Studying maltreated children can provide crucial information about mechanisms that underlie emotional processing and, by extension, autobiographical memory. It has been proposed that, in certain respects, maltreated children process emotional cues differently than do nonmaltreated children (see Pollak, 2003, for review). Specifically, Pollak (2003) contends that the processing of negative emotions can be heightened in maltreated children because negative emotional signals in their home environments are prominent and/or because children perceive evidence of threats in such signals. According to this view, abused children develop patterns of information processing that reflect priming for negative emotions.
Evidence of heightened attention to threat cues comes from a study by Pollak and Tolley-Schell (2003). These authors found that physically abused children disengaged from anger cues less easily than did non-abused children, although no differences were found for happy cues. This demonstrates a bias toward anger; maltreated children find it difficult to disengage their attention upon seeing anger cues.
Research has also shown that maltreated children, compared to non-maltreated children, are aroused for longer periods of time by anger in background situations (Pollak, Vardi, Bechner, & Curtin, 2005). Specifically, although physically abused and nonabused children did not significantly differ in initial arousal (as measured by skin conductance) to a background argument, once the anger was resolved, abused children's arousal did not return to baseline as quickly as did that of nonabused children. Pollak and colleagues interpreted this continued state of arousal as resulting from maltreated children's monitoring of the environment for any indications that the anger may continue. Physically abused children were also more concerned than nonabused children when the situation was unresolved and ambiguous, which may, in the abused children's home environments, foreshadow frightening events (Pollak et al., 2005). Overall, it appears that physically abused children are particularly primed to attend to expressions of anger and consequently devote considerable attentional resources to these emotional signals.
If physically abused children are primed to anger cues in the environment, one would expect to see heightened reactions in the brain to emotional stimuli. To examine this possibility, Pollak et al. (1997) examined the event-related potential component P300, which may reflect cognitive processes activated when an individual is presented with salient emotional stimuli. The authors found differences in P300 amplitude between maltreated and nonmaltreated children who were asked to attend to faces reflecting various emotional expressions (e.g., a happy expression, an angry expression): maltreated children showed a larger amplitude when asked to attend to angry (versus happy) faces. Pollak et al. (1997) suggest that this could be due in part to how maltreated compared to nonmaltreated children process negative versus positive emotional expressions. In another study, Pollak et al. (2001) found larger P3b amplitudes for maltreated children compared to nonmaltreated children when shown angry faces, although significant differences were not found for happy or fearful expressions. (The term "P3b" is used interchangeably with P300, and thus it is also an event-related potential of the brain that is associated with the valence of a stimulus.) This study supports Pollak's previous findings of differences in brain activity when maltreated children are shown angry faces and also supports the hypothesis that maltreated children's attention and sensitivity to negative emotions has become accentuated as a result of their traumatic experiences.
Although Pollak's studies mainly concern child physical abuse, similar findings would be expected for child sexual abuse. Specifically, victims of child sexual abuse may be particularly attuned to safety and threat cues relevant to sexuality and/or sexual aggression. This should result in greater attention to or less ability to withdraw attention from such cues.
To our knowledge, studies comparable to Pollak's (e.g., on abuse-induced hypervigilance) have not been carried out with child sexual abuse victims. However, it is possible that a related phenomenon is tapped by the modified Stroop task (also called the emotional Stroop task), which has been used to examine attentional processes of clinical and traumatized populations including child sexual abuse victims. Similar to the standard Stroop task (Stroop, 1935), the modified Stroop task measures participants' ability to inhibit prepotent responses, which has been linked to attention and executive function (W right, Waterman, Prescott, & Murdoch-Eaton,
2003). However, the modified Stroop task does so by presenting emotional words in different colors of ink. The participant is asked to name as fast as possible the color in which each word is printed instead of reading the actual word. The extent to which the meaning of the word slows naming of the color is taken as a measure of interference, caused for instance by being unable to inhibit the prepotent response, by an inability to withdraw attention quickly, and/or by automatic processing of the word to the semantic level (see Algom, Chajut, & Lev, 2004, for alternative views).
In maltreated and other traumatized populations, the modified Stroop interference effect is especially pronounced for words that are highly associated with their specific trauma. For example, words related to sexual abuse caused greater interference than nonsexual abuse words for adults who as children suffered sexual abuse (Field et al., 2001). These findings, like Pollak's, appear to indicate priming connected to child maltreatment. In the case of the modified Stroop task, reactivated memory "networks" may contribute to a delay in response (Field et al., 2001). This research suggests that earlier traumatic experiences guide attention in adults even at the unconscious level, leading to heightened attention to threat or trauma-related cues (Michael, Ehlers, & Halligan, 2005).
Thus, the overattention to threat cues in children as described by Pol-lak appears to carry forward into adulthood. This persistence into adulthood seems to be particularly apparent if the individual develops PTSD. In modified Stroop tasks, it has been consistently found that adults with a diagnosis of PTSD (including if the PTSD symptoms are associated with past child sexual abuse) exhibit more interference for trauma-related than non-trauma-related words. Adults with PTSD take longer to color-name trauma-related words (versus neutral, positive, or negative words), demonstrating an attentional bias in favor of such words (McNally, Kaspi, Riemann, & Zeitlin, 1990). Further, McNally, Clancy, Schacter, and Pitman (2000) found that PTSD severity was the strongest predictor of modified Stroop interference for adults with self-reported child sexual abuse histories. Similar attentional bias effects have also been found for individuals with PTSD subsequent to war/combat (e.g., McNally, English, & Lipke, 1993), rape (e.g., Cassidy, McNally, & Zeitlin, 1992; Foa, Feske, Murdock, Kozak, & McCarthy, 1991), and traffic accidents (e.g., Harvey, Bryant, & Rapee, 1996).
Few studies have investigated performance on the modified Stroop test in children with trauma histories or trauma-related psychopathology (e.g., depression). Dubner and Motta (1999) were perhaps the first to show that adolescents with trauma histories, specifically histories of child abuse, show attentional bias to trauma-related words on the modified Stroop test. However, when Doost, Taghavi, Moradi, Yule, and Dalgleish (1997) examined Stroop interference in 9- to 18-year-olds with depression, using depression-related, trauma-related, threat, happy, and neutral words, although depressed children were slower at color-naming of words overall, contrary to expectation, there was no significant difference in reading times for the depression- and trauma-related words as compared to happy and neutral words. In contrast, in a study on children (9 to 17 years of age) with PTSD, Moradi, Taghavi, Neshat-Doost, Yule, and Dalgleish (2000) found a significant difference in color-naming times for trauma-related compared to neutral words. Children with PTSD took longer to color-name trauma-related compared to neutral words. Also, children with PTSD showed more trauma-related interference than did children with no history of trauma or PTSD.
Overall, findings on attentional processing in child maltreatment victims generally (though not invariably) support the hypothesis that information that is particularly relevant to one's life concerns may attract more attentional resources or more automatic processing than information that is less relevant. Maltreated children's attention to, and possible priming for, emotionally salient information in the environment may facilitate better encoding of such information. The cognitive resources devoted to detecting and processing trauma-related emotional experiences, as demonstrated in brain activation, may lead to better autobiographical memory of these experiences.
Was this article helpful?