The Andrea Yates Effect: Priming Mental Illness Stereotypes Through Exemplification of Postpartum Disorders

ABSTRACT

In a randomized between-subjects design, participants (N = 80) were assigned to one of four conditions, 2 (pregnant, not pregnant) × 2 (extreme prime, moderate prime). It was hypothesized that primes involving moderate mental illness would be positively associated with increased perceived risk of developing postpartum depression. Hayes and Preacher’s bootstrapping procedure was used to test the direct, indirect, and conditional indirect effects related to the hypothesized model. In addition, further analyses evaluated whether implicitly activated goals (to be healthy or to be a good mother) were positively associated with increased perceptions of risk and engagement of downstream avoidance behavioral intentions. Findings show that for pregnant participants, the effect of the prime condition on perceived personal risk of developing postpartum depression was mediated by perceptions about the target character’s sanity. However, activated “healthy” and “good mother” goals are not influencing behavioral intentions.

Funding

This research was made possible by grants from the Triad Foundation, the Appalachian State University Research Council, and the Minnie S. and Eli A. Rubinstein Research Award.

Notes

¹ Indeed, a content analysis of 11 years of media coverage of postpartum disorders found that exemplars overall were negative (58.4%), with 70.9% of newspapers and 33.8% of magazines presenting negative exemplars. In this study it was predicted that in their coverage of postpartum disorders, journalists would use exemplars of women who have committed violence against their children significantly more than exemplars of women who are simply suffering from a disorder, and found that 50.5% of the articles featured a story about a mother who killed her children or husband (one case). Chi-squared analysis supported this hypothesis, even when the file was split to compare stories before the Andrea Yates case and those written after the June 20, 2001, murders. Of the 37 stories about postpartum disorders before Andrea Yates murdered her children, 22 featured women who killed and only 3 were stories about regular mothers who suffered from a postnatal disorder, and 7 were stories general stories about postpartum disorders (χ² = 52.351, p < .001). Of the 165 stories about postpartum disorders written after Yates drown her five children, 80 were stories about women who had committed violence against their children or husband, 24 were about regular mothers, and 25 were general stories about postpartum disorders (χ² = 173.164, p < .001). It can be argued that the case of Andrea Yates served as a focusing event, where many were introduced to the issue of postpartum disorders for the first time. Unfortunately, many reports misrepresented this story as a case of the much more common PPD, and as such, this case has served as an exemplar of that particular postpartum disorder rather than postpartum psychosis (Holman, 2011).

² In particular, when participants were primed with a “progress” goal toward a healthy lifestyle through a weight-loss comparison scale that was smaller (thus resulting in the illusion of more progress being made), these participants were more likely to accept a chocolate bar over a healthy snack.

³ Some pediatricians recommend avoiding breastfeeding when taking certain medications, like Prozac, as they aren’t tolerated by the infant as well as other medications. Many new mothers worry about taking antidepressants, especially if they are nursing, because medication does get into breast milk. But most experts say such fears are generally exaggerated. “Infant exposure of antidepressants through breast milk is generally low to very low. We consider that when antidepressant treatment is indicated in women with postpartum depression, they should not be advised to discontinue breastfeeding. Paroxetine and sertraline are most likely suitable first-line agents. Although some concern has been expressed for fluoxetine, citalopram and venlafaxine, we nevertheless consider that if the mother has been treated with one of these drugs during pregnancy, breast-feeding could also be allowed during continued treatment with these drugs in the postpartum period. However, an individual risk-benefit assessment should always be performed” (Berle & Spigset, 2011, p. 28). “At present, there is little evidence that exposure to antidepressants through breast milk has any serious adverse effects in infants; however, long-term neurodevelopmental effects have not been adequately studied. There are many benefits of treating postpartum depression and advantages of breastfeeding, for both the mother and the infant. Therefore, if maternal depression necessitates treatment with pharmacotherapy, then breast-feeding need not be avoided, and the antidepressant that would be most effective for the mother should be considered” (Chad, Pupco, Bozzo, & Koren, 2013, p. 634).

⁴ Goal-priming effects (a) involve value, (b) involve postattainment decrements in motivation, (c) involve gradients as a function of distance to the goal, (d) are proportional to the product of expectancy and value, (e) involve inhibition of conflicting goals, (f) involve self-control, and (g) are moderated by equifinality and multifinality (Forster et al., 2007).

⁵ Several additional variables (empathy, sympathy, depression, need for cognition) were assessed as potential confounds to rule out plausible alternative explanations to the study’s findings. Among these measures, involvement was the only variable significantly associated with the hypothesized mediator and the key dependent measure in the study, and was therefore controlled for in all analyses to account for potentially spurious associations that were not explicitly modeled. Details regarding the discarded covariates are listed here:

Empathy items included a two-item index for emotional contagion (“I cannot continue to feel okay if others around me are feeling depressed,” “I become nervous if others around me seem nervous”) (r = .686, p < .001); a four-item index for perspective taking (i.e.,“Before criticizing someone, I try to imagine how I would feel in their place”) (α = .778); a six-item index measuring empathic concern (i.e., “I am the type of person who is concerned when other people are unhappy”) (α = .833); and a five-item index assessing fictional involvement (i.e., “I really get involved with the feelings and characters in a novel or film”) (α = .848).

Sympathy was measured with a four-item scale (“I am concerned about others,” “I feel sympathy for those who are worse off than myself,” “I take an interest in other people’s lives,“I take an interest in, which showed a high degree of internal consistency (α = .866).

Depression. All participants completed pretest measures assessing their emotional status in the past month on a 9-point scale (1 = strongly disagree, 9 = strongly agree) using the 20-item self-report Center for Epidemiologic Studies Depression Scale (Radloff, 1977) (M = 3.38, SD = 1.51, Cronbach’s α = .925).

Need for cognition. This 18-item scale (Cacioppo et al., 1984) was designed to measure participants’ deliberation (“I find satisfaction in deliberating hard and for long hours”) and intellectual motivation (“I would prefer a task that is intellectual, difficult, and important to one that is somewhat important but does not require much thought”) by having them rate how characteristic the descriptions were based on a 9-point scale (1 = extremely uncharacteristic, 9 = extremely characteristic) (M = 5.70, SD = 1.42, Cronbach’s α = .918).

⁶ Depressed mothers tend to have a higher risk of premature delivery. In addition, if a woman has prenatal depression, she has a 50% chance of developing postpartum mood disorder. After birth, these babies have higher levels of stress hormones such as cortisol and norepinephrine and lower levels of mood-enhancing neurotransmitters such as serotonin and dopamine. Babies, like their depressed mothers, experience greater activity in the right frontal lobe, which processes emotional stimuli. These effects often lead to more irritable and less attentive and animated infants (Colino, 2002).

Additional information

Funding

This research was made possible by grants from the Triad Foundation, the Appalachian State University Research Council, and the Minnie S. and Eli A. Rubinstein Research Award.