Additional evidence consistent with the emotion regulation hypothesis comes from a secondary analysis of the autobiographical memories produced at Year 6, in which Greenhoot, Johnson, and McCloskey (2005) examined the teens' use of internal states language, particularly emotional language. We were interested in the representation of internal states in participants' recollections because recalling how one felt during a past experience is seen as one way in which people make sense of and find personal significance in past experiences (see Chapter 6 in this volume; Bauer, Stennes, & Haight, 2003; Fivush, Berlin, Sales, Mennuti-Washburn, & Cassidy, 2003; Pennebaker, Mayne, & Francis, 1997), and little is known about how individuals with trauma histories evaluate and make sense of their pasts. This investigation involved recoding the narratives produced in response to each cue in the AMT for the frequency of emotional language and other internal-states language. Specifically, we coded words related to internal states into four categories using an adaptation of a coding scheme developed by Bauer et al. (2003): emotion terms (e.g., happy, scared), cognition terms (e.g., understand, know), perception terms (e.g., hear, smell), and physiological states terms (e.g., hurt, tired). In addition, emotion terms were further classified according to valence (i.e., positive terms, such as "excited"; negative terms, such as "worried"; and neutral terms, such as "didn't care"), explicitness (i.e., whether an emotion was explicitly referred to or implied), and experiencer (i.e., self-referent or referring to others).
We related the frequency of internal-states language in childhood recollections on the AMT to measures of childhood (Year 1) and recent (Year 6) abuse. Because we were interested in whether the effect of abuse on internal-states language might vary as a function of the type of memory cue (positive, negative, or neutral), we constructed dichotomous indicators of abuse at Year 1 and Year 6. This approach facilitated examination of interactions between cue valence and abuse exposure, and also permitted us to collapse across all forms of abuse, including mother-directed violence, child-directed violence, and childhood sexual abuse. For this analysis, we selected participants who at Year 1 reported either high levels of exposure to these forms of abuse (the Abuse Exposure group) or no exposure to abuse during childhood (the No Exposure group). Only children who reported frequent (i.e., 35 to 150 incidents) or severe (e.g., beating, choking) abuse and whose mothers corroborated their reports were included in the Abuse Exposure group (n = 47). The No Exposure group (n = 24) included children exposed to no family aggression whatsoever, according to either the child's or mother's Year 1 report. Participants who reported moderate levels of domestic violence or whose mothers failed to corroborate their Year 1 reports were excluded from the analyses. The resulting 71 participants were also classified according to their abuse exposure at Year 6 (Abuse Exposure and No Exposure); 21 adolescents reported abuse exposure at Year 6, and this included adolescents who were exposed to moderate (rather than only severe) violence, due to the overall drop in violence exposure at Year 6. Although we used this classification approach to simplify presentation of the findings, it should also be noted that the patterns of results were quite similar when we conducted the analyses using continuous measures of abuse exposure and included participants with moderate exposure.
The results revealed that patterns of emotional language in the teens' recollections differed as a function of childhood abuse exposure, whereas other forms of internal-states language did not. Specifically, there was an interaction between childhood abuse history and cue type in the model predicting the frequency of emotion terms, F(2, 62) = 5.23, p = 0.008, controlling for overall narrative length, as well as gender and age. As shown in Figure 4.1, adolescents with no childhood abuse histories used more emotional language in recollections related to negative cues than in memories elicited by positive or neutral cues. It is important to note that analyses of the content of the teens' recollections indicated that almost all of the memories generated in response to the negative cues were negative and highly related to the cues themselves (i.e., involved punishment or conflict), whereas almost none of the memories prompted by positive and neutral cues referred to such events. Thus the pattern observed among nonabused teens is consistent with the findings of other investigations suggesting that emotions may be more salient and relevant in children's recollections of stressful events than nonstressful events (Fivush, Hazzard, et al., 2003). Yet Figure 4.1 also illustrates that adolescents with childhood abuse histories showed no such increase in the use of emotion terms in their responses to negative cues, in spite of the fact that their childhood experiences related to arguing and punishment were probably even more emotionally arousing than those of adolescents without childhood abuse histories. Indeed, teens in the Year 1 Abuse Exposure group used significantly fewer emotion words in their narratives elicited by negative cues than teens in the Year 1 No Exposure group. These patterns, moreover, were observed across all categories of emotions terms; when discussing
□ No Childhood Abuse Exposure ■ Childhood Abuse Exposure
Was this article helpful?
Parenting is a challenging task. As a single parent, how can you juggle work, parenting, and possibly college studies single handedly and still manage to be an ideal parent for your child? Read the 65-page eBook Single Parenting Becoming The Best Parent For Your Child to find out how. Loaded with tips, it can inspire, empower, and instruct you to successfully face the challenges of parenthood.