Recognition of dynamic and static facial expressions of emotion among older adults with major depression

Introduction: The recognition of facial expressions of emotion is essential to living in society. However, individuals with major depression tend to interpret information considered imprecise in a negative light, which can exert a direct effect on their capacity to decode social stimuli. Objective: To compare basic facial expression recognition skills during tasks with static and dynamic stimuli in older adults with and without major depression. Methods: Older adults were selected through a screening process for psychiatric disorders at a primary care service. Psychiatric evaluations were performed using criteria from the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5). Twenty-three adults with a diagnosis of depression and 23 older adults without a psychiatric diagnosis were asked to perform two facial emotion recognition tasks using static and dynamic stimuli. Results: Individuals with major depression demonstrated greater accuracy in recognizing sadness (p=0.023) and anger (p=0.024) during the task with static stimuli and less accuracy in recognizing happiness during the task with dynamic stimuli (p=0.020). The impairment was mainly related to the recognition of emotions of lower intensity. Conclusions: The performance of older adults with depression in facial expression recognition tasks with static and dynamic stimuli differs from that of older adults without depression, with greater accuracy regarding negative emotions (sadness and anger) and lower accuracy regarding the recognition of happiness.


Introduction
Major depression can affect up to 7% of the population of older adults. 1 This chronic, recurring condition is associated with functional incapacity and a reduction in quality of life. 2 Depression often leads to social isolation due to difficulties in maintaining social interactions. 3 Basic emotions are involuntary physiological responses that are universally shared by our species, visually distinguishable and pre-established by stimuli that have influenced our evolution and can also be molded by life experiences. 4 Most studies establish emotions like happiness, sadness, anger, disgust, surprise and fear as basic or primary, however the conceptualization of basic emotion as well as which emotions would be part of this set are still much debated and controversial. 5 The recognition of facial expressions of basic emotions is an extremely important skill in society, as it is related to the capacity to interpret the feelings and emotions of others. Facial emotion recognition is associated with emotional regulation, enabling individuals to use emotions adaptively. 5 However, the decoding of these stimuli may be altered in individuals with depression, who tend to interpret information considered imprecise in a negative light, which can affect their capacity to decode some social stimuli. 6,7 A recent meta-analysis has found that individuals with depression have impaired facial expression recognition for all basic emotions, except sadness. 8 Moreover, such individuals seem to exhibit an increase in vigilance and selective attention for faces with sad expressions compared to other emotions. 9 However, a limitation that should be considered in studies involving facial expression recognition regards the use of different tasks and procedures employed in investigations addressing depression 9  preferably selected for the control group. Figure 1 displays the flowchart of the selection process.
In the group of 23  Differences in the number of participants who performed the tasks within each group were due to the fact that some older adults did not complete all tasks. Individuals with severe vision or hearing impairment that could hinder their comprehension during the interview and tests and those with severe clinical comorbidity, major neurocognitive disorder or psychotic disorder were excluded from the study.

Patient Health Questionnaire-2 (PHQ-2)
The PHQ-2 is used to screen and quantify depression and is composed of two items referring to the previous two weeks. 15 Response options range from 0 (not at all) to 3 (nearly every day) for each item. The total score is determined from the sum of the points and may range from 0 to 6. The PHQ-2 has been validated for the Brazilian population. 16,17 Trends Psychiatry Psychother. 2019;00(0) -3 Facial emotion and depression in older adults -Bomfim et al.

Mini Mental State Examination (MMSE)
The MMSE is the test most widely employed to screen for cognitive decline. It has the following domains: recall (registration and evocation), temporal and spatial orientation, attention and calculation, language (naming, repetition, reading and writing, comprehension and praxis) and visuoconstructive capacity. The total MMSE score may range from 0 to 30, and the cutoff point depends on the individual's level of schooling, regardless of age, with lower scores denoting greater cognitive decline. 18 Task with dynamic stimuli The emotion recognition task normalized by Kessels et al. 19

Data analysis
Descriptive analysis was performed to characterize the social-demographic profile of the groups. The Shapiro-Wilk test was used to determine the normality of the data.
Differences in the response pattern between the groups were determined using the chi-square test. Either the Student's t-test or the Kolmogorov-Smirnov Z test was used to determine differences between groups regarding the number of correct responses, emotion intensity and reaction time, depending on the distribution of the data.
Cohen's d effect sizes were used to estimate differences between groups and were defined as the difference between the mean obtained in the depression and control groups divided by a standard deviation of the full sample.

Results
The clinical-demographic data of the two groups are displayed in Table 1. No significant differences were found regarding age (p=0.649), sex (p=1.000), schooling (p=1.000) or cognitive status based on the MMSE (p=0.275). In contrast, a statistically significant difference was found between the groups regarding the PHQ-2 score (p<0.001), with a higher mean score found in the group with depression compared to the control group.
Regarding the number of correct responses per emotion on the task with dynamic stimuli, a significant difference between the groups was found for the recognition of happiness (p=0.020), with a higher mean number of correct answers in the control group compared to the group with depression. No other significant differences between the groups were found in the number of correct responses per intensity or at total in the same task (Table 2).
Regarding the number of correct responses per emotion on the task with static stimuli, statistically significant differences between the groups were found for the recognition of sadness (p=0.023) and anger (p=0.024), with higher means of correct answers in the group with depression compared to the control group (Table 3).
On the task with dynamic stimuli, the individuals in both groups had difficulty discriminating fear and happiness from surprise. The chi-square test demonstrated differences in the response pattern given by the two groups on the dynamic task (χ 2 =25.81; p<0.001). The control group mistook surprise for happiness in 57.5% of the trials (Figure 2).
On the task with static stimuli, both groups had difficulty discriminating negative emotions, such as disgust, sadness and anger. The control group also demonstrated difficulty discriminating anger from fear and from neutral expressions. The response pattern also differed significantly between the groups on the static task (χ 2 =22.89; p<0.001) (Figure 3).

Discussion
The present results indicate greater accuracy among older adults with major depression in recognizing sadness and anger on the task with static stimuli and less accuracy in recognizing happiness on the task with dynamic stimuli compared to the control group without depression. Moreover, differences in response bias were found between the two groups. The data indicate differences in the recognition pattern depending on the task performed.   Some studies report difficulty in discriminating surprise from fear and happiness. 34,35 This seems to be an important aspect of facial expression recognition tasks, as characteristics of emotions that can be considered in an ambiguous manner could help better discriminate between different study groups. 36 Problems among individuals with major depression in recognizing facial expressions of emotion are associated with dysfunctions in the limbic system, paralimbic cortex and prefrontal areas of the brain, which are associated with altered functioning in emotional processing. 37 Some authors propose that antidepressants normalize the processing of emotions and constitute an initial step toward the treatment of depression. 38,39 Other characteristics of depression, such as slower cognitive processing and difficulty making decisions, which are often found in older adults, are also associated with a worse performance in the recognition of facial expressions of emotion. 40

Conclusions
Major depression in older adults may be related to the performance of these individuals on facial emotion recognition tasks, with differences depending on the task and stimuli used. In general, older adults with depression seem to recognize the emotions of sadness and anger with more and of happiness with less accuracy. Emotion variables, response bias, type of stimuli (static or dynamic) and emotion intensity may exert an influence on the results and should, therefore, be considered.