Deffenbacher, Bornstein, Penrod, and McGorty (2004) conducted a meta-analysis to determine the effects of high levels of stress on eyewitness memory. It has been argued that much of the confusion in determining the effect of stress on memory is because many studies do not actually push participants to a high enough level of stress. This review excluded all studies that did not elevate participants' stress levels to elicit a defensive response or activation mode of attention control. This defensive response as defined by Deffenbacher et al. is characterized by a pronounced change in physiological measures of stress. These measures include an increase in heart rate, blood pressure and muscle tone (2004). Two meta-analyses were conducted in this review. One of them attempted to determine whether stress had an impact on an individual's facial recognition ability. Moderating variables that may impact the results of these studies were also examined. The second analysis examined the accuracy of eyewitness memories and the impact of stress. Additionally, the effect of other variables that may have influenced memorization of details was also examined. The conclusion of the first analysis was that stress actually had a negative effect on eyewitness identification accuracy (Deffenbacher et al., 2004). This is the expected result, but many other variables have been found to impact the impact of stress on eyewitness memory. These variables were: lineup type, research paradigm, presence or absence of a staged crime, and age of the witness. The type of queue, i.e., either target present (TP) or target absent (TA), was found to moderate the effect of stress on memory. Target's current lines had middle of paper...... something that has been seen in studies using other methods to induce stress. Deffenbacher et al. There were strong findings that stress had a negative impact on memory, both in terms of identification and the ability to remember details. However, only studies causing a particular type of stress were examined. Studies that produced only orienting responses in participants were included, instead studies producing a defensive response were used. Included (2004). Furthermore, stress had to be controlled directly by the experimenter. Naturalistic studies were not included. Finally, the studies had to prove that the stress had indeed been induced and could not simply be assumed to have occurred. Works cited: Deffenbacher, KA, Bornstein, BH, Penrod, SD and McGorty, EK (2004). A meta-analytic review of the effects of high stress on eyewitness memory. Publication of the Faculty of Psychology