An experimental analysis of couple aggression using a response choice paradigm

Research suggests that a majority of the violence reported by couples involves mutual, low-level acts of aggression; however, there is a dearth of research examining this "common couple violence" using a true experimental paradigm. The current study was designed to more closely approximate a naturalistic situation involving common couple violence by allowing participants to choose whether to retaliate in the face of provocation by their partner. Couples were randomly assigned to four conditions representing different patterns of provocation. Based on the assigned condition, participants received varying amounts of bad tasting juice allegedly poured for them by their partners across 5 experimental trials. Building on the response choice methodology of Zeichner, Parrott, and Frey (2003), participants had the option to respond to their partners' aggression by pouring either a neutral flavored beverage (water) or the bad juice. Of interest was the number of trials that would elapse prior to an individual's decision to retaliate (flashpoint latency) as well as the amount of bad juice poured (flashpoint intensity). Individual and relationship variables were examined as they related to a participant's decision to aggress and the intensity of the aggressive response. Flashpoint latency did not vary as a function of condition. Gender differences were not found with regard to overall aggression, although male participants aggressed earlier than female participants. Participants in the Decreasing Provocation condition poured more juice on the flashpoint trial than those in the Increasing Provocation Condition. Partial support was obtained for the prediction that participants would respond in kind to the level of provocation received. Flashpoint latency did not vary as a function of individual or relationship variables. Increased irritability was related to increased flashpoint aggression. In addition, the greater the degree of irritability and emotional susceptibility reported by participants, the more aggression they displayed over the course of the experiment. Hypotheses regarding the relationship between trait anger and aggression were not supported. Flashpoint behavior did not vary as a function of relationship commitment or aggression levels. Implications of these and other findings, as well as methodological limitations and directions for future research are discussed.