Self-reported assessments of sexual behavior are commonly used to gauge levels of HIV risk in target populations and to determine the effects of prevention interventions on sexual risk behavior change. However the valid and reliable measurement of sexual behaviors via self-report has been hampered by methodological concerns related to individuals' inaccurate memory for previous sexual behaviors. Nevertheless, vitually no research has examined the cognitive mechanisms by which individuals recall sexual behaviors. Such information is crucial for pinpointing the types of biases that affect self-reports, which in turn could inform the design of more sexual behavior assessments. Support is requested for a methodological study that will use protocol analysis techniques to identify the cognitive strategies underlying individuals' responses to commonly- used measures of past sexual behaviors. One hundred men who have sex with men, 100 heterosexual men, and 100 heterosexual women who are at high risk for HIV will be recruited from urban settings. Participants will be asked to verbalize what they are thinking as they estimate frequencies of sexual behaviors, and numbers of partners for two different time periods (the past two weeks and the past three months). Participants' responses will be audio tapes, transcribed and analyzed with protocol analysis, which is a tool developed by cognitive psychologists to uncover the cognitive mechanisms used for recall and problem-solving. It is predicted that enumerations strategis, in which each separate occurrence of the event is recalled and then counted, will be used to recall low-frequency behaviors and behaviors in the past two weeks. In contrast it is predicted that estimation strategies, in which a rate of sexual behavior is recalled and then adjusted to estimate behavioral frequencies fort he time period assessed, will be used to recall high-frequency behaviors in the past three months. Further, it is expected that general impression strategies, in which a vague impression about the contents of memory is used to estimate frequencies, will be used to recall extremely high frequencies of sexual behavior. This research will identify the conditions under which different types of cognitive strategies and heuristics are used to recall sexual behaviors and sexual partners. Information about these cognitive strategies will ultimately be used to develop more reliable and valid assessments of sexual behavior that incorporate interviewers probes, cues, and contexts to aid recall.
| Delahanty, D L; Bogart, L M; Figler, J L (2004) Posttraumatic stress disorder symptoms, salivary cortisol, medication adherence, and CD4 levels in HIV-positive individuals. AIDS Care 16:247-60 |
