This study investigates whether mothers who are neglectful and at high risk for child physical abuse present a deficit in empathy. Participants were neglectful mothers (n=37), mothers at high risk for child physical abuse (n=22), and nonmaltreating mothers (n=37). The Interpersonal Reactivity Index, a self-report measure assessing specific dimensions of empathy, was used to assess dispositional empathy. No differences between neglectful and non-neglectful mothers were found for perspective taking, empathic concern, and personal distress. High-risk mothers reported less perspective taking and more personal distress than nonmaltreating mothers. No difference between groups was found for empathic concern. The present study supported the hypothesis that parents at high risk for child physical abuse show a deficit in particular aspects of dispositional empathy: personal distress and perspective taking. However, no differences were found between neglectful and nonmaltreating mothers in any dimension of dispositional empathy.

Exposure of an aggressor to the suffering of his/her victim generally inhibits subsequent attacks [e.g. Baron, 1971a] presumably because of an empathic process. Physically abusive parents and individuals at high risk for child physical abuse are thought to present a deficit in empathy [e.g. Milner, Halsey and Fultz, 1995]. The present research was designed to investigate whether highrisk, compared to low-risk, subjects for child physical abuse select more aggressive responses and aggress with more intensity toward a supposed child whose behavior is adequate, inadequate or ambiguous in the presence of the child's pain cues. A second objective was to explore whether high-risk, compared to low-risk, subjects for child physical abuse fail to adequately integrate mitigating information and, therefore, do not inhibit their aggressive behavior following the receipt of mitigating information before or during an inadequate child behavior. Participants were engaged in an experimental task that involved attempting to teach a child how to proceed accurately through a computer-displayed maze and were given the opportunity to send positive (pleasant sounds and happy faces) and negative feedback (mad faces and unpleasant sounds) to the child. Two hundred and fifty undergraduate students participated in the experiment. A 2 Â 5 factorial design based on two levels of the participant's risk status (high, low) and five conditions based on the child's behavior and the introduction of mitigating information was employed. As expected, low-risk, compared to high-risk, subjects showed less aggression when mitigating information was provided.