Despite the recognized benefits of regular physical activity (PA), mid-life and older ethnic minority adults are the most inactive segment of the U.S. population, putting them at increased risk for a number of chronic diseases and conditions including some forms of cancer and weight gain. Few interventions are available in low-income, ethnic minority communities due to cost, personnel, service delivery, and language/cultural barriers. Computer-based counseling programs that are culturally tailored, require minimal amounts of reading and computer literacy, and could be made available in trusted and regularly used community venues offer a promising though currently under-explored alternative for delivering health promotion advice and support to underserved groups. One such promising automated tool is the embodied conversational agent (ECA), an innovative computer-based communication medium that maximizes both visual and aural conversation elements while allowing for individualized message content and a user-friendly interface. The objective of this R-21 exploratory grant is to culturally adapt and pilot-test a theoretically-based, empirically supported ECA-PA program for use with older, low-income Latino adults, who represent the fastest-growing ethnic group in the U.S. Program delivery will occur at a trusted, frequently used community senior center. During Year 01, we will use participatory qualitative and quantitative research methods to culturally and linguistically adapt and pre-test the ECA-PA program with the target population. In Year 02, we will conduct an experimental 4-month pilot study to test the feasibility and preliminary efficacy of the ECA-PA program in promoting initial PA increases (documented via age- and language-appropriate questionnaires, logs, and accelerometry) in the target group. A total of 60 Latino adults ages 55 and over will be randomized to either the ECA-PA arm or a standard care control arm. We will follow those assigned to the ECA-PA program for an additional 4 months to ascertain preliminary behavioral maintenance effects of the program. As part of this study we will explore potentially important theoretically and empirically supported baseline moderators (e.g., health literacy, computer attitudes, acculturation, gender) and mediators associated with PA change (e.g., changes in PA self-efficacy, working alliance with the embodied conversational agent). The effect-size, process, and outcomes information generated by the feasibility study wlll inform a larger-scale trial aimed at evaluating the utility and cost-effectiveness of culturally adapted ECA programs as a means of promoting PA among a fast-growing though sedentary and under-served segment of the U.S. population. ? ? ?