Major depression is associated with impaired processing of emotion in music as well as in facial and vocal stimuli

Abstract

Background:

The processing of emotional stimuli is thought to be negatively biased in major depression. This study investigates this issue using musical, vocal and facial affective stimuli.

Methods:

23 depressed in-patients and 23 matched healthy controls were recruited. Affective information processing was assessed through musical, vocal and facial emotion recognition tasks. Depression, anxiety level and attention capacity were controlled.

Results:

The depressed participants demonstrated less accurate identification of emotions than the control group in all three sorts of emotion-recognition tasks. The depressed group also gave higher intensity ratings than the controls when scoring negative emotions, and they were more likely to attribute negative emotions to neutral voices and faces.

Limitations:

Our in-patient group might differ from the more general population of depressed adults. They were all taking anti-depressant medication, which may have had an influence on their emotional information processing.

Conclusions:

Major depression is associated with a general negative bias in the processing of emotional stimuli. Emotional processing impairment in depression is not confined to interpersonal stimuli (faces and voices), being also present in the ability to feel music accurately.

Introduction

Accurate affective information processing is a necessary skill when responding to environmental demands and social interactions (Boyatzis and Satyaprasad, 1994, Carton et al., 1999, Gross, 1998, Philippot and Feldman, 1990). A disrupted capacity for social functioning, which is characteristic of mood disorders, has been associated with impaired emotional information processing (Harmer et al., 2002).

The impact of depression on the processing of emotions in faces and voices has already been investigated and is briefly reviewed below. To the best of our knowledge, the effect of depression on the recognition of emotion in music has not previously been studied. As the recognition of emotion in music is not associated with interpersonal contexts (as are emotion in faces and voices), it could yield different results. Negative biases when decoding emotions in faces and voices could be due primarily to a cognitive bias (i.e. to the feeling that people are hostile and rejecting). If this is correct, there would be no reason for the negative bias to extend to music as well. However, if the negative bias reflects a general perception bias in the processing of emotion, similar results should be observed whatever the source of the emotion, be it interpersonal (faces and voices) or impersonal (music).

Depressed subjects misrecognize facial emotions in general compared to controls, and display a greater sensitivity to negative cues in particular (Feinberg et al., 1986, Zuroff and Colussy, 1986, Cooley and Norwicki, 1989, Persad and Polivy, 1993, Gur et al., 1992, Mikhailova et al., 1996, Surguladze et al., 2004, Weniger et al., 2004). However a few studies have not found this effect (Archer et al., 1992, Gaebel and Wöelwer, 1992, Bradley et al., 1997, Gotlib et al., 2004). A processing impairment for neutral faces has been observed in depressed patients, who are more likely to interpret neutral stimuli as negative (Gur et al., 1992, Bouhuys et al., 1995, Bouhuys et al., 1997, Hale et al., 1998, Bouhuys et al., 1999, Leppänen et al., 2004, Gollan et al., 2008). Findings about intensity evaluations are controversial: some suggest that severe depressive symptoms are associated with lower ratings of the perceived intensity of affective stimuli in general (Hale et al., 1998, Gur et al., 1992), whereas others indicate that depressed patients are inclined to perceive negative cues as more intense than healthy controls (Gollan et al., 2008).

Emotion recognition in voices has been studied less than emotion recognition in faces among depressed subjects. However, some behavioural data suggest that depressed patients recognize auditory stimuli displaying an emotional prosody less accurately than healthy controls. Depression is associated with difficulties in processing emotion in voices, the impairments being significantly related to cognitive deficits (Uekermann et al., 2008, Emerson et al., 1999).

At the neuro-anatomical level, neurobiological studies investigating the emotional processing of faces and voices suggest a key role for the amygdalar regions. Amygdala are involved in the processing of fearful, sad, happy and ambiguous facial stimuli (Adolphs et al., 1999, Morris et al., 1996, Blair et al., 1999, Breiter et al., 1996, Whalen, 1998). The processing of emotion in voices has also been linked to the functioning of the amygdala (Scott et al., 1997, Morris et al., 1999, Phillips et al., 1998), although not consistently so (Anderson and Phelps, 1998, Royet et al., 2000, Adolphs et al., 2001).

Facial and vocal affective information have thus been extensively studied in depression, but this is not true for the processing of emotion in music. A study by Bodner et al. (2007) suggests that patients with major depressive disorder react more intensely to sad music than to music portraying other emotions but did not explore recognition accuracy. This lack of studies is astonishing as music has long been considered an important emotional inductor in human beings (Mithen, 2005). In religious rituals, political gatherings and military marches, to cite but a few examples, music is used to synchronize the mood of the participants. Music is also an efficient research tool for studying emotions as it allows the laboratory induction of strong, reproducible, positive and negative emotions (Västjäll, 2002, Krumhansl, 1997, Gilet, 2008).

Furthermore, music is of great interest in clinical practice: it has been shown to improve cognitive abilities (Rauscher et al., 1993, Rauscher et al., 1995, Rauscher et al., 1997, Hetland, 2000), possibly by inducing a temporary mood change (Chabris, 1999, Waterhouse, 2006). Music-therapy has also been shown to be a useful tool in treating mood disorders (Gold et al., 2005). In particular, it has been shown to improve depressed people's moods (Maratos et al., 2008) and to lower their stress levels (Nakayama et al., 2009).

The goal of the present study is to investigate emotional recognition in music in depressed patients. As music is not directly linked to interpersonal communication, comparing a musical task with a facial and a vocal one will allow us to determine whether the impaired processing of emotional stimuli in depression is limited to interpersonal contexts. The results could therefore contribute to our understanding of the nature of the deficit involved in the processing of emotions in depression.