s of 3rd International Congress of the Association of Sleep Medicine (WASM) / Sleep Medicine 10, Suppl. 2 (2009) S1–S83 S19 Objectives: The aim of this study was to test the hypothesis that ongoing hyperarousal processes during sleep, as indexed by EEG beta activity, mediates previously shown deficits in memory consolidation during sleep in patients with primary insomnia. Methods: General neurocognitive and memory performance (procedural mirror-tracing task, declarative visual and verbal learning task) was assessed before and after one night of polysomnographic monitoring in 18 patients with primary insomnia (7 men, aged 45.5±4.5 years) and 34 sex, age and IQ matched healthy subjects. EEG spectral analysis was performed for epochs of NREM sleep. Results: Insomnia patients showed significantly increased EEG beta activity during NREM sleep (16-32 Hz), decreased overnight procedural memory consolidation (MANOVA, p 36 hours), and/or focused solely on global assessments of negative mood. Objective: To determine the effects of experimental sleep restriction and sleep disruption, two common and distinct forms of sleep loss, on negative and positive mood. Methods: Fifty-three healthy adults (62.2% female, age 26±6.1) spent five consecutive nights on an inpatient research unit. After a baseline sleep night, participantswere randomized to one of three conditions: Forced Awakenings (FA), Restricted Sleep (RS), or Control (C) for the following three nights. FA underwent one forced awakening each hour for an 8-hour sleep opportunity (seven random 20-min. and one 60-min. awakening). RS slept undisturbed for no more than 280 mins. C slept undisturbed for 8 hrs. The Profile of Mood States (POMS) was given prior to bedtime each night. POMS total negative (POMS-N) and positive (POMS-P) mood scales were utilized by aggregating responses to negative and positive valence items. Results: Independent 3 Condition (C, FA, RS) × 4 Time (Baseline, Night 1, Night 2, Night 3) mixed-model ANOVAs were conducted with POMS-N and POMS-P as DVs. Follow-up single-df comparisons focused on changes in POMS-N and POMS-P across nights for RS vs. Control and FA vs. Control. The omnibus Condition × Time effect was significant (p<0.05) for POMS-N and POMS-P. Comparisons for POMS-N revealed that FA and RS yielded greater increases in POMS-N than C (p<0.05). For POMS-P, FA yielded greater decreases in POMS-P than C (p<0.05). A post hoc comparison showed that FA tended to yield greater decreases in POMS-P than RS (p=0.06). Both RS and FA demonstrated equivalent total sleep time. Conclusion: Partial sleep loss by disruption or curtailment has detrimental effects on mood. Moreover, fragmented sleep contributed to greater decrements in positive mood than restricted sleep. These data have important implications for the well-being of individuals who have repeated awakenings or voluntarily restrict their sleep. The mechanisms of these effects, mental health consequences, and the novel finding that fragmented (but not restricted) sleep contributes robustly to anhedonia require additional inquiry. Support: NINDS R21NS051771 (MTS) and MH075884 (PJQ) 070 BODY MASS INDEX, EATING ATTITUDES AND SLEEP DISTURBANCE IN YOUNG ADULTS M.J. Soares1, S.C. Bos1, B. Maia1, A.T. Pereira1, M. Marques1, J. Valente1, A.A. Gomes2, A. Macedo1, M.H. Azevedo1. 1Institute of Medical Psychology, Faculty of Medicine, University of Coimbra, Coimbra, Portugal; 2Educational Sciences Department, University of Aveiro, Aveiro, Portugal Objective: To examine the association between body mass index, eating attitudes, and self-reported sleep disturbances in university students. Methods: Self-reported current body weight and height were used to calculate BMI (BMI; KG/M2). The Eating Attitudes Test-40 (EAT40 Garner & Garfinkel, 1979) was used to measure eating attitudes and behaviors (Diet Concerns, DC; Bulimic Behaviors, BB, Social Pressure to Eat, EAT). Sleep disturbance was assessed with two items: (1) I have difficulty in falling asleep -DIS and (2) I wake up many times during the night -DMS. Each item was scored on a 5-point scale ranging from 0 (never) to 5 (always). An overall index of sleep disturbance score (SDI) was calculated from the sum of the scores of the individual items. The sample consisted of 544 female (M age=19.5 years, SD=1.56) and 326 male (M age=19.8 years, SD=1.68) undergraduate students. Results: In the total sample, subjectswith a BMI lower than 19 or between 19 and 22 had significantly more difficulties initiating sleep (M=1.83; SD=1.04; p<.05 and M=1.77; SD=1.07; p<.05, respectively) than subjects with a BMI higher than 24 (M=1.50; SD=1.04). Eating disorder behavior (DC, BB, SPE and EAT total score) was positively and significantly associated with DIS, DMS and SDI (values ranging from r=0.09 to r=0.21; p<0.05). The exception was that DC was not associated to sleep measures in males, while in females it was not associated to DMS. Insomniacs (subjects who often, very often or