Using lexical content analysis (linguistic inquiry and word count), the hypotheses that social detachment and impaired cognitive processing are typical for alexithymia are investigated. Based on clinical interviews with 32 outpatients (mixed diagnoses), we found support for the hypotheses for the externally oriented thinking facet of alexithymia only.
Following Taylor et al. (1997, p. 29), alexithymia is usually defined as a combination of (1) difficulties identifying feelings (DIF); (2) difficulties describing feelings to other people (DDF); (3) constricted imaginal processes (CIP); and (4) an externally oriented thinking style (EOT). This definition puts forward the ‘salient features’ of the construct (Taylor et al., 1997, p. 29). Clinically speaking, alexithymia should be understood within a broader process of affect regulation (Taylor et al., 1997 and Vanheule, 2008). This study focusses on two aspects of broader alexithymic functioning: investment in social relationships and cognitive processing.
In terms of social functioning, previous research shows that alexithymic persons show a tendency towards distance taking (Vanheule et al., 2010). This might reflect two different types of social disposition: detachment or avoidance. Working within the framework of psychoanalytic theory, Marty (1980) described individuals with an externally oriented cognitive style as detached, and stated that their relation to others is ‘devitalised’ and ‘decathected’ (p. 62). Marty suggested that these individuals show poor investment in relationships and a lack of concern for others (detachment thesis). However, distance taking could also indicate excessive concern for others, leading to the avoidance of painful confrontation. In this case, investment in close relationships has taken place, yet actual interpersonal contact evokes unpleasant affects, eliciting avoidance behaviours (avoidance thesis). In this article, we consider both hypotheses.
Second, we examine cognitive processing. Taylor et al. (1997, p. 31) consider alexithymia as an “inability to modulate emotions through cognitive processing,” and as a problem in the cognitive appraisal of events that evoke affects. In alexithymia “linking their inner states with the events causing them” is problematic (Dimaggio et al., 2007, pp. 17–18). If this proposition holds true, we expect less cognitive processing with increasing alexithymia scores. In this article, we present a general test of the relationship between cognitive processing and alexithymia.
Investment in social relationships and cognitive processing are tested implicitly through lexical content analysis. Lexical analysis holds the assumption that natural word use serves as a psychological marker, and that “the words people use convey psychological information over and above their literal meaning and independent of their semantic context” (Pennebaker et al., 2003, p. 550). The method builds on word counts, and maps the number of words from predefined thematic or grammatical categories.
Our lexical analyses use frequency counts (weighted amount of words used from a thematic category), which reflect the degree of preoccupation with a theme (higher-frequency scores indicate more preoccupation or concern), and complexity counts (weighted amount of different words used from a thematic category), which reflect how differentiated the representation of a theme is (higher complexity scores indicate more differentiation).
Investment in social relationships is examined by mapping words referring to social processes, which indicate involvement in social relationships (Pennebaker et al., 2003). We hypothesise that the detachment thesis holds true, and expect (a) less preoccupation with social process word use to be associated with higher alexithymia scores, and (b) less differentiation in social process word use. This would indicate that, with increasing alexithymia scores, mental representations of social relationships are less comprehensive. If the opposite is true, the results support the avoidance thesis.
Cognitive processing is examined by mapping cognitive word use indicative of the degree of cognitive processing (Mehl, 2005). We hypothesise weaker cognitive processing, as reflected in less preoccupation and less differentiation, with increasing alexithymia scores.
As it has been documented that both depression and the level of education may affect alexithymia scores (Mattila et al., 2010) and lexical style (Pennebaker et al., 2003), these variables are controlled for in our analyses.
