The relationship between social cognition and executive function in Major Depressive Disorder in high-functioning adolescents and young adults

Highlights

• Executive function is associated with social cognition only during current depression.

• Enhanced planning ability might compensate for social cognitive dysfunction.

• Reduced cognitive flexibility might lead to a rigid interpretation of social stimuli.

• The results cannot be attributed to age or chronicity of depression.

• The method investigates prosody, facial affect recognition and theory of mind.

Abstract

To understand how cognitive dysfunction contributes to social cognitive deficits in depression, we investigated the relationship between executive function and social cognitive performance in adolescents and young adults during current and remitted depression, compared to healthy controls. Social cognition and executive function were measured in 179 students (61 healthy controls and 118 patients with depression; M_age = 20.60 years; SD_age = 3.82 years). Hierarchical regression models were employed within each group (healthy controls, remitted depression, current depression) to examine the nature of associations between cognitive measures. Social cognitive and executive function did not significantly differ overall between depressed patients and healthy controls. There was no association between executive function and social cognitive function in healthy controls or in remitted patients. However, in patients with a current state of depression, lower cognitive flexibility was associated with lower performance in facial-affect recognition, theory-of-mind tasks and overall affect recognition. In this group, better planning abilities were associated with decreased performance in facial affect recognition and overall social cognitive performance. While we infer that less cognitive flexibility might lead to a more rigid interpretation of ambiguous social stimuli, we interpret the counterintuitive negative correlation of planning ability and social cognition as a compensatory mechanism.

Introduction

Social cognitive dysfunction has been often observed in distinct psychiatric conditions (e.g. autism spectrum disorder), but studies provide conflicting results on whether and how social cognitive performance is impaired within Major Depressive Disorder (MDD; Weightman et al., 2014). Most studies demonstrate that patients with MDD show significantly poorer social cognition compared to healthy controls (Csukly et al., 2009, Harkness et al., 2011, Langenecker et al., 2005, Surguladze et al., 2004, Szily and Kéri, 2009, van Wingen et al., 2011). However, other studies indicate that patients suffering from MDD as well as healthy controls may not differ on measures of social cognition, such as affective prosody (Kan et al., 2004) or the recognition of facial affect (Bertoux et al., 2012, Joormann and Gotlib, 2006, Suslow et al., 2010). The reasons for inconsistent findings of social cognitive impairment in patients suffering from MDD could be threefold: First, social cognitive impairment is more subtle in MDD than in autism spectrum disorder or schizophrenia, and therefore only observable in current and severe depression (Air et al., 2015). Studies using the same measure to test for social cognitive performance (e.g. the reading the mind in the eyes test, Baron-Cohen et al., 2001) in MDD found consistent results (Szily and Kéri, 2009, Wang et al., 2008) when tested in clinically comparable cohorts. Only one study did not find social cognitive impairment in MDD using the reading the mind in the eyes test (Wolkenstein et al., 2011). While Wolkenstein et al. (2011) investigated an outpatient sample with moderate symptom-severity, the studies by Szily and Kéri (2009) and Wang et al. (2008) investigated samples with severe and acute depression. Thus, the inconsistent results might in this case be a result of varying clinical characteristics across the studies.

Second, in almost all studies (Bertoux et al., 2012, Csukly et al., 2009, Joormann and Gotlib, 2006, Kan et al., 2004, Suslow et al., 2010, Szily and Kéri, 2009, van Wingen et al., 2011) the investigators used a different measure for social cognitive performance. Measurements for social cognitive performance are often characterized by varying underlying concepts and outcome measures, which might contribute to the heterogeneity of results reported so far. For example, Bertoux et al. (2012) used a test for social cognitive performance (the Social Cognition and Emotional Assessment - SEA) to discriminate between frontotemporal dementia and depression. The SEA comprises a high sensitivity towards detecting social cognitive impairment in frontotemporal dementia, but not in MDD patients compared to controls (Bertoux et al., 2012). Hence, variance in measurements and their psychometric properties could contribute to inconsistent results regarding social cognitive performance in depression across studies.

Third, social cognitive performance itself could be affected by other, more general cognitive impairments that occur in current and/or remitted depressive states. Since MDD has been previously associated with general cognitive impairments in samples with adolescents and young adults (Baune et al., 2014) and previous studies have found general cognitive impairment to be associated with social cognitive impairment (e.g. Uekermann et al., 2008), one could hypothesize that general cognitive impairment could affect social cognitive impairment or vice versa. Compensatory associations of social cognitive impairment and general cognition could mask social cognitive impairment. Because only few studies have considered general cognitive impairment in their analysis of social cognitive performance, this could have affected the results of the studies.

Social cognition describes the ability to identify, perceive and interpret socially relevant stimuli such as facial expressions, prosody and body language (Kandalaft et al., 2012). As an umbrella-term, social cognition comprises affect recognition and theory of mind. Theory of mind itself can be diverted in affective and cognitive theory of mind and describes the ability to infer another person's emotional state or their thoughts and beliefs (Wolkenstein et al., 2011). Social cognitive dysfunction might lead to poorer response to psychotherapy across psychiatric disorders. Particularly in depression, social cognitive dysfunction might hinder proper rehabilitation of relationships, working ability and return to work.

Social cognitive performance can be influenced by deficits in processing speed (Anselmetti et al., 2009, Antila et al., 2011, Dhar et al., 2010) and executive function (Kerns et al., 2008, Torralva et al., 2011) within various psychiatric conditions, such as ADHD and bipolar disorder. Patients with damage to the frontal lobe exhibit both, executive dysfunction and social cognitive impairment which supports the hypothesis that these constructs are related to each other or use the same neural substrates (Channon, 2004).

Since abnormal executive function and processing speed are often observed deficits in patients suffering from MDD (Baune et al., 2010, Fossati et al., 2002, Snyder, 2013, Tsourtos et al., 2002) one could hypothesize, those deficits could also relate or lead to social cognitive dysfunction (and vice versa). However, there is no consistent evidence of a relationship between these two domains in MDD (Ladegaard et al., 2014; Langenecker et al., 2005; Thoma et al., 2015, Thoma et al., 2011; Uekermann et al., 2008; Zobel et al., 2010):

Previous findings on social cognitive impairment indicate impact of the state of the illness: Patients with first-episode MDD, show impaired theory of mind and metacognitive abilities, irrespective of executive function (Ladegaard et al., 2014). In contrast, patients with chronic depression do not show an impairment of theory of mind after controlling for dysfunction of working memory and cognitive flexibility (Zobel et al., 2010). Similarly, cognitive flexibility as part of executive function predicted empathy in patients with mixed clinical characteristics of MDD, but not in healthy controls (Thoma et al., 2011).

Compared to this line of research on higher-order social cognition, studies focusing on affect perception show mixed results. Although facial affect recognition, inhibitory control and attention were impaired in partly remitted depressed patients, they were not related with each other (Langenecker et al., 2005). Similarly, severely depressed in-patients showed an impairment concerning the recognition of affective prosody that was not related to executive function. However, when semantic and affective content were not matching, executive function and affective prosody recognition were related in the group of depressed patients (Uekermann et al., 2008). In summary, although the previous literature points towards a relationship between executive function and social cognition in MDD, the current evidence remains inconclusive. Most studies focus only affect recognition or theory of mind when investigating this relationship. This approach fails to systematically analyze the relationship between social cognition and executive function in MDD.

There has been a lack of studies investigating cognitive impairments in MDD in adolescents and young adults. Due to the significant neurodevelopmental changes during this younger period of life (Blakemore and Choudhury, 2006), studies investigating cognition in adolescents may offer important information about cognitive development and hence, about the development of cognitive dysfunction within depression.

Our first question was whether there is a relationship between social cognitive performance (Affect Recognition, Theory of Mind) and executive function (Updating, Planning, Inhibition and Set-Shifting) in current and remitted depression vs healthy controls.

Based on previous literature, we hypothesized that social cognitive performance and executive function are related to each other only in depressed patients, since studies comparing social cognitive function between depressed patients and controls did not find an association of executive function and social cognition in healthy controls (Thoma et al., 2015, Uekermann et al., 2008, Zobel et al., 2010). Our second question was how this relationship was expressed in both current and remitted states of MDD. Most studies in acute or chronic states report a relationship between social cognition and executive function in depressed participants (Zobel et al., 2010) and to our knowledge there are no studies investigating this topic in remitted depressed patients. Langenecker et al. (2005) could not find an association between executive function and social cognition in a depressed sample. They hypothesized that no association was found because of the low symptom-severity of their depressed participants. Although the evidence in remitted participants is sparse, we hypothesized that there is no correlation between executive function and social cognition in remitted as opposed to current MDD. Our focus on the younger age group eliminates known confounders of cognitive functions such as chronic course of illness and ageing both of which have shown to impact on cognitive function in depression. The present research question contributes to the understanding of cognitive and social cognitive processes in younger MDD patients and healthy controls. The results of this study might have impact on the treatment of depression in psychotherapy and with medication. An association between social cognitive performance and executive function could yield information about new targets of treatment.