The development and psychometric investigation of the Perceived Support Network Inventory (PSNI), a measure of perceived social support, is described. A group of 146 introductory psychology students participated in a test-retest study design that collected reliability, internal consistency, and construct, convergent, and discriminant validity data on the scale. In addition, a contrast group of 28 Counseling Center clients was administered the PSNI. Test-retest reliability of the PSNI total score and subscale scores ranged from .72 to .88. Internal consistency for the PSNI was .77. Construct validity estimates ranged from .21 to .57. Convergent validity estimates varied from -.25 to .20. Discriminant validity estimates varied from -.11 to .19. Sample differences between introductory psychology student mean PSNI total and subscale scores and Counseling Center client mean PSNI total and subscale scores proved significant. Recommendations for subsequent research and future application are offered.

Conversational agents (CAs) are effective tools for health behavior change, yet little research investigates the mechanisms through which they work. In accordance with the Computer as Social Actors (CASA) paradigm, we suggest that agents are perceived as human-like actors and hence influence behavior much as human coaches might. As such, agents should be designed to resemble ideal interaction patterns-for example, by resembling their users. Our study will test this paradigm by testing the impact of customization on similarity and reciprocity, which in turn are hypothesized to improve perceptions of the agent and compliance with the agent's recommendations.

No abstract

Very little research has been done in applying modeling principles to the treatment of juvenile delinquents (Sarason, 1968; Sarason & Ganzcr, 1973). The present study was designed to assess the feasibility and effectiveness of using videotaped models plus role playing to improve the interpersonal adjustment of delinquents living in a group home. The major focus was on dayto-day interpersonal behavior. Since videotapes are technologically efficient and can be inexpensively duplicated, the use of videotaped models for treatment intervention has potential advantages over the use of live models.The subjects were eight male residents (six black, two white) ranging in age from 12 to 16. Six randomly selected boys were assigned to view modeling tapes and two (one white, one black) viewed control videotapes.Modeling subjects observed 14 S-minute tapes in which a white male actor portrayed a delinquent in various problematic interpersonal situations, 10 in a group home setting and 4 in a school setting. During each tape, the model was initially depicted as unsuccessfully coping with and then gradually mastering the situation. Modeled behaviors included expressing positive (e.g., empathic) feelings, taking problems to a staff member, and dealing with an accusation. After viewing a videotape, the subject role played the model's part with the experimenter (a white male graduate student), viewed the videotape again, and role played a second time. Control subjects observed lecture tapes emphasizing corresponding social skills.Four experimental phases were used in a multiple baseline design: baseline (2 weeks), presentation of home tapes (5 weeks), presentation of school tapes (2 weeks), and follow-up (2 weeks).Requests for reprints, an extended report, and copies of the rating scales should be sent to