Explicit identification and implicit recognition of facial emotions: I. Age effects in males and females across 10 decades

Abstract

A number of psychiatric and neurological disorders are characterized by impairments in facial emotion recognition. Recognition of individual emotions has implicated limbic, basal ganglionic, and frontal brain regions. Since these regions are also implicated in age-related decline and sex differences in emotion processing, an understanding of normative variation is important for assessing deficits in clinical groups. An internet-based test (“WebNeuro”) was administered to 1,000 healthy participants (6 to 91 years, 53% female) to assess explicit identification of basic expressions of emotion (happiness, sadness, fear, anger, disgust, neutral). A subsequent implicit recognition condition was based on a priming protocol, in which explicit identification provided the “study” phase. Responses were most accurate for happiness and slowest for fear in the explicit condition, but least accurate for happiness and fastest for fear in the implicit condition. The effects of age, by contrast, showed a similar pattern for both explicit and implicit conditions, following a nonlinear distribution in which performance improved from childhood through adolescence and early adulthood and declined in later adulthood. Females were better than males at explicit identification of fear in particular. These findings are consistent with the priority of threat-related signals, but indicate opposing biases depending on whether emotion processing is conscious or nonconscious. The lifespan trends in emotion processing over 10 decades point to an interaction of brain-based (maturation, stability, and then atrophy of cortical and subcortical systems) and experiential contributing factors. These findings provide a robust normative platform for assessing clinical groups.

Keywords:

• Facial emotion identification
• Facial emotion recognition
• Explicit processing
• Implicit processing
• Social cognition
• Age differences
• Sex differences
• 10 decades

This project is supported by an Australian Research Council (ARC) grant (DP0452237). L.M.W. is supported by an independent, competitive, and peer-reviewed Pfizer Senior Research fellowship. We acknowledge the support of the Brain Resource International Database (under the auspices of the Brain Resource, BR; www.brainresource.com) for use of data. We also thank the individuals who gave their time to participate in the database. Access to the database for scientific purposes is overseen by a scientific network (BRAINnet; www.brainnet.org.au), which is coordinated independently of the operations of BR. The facial stimuli and tasks were developed with support from National Institute of Mental Health (NIMH) Grant MH-60722 (R.C.G.).

Declaration of interest. L.M.W. is a small equity holder in the Brain Resource (BR) and has received fees from BR for tasks unrelated to this study. E.G. is the CEO of BR and holds significant equity and stock options in the company. However, scientific decisions about access to the Brain Resource International Database for scientific purposes are made by an independently administered scientific network of scientists; BRAINnet (www.brainnet.net). R.C.G. is on the Advisory Board of the BR, and R.E.G. is a consultant to this board.

Notes

¹Similarly, correct identification of happiness showed the least variability of RT, followed by neutral, sadness, anger, disgust, and fear as the most variable (Figure 1), confirmed in significant contrasts (p < .05) and suggesting that variability of RT varies in a consistent manner with mean RT.

²The variability of RT for implicit recognition also followed the trends for mean RT. Variability was least for fear, followed by disgust, sadness, anger, happiness, and neutral as the most variable. Pairwise contrasts were significant (p < .05) with the exception of happiness versus anger and neutral.

³Variability of RT followed the same pattern over age as mean RT, with variability least in middle decades. Given this correspondence, variability of RT was not reported further in the body of this paper or in figures.

⁴Within each decade, we first confirmed that there were no interactions between emotion and year, indicating that a finer grained categorization than decade-based grouping would be preferable.

⁵As for age, variability of RT showed the same pattern of differences between sexes as did mean RT. It was also not impacted by the interaction of age and sex. Thus, it is not reported in further detail in the main text or figures.