In this work, we aimed to explore sex-specific effects in a sample of healthy adults with history of childhood poverty. We examined sex-specific effects in implicit and explicit emotional reactivity, using EFAT and ERT accordingly. We also examined sex-specific effects of childhood poverty on implicit and explicit emotional regulation, using appraisal condition in SEAT and reappraisal condition in ERT, accordingly. Our findings suggest that the effects of poverty on implicit emotional reactivity in amygdala are mainly seen in females. The previously observed negative bias toward social fear cues (fearful faces) and away from positive cues (happy faces) in the EFAT task was derived by the female group. This effect was specifically more pronounced on the right side. Interestingly, this effect was not present during explicit emotional reactivity (ERT Maintain > baseline contrast). On the other hand, deficits in brain activation in explicit reappraisal-related emotion regulation were an effect of male participants.
Although human studies of sex-specific effects of early life stress on brain anatomy and function are scarce, animal research provides intriguing evidence for sex-specific effects. Hypothalamus Pituitary Axis (HPA) reactivity is more susceptible to prenatal stress in female rats such that adult female rats show larger corticosterone level at baseline and in response to stress [33
]. Female rats with early life stress exhibit greater nociceptive responses than males with early life stress [35
]. Even central administration of female hormones can induce visceral hypersensitivity in female rats [37
]. In humans, the social stress test, which triggers a robust HPA response and increases cortisol level, shows sex-specific effects on fear conditioning. This effect is very much in concert with our findings and suggests reduced prefrontal activation in males, and increased amygdala activation in females in response to the fear-conditioned stimulus [24
]. Also, administration of cortisol in a fear-conditioning study led to reduced activation in prefrontal regulatory areas in men, and increased activation in these areas in women in response to conditioned stimulus [23
]. Our findings suggest sensitized implicit amygdala reactivity to emotional faces as an effect of childhood poverty in adult females. This is in line with previous reports of larger amygdala response to threat pictures in healthy females [26
], and emotional faces in female patients with anxiety disorders ([27
]. Indeed, prevalence of anxiety disorders is higher in adolescents with history of childhood poverty (for a review see [5
]). It is thus possible that a more sensitive salience detection in female amygdala in response to social cues [28
] may lead to sensitization to negative social signals as an effect of early life stress, as well as to subsequent development of anxiety disorders with heightened threat detection. Explicit awareness of the emotional response to aversive stimuli (ERT Maintain task) did not show the sex-specific effects of poverty in amygdala activity, which were observed during implicit emotional reactivity (EFAT task). Emotional faces processed implicitly might be a more sensitive probe of amygdala reactivity than explicit processing of emotional pictures that might involve some measure of cognitive processing. Interaction effects in amygdala were lateralized, and seen on the right side and in females. Multiple studies have previously shown left amygdala response to emotional pictures, or recall of emotional memories in females, and right amygdala response in males [26
]. Absence of similar findings in the left amygdala in our study (B
= 0.565, t
(48) = 1.53, p
= 0.13) could be due to the small sample size and low power. On the other hand, right amygdala response could be more vulnerable to the effects of poverty.
Our findings of impaired function in the emotion regulatory areas as an effect of poverty in males is consistent with some earlier findings reported in the animal and human literature. Several animal studies have shown that adult male, but not female, rats with prenatal or infancy exposure to stress had decreased dendritic complexity in medial prefrontal cortex (mPFC) and prelimbic cortex [40
]. Blaze and colleagues [41
] showed sex-specific changes in methylation of BDNF
genes in the mPFC in male mice maltreated during infancy. Their findings suggest sex-specific epigenetic changes in expression of genes with a role in brain development and synaptic plasticity in emotion regulatory areas in male mice. Similarly in human studies, [27
] cortisol reduced mPFC activation in fear-conditioning paradigm in males, but not females. This indirect evidence is consistent with our findings, and together suggest that structure and function of the prefrontal emotion regulatory areas in males might be more sensitive to developmental effects in general and to stress effects in particular.
Several limitations of the reported findings have to be acknowledged. Our low sample size might have led to limited power in detection of gender by childhood income-to-need interaction. The imaging tasks were not designed to probe differences in emotional processing between males and females, thus more gender-specific paradigms and analyses (e.g., differential response to female or male threat faces, or gender-specific negative images) could provide more specific data for possible threat-specific differences in neural responses of men and women. Secondly, while the absence of sex-specific effects in amygdala reactivity in the ERT task and the presence of these effects in the EFAT could be, as we suggested, due to difference between explicit and implicit emotional reactivity, other between-task differences could have contributed to these “inconsistencies.” ERT tasks involved presentation of aversive IAPS images, and not faces, and some previous studies reported higher amygdala activation in response to negative emotional faces than to negative IAPS images [42
]. Weaker responses to pictures compared to faces may have reduced our power of detecting possible effects in amygdala in response to IAPS pictures. Furthermore, as is inherent to the majority of laboratory studies, complexity and intensity of the emotional stimuli are lower than in real-life experiences. Hence, expanding the results into in vivo emotion processing should be done cautiously. Finally, our work does not address the question of which mechanisms accompanying early life experience of poverty can lead to differential effects in neurocircuitry of emotion processing between males and females.