Previous studies have demonstrated the existence of adaptation aftereffects for facial expressions. Here we investigated which aspects of facial stimuli contribute to these aftereffects. In Experiment 1 we examined the role of local adaptation to image elements such as curvature, shape and orientation, independent of expression, by using hybrid faces constructed from either the same or opposing expressions. While hybrid faces made with consistent expressions generated aftereffects as large as that with normal faces, there were no aftereffects from hybrid faces made from different expressions, despite the fact that these contained the same local image elements. In Experiment 2 we examined the role of facial features independent of the normal face configuration, by contrasting adaptation with whole faces to adaptation with scrambled faces. We found that scrambled faces also generated significant aftereffects, indicating that expressive features without a normal facial configuration could generate expression aftereffects. In Experiment 3 we examined the role of facial configuration by using schematic faces made from line elements that in isolation do not carry expression-related information (e.g. curved segments and straight lines), but that convey an expression when arranged in a normal facial configuration. We obtained a significant aftereffect for facial configurations but not scrambled configurations of these line elements. We conclude that face expression aftereffects are not due to local adaptation to image elements but due to high-level adaptation of neural representations that involve both facial features and facial configuration.

No abstract

Purpose Little is known about the potential adverse effects of interventions to reduce dietary fat. We examined the physical and emotional health effects, and social consequences experienced by women at high risk for breast cancer who had participated in a low-fat diet intervention, randomized, controlled trial for at least 5 years. Methods Participants in the Canadian Diet and Breast Cancer Prevention Trial from British Columbia were mailed a survey questionnaire that included the validated Medical Outcomes Study 36-item Short Form Health Survey (SF-36) and Women's Health Questionnaire (WHQ), and a series of questions on health-related and social constructs. Responses were compared between the diet intervention and control groups by menopausal status. Results Completed questionnaires were returned by 359 women in the diet intervention group and 382 in the control group. No significant differences were found between these groups for SF-36 and WHQ health outcomes, hair/nail changes, physical activity levels, family/friend support levels, and doctor visits. Significantly more women in the intervention group reported taking products for arthritis (other than pain medication), greater difficulty in maintaining eating habits in social situations and at work, greater stress, and guilt related to personal eating habits. These findings persisted for both premenopausal and postmenopausal women. Conclusion Changes resulting from a low-fat diet intervention can be incorporated into women's daily lives with limited long-term negative effects.