Elsevier

Acta Psychologica

Volume 163, January 2016, Pages 97-106
Acta Psychologica

Emotion biases voluntary vertical action only with visible cues

https://doi.org/10.1016/j.actpsy.2015.11.003Get rights and content

Highlights

  • The effect of emotion on free positioning was examined under interocular masking.

  • Participants freely placed a dot via joystick after viewing emotional images.

  • The dot was placed higher in the positive condition than the negative condition.

  • This effect occurred only when the emotional images were unmasked.

  • Conscious emotional information activates representations of vertical directions.

Abstract

Emotional information influences our bodily experiences according to the space–valence metaphor (positive/negative is up/down). In the present study, we examined whether visible and invisible emotional stimuli could also modulate voluntary action. After observing an emotional image (e.g., positive, neutral, or negative), participants used a joystick to arbitrarily position a dot stimulus in a display. The emotional image was either masked (masked condition) or not (unmasked condition) via a continuous flash suppression technique, i.e., dynamic interocular masking. We found that in the unmasked condition, the placed position of the dot was significantly higher after observing the positive image compared with the negative image, but this difference was not present in the masked condition. Our findings suggest that conscious emotional information is necessary for activating sensorimotor representations of vertical directions, and voluntary action is performed based on these activations.

Introduction

A baseball player assumes a victory pose (raising his fist over his head in triumph) after he/she hits a home run. A woman jumps for joy because she receives an unexpected gift from her sweetheart. Conversely, a boy's shoulders will sink when he learns that his close friend has to transfer to another school. These examples illustrate how emotion is reflected in the movement of our bodies. In particular, positive and negative emotions are associated with upward and downward actions, respectively.

Previous studies have established a relationship between emotion and vertical position. For example, when positive or negative words are presented at the top or bottom of a display, respectively, participants evaluate the valence of words more quickly than when the words are presented in the opposite position (Meier & Robinson, 2004). Casasanto (2009) demonstrated that participants tended to position objects representing positive abstract concepts closer to the top of a piece of paper. Moreover, the word “up” is evaluated more positively than the word “down” (Marmolejo-Ramos, Elosúa, Yamada, Hamm, & Noguchi, 2013). All of these phenomena are based on the space–valence metaphor, which is an associative link between representations of emotional valence and vertical position.

There are two classical models of the origin of the metaphor (Slepian, M. L., 2015, Slepian, M. L. and Ambady, N., 2014): Conceptual Metaphor Theory (CMT; Lakoff & Johnson, 1980) and the Perceptual Symbol System (PSS; e.g., Barsalou, 1999). In CMT, metaphor is regarded as the function that renders abstract concepts concrete. That is, we understand abstract concepts by translating them into well-understood concrete experiences. According to Lee and Schwarz (2012), although CMT tends to be misinterpreted as assuming only the effect of sensorimotor experience on conceptual processing because of the one-way mapping, a bi-directional effect (i.e., sensorimotor experience and conceptual processing influencing each other) could be assumed owing to the dynamic online interaction between sensorimotor experiences and psychological states. On the other hand, the PSS assumes that the mental representations of abstract concepts comprise multimodal sensory experiences associated bi-directionally with such concepts, and metaphors stem from these associations. Thus, an important premise of the PSS is that sensorimotor experience and conceptual activation are closely correlated. Together, these theories suggest that sensorimotor experience and abstract concepts associate bi-directionally and this association is shaped by the dynamic online interaction or co-occurrence of sensorimotor experiences and abstract concepts1.

In line with these theories, previous researches have indicated that bodily experience and motor action influence emotional processing according to the space–valence metaphor. For instance, one study showed that an upright posture induced more positive emotions compared to a slouched posture (Stepper & Strack, 1993). Another study revealed that positive or negative memories were more likely to be recalled while participants moved their arm upward or downward, respectively (Casasanto & Dijkstra, 2010). Similar findings of biased recall have been shown for illusory self-motion (Seno, Kawabe, Ito, & Sunaga, 2013): Positive memories were more frequently recollected during upward vection compared with downward vection. Seno et al. found that participants reported feeling more positive emotions after experiencing upward vection compared with downward vection and concluded that this recollection bias stemmed from the mood-congruency effect (Blaney, P. H., 1986, Bower, G. H., 1981). These studies indicate that upward (downward) bodily experiences induce positive (negative) emotions.

There are also several studies revealing that emotional processing can also modulate sensorimotor experience. Sasaki and colleagues demonstrated that positive sounds (e.g., a baby's laugh) promoted upward vection and inhibited downward vection (Sasaki, Seno, Yamada, & Miura, 2012). The authors concluded that this phenomenon was due to the association between the ‘upward’ direction of self-motion and the positive emotions induced by the semantic meanings of sounds. Moreover, another study reported that positive words were placed by participations at a higher position on a display compared to negative words (Marmolejo-Ramos et al., 2013), even though movements were made using a mouse and thus words were moved vertically by manipulating the mouse backward and forward. Additionally, saccade trajectories tend to deviate upward when participants view words with a positive valence (Gozli, Chow, Chasteen, & Pratt, 2013). Recently, Dudschig and colleagues demonstrated that emotional words affected the motor responses when the participants were required to evaluate the emotional valence of words (Dudschig, de la Vega, & Kaup, 2015). In Experiments 1 and 2 of their study, participants observed an emotional word with their fingers pressing a key on the middle of the keyboard. Then, participants were asked to press a key on the top or bottom of the keyboard as soon as possible depending on the color of the words. For example, when the font color was orange or blue, the participants released the middle key and pressed either the top or bottom key, respectively (i.e., they performed upward or downward arm movements). The results showed that emotional words modulated motor action in accordance with the space–valence metaphor: When the emotional valence of the word was positive, upward movements were faster, and when the emotional valence of the word was negative, downward movements were faster. In these studies, emotional stimuli were presented during a task (Dudschig, C., et al., 2015, Marmolejo-Ramos, F., et al., 2013, Sasaki, K., et al., 2012) or a rapid response was required immediately after emotional stimuli disappeared (Gozli et al., 2013). In contrast to such online processing, the present study examined whether emotional information could influence motor action when emotional stimuli were absent during the task and when rapid motor responses were not required (i.e., offline processing2).

In the present study, we examined the effect of emotional information on voluntary action. We employed a novel and simple cursor-positioning task, which is similar to the word-allocation task (Marmolejo-Ramos et al., 2013) and the word-priming task (Ansorge, Khalid, & König, 2013). In our task, participants were asked to observe an emotional image and then move a neutral object (a dot cursor) to an arbitrary position on a display using a joystick fixed on a vertical surface. Fixing the joystick on the vertical surface enabled us to directly examine the effect of emotional information on voluntary vertical actions, unlike Marmolejo-Ramos et al. Moreover, by using the neutral object during the task, we could completely eliminate the ongoing effect of emotional information during the task, and thus this is adequate for investigating the effect of emotional information on subsequent sensorimotor experience. If emotional information influences subsequent action in offline processing, then the dot cursor would be placed more in the upper or lower areas of the display after viewing positive or negative images, respectively. On the other hand, if emotional information does not influence motor action in offline processing, the placed position of the dot cursor will not significantly differ according to emotional valence.

There are some advantages of using the cursor-positioning task and visual images as emotional stimuli. We could rule out the possibility that the effect of emotional information on sensorimotor experience reflects concrete lexical association. Previous studies have repeatedly shown that emotional words influence sensorimotor experience (Brookshire, G., et al., 2010, Dudschig, C., et al., 2015, Gozli, D. G., et al., 2012, Gozli, D., et al., 2013, Marmolejo-Ramos, F., et al., 2013, Meier, B. P. and Robinson, M. D., 2004, Santiago, J., et al., 2012, Weger, U. W., et al., 2007, Xie, J., et al., 2015) or word classification (Ansorge et al., 2013). However, when linguistic stimuli are used as in these previous studies, it is not clear which of linguistic and non-linguistic association between vertical space and emotional valence is involved with tasks (see Lynott & Coventry, 2014). Using emotional images and the cursor-positioning task could solve this problem. Moreover, most of the previous studies measured reaction time until completing a task (Brookshire, G., et al., 2010, Dudschig, C., et al., 2015, Gozli, D. G., et al., 2012, Meier, B. P. and Robinson, M. D., 2004, Santiago, J., et al., 2012, Weger, U. W., et al., 2007, Ansorge, U., et al., 2013) and reaction time reflects processing fluency and efficiency. On the other hand, we examined the effect of emotional images on the bias of manual localization. Because participants moved the cursor by the joystick fixed on vertical surface, we could directly test the effect of emotional information on vertical and voluntary action.

We also examined whether invisible emotional images could modulate motor action. Although a previous study examined whether the emotional valence of an invisible word affected subsequent classification of spatial words (Ansorge et al., 2013), their results might reflect concrete lexical association between vertical space and emotional valence. Thus, the present study examined this issue with non-linguistic stimuli. By addressing this issue, we hoped to identify the stage of mental processing that contributes to the space-valence metaphor. To eliminate emotional information from awareness, we adopted a continuous flash suppression technique (Tsuchiya & Koch, 2005) in which an image stimulus presented to one eye is interocularly masked by a dynamic random pattern continuously presented to the other eye. If unconscious emotional information influences motor action according to the space–valence metaphor, then we would expect to observe emotional-valence modulation of the cursor positioning for both unmasked and masked emotional images.

Section snippets

Ethics statement

The ethics committees of Kyushu University approved the protocol of the present study. The experiment was conducted according to the guidelines established by the Helsinki Declaration. We obtained written informed consent from all participants.

Participants

Sixteen right-handed volunteers participated in the experiment. All were unaware of the purpose of the experiment and reported normal vision and motor function.

Apparatus

The experiment was conducted in a dark room. Stimuli were presented on a 22-inch CRT display

Experiment 2

The results of Experiment 1 demonstrated that emotional images influenced cursor positioning when the emotional images were visible; however, this modulation of cursor positioning disappeared when the emotional images were invisible. One may argue that, as the continuous flashes were presented only in the masked condition, the flashes may have attenuated the effect of the emotional information. Moreover, because the continuous flashes in Experiment 1 were composed of more squares than in

Experiment 3

In Experiment 2, the presentation characteristics of the continuous flashes differed between the masked and unmasked conditions. In the masked condition, the center of the continuous flashes corresponded to the fixation point of each eye (i.e., the participant saw continuous flashes presented at the center of the display). On the other hand, in the unmasked condition, the centers of the two groups of continuous flashes were above and below the horizontal sides of the square. Thus, it is

Experiment 4

In Experiments 1 and 2, participants placed the dot higher following positive images and lower following negative images, but only when the emotional images were visible (unmasked). One could argue that this modulation occurred because participants' attention was drawn to specific salient features in the images; for example, key features might be found in the lower or upper portions of the negative and positive images, respectively, and thus these salient features (and not emotional valance)

General discussion

The aim of the present study was to examine (a) whether emotional valence of a visual image could influence subsequent cursor positioning and (b) whether this effect requires conscious access to emotional information. The results of Experiments 1 and 2 demonstrated that emotional image can induce vertical positioning bias in accordance with the space–valence metaphor (Ansorge, U., et al., 2013, Casasanto, D., 2009, Casasanto, D. and Dijkstra, K., 2010, Gozli, D., et al., 2013, Marmolejo-Ramos,

Competing financial interests

The authors declare no competing financial interests.

Acknowledgments

The authors would like to thank Atsunori Ariga for his valuable suggestions on an earlier version of the manuscript. The present study was supported by a Grant-in-Aid for JSPS Fellows (#266025) given to K.S., a Grant-in-Aid for Challenging Exploratory Research (#26540067) given to Y.Y., a Grant-in-Aid for Scientific Research (S) (#15H05709) given to Y.Y., and Kyushu University Interdisciplinary Programs in Education and Projects in Research Development (#26806, #27822, and #26307) given to Y.Y.

References (61)

  • I. de la Vega et al.

    Keep your hands crossed: The valence-by-left/right interaction is related to hand, not side, in an incongruent hand-response key assignment

    Acta Psychologica

    (2013)
  • Y. Yamada et al.

    Emotion colors time perception unconsciously

    Consciousness and Cognition

    (2011)
  • Y. Yamada et al.

    One's own name distorts visual space

    Neuroscience Letters

    (2012)
  • Y. Yamada et al.

    Scents boost preference for novel fruits

    Appetite

    (2014)
  • U. Ansorge et al.

    Space–valence priming with subliminal and supraliminal words

    Frontiers in Psychology

    (2013)
  • L.W. Barsalou

    Perceptual symbol systems

    Behavioral and Brain Sciences

    (1999)
  • R. Blake et al.

    Visual competition

    Nature Reviews. Neuroscience

    (2002)
  • P.H. Blaney

    Affect and memory: a review

    Psychological Bulletin

    (1986)
  • G.H. Bower

    Mood and memory

    American Psychology

    (1981)
  • D.H. Brainard

    The psychophysics toolbox

    Spatial Vision

    (1997)
  • G. Brookshire et al.

    Modulation of motor-meaning congruity effects for valenced words

  • D. Casasanto

    Embodiment of abstract concepts: good and bad in right- and left-handers

    Journal of Experimental Psychology: General

    (2009)
  • D. Casasanto et al.

    When left is “right”. Motor fluency shapes abstract concepts

    Psychological Science

    (2011)
  • C.B. Cave et al.

    Binocular rivalry disrupts visual priming

    Psychological Science

    (1998)
  • L.E. Crawford et al.

    “Good is up” is not always better: A memory advantage for words in metaphor-incompatible locations

    PLoS ONE

    (2014)
  • L.E. Crawford et al.

    Affect biases memory of location: Evidence for the spatial representation of affect

    Cognition & Emotion

    (2006)
  • L. Damjanovic et al.

    Contrasting vertical and horizontal representations of affect in emotional visual search

    Psychonomic Bulletin and Review

    (2015)
  • C. Diemand-Yauman et al.

    Fortune favors the bold (and the italicized): Effects of disfluency on educational outcomes

    Cognition

    (2011)
  • C. Eliasmith

    How to build a brain: A neural architecture for biological cognition

    (2013)
  • D.G. Gozli et al.

    The cost and benefit of implicit spatial cues for visual attention

    Journal of Experimental Psychology: General

    (2012)
  • Cited by (11)

    • Upward action promotes selective attention to negative words

      2021, Heliyon
      Citation Excerpt :

      Second, the apparatus in the present study might not have been optimized. We referred to previous studies (Kato et al., 2018; Sasaki et al., 2016) to develop the present experimental settings. However, it is unclear whether the amount or intensity of action was appropriate.

    • Exocentric coding of the mapping between valence and regions of space: Implications for embodied cognition

      2021, Acta Psychologica
      Citation Excerpt :

      The relationship between conceptual processing and regions of space has generated a considerable amount of research interest (Globig et al., 2019; Meier & Robinson, 2004; Sasaki et al., 2016).

    • On the localization of tastes and tasty products in 2D space

      2019, Food Quality and Preference
      Citation Excerpt :

      Based on the transitivity hypothesis of crossmodal correspondences, one may expect that if a dimension A in one modality (taste quality) corresponds with a dimension B in another modality (auditory pitch), and dimension B corresponds with dimension C in a third modality (elevation in space, which is not strictly speaking referring to another modality but rather another perhaps, amodal, dimension), people will link dimensions A and C in a predictable manner (see Deroy, Crisinel, & Spence, 2013, for a review). On the other hand, the research also shows that people represent affective stimuli spatially, in particular, along the vertical plane, such that positively-valenced objects and words tend to be located higher relative to their negatively-valenced counterparts (Damjanovic & Santiago, 2016; Machiels & Orth, 2017; Marmolejo-Ramos, Elosúa, Yamada, Hamm, & Noguchi, 2013, 2017; Meier & Robinson, 2004; Sasaki, Yamada, & Miura, 2016). Moreover, in a recent study, Marmolejo-Ramos, Tirado, Arshamian, Vélez, and Arshamian (2018) provided some evidence for the idea that in terms of 3D space, participants consistently place positive (negative) concepts in higher (lower) locations and nearer to (farther from) their body (see also Piqueras-Fiszman, Kraus, & Spence, 2014) with no preferential associations with locations in the horizontal plane (though note that previous research has, on occasion, documented a weaker pitch/left-right association, see Mudd, 1963).

    View all citing articles on Scopus
    View full text