ArticlePDF Available

Abstract and Figures

In everyday language, abstract concepts are described in terms of concrete physical experiences (e.g., good things are “up”; the past is “behind” us). Stimuli congruent with such conceptual metaphors are processed faster than stimuli that are not. Since ease of processing enhances aesthetic pleasure, stimuli should be perceived as more pleasing when their presentation matches (rather than mismatches) the metaphorical mapping. In six experiments, speakers of English (Experiment 1-3a) and Farsi (Experiment 3b and 4) viewed valence- and time-related photos in arrangements congruent and incongruent with their metaphorical mapping. Consistent with the valence-verticality metaphor in both languages, English and Farsi speakers preferred visual arrangements that placed the happy photo above the sad photo. In contrast, participants’ preferences for time-related photos were moderated by the direction of writing. English speakers, who write from left to right, preferred arrangements that placed past-themed photos to the left of modern-themed photos; this was not observed for Farsi speakers, who write from right to left as well as left to right. In sum, identical stimuli enjoy an aesthetic advantage when their spatial arrangement matches the spatial ordering implied by applicable conceptual metaphors.
Content may be subject to copyright.
Running head: METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 1
Journal of Experimental Social Psychology, in press
Conceptual Metaphors, Processing Fluency, and Aesthetic
Preference
Lynn Zhang1
Mohammad Atari1
Norbert Schwarz1
Eryn J. Newman2
Reza Afhami3
1University of Southern California,
2Australian National University
3Tarbiat Modares University
Note. This is the accepted manuscript after peer review but prior to copy-editing by the
publisher. The final version of record may therefore differ in details, but not substance.
Address correspondence to Lynn Zhang (zhan306@usc.edu) or Norbert Schwarz
(norbert.schwarz@usc.edu)
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 2
Highlights
Metaphor-congruence increases fluency and fluency enhances aesthetic pleasure.
Experiments with visual arrangements in the U.S. and Iran support this prediction.
Arrangements matching a culture’s space-valence metaphor are preferred over
mismatches.
Arrangements matching a culture’s space-time metaphor are preferred over mismatches.
Abstract
In everyday language, abstract concepts are described in terms of concrete physical experiences
(e.g., good things are “up”; the past is “behind” us). Stimuli congruent with such conceptual
metaphors are processed faster than stimuli that are not. Since ease of processing enhances
aesthetic pleasure, stimuli should be perceived as more pleasing when their presentation matches
(rather than mismatches) the metaphorical mapping. In six experiments, speakers of English
(Experiment 1-3a) and Farsi (Experiment 3b and 4) viewed valence- and time-related photos in
arrangements congruent and incongruent with their metaphorical mapping. Consistent with the
valence-verticality metaphor in both languages, English and Farsi speakers preferred visual
arrangements that placed the happy photo above the sad photo. In contrast, participants’
preferences for time-related photos were moderated by the direction of writing. English speakers,
who write from left to right, preferred arrangements that placed past-themed photos to the left of
modern-themed photos; this was not observed for Farsi speakers, who write from right to left as
well as left to right. In sum, identical stimuli enjoy an aesthetic advantage when their spatial
arrangement matches the spatial ordering implied by applicable conceptual metaphors.
Keywords: Conceptual metaphor, Processing fluency, Aesthetic pleasure, Language
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 3
Conceptual Metaphor, Processing Fluency, and Aesthetic Preference
Is beauty in the eye of the beholder or in features of the beholden? Experimental research
indicates that attributes of the beholden (e.g., symmetry, contrast, and clarity) as well as attributes
of the beholder (e.g., prior exposure and implicit learning) influence perceptions of beauty through
facilitating or impairing ease of processing (for reviews, see Reber, Schwarz, & Winkielman,
2004; Schwarz, 2018). From this perspective, beauty is a function of the perceiver’s processing
experience, which depends on the interplay of object, perceiver, and context characteristics.
Supporting this view, research has shown that object attributes that enhance aesthetic pleasure --
such as high figure-ground contrast (e.g., Checkosky & Whitlock, 1973), clarity (Whittlesea,
Jacoby, & Girard, 1990), symmetry (Cárdenas & Harris, 2006; Enquist & Arak, 1994; Garner,
1974), and prototypicality (Winkielman, Halberstadt, Fazendeiro, & Catty, 2006)-- also facilitate
efficient processing and fast recognition (for a review, see Reber et al., 2004). So do perceiver
variables, such as prior exposure to the object (Haber & Hershenson, 1965; Jacoby & Dallas,1981),
to some of its attributes (Reber, Winkielman, & Schwarz, 1998), or to related semantic concepts
(Winkielman, Schwarz, Fazendeiro, & Reber, 2003). Similarly, context variables, from ambient
lighting to noise and the presence of materials that can serve as semantic primes, can influence
processing fluency and pleasure (for a review, see Reber et al., 2004). Independent of which
variable facilitates ease of processing, easy processing elicits a sense of familiarity (Whittlesea,
1993; Kinder, Shanks, Cock, & Tunney, 2003) and positive affect (Winkielman et al., 2003;
Winkielman et al., 2006). When asked to evaluate how much they like an object, how beautiful
they find it, or which of several objects they prefer, people draw on their concurrent subjective
experiences and provide more favorable evaluations of fluently processed objects (for reviews, see
Reber et al., 2004; Schwarz, 2018; Schwarz, Jalbert, Noah, & Zhang, 2021). The underlying
process is consistent with feelings-as-information theory (Schwarz, 1990, 2012; Schwarz & Clore,
1983), which conceptualizes the use of subjective experiences (including moods, emotions,
metacognitive, and bodily experiences) as a source of information in human judgment.
Building on this work, we examine the influence of conceptual metaphors (Lakoff &
Johnson, 1980). In everyday language, abstract concepts are often described in terms of concrete
physical experiences. For example, good things are “up”; the past comes “before” the present;
important things are “heavy”; and nice people are “warm”. Numerous studies demonstrated that
such metaphors are not just figures of speech, but cognitive tools people use to conceptualize
abstract concepts (Lakoff & Johnson, 1980, 1999; for a review see Landau, 2017). Hence,
processing of information should be easier when the information is consistent with the implications
of conceptual metaphors than when it is not. Empirically, this is the case, as reviewed in the next
section. Accordingly, we predict that metaphor-congruent materials are also more aesthetically
pleasing. We test this hypothesis with two different metaphors. One pertains to the relationship
between valence and vertical location in space and places the good “above” the bad; the other
pertains to the relationship between time and horizontal location in space and places the past
“before” the present and future. Below, we elaborate on these metaphors and cultural and linguistic
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 4
variations in their expression before we report experimental tests conducted with native speakers
of English (in the United States) and Farsi (in Iran).
Valence and Verticality
Numerous expressions entail that good things are “up” and bad things are “down” -- feeling
“on top of the world” is preferred over feeling “down”, a “thumbs up” signal is more favorable
than a “thumbs down” signal, and good people go “up” to heaven, whereas bad people go “down”
to hell. This metaphorical link between valence and verticality can be observed in many languages,
(Kövecses, 2000; Lakoff & Johnson, 1980; Li, 2010; Niksiyar, 2018; Sutton-Spence, 2010),
including English and Farsi, and may be universal. Research examining the impact of vertical
placement on evaluative judgment has shown that things that are positioned “up” are perceived
more positively than things that are positioned “down”. For example, fictional cities that are
positioned in the top vs. bottom section of a map are perceived as being more desirable to live in
(Meier, Moller, Chen, & Riemer-Peltz, 2011) and survey items are evaluated more positively when
presented near the top of the screen than when presented further down (Tourangeau, Couper, &
Conrad, 2013).
Since people use verticality as a way to process valence-related concepts, material that is
congruent with the metaphor should be easier to process than material that is incongruent with it.
Empirically, this is the case. Meier and Robinson (2004, Experiment 1; see also Meier, Sellbom,
& Wygant, 2007) found that positive words were evaluated faster when presented at the top of the
computer screen, above a fixation point, whereas negative words were evaluated faster when
presented at the bottom of the screen, below the fixation point. Moreover, exposure to a positive
word in the middle of the screen facilitated the subsequent identification of a neutral visual
stimulus shown in the upper region of the screen, whereas exposure to a negative word did so for
stimuli shown in the lower region of the screen (Meier & Robinson, 2004, Experiment 2).
Facilitative effects of metaphor congruence can also be observed across modalities, with exposure
to positive words facilitating the identification of high-pitched tones and exposure to negative
words facilitating the classification of low-pitched tones (Weger, Meier, Robinson, & Inhoff,
2007). Such findings indicate that stimuli that are congruent with the metaphorical valence-
verticality link are processed more easily than stimuli that are not. Given that ease of processing
is a key determinant of aesthetic pleasure (Reber et al., 2004; Schwarz, 2018), we predict that
visual arrangements of valenced photographs that are congruent with the up-is-good/down-is-bad
metaphor will be preferred over arrangements that are not. We further predict that this preference
will be observed for speakers of English as well as Farsi, given that their cultures share the same
valence-verticality metaphor.
Time and Space
The progression of time is often described in terms of locomotion through space
(Boroditsky, 2000; Tenbrink, 2011; for a review, see Boroditsky, 2011). Metaphorically, the past
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 5
is “behind” us and the future is “ahead” of us; we move “forward” towards the future, leaving our
past “behind”. The grounding of time in space has been observed in many of the world’s languages
(Haspelmath, 1997; Radden, 2004). Supporting the assumption that metaphor congruence
facilitates fluent processing, Torralbo, Santiago, and Lupianez (2006, Experiment 1) found that
the spatial placement of a word influences how fast its temporal meaning can be identified. In an
ingenious arrangement, they showed past- or future-oriented words next to a side-looking head
silhouette and asked participants whether the word refers to the past or the future. The temporal
reference of future-oriented words was identified faster when the words were presented in front of
the silhouette (as if the person were looking at them), whereas the temporal orientation of past-
oriented words was identified faster when the words were presented in the back of the silhouette.
Although a general grounding of time in space may be universal (Boroditsky, 2011), its
specific implementation shows cultural variation when the relationship is mapped onto a two-
dimensional space. Not surprisingly, whether the letter X comes “before” or “after” the letter Y in
the sequence “X-Y” depends on whether the perceiver reads from left to right (as you will just
have done, reading an English language article) or from right to left. Empirically, the forward-
backward spatial representation of time follows the direction of writing and reading (Boroditsky,
2011; Tversky, Kugelmass, & Winter, 1991). Hence, speakers of languages that are written from
left to right (e.g., English) project the past to the left and the future to the right (Ouellet, Santiago,
Funes, & Lupáñez, 2010; Santiago, Lupáñez, Pérez, & Funes, 2007), whereas the reverse has been
observed for speakers of languages that are written from right to left, such as Hebrew (Fuhrman &
Boroditsky, 2010) and Arabic (Maass & Russo, 2003). Further supporting the hypothesis that
metaphor congruence facilitates processing, Ouellet et al. (2010, Experiment 1) found that the
mere activation of temporal concepts is sufficient to shift the focus of spatial attention. Holding
past-related concepts in mind directed the attention of native Spanish speakers to the left, whereas
holding future-related concepts in mind directed their attention to the right, resulting in enhanced
performance on a spatial orientation task when the target location was congruent with the spatial
implication of the activated time concept. Related work suggests that these facilitation effects are
not limited to visual stimulus presentations but also observed under auditory conditions (Ouellett,
Santiago, Israeli, & Gabay, 2010). Native speakers of Spanish (who read and write from left to
right) were faster responding to orally presented past-words with their left hand and to orally
presented future-words with their right hand; this pattern reversed for native speakers of Hebrew
(who read and write from right to left).
Given that ease of processing influences aesthetic preference (Reber et al., 2004; Schwarz,
2018), we hypothesize that visual arrangements of time-related photographs that are consistent
with the writing direction of the perceiver’s language will be preferred to arrangements that are
not. This predicts that speakers of languages that are written from left to right (e.g., English, Dutch,
Spanish) prefer arrangements that place past-themed photos to the left. This preference should not
be observed for speakers of languages that are written from right to left (e.g., Hebrew, Arabic,
Farsi), who should instead prefer arrangements that place past-themed photos to the right. We
further conjecture that the observed effects may be more pronounced for speakers of left-to-right
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 6
languages than for speakers of right-to-left languages. Most notably, left-to-right languages do not
include elements that are written right-to-left, whereas right-to-left languages write all numerical
expressions from left-to-right, thus reversing the direction of writing between verbal and numerical
expressions. Moreover, the cultural dominance of Western media and smartphones makes it more
likely that speakers of Farsi are exposed to some material that flows left-to-right than that English
speakers are exposed to material that flows right-to-left.
The Present Research
To test the prediction that metaphor congruence enhances aesthetic experience, we draw
on the two metaphors discussed above, which relate valence to vertical position in space and time
to horizontal position in space. Metaphor congruence and incongruence is implemented through
the spatial arrangement of photographs, as illustrated in Figure 1. Displays that place a positively
valenced photo above a negatively valenced photo are congruent with the valence-verticality
metaphor, whereas the reverse arrangement is incongruent with this metaphor. Displays that place
a past-themed photo “before” a modern-themed photo are congruent with the time-space metaphor,
whereas the reverse arrangement is incongruent with this metaphor. As discussed, what counts as
“before” or “after” in a horizontal display depends on the perceiver’s direction of reading, giving
rise to cultural variation.
Because a choice between two simultaneously presented stimuli imposes no memory load
and forced-choice tasks show minimal response bias effects (Palmer, Schloss, Sammartino, 2013),
we assess aesthetic preference by using a two-alternative forced-choice paradigm, asking
participants to select their preferred arrangement. Previous research found that judgments of
preference, liking, and beauty show similar patterns (Bornstein, 1989; Reber et al., 2004) and we
vary the wording of the choice task across experiments. Throughout, our interest is in the
perceiver’s aesthetic preference, not in the perceiver’s evaluation of the object’s artistic value.
Many aesthetically pleasing objects lack high artistic value (e.g., a photo of a sunset); conversely,
many objects of high artistic value are not aesthetically pleasing (e.g., Marcel Duchamp’s
Fountain, a urinal altered to resemble a drinking fountain).
In all experiments, we present valenced photographs in arrangements that are congruent or
incongruent with the good-is-up/bad-is-down metaphor and past- vs. present-themed photographs
in arrangements that are congruent or incongruent with the past-before-present metaphor.
Consistent with the theoretical rationale of fluency research and robust findings across classic
fluency tasks, all manipulations are within-participants. People are more sensitive to changes in
their processing experience than to stable states, which makes within-participant manipulations
more powerful than between-participant manipulations, where some participants are only exposed
to easy-to-process and others only to difficult-to-process material (Wänke & Hansen, 2015).
Hence, classic fluency effects, including Zajonc’s (1968) mere exposure effect and Hasher and
colleagues’ (1977) illusory truth effect, are reliably observed in within- but not in between-
participant designs (Dechêne, Stahl, Hansen, & Wänke, 2009; Hansen, Dechêne, & Wänke, 2008).
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 7
This makes within-participant manipulations more appropriate for testing fluency predictions than
between-participant manipulations.
Experiment 1 examines aesthetic preference by asking native speakers of English to select
the arrangement they think is better. Experiments 2a-b replicate Experiment 1 by asking native
speakers of English to select the arrangement they like more. Experiments 3a-b investigate cultural
differences of preference between English and Farsi speakers. Experiment 4 examined the effect
of familiarity with English on Farsi speakers’ preference. Finally, a single paper meta-analysis
(Bornstein, Hedges, Higgins, & Rothstein, 2013) examined the consistency of results across
experiments and is presented in Appendix A of the supplementary materials. All materials and
data are available at https://osf.io/xystg/?view_only=bf82a309fa3e4c7abdc5a06a626e5a33.
Experiment 1
Method
Participants. Based on the effect size of Reber and colleagues’ (1998) experiment 1
(Cohen’s d=0.24), a sample of 139 participants is required to achieve a power of 0.80 at alpha =
0.05. To ensure sufficient power after excluding non-native speakers of English, we recruited 190
undergraduate students from the University of Southern California, who completed the experiment
online. The exclusion of 28 non-native speakers of English left a total of 162 participants for
analysis.
Design and materials. Eight pairs of time-related photos (past vs. modern) and eight pairs
of valence-related photos (happy vs. sad) were presented to each participant, along with eight pairs
of filler photos (e.g., animals and landscape). For the time-related and valence-related photos, half
of the pairs were arranged horizontally and half vertically (Figure 1). This results in four within-
subject combinations of photos x spatial arrangement. Two of these combinations (time-horizontal,
valence-vertical) bear on the theoretical predictions and two (time-vertical, and valence-
horizontal) are exploratory. Next, we describe these combinations.
Metaphor conditions. The combinations of interest pertain to the horizontal placement of
time-related photos and the vertical placement of valence-related photos. For speakers of English
and other languages that write from left to right, the past-before-future metaphor implies that the
past is on the left and the future on the right when presented in two-dimensional space. Hence, a
horizontal visual arrangement that presents past-themed photos to the left of modern-themed
photos is metaphor congruent, whereas the reverse arrangement is metaphor incongruent. For the
good-is-up/bad-is-down metaphor, a vertical visual arrangement that presents positively valenced
photos above negatively valenced photos is metaphor congruent, whereas the reverse arrangement
is metaphor incongruent.
Exploratory conditions. The remaining combinations are silent on the role of metaphor
congruence in aesthetic preference because the metaphors do not bear on the vertical location of
time or the horizontal location of valence. We included these combinations for exploratory
purposes (Appendix B of the supplementary materials).
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 8
Procedure. All participants saw all 16 pairs of photos along with 8 pairs of filler photos.
Half of the participants saw all horizontal arrangements first and half saw all vertical arrangements
first. The order of presentation was randomized within the horizontal and vertical conditions. For
each pair of photos, participants were asked to choose the arrangement that they think is better. At
the end of the experiment, demographic information was collected.
Figure 1. An example of horizontal arrangements of time-related photos (left) and vertical
arrangements of valence-related photos (right).
Results and Discussion
Trials where the participant chose the congruent arrangement were coded “1” and trials
where the participant chose the incongruent arrangement were coded “0”. For each metaphor
condition, the proportion of times participants chose the metaphor congruent over the metaphor
incongruent arrangement was obtained by averaging responses from each trial, which was then
compared against chance (0.5) with a two-tailed one-sample t-test. A sensitivity power analysis
using G*Power indicates 80% power to detect a minimum effect size of Cohen’s d = 0.196.
As predicted, participants preferred the metaphor-congruent arrangement over the
metaphor-incongruent arrangement for both time and valence metaphors (see Figure 2 and Table
1). The proportion of trials participants chose happy-above-sad arrangements over sad-above-
happy arrangements was 0.613, 95% CI [0.563, 0.662], t(161) = 4.445, p < 0.001. The preference
for the happy-above-sad arrangements was observed in seven out of eight pairs of valence-themed
photos, with five of them being significant. The proportion of times participants chose past-on-
the-left arrangements was 0.664, 95% CI [0.617, 0.710], t(161) = 6.928, p < 0.001. The preference
for past-on-the-left arrangements was observed for all of the eight time-themed photo pairs, with
six of them being significant. These results provide first evidence that metaphor-congruent
arrangements are perceived as “better” than metaphor-incongruent arrangements. Experiments 2a-
b replicate this effect by asking participants to select the arrangement they “like more”.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 9
Experiment 2a-b: Liking
Method
Participants. Based on the smallest effect size observed in Study 1 (valence-verticality,
Cohen’s d = 0.349), G-power software (Faul, Erdfelder, Lang, & Buchner, 2007) indicates that a
total sample size of 67 is needed for a two-tailed one-sample t-test at an alpha level of .05 and
power of .80. We recruited 80 participants to allow for the exclusion of non-native speakers of
English (see https://aspredicted.org/pp59i.pdf for the pre-registration).
Experiment 2a was conducted on Amazon Mechanical Turk and Experiment 2b in the
subject pool of the University of Southern California; eighty time slots were posted for each
experiment. For Experiment 2a, participation was limited to those with United States IP addresses
and approval ratings of 95% or higher for previous HITS. Participants were compensated with
$0.50; N = 80 participants completed the study. For Experiment 2b, undergraduates who are native
speakers of English were recruited. All participants in Experiment 2a and 2b reported being native
speakers of English and none were excluded from data analysis.
Design and material. The design of Experiment 2a and 2b was identical to Experiment 1,
except that half of the photographs used in Experiment 1 were replaced with new stimuli, in order
to 1) make the photos in each pair resemble each other more in terms in terms of lighting and
orientation, and 2) to use more culturally appropriate photos for an Iranian sample that we planned
to recruit in Experiment 3.
Procedure. Experiments 2a and 2b followed the same procedures as Experiment 1, except
that the wording of the selection task now read, “Which arrangement do you like more?”.
Results and Discussion
The proportion of times participants chose the metaphor congruent arrangement over the
incongruent one was again compared to chance (0.5) with a two-tailed one-sample t-test. A
sensitivity power analysis using G*Power indicates each experiment has 80% power to detect a
minimum effect size of Cohen’s d = 0.280.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 10
Figure 2. Results of Experiment 1, 2a, and 2b. The error bars represent 95% confidence intervals.
Valence: happy is “up”. As shown in Figure 2 and Table 1, participants selected the
metaphor-congruent happy-above-sad arrangements more frequently than the metaphor-
incongruent sad-above-happy arrangements. However, the observed differences were small and
not significant; M = 0.550, 95% CI [0.464, 0.636], t(79) = 1.139, p = 0.258 for Experiment 2a, and
M = 0.553, 95% CI [0.479, 0.627], t(79) = 1.411, p = 0.162 for Experiment 2b. The preference for
happy-above-sad arrangements was observed in six of the eight pairs in Experiment 2a, and seven
out of the eight pairs in Experiment 2b. However, none reached significance. A single-paper meta-
analysis (Bornstein et al., 2013) including these nonsignificant effects confirmed the overall
reliability of the consistent patterns across studies (see Appendix A of supplementary materials).
Time: past “before” future. Replicating Experiment 1, participants preferred the
metaphor congruent past-on-the-left arrangements over the metaphor incongruent past-on-the right
arrangements in both studies; M = 0.700, 95% CI [0.629, 0.771], t(79)=5.519, p < 0.001, for
Experiment 2a, and M = 0.631, 95% CI [0.563, 0.700], t(79)=3.749, p < 0.001, for Experiment 2b.
This preference for past-on-the-left arrangements was observed in all of the eight time-themed
photo pairs in Experiment 2a, and seven out of the eight pairs in Experiment 2b, with five pairs
being significant in each experiment.
In sum, Experiments 1, 2a, and 2b show the predicted preference for metaphor congruent
over metaphor incongruent displays of visual stimuli. Among native speakers of English, the
observed differences are larger and more reliable for time-related stimuli than for valence-related
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 11
stimuli. Next, we turn to the moderation of metaphor congruence effects by culturo-linguistic
differences in the direction of writing.
Experiment 3a-b: Cultural Differences
At the conceptual level, good things are “up” and the past comes “before” the present across
many cultures and languages. However, whether X “precedes” or “follows” Y in an X-Y
arrangement may depend on the direction of reading -- X precedes Y when reading from left to
right, but follows Y when reading from right to left. Hence, the direction in which perceivers read
should moderate aesthetic preference for time-related materials but not for valence-related
materials. Experiment 3a and b test this prediction with native speakers of English in the U.S., who
read from left to right, and native speakers of Farsi in Iran, who write from right to left.
Because good things are metaphorically “up” in both English and Farsi, happy-above-sad
arrangements should be preferred over sad-above-happy arrangements for both U.S. and Iranian
participants. In Farsi, there are numerous metaphoric expressions indicating that good, important,
sacred, and valuable things or beings are “up”, while deplorable, trivial, evil, and worthless things
are “down”. Much of this metaphoric language is reflected in ancient and contemporary Persian
poetry (Niksiyar, 2018). However, English and Farsi speakers should differ in their preference for
the horizontal arrangement of time-themed materials. Past-on-the-left arrangements should be
preferred over past-on-the-right arrangements by U.S. participants, whereas Farsi speakers in Iran
should show the reverse pattern because Farsi text reads from right to left, although numbers read
from left to right in Farsi. Consistent with this difference in the direction of text and numbers,
Iranians prefer the right-to-left arrangements when objects are labeled in Farsi, but prefer left-to-
right arrangements when objects are labeled numerically (Matoori, Gorjian, Veysi, & Memari,
2020).
Method
Participants. We preregistered to recruit 80 participants in each country. Due to a
discrepancy between the recruitment platforms and Qualtrics, 84 U.S. students at the University
of Southern California (Experiment 3a) and 90 Iranian participants (Experiment 3b) from Tarbiat
Modares University (Tehran, Iran) completed the experiments online. We pre-registered to exclude
participants who do not speak English as their native language for Experiment 3a
(https://aspredicted.org/pv2py.pdf) and participants who indicate a native language that writes
from left to right for Experiment 3b (https://aspredicted.org/74qp3.pdf). This resulted in the
exclusion of one U.S. participant in Experiment 3a (leaving N = 83 for analysis) and five Iranian
participants in Experiment 3b (leaving N = 85 for analysis).
Design, material, and procedure. The design and material of Experiment 3a and 3b were
identical to Experiment 2a and 2b. The procedure was identical to Experiment 1, where participants
were asked to select the arrangement that they think is “better”. For Iranian participants
(Experiment 3b), all instructions and questions were presented in Farsi.
Result and Discussion
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 12
We again computed a two-tailed one-sample t-test for each metaphor to compare the
proportion of times participants choose the metaphor-congruent arrangement to chance (0.5).
Figure 3 depicts the results. A sensitivity power analysis using G*Power indicates Experiment 3a
and 3b have 80% power to detect a minimum effect size of Cohen’s d = 0.275 and 0.272,
respectively.
Valence: happy is “up”. Consistent with the hypothesis, native speakers of English as
well as Farsi preferred the happy-above-sad arrangements; M = 0.584, 95% CI [0.517, 0.652],
t(82) = 2.455, p = 0.016 for the English speaking U.S. participants, and M = 0.629, 95% CI [0.559,
0.700], t(84) = 3.605, p = 0.001 for the Farsi-speaking Iranian participants. This preference for
happy-above-sad arrangements was observed in all of the eight pairs for English speakers and in
seven out of eight pairs for Farsi speakers, although only one and three pairs reached significance,
respectively. A logistic regression (P(Preferenceij =1) = logistic(γ00 + γ01 Countryj + u0j)) with
country (0 = US, 1 = Iran) as a predictor yielded similar results, Mean Predicted Probability =
0.591, 95% CI [0.517, 0.664], p = 0.017 for the English speaking U.S. participants, and Mean
Predicted Probability = 0.641, 95% CI [0.569, 0.712], p < 0.001, for Farsi speaking participants,
with no significant difference in the mean predicted probabilities between the two countries, OR
= 1.288, 95% CI [0.771, 2.175], p = 0.331.
Time: past “before” future. Consistent with our predictions, native speakers of English
again preferred the past-on-the-left arrangements, M = 0.690, 95% CI [0.622, 0.758], t(82)=5.467,
p < 0.001. This preference was observed in all eight pairs of time-themed photos, with five of them
being significant. Also consistent with predictions, this preference was not observed for speakers
of Farsi; for Farsi speakers, the preference for past-on-the-left arrangements was only observed in
two out of eight time-themed photos, and none was significant. However, contrary to predictions,
the preference for past-on-the-left arrangements was only eliminated and not fully reversed, M =
0.477, 95% CI [0.409, 0.544], t(84) = -0.679, p = 0.499. A logistic regression (P(Preferenceij =1)
= logistic(γ00 + γ01 Countryj + u0j)) with country (0 = US, 1 = Iran) as a predictor yielded similar
results, Mean Predicted ProbabilityPast-Left = 0.706, 95% CI [0.637, 0.773], p < 0.001 for the English
speaking U.S. participants, and Mean Predicted Probability Past-Left = 0.475, 95% CI [0.400, 0.549],
p = 0.504 for the Farsi speaking participants. This is reflected in a significant difference between
the two countries, with a 0.692 decrease in the odds of Iranian participants choosing the past-on-
the-left arrangements compared to U.S. participants, OR=0.308, 95% CI [0.176, 0.518], p < 0.001.
In sum, these findings replicate Experiments 1, 2a, and 2b for English speakers in the U.S.
and allow comparisons with Farsi speakers in Iran. As predicted, both groups preferred the
metaphor-congruent happy-above-sad arrangements over the metaphor-incongruent sad-above-
happy arrangements. This is consistent with the use of the same metaphor in both cultures. Both
cultures also share the past-before-future metaphor, but the representation of this metaphor in two-
dimensional space differs as a function of the direction of writing. As predicted. English speakers
preferred the metaphor congruent past-on-the-left arrangements over the metaphor-incongruent
past-on-the-right arrangements, whereas Farsi speakers did not. In contrast to predictions,
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 13
however, Farsi speakers did not show a reverse preference for past-on-the right arrangements.
Experiment 4 addresses possible reasons for this observation.
Figure 3. Results of Experiment 3a and 3b. The error bars represent 95% confidence intervals.
Experiment 4: Iranian Replication
As noted, Farsi is written from right to left, whereas numbers are written from left to right,
which makes Farsi speakers familiar with both directions of writing and reading. Moreover,
modern technology has to some extent disrupted the traditional flow of writing: when texting on
mobile phones, many Farsi speakers write Farsi with English letters in a left-to-right direction.
Some Iranian scholars have referred to this manner of writing as “Finglish”, a portmanteau coined
from the combination of the words “Farsi” and “English” (e.g., Alipour, Aghayoosefi, &
Abaszade, 2013). Finglish is particularly popular among young people in Iran. Moreover, the Farsi
speakers of the university sample used in Experiment 3b were most likely familiar with English as
a second language. To address these possibilities, we collected information on the frequency of
texting in “Finglish” and familiarity with English as a second language in Experiment 4, which is
a replication of Experiment 3b in all other respects.
Method
Participants. We pre-registered to recruit 260 participants and to exclude those who
indicate a native language that writes from left to right (see https://aspredicted.org/3cy83.pdf for
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 14
the pre-registration). Due to a discrepancy between the recruitment platforms and Qualtrics, 269
participants at Tarbiat Modares University completed the experiment. One participant who
reported English as their native language was excluded from data analysis.
Design, material, and procedure. Experiment 4 used the same material and design as
Experiment 2a-b and 3a-b. Participants completed the experiment in Farsi, and were asked to select
the arrangement that they think is better. Participants were asked to self-report their English
proficiency on a 1-7 scale, where 1 indicates low and 7 indicates high English proficiency.
Participants were also asked to report the frequency with which they use “Finglish” (using English
letters to type Farsi in the left-to-right direction).
Results and Discussion
A two-tailed one-sample t-test was computed for each metaphor. A sensitivity power
analysis using G*Power indicates the experiment has 80% power to detect a minimum effect size
of Cohen’s d = 0.152. In addition, English familiarity scores were obtained by averaging
participants’ self-reported English proficiency and Fenglish usage. The effect of English
familiarity on preference for past-on-the-left arrangements was then tested by performing a one-
sample t-test separately for those with high vs. low English familiarity. The median of the
composite English familiarity was 3 out of 6. A score smaller than 3 was categorized as having
low familiarity (n =111) and a score larger than 3 was categorized as high familiarity (n =166).
Replicating Experiment 3b, Iranian participants preferred the metaphor-congruent happy-
above-sad arrangements at above chance level, M = 0.622, 95% CI [0.581, 0.664], t(267) = 5.773,
p < 0.001. This preference was observed in all of the eight pairs of valence-themed photos, with
seven being significant. Also replicating Experiment 3b, Iranian participants did not show a
preference for past-on-the-left arrangements, nor did they show a significantly reversed preference
for past-on-the-right arrangements, M = 0.496, 95% CI [0.460, 0.533], t(267) = -0.200, p = 0.842.
This is reflected in the item analysis. Iranian participants preferred past-on the right for 3 of 8 pairs
and past-on-the left for another 3 pairs, with none of the differences significant; they had no
directional preference for the remaining 2 pairs. Similar results were obtained through logistic
regressions (P(Preferenceij =1) = logistic(γ00 + u0j )) with Mean Predicted ProbabilityHappy-
Up=0.636, 95% CI [0.591, 0.682], p < 0.001, for the valence metaphor, and Mean Predicted
ProbabilityPast-Left = 0.496, 95% CI [0.458, 0.534], p = 0.83, for the time metaphor. Contrary to
expectations, participants’ preference for the time-themed materials was not moderated by self-
reported familiarity with English. The proportion of time participants preferred the past-on-the-
left arrangements was 0.473, 95% CI [0.416, 0.530], t(110)= - 0.931, p = 0.354 for participants
with low English and 0.517, 95% CI [0.461, 0.573], t(115)= 0.605, p = 0.546, for participants with
high English familiarity.
An exploratory logistic regression (P(Preferenceij =1) = logistic(γ00 + γ10 Eng_Famij + u0j))
with English familiarity as a continuous variable yielded similar results. When English familiarity
is at 0 (i.e., not familiar with English at all), the mean predicted probability of preferring past-on-
the-right arrangements was 0.465, 95% CI [0.352, 0.582], p = 0.559, and with a one unit increase
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 15
in English familiarity, the mean predicted probability increased to 0.475, 95% CI [0.392, 0.560].
However, similar to the pre-registered median split analysis, this was not significant, OR = 1.042,
95% CI [0.902, 1.204], p = .574.
In combination, Experiments 3b and 4 suggest that the horizontal ordering of time-related
stimuli does not influence the aesthetic preferences of native speakers of Farsi. This may reflect
that Farsi includes writing from right to left (for verbal material) as well as writing from left to
right (for numbers). In related research, Matoori et al. (2020) found that Iranians prefer left-to-
right arrangements when objects are labeled numerically. In addition, exposure to Western media
and culture, such as its left-to-right presentation of chronology, may have contributed to a more
flexible representation of the time-space relationship for Iranian participants.
Table 1. Results of Experiment 1-4.
Experiment
Mean
t
Valence:
Happy is "up"
1: US
0.613***
4.445
2a: US
0.550
1.139
2b: US
0.553
1.411
3a: US
0.584*
2.455
3b: Iran
0.629**
3.605
4: Iran
0.622***
5.773
Time:
Past "before" future
1: US
0.664***
6.928
2a: US
0.700***
5.519
2b: US
0.631***
3.749
3a: US
0.690***
5.467
3b: Iran
0.524
0.679
4: Iran
0.504
0.200
1. The mean represents the mean proportion of congruent arrangements being chosen.
2. The congruent arrangement for Valence is happy-above-sad for both US and Iranian
samples.
3. The congruent arrangement for Time is past-on-the-left for US samples and past-on-the-right
for Iranian samples.
(Note that the text of the paper reports the mean proportion of past-on-the-left arrangements
being chosen for both US and Iranian samples instead).
4. * represents p < 0.05, ** represents p < 0.01, *** represents p < 0.001.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 16
General Discussion
Our six experiments show that people prefer visual arrangements that are congruent with
applicable conceptual metaphors over arrangements that are not. This observation connects
theories of embodied cognition with theories about the metacognitive basis of aesthetic preference.
According to conceptual metaphor theory (Lakoff & Johnson, 1980, 1999; for a review see
Landau, 2017), representations of abstract concepts (e.g., valence, time) are grounded in
sensorimotor experiences with the physical world as reflected in the mappings of valence on
vertical and time on horizontal space. Stimuli that are congruent with applicable metaphors are
processed faster than stimuli that are metaphor incongruent as has been observed for valence (e.g.,
Meier, Hauser, Robinson, Friesen, & Schjeldahl, 2007; Meier & Robinson, 2004) and time (e.g.,
Ouellet et al, 2010 a,b).
Ease of processing, in turn, is a key input into judgments of aesthetic preference (Reber et
al., 2004). It underlies the influence of object (e.g., symmetry, contrast, clarity, prototypicality)
and perceiver variables (e.g., exposure history, expertise) that have long been the focus of
empirical aesthetics and predicts systematic effects of variables that are outside the scope of
traditional theories of aesthetics, including visual (e.g., Reber et al., 1998) and conceptual (e.g.,
Winkielman et al., 2003) primes. A processing fluency account of aesthetic pleasure thus provides
a parsimonious mechanism that connects variables that would otherwise have been considered in
isolation, with each requiring separate explanations. Going beyond previous observations, the
present experiments identify the influence of a variable that qualifies as a joint characteristic of the
object, the perceiver, and the cultural context, namely metaphor congruence: perceivers prefer
stimuli whose characteristics are congruent with the form in which a conceptual metaphor is
expressed in the perceiver’s culture over objects that are not. This observation highlights that
beauty is neither in the beholden nor in the eye of the beholder, but in the perceiver’s processing
experience, which is a joint function of object, perceiver, and context variables (Reber et al., 2004;
Schwarz, 2018).
The current findings also suggest cross-cultural similarities in aesthetic preference to the
extent that an applicable conceptual metaphor is shared across cultures. In the present experiments,
both English and Farsi speakers share the valence-verticality metaphor, such that good things are
“up” and bad things are “down”, which results in a consistent preference for metaphor congruent
happy-above-sad arrangements. We expect similar effects for other widely shared conceptual
metaphors. For example, the conceptualization of power is also grounded in verticality across
many cultures, with high power represented higher in vertical space than low power (e.g.,
Schubert, 2005; Tang, Zhou, & Zhang, 2018; Wu, Jia, Wang, Du, Wu, & Dang, 2016). Hence,
stimuli associated with high power (e.g., photographs of influential world leaders) should have
more aesthetic appeal when placed high in space, whereas artworks associated with low power
(e.g., photographs depicting poverty) should have more aesthetic appeal when placed low in space.
Similarly, as brightness is associated with valence (e.g., Meier, Robinson, & Clore, 2004; Meier,
Robinson, Crawford, & Ahlvers, 2007), the appeal of positively (vs. negatively) valenced artworks
should increase with their brightness.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 17
Conversely, cultural differences in conceptual metaphors should result in cultural
differences in aesthetic preference. Although both English and Farsi speakers share the time-space
metaphor, they differ in their conceptualization of the direction in which time flows, which follows
the direction of writing and reading. English speakers consistently preferred arrangements that
placed a past-themed image before a modern-themed image. This preference was consistently
eliminated with Farsi speakers, although not fully reversed. As already discussed, Farsi speakers’
indifference to the spatial placement of time-oriented stimuli may reflect that Farsi speakers are
familiar with both directions of reading and writing, due to verbal materials written from right to
left and numbers from left to right.
Some caveats should be addressed. The present experiments used choice as a measure of
preference, consistent with a long tradition in behavioral science and decision research. A choice
format, in which both stimuli are presented simultaneously, also has the advantage of providing a
sensitive test of the influence of subjective processing experiences. From sensory perception to
judgment, people are more sensitive to changes in subjective experience than to stable states;
hence, fluency effects are more reliably obtained in within-participant than between-participant
designs (Wanke & Hansen, 2015). However, such manipulations can also provide participants with
increased insight into the hypotheses studied. We therefore included an open-ended debriefing
question that asked participants to tell us their thoughts about the studies. Not a single participant
mentioned “metaphor” or “valence”, and only a few mentioned concepts related to “chronology”
(7.9%, 4.7%, 1.5%, 5.7% in Exp.1, Exp. 2a, Exp. 2b, and Exp. 3a respectively). This suggests that
participants had very limited insight into the hypotheses tested and renders concerns about demand
effects mute. Nevertheless, studies with more indirect indicators of preference -- assessing, for
example, participants’ spontaneous affective response with electromyography (Winkielman &
Cacioppo, 2001) -- would be welcome.
More importantly, the observed influence of metaphor congruence on aesthetic preference
should hold for any conceptual metaphor. Testing this prediction with a broad range of metaphors
and diverse aesthetic stimuli provides a promising avenue for future research. Finally, experienced
processing fluency serves as information for a wide range of judgments other than aesthetic
preference (for reviews, see Alter & Oppenheimer, 2009; Schwarz et al., 2021), including truth
(e.g., Reber & Schwarz, 1999), novelty (e.g., Jacoby, Woloshyn, & Kelly, 1989), and risk (e.g.,
Song & Schwarz, 2009), among others. Hence, the fluency enhancing effect of metaphor congruent
presentations may influence a broad range of judgments in ways that are not predicted by the
metaphor’s specific content.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 18
Open Practice
We have made the stimuli and data of all experiments publicly available at the Open
Science Framework: https://osf.io/xystg/?view_only=bf82a309fa3e4c7abdc5a06a626e5a33. We
confirm that the information and files uploaded to the Open Science Framework are sufficient for
an independent researcher to reproduce the reported methodology and results.
We have made the pre-registrations of Experiments 2a-b, 3a-b and 4 available at the
following links: https://aspredicted.org/pp59i.pdf, https://aspredicted.org/pv2py.pdf,
https://aspredicted.org/74qp3.pdf, https://aspredicted.org/3cy83.pdf. There were no additional
pre-registrations other than the ones reported, but due to a record-keeping error, the pre-registered
studies 2a-b, 3b, and 4 reference an erroneous pre-registration number #14284. The correct pre-
registration that should be referenced is #14487 (Experiment 3a).
There were no changes to the pre-registered analysis plan for the primary confirmatory
analysis in Experiments 2a-b and 3a-b. In Experiment 4, we pre-registered average reported 1)
English proficiency, 2) frequency of English usage, and 3) frequency of Finglish as a composite
indicator of English familiarity. However, due to a communication error, the frequency of English
usage was omitted in the translation process. Hence, the pre-registered analysis was performed by
averaging the responses of the other two variables. All analyses described in the pre-registered
plans are reported.
References
Alipour, A., Aghayoosefi, A., & Abaszade, A. N. (2013). The effect of reading Persian text written
with English letters (Finglish) on salivary cortisol in high school female students. Journal
of School Psychology, 2, 154-162.
Alter, A. L., & Oppenheimer, D. M. (2009). Uniting the tribes of fluency to form a metacognitive
nation. Personality and Social Psychology Review, 13(3), 219-235.
Bornstein, R. F. (1989). Exposure and affect: overview and meta-analysis of research, 19681987.
Psychological Bulletin, 106(2), 265.
Borenstein, M., Hedges, L. V., Higgins, J. P., & Rothstein, H. R. (2010). A basic introduction to
fixed‐effect and random‐effects models for meta‐analysis. Research Synthesis Methods,
1(2), 97-111.
Borenstein, M., Hedges, L., Higgins, J., & Rothstein, H. (2013). Comprehensive Meta-Analysis
(Version 3) [Computer Software]. Englewood, NJ: Biostat.
Boroditsky, L. (2000). Metaphoric structuring: Understanding time through spatial metaphors.
Cognition, 75(1), 1-28.
Boroditsky, L. (2011). How languages construct time. In S. Dehaene & E. Brannon (Eds.), Space,
time and number in the brain: Searching for the foundations of mathematical thought (pp.
333-341). New York: Academic Press.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 19
Cárdenas, R. A., & Harris, L. J. (2006). Symmetrical decorations enhance the attractiveness of
faces and abstract designs. Evolution and Human Behavior, 27(1), 1-18.
Checkosky, S. F., & Whitlock, D. (1973). The effects of pattern goodness on recognition time in
a memory search task. Journal of Experimental Psychology, 100, 341348.
Dechêne, A., Stahl, C., Hansen, J., & Wänke, M. (2010). The truth about the truth: A meta-analytic
review of the truth effect. Personality and Social Psychology Review, 14(2), 238-257.
Enquist, M., & Arak, A. (1994). Symmetry, beauty and evolution. Nature, 372(6502), 169-172.
Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical power
analysis program for the social, behavioral, and biomedical sciences. Behavior Research
Methods, 39, 175-191.
Fuhrman, O., & Boroditsky, L. (2010). Cross‐cultural differences in mental representations of
time: Evidence from an implicit nonlinguistic task. Cognitive science, 34(8), 1430-1451.
Garner, W. R. (1974). The stimulus in information processing. In H.R. Moskowitz, B. Scharf, J.C.
Stevens (Eds.), Sensation and measurement (pp. 77-90). Springer, Dordrecht.
Haber, R. N., & Hershenson, M. (1965). The effects of repeated brief exposures on growth of a
percept. Journal of Experimental Psychology, 69, 4046.
Hansen, J., Dechêne, A., & Wänke, M. (2008). Discrepant fluency increases subjective truth.
Journal of Experimental Social Psychology, 44(3), 687-691.
Hasher, L., Goldstein, D., & Toppino, T. (1977). Frequency and the conference of referential
validity. Journal of Verbal Learning & Verbal Behavior, 16, 107-112.
Haspelmath, M. (1997). From space to time: Temporal adverbials in the world’s languages.
Newcastle, UK: Lincom Europa.
Jacoby, L. L., & Dallas, M. (1981). On the relationship between autobiographical memory and
perceptual learning. Journal of Experimental Psychology: General, 110, 306340.
Jacoby, L. L., Woloshyn, V., & Kelley, C. (1989). Becoming famous without being recognized:
Unconscious influences of memory produced by dividing attention. Journal of
Experimental Psychology: General, 118(2), 115.
Kinder, A., Shanks, D. R., Cock, J., & Tunney, R. J. (2003). Recollection, fluency, and the
explicit/implicit distinction in artificial grammar learning. Journal of Experimental
Psychology: General, 132(4), 551.
Kovecses, Z. (2000). The scope of metaphor. Topics in English Linguistics, 30, 79-92.
Lakoff, G., & Johnson, M.. (1980). Metaphors we live by. Chicago, IL: The University of Chicago
Press.
Lakoff, G., & Johnson, M.(1999). Philosophy in the flesh: The embodied mind and its challenges
to western thought. New York: Basic Books.
Landau, M. J. (2017). Conceptual metaphor in social psychology: The poetics of everyday life.
New York: Psychology Press.
Li, X. (2010). Conceptual metaphor theory and teaching of English and Chinese idioms. Journal
of Language Teaching and Research, 1(3), 206-210.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 20
Maass, A., & Russo, A. (2003). Directional bias in the mental representation of spatial events:
Nature or culture? Psychological Science, 14(4), 296-301.
Matoori, H., Gorjian, B., Veysi, E., Memari, M. (2020). Temporal frames of reference in Persian
speakers language based on conceptual metaphors theory. Advances in Cognitive Sciences,
22, 102-115.
Meier, B. P., Hauser, D. J., Robinson, M. D., Friesen, C. K., & Schjeldahl, K. (2007). What's " up"
with God? Vertical space as a representation of the divine. Journal of Personality and
Social Psychology, 93(5), 699.
Meier, B. P., Moller, A. C., Chen, J. J., & Riemer-Peltz, M. (2011). Spatial metaphor and real
estate: North-south location biases housing preference. Social Psychological and
Personality Science, 2(5), 547-553.
Meier, B. P., & Robinson, M. D. (2004). Why the sunny side is up: Associations between affect
and vertical position. Psychological Science, 15(4), 243-247.
Meier, B. P., Robinson, M. D., & Clore, G. L. (2004). Why good guys wear white: Automatic
inferences about stimulus valence based on brightness. Psychological Science, 15(2), 82-
87.
Meier, B. P., Robinson, M. D., Crawford, L. E., & Ahlvers, W. J. (2007). When" light" and" dark"
thoughts become light and dark responses: Affect biases brightness judgments. Emotion,
7(2), 366.
Meier, B. P., Sellbom, M., & Wygant, D.P. (2007). Failing to take the moral high ground:
Psychopathy and the vertical representation of morality. Personality and Individual
Differences, 43, 757 767.
Niksiyar, Z. (2018). Articulating cognitive schemas and metaphors of motion in the poem Mosafer
of Sohrab Sepehri. Journal of Linguistic and Rhetorical Studies, 9, 261-286.
Ouellet, M., Santiago, J., Funes, M. J., & Lupiánez, J. (2010a). Thinking about the future moves
attention to the right. Journal of Experimental Psychology: Human Perception and
Performance, 36(1), 17.
Ouellet, M., Santiago, J., Israeli, Z., & Gabay, S. (2010b). Is the future the right time? Experimental
Psychology, 57, 308-314.
Palmer, S. E., Schloss, K. B., & Sammartino, J. (2013). Visual aesthetics and human preference.
Annual Review of Psychology, 64, 77-107.
Radden, G. (2004). The metaphor TIME AS SPACE across languages. In: N. Baumgarten, C.
Boettger, M. Motz, & J. Probst (Eds.), Uebersetzen, interkulturelle Kommunikation,
Spracherwerb und Sprachvermittlung - Das Leben mit mehreren Sprachen: Festschrift fuer
Juliane House zum 60. Geburtstag (pp. 225238). Bochum: Aks-Verlag.
Reber, R., Schwarz, N., & Winkielman, P. (2004). Processing fluency and aesthetic pleasure: Is
beauty in the perceiver's processing experience? Personality and Social Psychology
Review, 8(4), 364-382.
Reber, R., Winkielman, P., & Schwarz, N. (1998). Effects of perceptual fluency on affective
judgments. Psychological Science, 9(1), 45-48.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 21
Santiago, J., Lupáñez, J., Pérez, E., & Funes, M. J. (2007). Time (also) flies from left to right.
Psychonomic Bulletin & Review, 14(3), 512-516.
Schwarz, N. (1990). Feelings as information: Informational and motivational functions of affective
states. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of motivation and cognition:
Foundations of social behavior (Vol. 2; pp. 527-561). New York, NY: Guilford Press.
Schwarz, N. (2012). Feelings-as-information theory. In P. A. M. Van Lange, A. W. Kruglanski, &
E. T. Higgins (Eds.), Handbook of theories of social psychology (pp. 289308). Sage
Publications Ltd.
Schwarz, N. (2018). Of fluency, beauty, and truth: Inferences from metacognitive experiences. In
J. Proust & M. Fortier (Eds.), Metacognitive diversity: An interdisciplinary approach (pp.
2546). New York: Oxford University Press.
Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being:
Informative and directive functions of affective states. Journal of Personality and Social
Psychology, 45(3), 513.
Schwarz, N., Jalbert, M., Noah, T., & Zhang, L (2021). Metacognitive experiences as information:
Processing fluency in consumer judgment and decision making. Consumer Psychology
Review.
Schubert, T. (2005). Your highness: Vertical positions as perceptual symbols of power. Journal of
Personality and Social Psychology, 89 1, 1-21.
Song, H., & Schwarz, N. (2009). If it's difficult to pronounce, it must be risky: Fluency, familiarity,
and risk perception. Psychological Science, 20(2),135-138.
Sutton-Spence, R. (2010). Spatial metaphor and expressions of identity in sign language poetry.
Metaphorik. de, 19, 1-40.
Tang, F., Zhou, H., Zhang, H., & Zhu, L. (2018). Development of spatial representation of power
in children. Journal of Experimental Child Psychology, 173, 239-249 .
Tenbrink, T. (2011). Reference frames of space and time in language. Journal of Pragmatics,
43(3), 704-722.
Torralbo, A., Santiago, J., & Lupiáñez, J. (2006). Flexible conceptual projection of time onto
spatial frames of reference. Cognitive Science, 30(4), 745-757.
Tourangeau, R., Couper, M. P., & Conrad, F. G. (2013). “Up means good” the effect of screen
position on evaluative ratings in web surveys. Public Opinion Quarterly, 77(S1), 69-88.
Tversky, B., Kugelmass, S., & Winter, A. (1991). Cross-cultural and developmental trends in
graphic productions. Cognitive Psychology, 23(4), 515-557.
Wänke, M., & Hansen, J. (2015). Relative processing fluency. Current Directions in Psychological
Science, 24, 195 - 199.
Weger, U. W., Meier, B. P., Robinson, M. D., & Inhoff, A. W. (2007). Things are sounding up:
Affective influences on auditory tone perception. Psychonomic Bulletin & Review, 14(3),
517521.
Whittlesea, B. W. A. (1993). Illusions of familiarity. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 19, 12351253.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 22
Whittlesea, B. W. A., Jacoby, L. L., & Girard, K. (1990). Illusions of immediate memory:
Evidence of an attributional basis for feelings of familiarity and perceptual quality. Journal
of Memory and Language, 29, 716732.
Winkielman, P., & Cacioppo, J. T. (2001). Mind at ease puts a smile on the face:
psychophysiological evidence that processing facilitation elicits positive affect. Journal of
Personality and Social Psychology, 81(6), 989.
Winkielman, P., Halberstadt, J., Fazendeiro, T., & Catty, S. (2006). Prototypes are attractive
because they are easy on the mind. Psychological Science, 17(9), 799-806.
Winkielman, P., Schwarz, N., Fazendeiro, T., & Reber, R. (2003). The hedonic marking of
processing fluency: Implications for evaluative judgment. The Psychology of Evaluation:
Affective Processes in Cognition and Emotion, 189-217.
Wu, X., Jia, H., Wang, E., Du, C., Wu, X., & Dang, C. (2016). Vertical position of Chinese power
words influences power judgments: Evidence from spatial compatibility task and event-
related Potentials. International Journal of Psychophysiology, 102, 55-61.
Appendix A: Single Paper Meta-Analysis
To calculate more precise effect size estimates, we conducted a single-paper meta-
analysis using the Comprehensive Meta-Analysis Software (Version 3.0; Borenstein, Hedges,
Higgins, & Rothstein, 2013). Figure A1 shows a forest plot of the observed and estimated raw
differences between the proportion of congruent arrangements chosen and chance (0.50) for each
metaphor and language group. For the past-before-future metaphor, shown in the left-hand panel,
we coded past-on-the-left as metaphor-congruent for English speakers and past-on-the-right as
metaphor-congruent for Farsi speakers. For the good-is-up metaphor, shown in the right-hand
panel, we coded the happy-above-sad arrangement as metaphor-congruent for English and Farsi
speakers. Due to the small number of studies, tau-squared was pooled across studies, following
recommendations by Borenstein, Hedges, Higgins, and Rothstein (2010). A random effects
model was used and effect sizes were fixed across subgroups. Effect sizes were corrected for
small sample biases (Borenstein et al., 2010).
For the time metaphor, the overall raw effect size was 0.169, 95% CI [0.139, 0.199] for
English speakers, and 0.009, 95% CI [-0.024, 0.041] for Farsi speakers. The overall effect size
across the two language groups was 0.094, 95% CI [0.072, 0.116]. For the valence metaphor, the
overall effect size was 0.086, 95% CI [0.053, 0.118] for English speakers, and 0.124, 95% CI
[0.088, 0.160] for Farsi speakers. The overall effect size across both language groups was 0.103,
95% CI [0.079, 0.127].
In addition, an analysis was run for English speakers (Experiment 1-3a) to examine if the
question wordings “Which arrangement do you think is better?”(Experiment 1 and 3a) vs.
“Which arrangement do you like more?” (Experiment 2a and 2b) lead to different patterns of
results. There was no significant difference in the patterns of results between the two wordings,
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 23
for both time- and valence-related conditions, Q (1) = 0.061, p = .804, and Q (1) = 2.407, p
= .121, respectively.
Figure A1. Results from the single-paper meta-analysis. The x-axis represents the raw
differences between the proportion of congruent arrangements and chance (0.50). Note that for
the time metaphor, the congruent arrangement is past-on-the-the-left for English speakers and
past-on-the-right for Farsi speakers. The effect sizes of each individual study are represented by
the diamond shape and the total effect sizes are represented by the triangle shape.
Appendix B: Results of Exploratory Conditions from Experiment 1-4
Table B1. Results of Exploratory Conditions
Experiment
Mean
95% CI
t
p
Past-on-top
1
0.625
[.577, .673]
5.102
<0.001
2a
0.684
[.616, .753]
5.276
<0.001
2b
0.616
[.551, .680]
3.507
0.001
3a
0.642
[.571, .712]
3.943
<0.001
3b
0.544
[.479, .609]
1.327
0.188
4
0.545
[.510, .580]
2.500
0.013
Happy-on-the-left
1
0.548
[.496, .599]
1.818
0.071
2a
0.556
[.470, .643]
1.271
0.207
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 24
2b
0.469
[.394, .544]
-0.815
0.418
3a
0.584
[.513, .655]
2.331
0.022
3b
0.453
[.383, .523]
-1.312
0.193
4
0.344
[.303, .386]
-7.360
<0.001
Note: The mean represents the mean proportion of arrangement chosen.
As table B1 shows (see also figure B1 and B2), participants consistently showed a
preference for past-on-top arrangements across the experiments. This may reflect the “past
before future” metaphor as both English and Farsi speakers read from top to bottom, despite the
horizontal direction of reading and writing of the two languages being different (i.e., English
speakers read from left to right, top to bottom, whereas Farsi speakers read from right to left, top
to bottom).
The preference for happy-on-the-left arrangements was inconsistent across experiments.
In experiment 1 and experiment 3a, English speaking participants preferred the happy-on-the-left
arrangement, potentially reflecting the fact that the sequence “happy and sad” is more common
“sad and happy” in the English language (820 vs. 196 occurrences, respectively, according to the
News on the Web corpus: https://www.english-corpora.org/now/). This pattern was reversed for
Farsi speakers, although inconsistently. Future research may fruitfully investigate whether
linguistic sequencing plays a role in aesthetic preference.
Figure B1. Results of exploratory conditions in Experiments 1, 2a, and 2b. The error bars
represent 95% confidence intervals.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 25
Figure B2. Results of exploratory conditions in Experiments 1, 2a, and 2b. The error bars
represent 95% confidence intervals.
Appendix C: Trial Analysis of Experiments 1, 2a&b, and 3a
Given our use of a choice task to assess aesthetic preference, some may wonder whether
our effects depended on whether participants developed insight into the purpose of the
experiments as they see more juxtapositions of photos. To test this, trial analyses were conducted
for Experiments 1, 2a&b, and 3a and the results revealed that the likelihood of preferring
metaphor-congruent arrangements did not change from trial to trial as participants progressed
further in the study.
Given Experiment 2a and 2b used identical stimuli and procedures, we combined
Experiment 2a and 2b for the trial analysis. In each experiment, there were a total of four target
trials per participant for each metaphor. The proportion of participants that had chosen the
congruent arrangement on each trial were calculated and the detailed descriptive statistics are
shown in Table C1. Figure C1 and C2 show that the preference for metaphor-congruent
arrangements did not vary from trial to trial for the valence-verticality (rs = -0.32, p = 0.68 for
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 26
Exp.1; rs = 0, p > 0.999 for Exp. 2a&b; rs = -0.20, p = 0.92 for Exp. 3a) or the time-space
metaphor (rs = -0.40, p = 0.75 for Exp.1; rs = 0, p > 0.999 for Exp. 2a&b; rs = 0.40, p = 0.75 for
Exp. 3a). Further logistic regressions with trial number as a predictor and preference for the
metaphor-congruent arrangement as the outcome also revealed no significant relationship
between the two variables for either metaphor in all of the experiments. See Table C2 for a
detailed description of the results from the logistic regressions.
In sum, these results suggest that people prefer metaphor-congruent arrangements
throughout trials for both metaphors, and the size of the effects did not change as participants
became accustomed to the within-subject manipulations.
As noted in the general discussion, we also asked participants what they thought about
the studies in an open-ended format. Not a single participant mentioned anything containing
“metaphor” or “valence”, and only a few mentioned concepts related to “chronology” (7.9%,
4.7%, 1.5%, 5.7% in Exp.1, Exp. 2a, Exp. 2b, and Exp. 3a respectively). This suggests that
participants had very limited insight into the hypotheses tested and renders concerns about
demand effects mute.
Figure C1. Mean proportion of congruent arrangements chosen for valence (happy-above-sad) by
trial. The fitted lines represent the relationship between trial number and proportion of
congruent arrangements chosen (green, red, and purple for Exp.1, Exp. 2a&b, and Exp.3a,
respectively).. The shaded areas surrounding the fitted lines represent 95% CIs. R indicates
spearman correlation coefficients, followed by their respective p values.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 27
Figure C1. Mean proportion of congruent arrangements chosen for time (past-on-the-left) by
trial. The fitted lines represent the relationship between trial number and proportion of congruent
arrangements chosen (green, red, and purple for Exp.1, Exp. 2a&b, and Exp.3a, respectively)..
The shaded areas surrounding the fitted lines represent 95% CIs. R indicates spearman
correlation coefficients, followed by their respective p values.
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 28
Table C1. Proportion of Congruent Arrangements Chosen by Trial.
Metaphor
Exp.
Trial
n
Mean
SD
Valence
1
1
162
0.617
0.488
1
2
162
0.617
0.488
1
3
162
0.630
0.484
1
4
162
0.586
0.494
2ab
1
160
0.550
0.499
2ab
2
160
0.538
0.500
2ab
3
160
0.575
0.496
2ab
4
160
0.544
0.500
3a
1
83
0.614
0.490
3a
2
83
0.554
0.500
3a
3
83
0.566
0.499
3a
4
83
0.602
0.492
Time
1
1
162
0.691
0.463
1
2
162
0.648
0.479
1
3
162
0.630
0.484
1
4
162
0.685
0.466
2ab
1
160
0.675
0.470
2ab
2
160
0.650
0.478
2ab
3
160
0.681
0.467
2ab
4
160
0.656
0.476
METAPHOR, FLUENCY, AND AESTHETIC PREFERENCE 29
Table C2. Results from Logistic Regression with Trial Number as a Predictor.
Metaphor
Exp.
β (SE)
95% CI for OR
Lower
OR
Upper
Valence
1
Constant
0.543 (0.198)**
Trial
-0.034 (0.072)
0.839
0.967
1.114
2ab
Constant
0.188(0.195)
Trial
0.008(0.071)
0.876
1.008
1.158
3a
Constant
0.365(0.273)
Trial
-0.010(0.100)
0.814
0.99
1.204
Time
1
Constant
0.721(0.204)***
Trial
-0.017(0.074)
0.85
0.984
1.138
2ab
Constant
0.717(0.206)***
Trial
-0.011(0.075)
0.854
0.989
1.145
3a
Constant
0.534(0.287)
Trial
0.107(0.107)
0.904
1.113
1.373
Note. ** indicates p < .01; *** indicates p < .001.
3a
1
83
0.614
0.490
3a
2
83
0.759
0.430
3a
3
83
0.663
0.476
3a
4
83
0.723
0.450
... In Japan, at least 20 billion yen per year were considered to be avoidable losses two decades ago if medicines as a target of the circular economy (Viegas et al., 2019;Alshemari et al., 2020) and sharing (Viegas et al., 2021) or even informal economies running in the public sector locally (Viegas et al., 2022). This new perspective opens space to critically depict the economies of returns -Circular Economies (CE) and Sharing/Collaborative Economies (SCE) -as metaphors for medicines reuse, since the idea of a metaphor is taken as a form of representing the world through shifting the interpretive status (Tseng and Chuang, 2022;Zhang et al., 2022). ...
Article
This paper provides a literature review on medicines reuse explaining how this phenomenon has evolved and has being re-interpreted in academic research. Literature review in combination with a descriptive and a qualitative thematic analysis was employed to answer two questions: (i) How the medicines returns phenomenon has been documented and discussed? (ii) To what extent the critical literature on circular and sharing/collaborative economies can be taken as metaphors for the advancement of knowledge to better understand the complexities of medicines returns? After the selection of 125 papers, 10 hubs of collaboration were identified in this research field with early work on reuse stemming from the UK. From the thematic analysis, three outcomes emerged. The first relates to the problem of wastage versus costs and affordability, and the role of stockpiles and misplacement of leftover medicines. The second arises from a chronological description of the selected studies. Using a timeline, recurring debates were identified: prescription costs/sharing; reuse in evolutionary perspective of rational use; affordability and avoidance of wastes; and the role of pharmacists and care in the community/solidary pharmacies. The third outcome delivers metaphors linking critical aspects of circular and sharing/collaborative economies with concepts and situations of medicines reuse. It was found that the circular economy, which is still far from consumer reality, offers a limited contribution as a metaphor for medicines reuse, while sharing/collaborative economy reflects current practices that can be understood as access giving, thus interpreted as a form of affordability.
... Similar to the finding that perceiving color word and color stimuli can enhance women's beauty judgments on men (Pazda & Elliot, 2017;Roberts et al., 2010), the present study demonstrated that such metaphorical associations can extend to and have a further effect on aesthetic perception and judgment of abstract oracle bone scripts, providing further evidence for the overlapping representation between abstract language and concrete perceptual stimuli. We can interpret this match metaphorical association may be an enhanced processing fluency and has an impact on the aesthetic appraisal (Zhang et al., 2022). ...
Article
Full-text available
Ancient Chinese characters were produced using two systems, one were produced according to man-made conventions to convey abstract meanings (oracle bone script), and another were generated by outlining the shape of object (pictograph). Whether these two kinds of ancient Chinese characters elicit different aesthetic appraisals of font structure depend on the referential meaning or object? And whether this aesthetic appraisal vary by color metaphors? In this present study, participants recruited from the Han, Bai, and Yi ethnic groups were unacquainted with the ancient Chinese characters. Experiment 1a, 1b, and 1c used abstract oracle bone scripts as materials, and Experiment 2a, 2b, and 2c used concrete pictographs as materials. Across three Chinese ethnic groups, the results showed the font structure of oracle bone script or pictograph was more likely to be judged as beautiful when its referential meaning was positive or its referential object was beautiful. Whereas, when its referential meaning was negative or its referential object was ugly, the font structure was more likely to be judged as ugly. Moreover, beauty judgment was facilitated when a positive oracle bone script was presented in the ethnic preferred color, and ugliness judgment was facilitated in the ethnic disliked color. However, this modulation effect was not obvious in the aesthetic appraisal of pictograph. The results indicated that aesthetic appraisal of ancient Chinese characters may depend on the referential meaning or object, and this sense of beauty or ugliness is also modulated by the color preference and metaphors in different Chinese ethnic cultures.
Article
Full-text available
People use space (e.g., left–right, up-down) to think about a variety of non-spatial concepts like time, number, similarity, and emotional valence. These spatial metaphors can be used to inform the design of user interfaces, which visualize many of these concepts in space. Traditionally, researchers have relied on patterns in language to discover habits of metaphorical thinking. However, advances in cognitive science have revealed that many spatial metaphors remain unspoken, shaping people’s preferences, memories, and actions independent of language – and even in contradiction to language. Here we argue that cognitive science can impact our everyday lives by informing the design of physical and digital objects via the spatial metaphors in people’s minds. We propose a simple principle for predicting which spatial metaphors organize people’s non-spatial concepts based on the structure of their linguistic, cultural, and bodily experiences. By leveraging the latent metaphorical structure of people’s minds, we can design objects and interfaces that help people think.
Article
Full-text available
We attempted to evaluate the replicability and generalizability of ten influential research on sensory marketing by conducting a high-powered and pre-registered replication in online settings in non-WEIRD consumers. The results revealed that only 20% of findings could be successfully replicated, and their effect sizes are as half as the original ones. Two successful studies had relatively larger sample sizes, used sound symbolism, and employed within-participants manipulation of senses. No studies involving visual factors, between-participant manipulation of senses, or interactions between factors could be replicated. Our findings reveal an initial estimate of the replicability and generalizability of sensory marketing.
Article
Full-text available
Investigations of non-Western art forms and non-Western subjects are needed to understand cross-cultural empirical aesthetics. This study examined the aesthetic evaluation of Chinese calligraphy, a distinguished visual art in East Asia, through people’s preferences among five scripts. These scripts are representative of major artistic writing styles of Chinese characters. We conducted an experiment with Chinese (mean age = 22.64, SD = 2.90) and non-Chinese participants (International students from Asia, Africa, and Northern America; mean age = 26.76, SD = 2.35) and investigated through subjective measures and eye-tracking measures. We adopted Pleasure-Interest Model of Aesthetic Liking (PIA Model) and measured fluency processing, affective feelings, motivational evaluation, and overall likeness as subjective measures to capture the aesthetic evaluation. The two groups evaluated the scripts differently and showed distinctive preferences for scripts. Chinese participants perceived the stimuli to be more fluent than non-Chinese participants. Chinese rated cursive script most fluent, pleasant, interesting, and liked it most, whereas non-Chinese liked the official script the most and rated it most interesting. Regarding eye movement, Chinese participants had fewer fixations and shorter total fixation durations, relative to non-Chinese participants. Non-Chinese participants were more sensitive to the features of the scripts compared with Chinese participants. We suggest that culture could affect aesthetic evaluation of Chinese calligraphy. Experience of Chinese characters and aesthetic expectations of different cultures may contribute to the evaluation differences.
Article
Full-text available
Thinking is accompanied by metacognitive experiences of ease or difficulty. People draw on these experiences as a source of information that can complement or challenge the implications of declarative information. We conceptualize the operation of metacognitive experiences within the framework of feelings-as-information theory and review their implications for judgments relevant to consumer behavior, including popularity, trust, risk, truth, and beauty.
Chapter
Full-text available
Chapter
Full-text available
To evaluate whether a claim is likely to be true, people attend to whether it is compatible with other things they know, internally consistent and plausible, supported by evidence, accepted by others, and offered by a credible source. Each criterion can be evaluated by drawing on relevant details (an effortful analytic strategy) or by attending to the ease with which the claim can be processed (a less effortful intuitive strategy). Easy processing favors acceptance under all criteria – when thoughts flow smoothly, people nod along. Ease of processing is also central to aesthetic appeal and easily processed materials are evaluated as prettier. This sheds new light on why beauty and truth are often seen as related, by poets and scientists alike. Because people are more sensitive to their feelings than to where their feelings come from, numerous incidental variables can influence perceived beauty and truth by influencing the perceiver’s processing experience.
Chapter
Full-text available
Review of research into the consequences of easy vs difficult processing; traces fluency effects to the hedonic quality of the processing experience.
Article
The affect system, in its position to monitor organismic-environmental transactions, may be sensitive to the internal dynamics of information processing. Hence, the authors predicted that facilitation of stimulus processing should elicit a brief, mild, positive affective response. In 2 studies, participants watched a series of neutral pictures while the processing ease was unobtrusively manipulated. Affective reactions were assessed with facial electromyography (EMG). In both studies, easy-to-process pictures elicited higher activity over the region of zygomaticus major, indicating positive affect. The EMG data were paralleled by self-reports of positive responses to the facilitated stimuli. The findings suggest a close link between processing dynamics and affect and may help understand several preference phenomena, including the mere-exposure effect. The findings also highlight a potential source of affective biases in social judgments.
Book
The now-classic Metaphors We Live By changed our understanding of metaphor and its role in language and the mind. Metaphor, the authors explain, is a fundamental mechanism of mind, one that allows us to use what we know about our physical and social experience to provide understanding of countless other subjects. Because such metaphors structure our most basic understandings of our experience, they are "metaphors we live by"--metaphors that can shape our perceptions and actions without our ever noticing them. In this updated edition of Lakoff and Johnson's influential book, the authors supply an afterword surveying how their theory of metaphor has developed within the cognitive sciences to become central to the contemporary understanding of how we think and how we express our thoughts in language.
Article
The present study used event-related potentials (ERPs) to explore the influence of vertical position on power judgments. Participants were asked to identify whether a Chinese word represented a powerful or powerless group (e.g., “king” or “servant”), which was presented in the top or bottom of the screen. The behavioral analysis showed that judging the power of powerful words were significantly faster when they were presented at the top position, compared with when they were presented at the bottom position. The ERP analysis showed enhanced N1 amplitude for congruent trials (i.e., the powerful words in the top and the powerless words in the bottom of the screen) and larger P300 and LPC amplitude for incongruent trials (i.e., the powerful words in the bottom and the powerless words in the top of the screen). The present findings provide further electrophysiological evidence that thinking about power can automatically activate the underlying spatial up-down (verticality) image schema and that the influence of vertical position on the power judgments not only occurs at the early perceptual stage of power word processing, but also at the higher cognitive stage (i.e., allocation of attention resources, conflict solving and response selection). This study revealed the neural underpinnings of metaphor congruent effect which have great significance to our understanding of the abstract concept power.