Canadian research shows neglectful mothers have information processing biases towards their children’s emotions and behaviour

Date Published
Source

Hildyard, K. & Wolfe, D. (2007). Cognitive processes associated with child neglect. Child Abuse & Neglect, 31, 895-907.

Reviewed by
Jonathan D. Schmidt
Summary

Neglect is one of the most common forms of child maltreatment and yet little research has been conducted on the potential pathways to neglect. Attachment research shows that parents with different attachment styles differentially perceive and attend to their children’s emotions (e.g., secure mothers are more accurate at labeling infants’ emotions). Theory suggests that mothers found to be neglectful by child welfare authorities have information processing breakdowns- they fail to see their child’s signal of need, misinterpret their child’s signal of need, fail to choose a behavioural response to their child’s need, or fail to implement their chosen response to their child’s need. This study appears to be the first to compare the perceptions of mothers in cases on neglect and mothers not found to be neglectful on socio-emotional cues of children’s needs, attributions regarding children’s behavior, and recall ability of child-rearing related information. Mothers in cases of neglect (n = 34) were referred from a child protection agency and had substantiated chronic neglect allegations (but no physical or sexual abuse allegations) of children under 3 years of age within the last 3 years; comparison mothers (n = 33) were referred from a community agency and were selected for child age, income below the poverty line, and not having had previous contact with CPS or endorsing items of maltreatment from the Conflict Tactics Scale Parent-Child version. Mothers in cases of neglect were found to have more serious histories of maltreatment as children when compared to the comparison mothers. In a standardized emotional perception task, mothers in cases of neglect were significantly less likely to see ‘interest’ in children and more likely to see ‘sad’, ‘shame’, ‘other’, and ‘disgust’ (though ‘disgust’ was associated with depressive symptoms). Mothers in cases of neglect were more likely to give non-normative responses (responses that fell outside of what 95% of a normative sample perceived) and to perceive emotions as simplistic and black-and-white. In another research task, mothers read four stories about a mother and child with varying content (attachment theme- baby reacting differently to mother than strangers, obvious risk- infant placing small objects in mouth, ambiguous risk- child cannot be soothed which may or may not suggest pain or illness, and ambiguous behavior- infant scratches mothers face), after which they evaluated their perceptions of what happened and how they think they might have responded. Mother groups were found to differ significantly in their perceptions of the scenarios. Specifically, on the clear- and ambiguous-risk scenarios, mothers in cases of neglect reported that they would get more upset about the situation, perceive greater maternal control of the situation, and make greater attributions that the child's behaviours were due to internal and stable factors (as opposed to situational and temporal factors) than the comparison mothers. On a subsequent information recall task, the mother groups did not differ after controlling for depression. Overall, mothers identified by child welfare authorities for neglect were found to have information processing biases which may be a risk factor for neglect. The authors conclude that cognitive-behavioural interventions may be promising in correcting these biases and increasing mothers’ emotional vocabulary.

Methodological notes

The study used a cross-sectional group comparison design. Mothers in cases of neglect were identified after the fact and therefore the significant differences may not be risk factors for maltreatment since it is unclear whether they were present prior to child maltreatment. Additionally, the study used photos and scenarios of infants so it is unclear whether these findings can be replicated for mothers found to be neglectful of older children. The study had a fairly small sample size (N= 67) and included a relatively large number of analyses, which increases the likelihood of Type I errors (i.e., concluding that there was a true effect when there actually was not); however, examination of the effect sizes by the authors (generally moderate to large) suggests these are true effects, though replication is needed.