This thesis investigated positive and negative sleep appraisals for excessively long and short sleep duration disturbances commonly experienced by those with bipolar spectrum mood swings. Chapter 1 provides a background of the literature and the proposal of the Integrative Cognitive Sleep Model (ICSM). The ICSM is a subset of the Integrative Cognitive Model and proposes that positive and negative sleep appraisals for excessively long and short sleep durations play a key role in the development and maintenance of insomnia, hypersomnia and reduced need for sleep. Specifically, endorsing more of these conflicting appraisals will drive sleep fluctuation. Chapter 1 outlines the aims and hypotheses designed to test this new model while Chapter 2 is a review of the methodologies undertaken in order to achieve the aims and hypotheses. Study 1 (Chapter 3) provides a scoping review of the literature to identify the available sleep cognition measures. This study highlighted a significant gap in the knowledge base since no measures were identified for hypersomnia or reduced need for sleep. Study 2 (Chapter 4) is a Delphi method study with research and clinical professionals in the field of BD to generate appraisal statements for a new measure that will aid in testing the proposed ICSM. The statements generated in Study 2 informed the new Positive and Negative Sleep Appraisal Measure (PANSAM). Study 3 (Chapter 5) is a series of validity and reliability statistical tests to establish initial psychometric robustness for the PANSAM. The results highlighted four expected subscales: positive and negative appraisals for both short and long sleep durations. Study 3 also evidenced good convergent validity with validated measures for mood, sleep, and mood and sleep cognitions. Study 4 (Chapter 6) was a clinical study with participants who met criteria for either bipolar spectrum, unipolar depression or were non-clinical. It was hypothesised the PANSAM would discriminate between all three participant groups, with higher scores for those in the bipolar spectrum group due to their increased vulnerability to mood and sleep disturbances. This hypothesis was only partially supported since there was no statistically significant difference between the clinical groups, but both had statistically significantly higher PANSAM scores than the non-clinical control group. Finally, the ICSM was tested more directly in several different ways. Study 3 (Chapter 5) conducted a hierarchical regression to test the incremental validity of the PANSAM on subjective poor sleep quality, over and above age, gender, bipolar tendency and a commonly used sleep cognition measure for insomnia. The results highlighted the subscale for positive appraisals for sleeping more (representing sleep as a safety behaviour) was significant. To test the ICSM more robustly, Study 5 (Chapter 7) was devised to test the hypothesis that the PANSAM would predict subjective (sleep diary) and objective (actigraphy) total sleep time variability. Additionally, a second hierarchical regression was conducted in this study to test the incremental validity of the PANSAM on the sleep variability variables over and above bipolar tendency and the same commonly used sleep cognition measure for insomnia. The results of Study 5 (Chapter 7) showed the PANSAM was statistically significantly predictive of objective total sleep time variability only but did not uphold incrementally. The findings from these studies suggest the PANSAM offers unique contributions to sleep cognition research. Clinical implications and future research ideas are discussed.