Age-related decrease in recognition of emotional facial and prosodic expressions.

Abstract

The recognition of nonverbal emotional signals and the integration of multimodal emotional information are essential for successful social communication among humans of any age. Whereas prior studies of age dependency in the recognition of emotion often focused on either the prosodic or the facial aspect of nonverbal signals, our purpose was to create a more naturalistic setting by presenting dynamic stimuli under three experimental conditions: auditory, visual, and audiovisual. Eighty-four healthy participants (women = 44, men = 40; age range 20-70 years) were tested for their abilities to recognize emotions either mono- or bimodally on the basis of emotional (happy, alluring, angry, disgusted) and neutral nonverbal stimuli from voice and face. Additionally, we assessed visual and auditory acuity, working memory, verbal intelligence, and emotional intelligence to explore potential explanatory effects of these population parameters on the relationship between age and emotion recognition. Applying unbiased hit rates as performance measure, we analyzed data with linear regression analyses, t tests, and with mediation analyses. We found a linear, age-related decrease in emotion recognition independent of stimulus modality and emotional category. In contrast, the improvement in recognition rates associated with audiovisual integration of bimodal stimuli seems to be maintained over the life span. The reduction in emotion recognition ability at an older age could not be sufficiently explained by age-related decreases in hearing, vision, working memory, and verbal intelligence. These findings suggest alterations in social perception at a level of complexity beyond basic perceptional and cognitive abilities.

Age and Musical Expertise Influence Emotion Recognition in Music

“…Most literature is based on facial expressions (e.g., Calder et al, 2003;Ruffman et al, 2008), but there is evidence of decrements in other domains as well, such as body postures, speech prosody, and nonverbal vocalizations (e.g., Lambrecht et al, 2012;Laukka & Juslin, 2007;Lima, Garrett, & Castro, 2013;Mill et al, 2009;Paulmann et al, 2008;Ruffman et al, 2009;Ruffman et al, 2008). Decrements for sadness and fear are a fairly robust finding in this literature, both in the visual (e.g., Ruffman et al, 2008) and auditory modalities (e.g., Paulmann et al, 2008).…”

“…One possibility is that they result from age-related cognitive and sensory losses, in domains such as attention, memory, vision, and hearing (e.g., Fozard & Gordon-Salant, 2011;Hedden & Gabrieli, 2004;Salthouse, 2009). There is some evidence, though, that these factors are poor predictors of age effects for facial expressions, speech prosody, and body postures (Keightley, Winocur, Burianova, Hongwanishkul, & Grady, 2006;Lambrecht et al, 2012;Mitchell, 2007;Orbelo, Grim, Talbott, & Ross, 2005;Ruffman et al, 2008;Ryan, Murray, & Ruffman, 2010;Sullivan & Ruffman, 2004). Thus, more specific mechanisms may be involved.…”

“…Findings that age effects are smaller for the recognition of happiness as compared to negative emotions are often interpreted within this framework (Laukka & Juslin, 2007;Mill et al, 2009;Mitchell et al, 2011;Riediger, Voelkle, Ebner, & Lindenberger, 2011;Williams et al, 2006). However, the interaction between valence and age is not consistently obtained (e.g., Isaacowitz et al, 2007;Lambrecht et al, 2012;Ruffman et al, 2008). Furthermore, Cacioppo, Berntson, Bechara, Tranel, and Hawkley (2011) suggested that the positivity effect may stem not from controlled mechanisms but from automatic processes linked to deterioration in the amygdala, which would dampen responses to negative input.…”

“…For happiness and surprise, changes were much smaller, and for disgust there was a trend for age-related improvements. Modalities other than facial expressions are less studied, but there is evidence for age-related decrements in emotion recognition in lexical stimuli (Isaacowitz et al, 2007), body postures (Ruffman, Halberstadt, & Murray, 2009), visual scenes (St. Jacques, Dolcos, & Cabeza, 2008), and in dynamic auditory cues, namely speech prosody (e.g., Lambrecht et al, 2012;Mitchell, 2007;Mitchell, Kingston, & Bouças, 2011) and nonverbal vocalizations (Lima, Alves, Scott, & Castro, 2013). Although typically young adults are compared with individuals aged 60 and over (''extreme age group'' design; Isaacowitz & Stanley, 2011), a few studies using finer gradations in age revealed that decrements may proceed linearly with advancing age, with significant differences already in middle-age (e.g., Isaacowitz et al, 2007;Lambrecht et al, 2012;Mill, Allik, Realo, & Valk, 2009;Paulmann, Pell, & Kotz, 2008).…”

“…A body of research, mostly on facial expressions, shows that advancing age is associated with decreased accuracy in the recognition of some emotions (e.g., Calder et al, 2003;Isaacowitz et al, 2007;Lambrecht, Kreifelts, & Wildgruber, 2012;Orgeta, 2010;Williams et al, 2009). Ruffman, Henry, Livingstone, and Phillips (2008) reviewed 17 data sets of (still) facial expressions in a meta-analysis, and concluded that older adults, around 70 years of age, are less accurate than younger ones (around 24 years) in the recognition of fear, anger, and sadness.…”