The why, when and where of selective attention to sleep in psychophysiological insomnia

Psychophysiological Insomnia (PI) is characterized by morbid fear of insomnia, mental arousal and heightened somatic tension in bed (ICSD- 2). The most widely reported epidemiological study reports a prevalence of insomnia as between 9% and 15% (Ohayon 2002) and most studies conducted to date looking at prevalence report similar numbers. Espie and colleagues (2006) in their review paper outlined a pathway into PI with three main components; selective attention to sleep, explicit intention to sleep and sleep effort. This model moves forward from Espie’s (2002) Psychobiological Inhibition Model of insomnia which considered both the psychological and physiological states of normal sleep and how these are affected in insomnia. The Attention-Intention-Effort (A-I-E) Model further addresses the loss of automaticity and flexibility in insomnia but moves towards specifically outlining the processes which are present in PI as compared to the good sleeper (GS). The first step in the A-I-E is selective attention to sleep, more commonly reported as attention bias to sleep. The University of Glasgow have pioneered the work establishing this attention bias towards sleep as an indicator in insomnia by using several cognitive probe paradigms presenting neutral and sleep related words and images to PI and GS (Jones et al 2005, Marchetti et al 2006, MacMahon et al 2006, Woods et al 2009). The various paradigms applied to understanding attention bias in PI have confirmed that PI will selectively attend to sleep related stimuli compared to neutral and to GS. We now find ourselves at the juncture of wanting to further understand the underlying mechanism to this attention bias as the previous research has mainly attributed it to an anxiety provoked response. This has its basis in Harvey’s (2002) cognitive model of insomnia which makes comparisons with insomnia and anxiety disorder as well as the absence of de-arousal, both physiological and cognitive, in insomnia as outlined in Espie’s (2002) Psychobiological Inhibition Model. This thesis aims to further our understanding and answer our questions regarding the underlying mechanisms of attention bias in insomnia by addressing the time course, specificity and valence of attention bias in insomnia. Four experiments are used to address these three factors. Firstly, the specificity of AB is examined and compares the performance of GS and those going through a period of acute insomnia on a modified pictorial Posner paradigm in Experiment 1. Experiments 2 and 3 move on from Woods et al (2009) looking at AB to sleep and day times presented on an alarm clock using another modified Posner paradigm. By adding day times into the experiment and adjusting the presentation time of the salient stimuli we address the time course and valence questions. Finally, in experiment 4, an eye tracking experiment, which is new to insomnia research, has been developed where positive sleep, negative sleep and neutral words are presented to PI and GS. . This definitive experiment addresses factors of time course and valence by experimentally manipulating the saliency of the stimuli presented as well as monitoring over a continuous period of presentation. Overall, the findings of this thesis confirm that PI will selectively attend to salient stimuli at shorter presentation times but this attention bias changes into a performance deficit as presentation time increases. This prompts consideration on how the nature of the tasks are exposing elusive performance impairments in insomnia. Also, the saliency of stimuli representing the day presented to PI opens discussion into the 24 hour nature of PI