Elsevier

Brain and Cognition

Volume 53, Issue 3, December 2003, Pages 452-463
Brain and Cognition

The valence-specific laterality effect in free viewing conditions: The influence of sex, handedness, and response bias

https://doi.org/10.1016/S0278-2626(03)00217-3Get rights and content

Abstract

The right hemisphere has often been viewed as having a dominant role in the processing of emotional information. Other evidence indicates that both hemispheres process emotional information but their involvement is valence specific, with the right hemisphere dealing with negative emotions and the left hemisphere preferentially processing positive emotions. This has been found under both restricted (Reuter-Lorenz & Davidson, 1981) and free viewing conditions (Jansari, Tranel, & Adolphs, 2000). It remains unclear whether the valence-specific laterality effect is also sex specific or is influenced by the handedness of participants. To explore this issue we repeated Jansari et al.’s free-viewing laterality task with 78 participants. We found a valence-specific laterality effect in women but not men, with women discriminating negative emotional expressions more accurately when the face was presented on the left-hand side and discriminating positive emotions more accurately when those faces were presented on the right-hand side. These results indicate that under free viewing conditions women are more lateralised for the processing of facial emotion than are men. Handedness did not affect the lateralised processing of facial emotion. Finally, participants demonstrated a response bias on control trials, where facial emotion did not differ between the faces. Participants selected the left-hand side more frequently when they believed the expression was negative and the right-hand side more frequently when they believed the expression was positive. This response bias can cause a spurious valence-specific laterality effect which might have contributed to the conflicting findings within the literature.

Introduction

Many studies have suggested that the right hemisphere (RH) of the brain has a greater role in the processing of emotional information than the left hemisphere (Christman & Hackworth, 1993; Levy, Heller, Banich, & Burton, 1983; Ley & Bryden, 1979). This research has used both brain-damaged and healthy participants and has examined the role of the RH in both the perception and expression of affect. For example, RH-damaged patients are more impaired than LH-damaged patients at recognising emotions conveyed by facial expressions (Adolphs, Damasio, Tranel, & Damasio, 1996). In addition, it is believed that emotions are expressed more intensely on the left side of the face because the RH has a greater role in the control of emotional expression (Heller & Levy, 1981; Kowner, 1995; Sackeim, Gur, & Saucy, 1978).

With healthy participants the majority of studies in this area have used the divided visual field technique to present faces with different emotional expressions. In accord with the findings from patient studies it has been found that there is a RH advantage (left visual field, LVF) in the perception and interpretation of emotional expressions (Ley & Bryden, 1979; Mandal & Singh, 1990; Stauss & Moscovitch, 1981). The view, however, that the RH is preferentially involved in the processing of all emotional information has been questioned on a number of occasions. Many studies have suggested that hemispheric biases in the processing of emotional information may depend on the valence of the emotion conveyed by that information (Ahern and Schwartz, 1979, Ahern and Schwartz, 1985; Davidson, 1992; Reuter-Lorenz & Davidson, 1981; Tucker, 1981). For example, Reuter-Lorenz and Davidson (1981) found that when happy faces were presented to the LH they were perceived more quickly compared with sad faces. Conversely sad faces presented to the RH were perceived faster than happy faces. They concluded that there was a valence-specific laterality effect, with a LH advantage for the perception of positive emotions and a RH advantage for the perception of negative emotions. Evidence in support of the valence hypothesis has been obtained from a range of other measures (Davidson, 1992, Davidson, 1993a, Davidson, 1993b) including the expression of facial affect, with the LH showing a greater involvement in the expression of positive emotions (Ross, Homan, & Buck, 1994), and the experience of emotions induced by films, with activation of anterior regions of each hemisphere being specific to the valence of the emotion experienced (Jones & Fox, 1992).

The validity of the valence hypothesis, however, continues to be a source of debate within the literature (Borod, Zgaljardic, Tabert, & Koff, 2001). Several early studies produced findings consistent with the right-hemisphere hypothesis rather than the valence hypothesis. For example, Ley and Bryden (1982), using a dichotic listening task, obtained a left ear advantage (RH) when participants were required to detect the affective tone of spoken sentences, regardless of whether the sentences had been spoken in a happy, sad, angry, or neutral tone of voice. Some studies have also suggested that both hemispheres are involved in the processing of positive emotions whereas the processing of negative emotions is lateralised in the RH (Asthana & Mandal, 2001; Mandal, Tandon, & Asthana, 1991). Moreover, a recent review of the literature (Borod et al., 2001) found little evidence for the valence-specific effect in facial emotion perception. Of the 20 studies reviewed, only one produced results consistent with the valence hypothesis and this was in women but not men (Burton & Levy, 1989). Of the 19 remaining studies, 17 produced results consistent with the right-hemisphere hypothesis and two showed no laterality effects.

It is important to note that two versions of the valence hypothesis have been proposed (Borod et al., 2001). One version suggests that the RH is specialised for negative emotions whereas the LH is specialised for positive emotions (e.g., Silberman & Weingartner, 1986). A second version makes the important distinction between the perception of an emotion and the experience of an emotion (Davidson, 1984). According to this theory the RH is dominant for the perception of both positive and negative emotions, but for emotional experience the RH is specialised for negative emotions and the LH for positive emotions. In accord with this view, Borod (1993) has proposed that posterior regions of the RH are specialised for the perceptual identification of an emotion regardless of the valence of the emotion. However, for emotional experience, it is suggested that anterior regions of the LH are preferentially involved in the experience of positive emotions whereas anterior regions of the RH are involved in the experience of negative emotions (Borod, 1993; Davidson, 1993a, Davidson, 1993b).

If valence-specific hemispheric asymmetries are present for emotional experience then valence effects may only emerge in tasks which cause participants to experience an emotion. The idea that valence effects may be dependent on the task used was suggested by Ley and Strauss (1986) who concluded from their review of the literature that evidence for the RH hypothesis came from tasks involving the rapid classification of emotions (e.g., Ley & Bryden, 1979; Suberi & McKeever, 1977) whereas evidence in favour of the valence-specific hypothesis came from studies where different stimuli were compared for affect (e.g., Reuter-Lorenz & Davidson, 1981; Reuter-Lorenz, Givis, & Moscovitch, 1983).

van Strien and van Beek (2000) made the same point and suggested that studies which have required participants to match emotional expressions of faces (e.g., Ley & Bryden, 1979) may show a RH advantage because these predominantly perceptual tasks are carried out by posterior regions of the RH. However, for tasks which require an evaluation of the intensity of an emotional expression, either directly (e.g., Davidson, Mednick, Moss, Saron, & Schaffer, 1987), or when identifying one face to be more expressive than another (e.g., Reuter-Lorenz & Davidson, 1981), then anterior regions of each hemisphere, involved in emotional experience, may be utilised. It is possible that participants, when evaluating subtle differences in the intensity of emotional expressions, use their own affective responses to facial expressions to aid them in reaching a decision (Jansari et al., 2000). This is plausible as it has been found that viewing an emotionally expressive face can elicit an experience of that emotion in the perceiver (Wild, Erb, & Bartels, 2001). Moreover, this may be one of the ways in which we decode the expressions of other people (Adolphs, 2002; Wild et al., 2001). If the perception of facial affect causes participants to experience the emotion expressed in the face, anterior regions of each hemisphere will be utilised, resulting in the emergence of valence-specific laterality effects (van Strien & van Beek, 2000).

If this account is correct then tasks involving a quick perceptual identification of facial affect are unlikely to produce valence-specific effects (van Strien & van Beek, 2000). However, if a task requires subtle gradations in emotional expression to be compared, perhaps causing participants to rely on the recruitment of their own affective responses to reach a decision (Jansari et al., 2000), then a valence-specific effect may be present. Partial support for this suggestion comes from a study conducted by Safer (1981) who found a LVF superiority when participants had to judge whether two facial emotions were the same or not. Prior to making the judgement, in one condition participants were instructed to think of a brief one or two-word label description of the expression, and in a second condition they were instructed to “empathize fully with the person in the slide, as if [they] were that person…” Importantly the instruction to empathise with the expressed emotion increased the RH advantage. This suggests that the lateralised perception of facial affect can change within the same task if emotional experience is recruited during the task.

It certainly seems to be the case that studies which have used tasks which require faces to be compared for affect are more likely to obtain valence-specific effects (Ley & Strauss, 1986). One such study by Jansari et al. (2000) provides further support for the valence-specific hypothesis. This experiment differed from previous work in that it examined laterality effects for emotional processing under free-viewing conditions. Thus, rather than the faces being presented very briefly the participants were able to examine the faces for as long as it took them to respond. The task was similar to that employed by Reuter-Lorenz and Davidson (1981), and consisted of presenting participants with two versions of the same male face, one of which displayed an emotion and one of which displayed a neutral expression. The faces were morphed so that differences in expression between the two faces ranged from very subtle to clear and they were presented on either side of a computer screen. In support of the valence hypothesis Jansari et al. (2000) found more accurate discrimination of faces with positive emotional expressions when they were presented on the right-hand side than when they were presented on the left-hand side. Conversely, greater accuracy in discrimination was found for faces with negative expressions when they were presented on the left-hand side rather than the right-hand side. These findings suggest that laterality effects in emotional processing are valence-specific and that they are present under free-viewing conditions.

It is also possible that evidence in support of the valence hypothesis has been inconsistent because most studies have ignored the influence of the participant’s gender. The majority of research which has examined differences in brain lateralisation between men and women has indicated that lateralisation of certain types of processing is more pronounced in men than in women. It appears that in men there is a RH advantage for non-verbal tasks and a LH advantage for verbal tasks, whereas women show less clear laterality effects on these tasks (Davidson, Cave, & Sellner, 2000; Hausmann & Gunturkun, 1999; Hellige, 1993; Iaccino, 1993; Inglis & Lawson, 1981).

Given that sex differences in lateralisation exist it is possible that hemispheric functioning differs between men and women for the processing of emotional stimuli (Eviatar, Hellige, & Zaidel, 1997; Sanz-Martin & Loyo, 2001). For example, Ladavas, Umiltà, and Ricci-Bitti (1980) found that women showed a LVF superiority in discriminating emotions whereas men showed no consistent asymmetries in emotional discrimination. In addition, Burton and Levy (1989) found that women, but not men, perceived faces with negative emotions fastest when presented in the LVF and faces with positive emotions fastest when presented to the RVF. These results indicate that women are more lateralised for the perception of emotion than are men.

Findings by van Strien and van Beek (2000) support this view. They examined the influence of sex and handedness on ratings of emotion in laterally presented cartoon faces. On each trial a face was presented briefly (150 ms) to the left or right visual field and participants were required to rate the intensity of the emotion expressed by the face. They found no effect of handedness but found that sex influenced ratings of emotion, with women rating neutral and mildly positive faces as more positive when they were presented in the RVF compared to when they were presented in the LVF. For the men the visual field did not affect how they rated the faces. These results, and those of Burton and Levy (1989), suggest that the valence hypothesis may depend upon the sex of the subject, with only women showing the valence-specific laterality effect. It is also possible that some of the conflicting results within the literature may have been caused by gender differences in emotional processing which went undetected because the effect of sex was not examined (e.g., Reuter-Lorenz & Davidson, 1981).

It should be noted, however, that reviews of the literature have found little evidence for sex differences in asymmetries of facial expression of emotion (Borod, Koff, Yecker, Santschi, & Schmidt, 1998) or the perception of facial affect (Borod et al., 2001). Moreover, van Strien and van Beek’s results conflict with those of Jansari et al. (2000) who found the valence-specific laterality effect in both male and female participants. Thus, the issue of whether the valence-specific laterality effect is sex specific is unresolved. One of the purposes of the present study was to determine whether the valence-specific effect applies only to female participants, when completing the same free-viewing task as used by Jansari et al. (2000), or is limited to restricted viewing conditions.

A further factor which may influence asymmetries in emotional processing is the handedness of participants. Differences in cortical organisation and processing between left- and right-handers have been reported and it is conceivable that these differences influence the lateralisation of emotional processing (Everhart, Harrison, & Crews, 1996). Many studies have avoided the handedness issue by selecting only right-handed participants. While findings have been inconsistent, there is some evidence to suggest that right- and left-handers may have different asymmetries for emotional processing. For example, Reuter-Lorenz et al. (1983) found that left-handers who did not show an inverted handwriting posture had an opposite pattern of valence asymmetry from right-handers and inverted left-handers. In addition, Everhart et al. (1996) showed that left-handers had a RVF bias for rating neutral expressions more negatively. This suggests that the processing of negative emotions is more strongly lateralised in the LH for left-handers. In contrast, however, van Strien and van Beek (2000) found no influence of handedness on the rating of emotional expressions as conveyed by cartoon faces. Thus, it is unclear whether handedness influences asymmetries in emotional processing. Moreover, it is likely that there are complex interactions between handedness, sex, and cortical organisation (Eviatar et al., 1997) which may only be clarified when these factors are studied in relation to one another.

To summarise, while some results indicate that the RH is predominantly involved in the processing of all emotions (Borod et al., 2001; Ley & Bryden, 1979), other evidence suggests that each hemisphere has a bias for processing emotions of a particular valence (Davidson, 1992, Davidson, 1993a, Davidson, 1993b; Jansari et al., 2000; Reuter-Lorenz & Davidson, 1981; van Strien & van Beek, 2000). It is also possible that there is a valence-specific laterality effect for women but not men (Burton & Levy, 1989; van Strien & van Beek, 2000), with women preferentially using the RH to process negative emotions and the LH to process positive emotions. Finally, the effect of handedness on the lateralised processing of emotional expressions has produced inconsistent findings, with some studies reporting an effect (Everhart et al., 1996; Reuter-Lorenz et al., 1983) and others reporting no effect (van Strien & van Beek, 2000).

In view of the contradictory findings within the literature the aim of the present study was to extend previous work and examine the effects of sex and handedness on the lateralised processing of emotional expressions. A primary purpose of the study was to determine whether the valence-specific laterality effect applies to both men and women in free viewing conditions using the task employed by Jansari et al. In addition, the influence of handedness on emotional processing was examined.

In accord with van Strien and van Beek’s results it was predicted that valence-specific effects would be restricted to female participants but not men. Moreover if valence effects emerge when a task requires subtle differences in emotional expression to be compared (van Strien & van Beek, 2000) then valence-specific effects may be largest when the discrimination of emotional expression is most difficult. As the effect of handedness on lateralised emotional processing has proved to be inconsistent we made no clear predictions regarding this factor.

Section snippets

Participants

Seventy-eight participants were used in this study. Thirty-five were right-handed (17 men and 18 women) and 43 were left-handed (19 men and 24 women). All participants were staff and students of the University of Abertay Dundee. All had normal colour vision and normal or corrected-to-normal visual acuity.

Hand preference

Hand preference was measured by a 27 item handedness questionnaire which consisted of 25 questions taken from Peters’ (1998) handedness inventory and a further two questions which enabled the

Results

A 2 × 2 × 2 × 2 × 5 [sex (male vs. female) handedness (left vs. right) side (left vs. right) valence (positive vs. negative) morphing (5, 10, 15, 20, 25%)] mixed-model ANOVA was conducted on the mean accuracy data. Sex and handedness were between-subject factors and side, valence, and morphing were within-subject factors.

Analyses revealed a significant main effect of valence, F(1,74)=79.66, p<.0001, with participants more accurately classifying positive emotions (mean=77%) compared to negative emotions

Control condition

For the control trials the two faces were identical but the emotion label differed and the participants were required to select the side that they thought the ‘emotional’ face was on. In this condition, the dependent variable was the percentage of times participants chose the left side (with the right side simply being its complement to 100%). As the faces were identical there should have been no systematic differences between the sides chosen, with participants selecting either side 50% of the

Discussion

The participants were more accurate at discriminating positive emotions than negative emotions. This replicates Jansari et al.’s (2000) results, and those of Hugdahl, Iverson, and Johnsen (1993), who also found participants to be more accurate at identifying positive emotions. This finding also corresponds to other work which suggests that some negative emotional expressions, such as sadness and fearfulness, may be more difficult to recognise than other emotions (Adolphs et al., 1996; Ekman &

Acknowledgements

We thank two anonymous reviewers for comments on an earlier version of the manuscript and Astrid Schepman for useful discussions.

References (60)

  • W. Heller et al.

    Perception and expression of emotion in right handers and left handers

    Neuropsychologia

    (1981)
  • K. Hugdahl et al.

    Laterality of facial expressions: Does the sex of the subject interact with the sex of the face?

    Cortex

    (1993)
  • N.A. Jones et al.

    Electroencephalogram asymmetry during emotionally evocative films and its relation to positive and negative affectivity

    Brain and Cognition

    (1992)
  • M. Kinsbourne

    The cerebral basis of lateral asymmetries in attention

    Acta Psychologica

    (1970)
  • R. Kowner

    Laterality in facial expressions and its effect on attributions of emotion and personality: A reconsideration

    Neuropsychologia

    (1995)
  • E. Ladavas et al.

    Evidence for sex differences in right-hemisphere dominance for emotions

    Neuropsychologia

    (1980)
  • R.G. Ley et al.

    Hemispheric differences in processing emotions in faces

    Brain and Language

    (1979)
  • R.G. Ley et al.

    A dissociation of right and left hemispheric effects for recognizing emotional tone and verbal content

    Brain and Cognition

    (1982)
  • M.K. Mandal et al.

    Right brain damage impairs recognition of negative emotions

    Cortex

    (1991)
  • R.C. Oldfield

    The assessment and analysis of handedness: The Edinburgh Inventory

    Neuropsychologia

    (1971)
  • P. Reuter-Lorenz et al.

    Differential contributions of the 2 cerebral hemispheres to the perception of happy and sad faces

    Neuropsychologia

    (1981)
  • P. Reuter-Lorenz et al.

    Hemispheric specialization and the perception of emotion: Evidence from right handers and from inverted and noninverted left handers

    Neuropsychologia

    (1983)
  • E.K. Silberman et al.

    Hemispheric lateralization of functions related to emotion

    Brain and Cognition

    (1986)
  • M. Suberi et al.

    Differential right hemispheric memory storage of emotional and non-emotional faces

    Neuropsychologia

    (1977)
  • J.W. van Strien et al.

    Ratings of emotion in laterally presented faces: Sex and handedness effects

    Brain and Cognition

    (2000)
  • B. Wild et al.

    Are emotions contagious? Evoked emotions while viewing emotionally expressive faces: Quality, quantity, time course and gender differences

    Psychiatry Research

    (2001)
  • R. Adolphs et al.

    Cortical systems for the recognition of emotion in facial expressions

    Journal of Neuroscience

    (1996)
  • R. Adolphs et al.

    A role for the human amygdalain recognizing emotional arousal from unpleasant stimuli

    Psychological Science

    (1999)
  • G.L. Ahern et al.

    Differential lateralization for positive versus negative emotions

    Neuropsychologia

    (1979)
  • M. Annett

    A classification of hand preference by association analysis

    British Journal of Psychology

    (1970)
  • Cited by (67)

    View all citing articles on Scopus
    View full text