SUSCEPTIBILITY TO COGNITIVE AND EMOTIONAL INTERFERENCE IN LATE-LIFE DEPRESSION

Abstract

Introduction Susceptibility to cognitive interference is common in older adults with major depression. Traditionally susceptibility to cognitive interference is measured with “cold” cognitive tasks such as the Stroop Color Word test. Etkin and colleagues developed an emotional version of the classic Stroop test included as a condition of the Cognitive and Emotional Conflict Test. There are two conditions for this test where participants are presented with photographs of emotional faces portraying either fear or happiness: 1) the “non-emotional task,” where participants identify the gender of the face while trying to ignore congruent or incongruent gender labels displayed across the faces; and 2) the “emotional task,” where participants identify the facial expression as happy to fearful while trying to ignore congruent or incongruent emotion labels displayed across the faces. There is some functional neuroimaging evidence from younger adults to suggest that these two conditions may activate distinct cognitive and emotional control systems. However, little if any behavioral data is available for this task in older adults with major depression. The aim of the current study was to examine the effect of late-life major depression and chronic worry / emotional instability (neuroticism) on the Cognitive and Emotional Conflict Test. We first contrasted test performance in older adults with major depression compared to non-depressed elders and hypothesized that depressed patients would perform worse on both test conditions. We next contrasted test performance in older adults with major depression with either high or low in neuroticism. Considering that adults with depression and neuroticism exhibit worse performance on “cold” cognitive tests and heightened sensitivity to negative stimuli, we hypothesized that elders with depression and high neuroticism would perform worse on both test conditions compared to depressed elders with low neuroticism. Finally, we explored whether the emotion valence expressed by the face in the test stimuli influenced the results. Methods The sample included 118 adults age 60+ with a current diagnosis of major depression based on DSM-V criteria and 43 non-depressed cognitively normal older adults (NC). All participants completed a comprehensive evaluation including a clinical interview with a geriatric psychiatrist and detailed subjective and objective assessment of cognition and mood. The measures relevant to this study included the Montgomery Asberg Depression Rating Scale (MADRS), The Cognitive and Emotional Conflict Test, and the neuroticism subscale of the NEO-PI-R. The main outcomes were differences between congruent and incongruent trials on the two conditions of the Cognitive and Emotional Conflict Test. Congruent trials are defined as having the word and picture match, while incongruent is when it is when they do not match. Multivariate analyses of covariance (MANCOVA) controlling for age and sex were used to test the main hypotheses. Results When contrasting patients and controls, depressed patients exhibited a significantly slower reaction time when completing the non-emotional task (F (1,158) = 4.90, p = .028). No differences were found between groups for reaction time on the emotional task (F (1,158) = .312, p = .577. Contrary to our hypotheses, when these analyses were repeated contrasting depressed patients with high and low neuroticism based upon median split (NEO-PI-R Neuroticism Raw Score = 111), there were no significant group differences on either the non-emotional (F (1,148) = 1.76, p = .187) or emotional (F (1,148) = .104, p = .747) condition. Separating the outcomes based upon emotional expression did not result in significant group differences when contrasting controls versus patients or high and low neurotic patients. Conclusions Compared with non-depressed cognitively normal subjects, older adults with major depression performed slower on a “non-emotional”, but not a presumed “emotional”, interference test. This finding supports evidence suggesting weakness in “cold” response inhibition (presumably dependent on the cognitive control network mediated by the dorsal lateral prefrontal cortex) is a prominent deficit in late-life depression. To better understand this test, future studies may wish to examine the correlation with other “hot” (e.g., Iowa Gambling) and “cold” (e.g., Stroop) cognitive tests as well as structural and functional imaging correlates in older adults. This research was funded by: Research was supported by the National Institute of General Medical Sciences of the National Institutes of Health under linked Award Numbers RL5GM118969, TL4GM118971, UL1GM118970; by National Institute of Mental Health grant R01 MH108578; and by the Leo and Anne Albert Charitable Trust.