The diathesis-stress model
proposed by Spielman and colleagues [8
], more commonly known as the “3-P” model, describes predisposing, precipitating, and perpetuating factors relevant to the development and maintenance of insomnia. Predisposing factors
include genetic, physiological, or psychological diatheses that confer differential susceptibility to individuals. Precipitating factors
include physiological, environmental, or psychological stressors which push an individual over a hypothetical insomnia threshold to produce acute symptoms. Perpetuating factors
include behavioral, psychological, environmental, and physiological factors that prevent the individual from re-establishing normal sleep. Most attention has focused on behavioral patterns adopted in an attempt to relieve insomnia symptoms, but which inadvertently worsen those symptoms. For example, many individuals with insomnia attempt to compensate for reduced sleep by spending increased time in bed, which may have the unintended effect of further fragmenting sleep. The 3-P model is useful heuristically and clinically to identify potential treatment targets. It permits an integrated and flexible view of insomnia, including factors ranging from genetic/familial predisposition to medical/psychiatric illnesses and volitional behaviors. However, its very breadth can also be regarded as a limitation. The 3-P model applies equally well to a wide variety of psychological-behavioral conditions besides insomnia; it does not specify neurobiological substrates or mechanisms; and its predictions are general.
The stimulus control model
proposed by Bootzin [9
] is based on classical conditioning principles. Sleep is viewed, in part, a conditioned response to the stimulus of the sleep environment. In insomnia, the bed/ sleep environment instead become stimuli for increased arousal, frustration, and wakefulness. This model forms the basis of stimulus control therapy, which has shown consistent efficacy for chronic insomnia [10
]. The therapy attempts to re-establish the sleep environment as a stimulus for sleep by restricting other activities in bed (e.g., lying awake, watching TV), and ensuring that sleep occurs only in bed. The stimulus control model is heuristically useful, is based on well-established principles, and logically leads to a treatment approach. Animal models based on classical conditioning, somewhat surprisingly, have not been exploited. Like the 3-P model, the stimulus control model does not specifically address the neurobiological substrate of insomnia.
of insomnia focus on thoughts, feelings, and beliefs that may interfere with sleep and lead to maladaptive behavioral patterns. Harvey [11
] has proposed that insomnia results from inappropriate worry about poor sleep and its daytime effects, which leads to increased physiological and psychological arousal, selective scanning of the internal milieu and environment for threat cues, and the development of counter-productive “safety behaviors” designed to maximize sleep and minimize the consequences of insomnia. A strength of cognitive models is that they lead to hypotheses that can be tested with specific therapeutic interventions directed at distorted thoughts or maladaptive behaviors. Despite their clinical utility, however, there is limited evidence to date that cognitive interventions alone are efficacious for the treatment of insomnia, or that changes in cognitions are necessary for improvement. Another relative weakness is that it is difficult to know whether specific thought patterns precede or follow the development of insomnia. If maladaptive cognitions arise from insomnia, then improving sleep directly (for example, with medications) should also be efficacious. Like more behaviorally-oriented models, cognitive models do not propose specific neural substrates.
Another cognitively-based model, the Psychobiological Inhibition Model
], addresses the role of selective attention in the development and maintenance of insomnia. Psychological and/or physiological stress is posited to lead to selective attention toward stressors, and inhibition of the “de-arousal” that normally accompanies sleep. Inappropriate arousal may then lead to selective attention to sleep-related cues (implicit or explicit) and increased explicit intention and effort to sleep, which further inhibit normal sleep-related de-arousal. The psychobiological inhibition model has good face validity and leads to specific, testable hypotheses regarding selective attention to sleep-related cues, which have received some support from empirical studies [13
]. Limitations include the difficulty of measuring intention and effort in a comparative sense, and the lack of specification of neural substrates.
Finally, the neurocognitive model
] builds on the diathesis-stress (3-P) model described above, but integrates neurobiological and neurophysiological observations. Specifically, the neurocognitive model proposes that insomnia leads to conditioned cortical arousal, manifest as increased high-frequency (beta and gamma) EEG activity during sleep. High-frequency EEG activity is thought to be associated with enhanced sensory processing, memory formation, and conscious perception. Thus, high-frequency EEG activity may underlie the phenomenon of “sleep state misperception” or “paradoxical insomnia,” in which subjective wakefulness is greater than that measured by concurrent PSG. Strengths of the neurocognitive model include good face validity and the incorporation of electrophysiological, neurocognitive, and clinical findings in chronic insomnia. For instance, high frequency EEG activity during non-rapid eye movement (NREM) sleep has been associated with subjective-objective sleep mismatch in some, if not all, studies [16
]. The model is still relatively nonspecific with regard to neural structures or circuits involved.