Journal of Police Strategies & Management 31(1): 77–92.

22.

Peña

Hancock

J.T.

Merola

N.A.

(2009) The priming effects of avatars in virtual settings. Communication Research 36(6): 838–856.

23.

Pope

. (1949) The Language of Drawing and Painting. Cambridge, MA: Harvard University Press.

24.

Schubert

T.W.

(2005) Your highness: Vertical positions as perceptual symbols of power. Journal of Personality and Social Psychology 89(1): 1–21.

25.

Schubert

T.W

Waldzus

Giessner

S.R.

(2009) Control over the association of power and size. Social Cognition 27(1): 1–19.

26.

Sullivan

(2006) Frame-based constraints on lexical choice in metaphor. Proceedings of the 32nd Annual Meeting of the Berkeley Linguistics Society, Berkeley, CA.

27.

Sullivan

(2013a) Frames and Constructions in Metaphoric Language. Amsterdam: John Benjamins.

28.

Sullivan

(2013b) A tangled knot of target domains: Assessing intelligence is brightness and goodness is brightness in an image rating task. Metaphorik.de 23: 7–18.

29.

Valenzuela

Soriano

(2007) Reading anger stories: A lexical decision task as a test for the existence of metaphorical representation. In: Language, Mind and the Lexicon. Frankfurt: Peter Lang, 281–303.

30.

Valenzuela

Soriano

(2008) Is ‘friendship’ more important than ‘money’: A psycholinguistic exploration of the important is big metaphor. VI AELCO Conference, Castellón, Spain.

31.

Valenzuela

Soriano

(2009) Is Control really up? A psycholinguistic exploration of a primary metaphor. In: Valenzuela

. (eds) Trends in Cognitive Linguistics: Theoretical and Applied Models. Frankfurt: Peter Lang, 31–50.

32.

Waugh

L.R.

. (2007) Multiple empirical approaches to a complex analysis of discourse. In: Gonzalez-Marquez

. (eds) Methods in Cognitive Linguistics. Amsterdam: John Benjamins.

33.

Williams

J.E

Morland

J.K.

(1976) Race, Color, and the Young Child. Chapel Hill: University of North Carolina Press.

34.

Zhong

C.B.

Leonardelli

G.J.

(2008) Cold and lonely: Does social exclusion literally feel cold? Psychological Science 19(9): 838–842.

35.

Zhong

C.B

Liljenquist

. (2006) Washing away your sins: Threatened morality and physical cleansing. Science 313: 1451–1452.