Sleep state misperception (SSM), historically termed paradoxical insomnia, is characterized by a marked discrepancy between subjective reports of little or no sleep and objective evidence of relatively preserved sleep. Traditional models emphasize hyperarousal, cognitive bias, and sleep–wake boundary instability, yet these frameworks do not fully explain why physiologically valid sleep is often experienced as sustained wakefulness.

This paper advances a conceptual reframing of SSM as a disturbance of perceptual inference grounded in predictive processing theory. Within this framework, the experience of sleep and wakefulness is understood as an inferential construction shaped by prior expectations, precision weighting of internal signals, attentional monitoring, and memory-based reconstruction. Under conditions of low sensory precision, such as light or transitional sleep, top-down expectations increasingly guide state classification. Hyperarousal amplifies internally generated signals, while monitoring behaviors bias sampling toward wake-like experiences. Retrospective memory further consolidates this bias, producing a coherent but inaccurate percept of wakefulness.

The model integrates clinical phenomenology, neurophysiological findings, and cognitive-behavioral mechanisms within a unified account and generates testable hypotheses regarding interoceptive precision, perceptual thresholds, and resistance to belief updating. Clinical implications include reframing treatment targets from sleep generation to sleep interpretation, with adaptations of cognitive behavioral therapy for insomnia aimed at reducing monitoring, modifying maladaptive priors, and increasing tolerance for ambiguity.

Conceptualizing SSM as a disorder of perceptual inference situates insomnia within contemporary theories of perception and provides a framework for understanding subjective–objective discrepancy across clinical contexts.

• Sleep State Misperception

• Perceptual Inference

• Insomnia

• Clinical Implications

Forman BD (2026) Sleep State Misperception as a Disorder of Perceptual Inference: A Predictive Processing Framework for Insomnia Phe nomenology and Treatment. Ann Psychiatry Ment Health 14(1): 1215.

Insomnia disorder is common, clinically consequential, and heterogeneous in both experience and biology [1,2]. Many individuals present with prolonged sleep onset, repeated nocturnal awakenings, and daytime fatigue. Yet a clinically important subset report an experience that is more perplexing than typical insomnia: they insist they have slept little or not at all, sometimes citing “zero hours” of sleep, even when polysomnography (PSG) or actigraphy demonstrates relatively preserved total sleep time and sleep architecture. This phenomenon is commonly referred to as sleep state misperception (SSM), historically termed paradoxical insomnia [3,4].

The clinical paradox is not that patients are exaggerating or misunderstanding their experience. Rather, the paradox is that sleep can occur physiologically without being registered as sleep subjectively. Individuals with SSM frequently report continuous awareness, persistent thinking, and a sense of having been awake throughout the night, while objective measures identify both non-rapid eye movement (NREM; light and deep sleep stages) and rapid eye movement (REM) sleep [3,5]. This discrepancy is clinically significant and is associated with heightened distress, repeated reassurance-seeking, increased diagnostic utilization, and often limited response to interventions that target sleep duration alone [5,6].

Modern behavioral sleep medicine emphasizes that insomnia is not solely a disorder of sleep generation, but also a disorder of how sleep is attended to, interpreted, and remembered [7,8]. In most individuals, sleep occurs with minimal monitoring and limited expectation of accurate recall. In SSM, by contrast, sleep becomes a monitored performance. The individual attends closely to internal states, detects moments of awareness, and uses these moments as evidence of wakefulness. Because consolidated sleep is largely amnestic, the resulting internal “data set” becomes biased toward remembered wake-like experiences [9]. Over time, this biased sampling produces a stable and compelling conclusion: “I was awake.”

Several models have attempted to account for this phenomenon. Hyperarousal models emphasize elevated cognitive and physiological activation that alters the subjective texture of sleep [10,11]. Cognitive models highlight worry, selective attention, and maladaptive beliefs about sleep [7]. Other accounts focus on instability at the sleep–wake boundary, suggesting that transitional states may be experienced as wakefulness due to preserved awareness [9]. While these frameworks identify important contributing processes, they do not fully explain why physiologically valid sleep is consistently experienced as sustained wakefulness, nor why this perception remains stable even when contradicted by objective data.

A central limitation of existing models is that they do not explicitly address how subjective experience is constructed under conditions of ambiguity. Sleep, particularly lighter stages and transitional states, is characterized by reduced sensory precision, diminished environmental input, and variable degrees of awareness. Under such conditions, the brain must infer whether the current state represents sleep or wakefulness. The discrepancy observed in SSM suggests that this inferential process may be systematically biased.

Contemporary theories of perception, particularly predictive processing frameworks, offer a powerful lens through which to understand this phenomenon. These models conceptualize perception as an active inferential process in which prior expectations and incoming signals are integrated to generate experience [12-14]. Perception is not a direct readout of physiological state, but a best-guess inference about the causes of sensory input. When sensory signals are weak or ambiguous, as they are during sleep, top-down expectations exert greater influence [15].

From this perspective, the experience of being awake or asleep is not simply detected, but inferred. Strong priors about poor sleep, heightened attention to internal signals, and biased memory processes may combine to produce a coherent but inaccurate percept of wakefulness. Importantly, because this percept arises from the same inferential processes that underlie all perception, it is experienced as real and is often resistant to correction [14].

The present paper advances a predictive processing framework for sleep state misperception, positioning it as a disorder of perceptual inference. This approach integrates clinical phenomenology, neurophysiological findings, and cognitive-behavioral models within a unified account. It further generates testable hypotheses regarding interoceptive precision, perceptual thresholds, and resistance to belief updating. Finally, it outlines clinical implications, suggesting that treatment may benefit from targeting perceptual processes such as monitoring, expectation, and tolerance for ambiguity alongside traditional behavioral interventions.

By situating SSM within a broader science of perception, this framework reframes subjective–objective sleep discrepancy not simply as a problem of measurement, but as an instance of how the brain constructs experience under conditions of uncertainty.

This paper employs a narrative conceptual review with integrative synthesis, designed to bring together empirical findings from sleep medicine with contemporary theoretical frameworks from cognitive neuroscience and perception science. This approach is appropriate given the heterogeneity of the sleep state misperception (SSM) literature, including variability in definitions, measurement methods, and study populations.

Across studies, SSM has been operationalized using differing thresholds for subjective–objective discrepancy, including absolute and relative differences in total sleep time, sleep onset latency, and wake after sleep onset. Measurement approaches also vary, with studies relying on polysomnography (PSG), actigraphy, sleep diaries, and retrospective self-report. In addition, samples differ across primary insomnia, comorbid insomnia, and non-clinical populations. These sources of variability limit the feasibility of formal meta-analysis and instead support the use of interpretive synthesis aimed at mechanistic understanding.

Narrative and integrative review approaches are widely used in behavioral sleep medicine when the goal is to synthesize across heterogeneous findings and develop conceptual models that cut across levels of analysis [1,8,2]. The present approach is consistent with these aims and is particularly suited to integrating clinical, neurophysiological, and perceptual frameworks.

A structured literature search was conducted using PubMed, PsycINFO, and Web of Science, covering publications from January 1995 through January 2026, with particular emphasis on studies published within the past decade.

Search terms included combinations of:

• “sleep state misperception”

• “paradoxical insomnia”

• “subjective objective sleep discrepancy”

• “sleep perception” • “insomnia phenotype”

• “hyperarousal insomnia” • “sleep monitoring”

• “interoception” AND “insomnia”

• “predictive processing” OR “predictive coding” AND “perception”

Additional sources were identified through:

• Reference list review of key articles and reviews

• Citation tracking of foundational papers

• Inclusion of seminal theoretical works in perception and cognitive neuroscience

Priority was given to:

1. Peer-reviewed empirical studies directly examining subjective–objective sleep discrepancy

2. Foundational theoretical models relevant to insomnia and perception

3. Recent reviews and meta-analyses that synthesize current knowledge

4. Translational work linking basic mechanisms to clinical phenomena

Inclusion Criteria and Evidence Prioritization

Studies were included if they met one or more of the following criteria:

• Direct examination of sleep state misperception or paradoxical insomnia

• Investigation of subjective–objective discrepancies in sleep

• Analysis of cognitive, attentional, or behavioral processes in insomnia

• Examination of neurophysiological correlates of sleep perception (e.g., EEG activity, arousal indices)

• Evaluation of behavioral or psychological treatments relevant to insomnia and sleep perception

In addition, the review incorporated theoretical and computational neuroscience frameworks, particularly predictive processing models [12-14], where these provided explanatory value for the observed phenomena.

Excluded were:

• Studies lacking relevance to subjective sleep experience

• Non-peer-reviewed sources without theoretical or empirical contribution

• Redundant publications without novel insight

Conceptual Integration Strategy

The primary aim of this review is conceptual integration rather than comprehensive enumeration of findings. To achieve this, the literature was organized across five domains:

1. Clinical Phenomenology Subjective experience of sleep and wakefulness, including reports of continuous awareness and perceived absence of sleep

2. Measurement and Discrepancy Differences between subjective reports and objective indices (PSG, actigraphy), and methodological influences on these discrepancies

3. Neurophysiological Processes Hyperarousal, sleep–wake boundary instability, and wake-like neural activity during sleep

4. Cognitive–Behavioral Mechanisms Monitoring, attentional bias, expectancy effects, and memory reconstruction

Predictive processing, precision weighting, and interoceptive inference Rather than treating these domains independently, this review identifies converging mechanisms across levels of analysis. For example, hyperarousal is considered not only as a physiological state, but also as a factor influencing the precision and salience of internally generated signals, thereby shaping perceptual inference.

Integration of Predictive Processing and Interoceptive Frameworks

To extend existing models of insomnia, this review incorporates predictive processing frameworks, which conceptualize perception as an inferential process shaped by prior expectations and sensory input [12-14].

This framework is particularly relevant to SSM for several reasons:

• Sleep occurs under conditions of reduced and ambiguous sensory input

• Internal signals (e.g., cognitive activity, bodily sensations) play a dominant role

• Classification of sleep versus wakefulness requires interpretation rather than direct detection

In parallel, research on interoception suggests that perception of internal bodily states is itself constructed through inferential processes [15,16]. These models provide a mechanism for understanding how internal signals associated with sleep may be interpreted as evidence of wakefulness.

By integrating these frameworks with clinical and empirical findings, the present review develops a mechanistic account of sleep state misperception as a disorder of perceptual inference.

Although this review does not follow formal systematic review procedures, its structure was informed by principles of transparent reporting. Search strategies, inclusion criteria, and conceptual organization are described explicitly to allow evaluation of the review process.

This approach reflects current best practices for conceptual and translational reviews, particularly in interdisciplinary areas where theoretical integration is a primary objective.

Use of Generative Tools

Generative AI tools were used to assist with language refinement and organizational clarity. All conceptual content, interpretation of findings, and selection and verification of references were conducted by the author. Care was taken to ensure that all cited sources correspond to authentic, peer-reviewed publications.

Clinical Phenomenology of Sleep State Misperception

Sleep state misperception (SSM) is characterized by a marked and often striking discrepancy between subjective reports of sleep and objective indices derived from polysomnography (PSG) or actigraphy. Individuals with SSM frequently report sleeping very little or not at all, sometimes describing “zero hours” of sleep, despite objective evidence indicating substantial total sleep time and relatively preserved sleep architecture [3,5]. In some cases, objective sleep parameters fall within normative ranges, further underscoring the divergence between subjective experience and physiological measurement.

The subjective experience of SSM differs qualitatively from that of typical insomnia. Rather than reporting fragmented or restless sleep, individuals often describe a continuous sense of awareness across the night. This may include ongoing cognitive activity, environmental monitoring, or a persistent sense of “being awake,” even during periods later identified as sleep. These reports are often internally consistent and accompanied by high confidence, contributing to the conviction that sleep is absent or severely impaired.

A defining feature of SSM is the stability of this perceptual conviction. Individuals frequently maintain their belief that they were awake despite exposure to contradictory evidence, including PSG recordings demonstrating substantial sleep. This resistance to updating is not adequately explained by misunderstanding or denial. Rather, it reflects the compelling nature of the subjective experience itself, which is coherent, memorable, and reinforced over time.

The phenomenology of SSM is also shaped by selective attention and monitoring. Many individuals report actively or implicitly tracking their internal state during the night, evaluating whether they are asleep, estimating time elapsed, or attending to bodily sensations and cognitive activity [7,17]. This monitoring increases the likelihood that transient awakenings, microarousals, or periods of light sleep are detected and encoded. Over time, this leads to an overrepresentation of wake-like experiences within the individual’s subjective account of the night.

Recent work has further highlighted the role of attentional bias toward internal signals in insomnia, with evidence suggesting that individuals with insomnia exhibit heightened sensitivity to sleep-related cues and internal states [18,19]. While these processes are not unique to SSM, their amplification may contribute to the persistent discrepancy observed in this subgroup.

Memory processes also play a central role. Sleep, particularly deeper stages, is associated with reduced encoding of experience, whereas wakefulness is associated with continuous awareness and memory formation. As a result, retrospective judgments of sleep are constructed from a dataset that is inherently incomplete and biased toward remembered periods of wakefulness [9]. In SSM, this bias appears to be magnified, with individuals relying heavily on remembered experiences as evidence for their conclusions about the night as a whole.

This asymmetry between experienced and remembered states contributes to a reconstructive process in which the absence of memory is interpreted as the absence of sleep. In other words, individuals may equate “not remembering sleep” with “not having slept,” a conclusion that is both intuitively compelling and empirically inaccurate.

Another characteristic feature of SSM is the presence of stringent criteria for what counts as sleep. Many individuals implicitly define sleep as a state devoid of awareness, thought, or sensation. Under this definition, any detectable mental activity is interpreted as evidence of wakefulness. Given that lighter stages of sleep frequently include residual cognitive activity, this criterion systematically biases classification toward wakefulness.

This phenomenon has been described in related work on sleep perception thresholds, suggesting that individuals vary in the criteria they use to label a state as sleep [6]. In SSM, this threshold appears to be elevated, requiring deeper or more discontinuous states before sleep is acknowledged subjectively.

Recent research has also begun to examine the role of interoceptive awareness in insomnia and sleep perception. Emerging findings suggest that individuals with insomnia may exhibit altered sensitivity to internal bodily signals, which in turn shapes subjective experience [20,21]. Although this literature is still developing, it supports the notion that internal signal processing is a critical component of sleep perception.

Clinically, SSM is associated with significant distress and functional impairment. Individuals often report frustration, anxiety, and a sense of loss of control related to their perceived inability to sleep. This may lead to repeated reassurance-seeking, increased reliance on sleep monitoring technologies, and pursuit of multiple diagnostic evaluations. In some cases, the discrepancy between subjective experience and objective findings can strain the therapeutic alliance, particularly when patients feel that their experience is being invalidated.

At the same time, SSM may not always be associated with the degree of daytime impairment expected given reported sleep loss. Some individuals who report “no sleep” demonstrate relatively preserved daytime functioning, suggesting that objective sleep may be sufficient despite subjective perception to the contrary. This dissociation further complicates clinical assessment and underscores the importance of understanding the mechanisms underlying subjective experience.

Taken together, the phenomenology of SSM is best understood not as a simple misestimation of sleep duration, but as a systematic pattern of perception, attention, and memory that produces a coherent but inaccurate representation of the night. This pattern includes persistent awareness, selective encoding of wake-like experiences, elevated thresholds for classifying sleep, and resistance to updating in the face of contradictory evidence.

These features suggest that SSM reflects a disturbance not only in sleep itself, but in the processes through which sleep is perceived and interpreted. Understanding these processes is essential for developing more comprehensive models of insomnia and provides a foundation for the application of predictive processing frameworks to this phenomenon.

The defining feature of sleep state misperception is the discrepancy between subjective reports of sleep and objective indices derived from polysomnography (PSG) or actigraphy. This discrepancy has been documented across multiple decades of research and remains one of the most clinically and theoretically significant aspects of insomnia [3,5].

Polysomnography is considered the gold standard for assessing sleep architecture, providing detailed information regarding sleep stages, continuity, and physiological markers such as electroencephalographic (EEG) activity, muscle tone, and eye movements. In individuals with SSM, PSG frequently reveals substantial total sleep time, often several hours greater than patient estimates, with sleep efficiency that may fall within normal or near-normal ranges [3,22]. These findings highlight that the discrepancy is not simply a matter of mild misestimation, but can represent a profound divergence between subjective experience and physiological state.

Actigraphy provides a complementary method for assessing sleep–wake patterns over extended periods in naturalistic settings. Although less precise than PSG in differentiating sleep stages, actigraphy consistently demonstrates that individuals with insomnia tend to underestimate total sleep time and overestimate wakefulness relative to objective indices [5]. Importantly, discrepancies observed via actigraphy often persist across multiple nights, suggesting that misperception is not limited to isolated laboratory conditions.

A key issue in interpreting these findings is that objective and subjective measures assess fundamentally different aspects of sleep. PSG defines sleep based on neurophysiological criteria, whereas subjective reports rely on conscious awareness and retrospective judgment. These represent distinct levels of analysis, each valid within its own domain but not necessarily expected to align perfectly.

This divergence becomes especially pronounced during lighter stages of sleep. Stage N1 sleep, which marks the transition between wakefulness and sleep, is characterized by low-amplitude EEG activity, reduced but not absent awareness, and increased susceptibility to environmental or internal stimuli. Even portions of stage N2 sleep may include ongoing cognitive activity or intermittent awareness. From a physiological perspective, these states are classified as sleep; from a subjective perspective, they may be experienced as wakefulness [9].

This mismatch underscores a central limitation of equating physiological sleep with perceived sleep. Subjective experience is not a direct reflection of EEG-defined states but is instead shaped by awareness, attention, and memory. As a result, discrepancies between subjective and objective measures are not inherently pathological but reflect differences in how sleep is defined and accessed.

Several explanatory models have been proposed to account for this discrepancy. One class of explanations emphasizes measurement limitations, suggesting that standard PSG criteria may not fully capture the subjective qualities of sleep. For example, the presence of wake-like cognitive activity during sleep may lead individuals to classify these periods as wakefulness, even when neurophysiological markers indicate sleep.

A second class of explanations focuses on attentional and cognitive processes. Individuals with insomnia may exhibit heightened monitoring of internal states, increasing the likelihood that brief awakenings or transitional states are detected and encoded [7]. This selective attention results in a disproportionate representation of wake-like experiences within subjective accounts.

A third explanation highlights memory processes. Sleep, particularly deeper stages, is associated with reduced encoding of experience, whereas wakefulness is associated with continuous awareness and memory formation. As a result, retrospective estimates of sleep are constructed from an inherently biased dataset in which wakefulness is overrepresented [9]. This bias is further amplified in SSM, where individuals rely heavily on remembered experiences as evidence for their conclusions about the night as a whole.

More recent work has emphasized the role of neurophysiological hyperarousal in shaping subjective sleep experience. Elevated high-frequency EEG activity during sleep, particularly in the beta and gamma ranges, has been associated with increased cortical activation and may contribute to the experience of wake-like awareness despite physiological sleep [11,10]. This “wake-like” neural signature provides a potential bridge between physiological and experiential accounts of insomnia.

However, while these models account for important components of the discrepancy, they do not fully explain why the resulting perception is often stable, compelling, and resistant to change. Individuals with SSM frequently maintain their perception of wakefulness even when presented with objective evidence to the contrary. This suggests that the discrepancy is not simply a failure of measurement, attention, or memory in isolation, but reflects a more fundamental process governing how sleep-related information is interpreted.

From a perceptual standpoint, the critical issue is not merely the presence of discrepancy, but how that discrepancy is resolved. When faced with ambiguous or conflicting information, the brain must determine whether the current state represents sleep or wakefulness. The persistence of misperception suggests that this process of state classification is systematically biased.

This observation points toward the need for a framework that can account for how subjective experience is constructed under conditions of uncertainty, rather than focusing solely on the accuracy of measurement. Predictive processing models of perception provide such a framework, offering a mechanistic account of how prior expectations, signal precision, and interpretive processes shape experience.

Predictive processing frameworks conceptualize perception as an active inferential process in which the brain continuously generates predictions about incoming sensory input and updates those predictions based on incoming signals [12-14]. Within this model, perception reflects the brain’s best estimate of the causes of sensory input rather than a direct representation of physiological or external reality. Perceptual experience emerges from the dynamic interaction between top-down expectations (priors) and bottom-up sensory evidence, with discrepancies between prediction and input (prediction errors) driving learning and updating.

Applied to sleep, this framework suggests that the experience of being awake or asleep is not a direct readout of neurophysiological state, but an inferential construction based on ambiguous internal signals. Sleep—particularly lighter stages and transitions—is characterized by reduced sensory precision, diminished environmental input, and variable degrees of awareness [9]. Under these conditions, the brain must infer whether the current state represents sleep or wakefulness, relying more heavily on internal models when sensory input is weak or uncertain [13].

A central principle of predictive processing is the minimization of prediction error [12]. In typical perceptual systems, discrepancies between expected and incoming signals lead to updating of priors, resulting in increasingly accurate representations. In sleep state misperception, however, this updating process appears to be altered.

Rather than revising the prior expectation that “I am awake,” ambiguous or conflicting signals may be reinterpreted in a manner consistent with that expectation. In this sense, the system resolves uncertainty by maintaining coherence rather than maximizing accuracy [14]. Prediction errors are minimized not through belief updating, but through biased interpretation of incoming data.

This mechanism helps explain the persistence of SSM despite contradictory evidence. Objective measures such as polysomnography may demonstrate substantial sleep, yet this information remains external to the ongoing inferential process. Because subjective experience is constructed from internally weighted signals and entrenched priors, external feedback may fail to generate sufficient prediction error to drive updating [14].

From a computational perspective, this pattern reflects a shift in the balance between priors and sensory evidence, with priors exerting disproportionate influence over perception. Such dynamics are consistent with predictive processing accounts of other conditions in which perceptual beliefs remain stable despite contradictory input, particularly under conditions of uncertainty.

Priors: Expectation as a Constraint on Perception

Within predictive processing models, priors function as constraints on perception, shaping how incoming signals are interpreted and determining the range of plausible perceptual states [14]. In SSM, individuals often develop strong priors such as “I do not sleep” or “I am awake most of the night,” consistent with cognitive models of insomnia that emphasize maladaptive beliefs and expectations [6,7].

These priors may arise from repeated experiences of perceived poor sleep, heightened concern about sleep, or prior episodes of insomnia. Over time, they become increasingly entrenched and begin to guide interpretation of ambiguous internal states. Transitional sleep, brief awakenings, or ongoing cognitive activity may all be interpreted as evidence of wakefulness because they are consistent with the prior expectation.

Importantly, priors influence perception at a prereflective level. The resulting experience is not interpreted as an inference but is experienced as reality. This helps explain why individuals with SSM often express high confidence in their reports, even when those reports conflict with objective data.

Predictive processing accounts emphasize not only the content of priors and sensory input, but also their relative precision, or weighting. Precision reflects the confidence assigned to a given signal and determines its influence on perceptual inference [12].

In SSM, hyperarousal may alter this balance. Elevated cognitive activity, physiological activation, and vigilance increase the salience of internally generated signals, including thoughts, bodily sensations, and moments of awareness [10,11]. These signals are treated as reliable indicators of state, even when they occur during physiological sleep.

At the same time, sensory input associated with sleep is inherently ambiguous and assigned lower precision. The combination of high-precision internal signals and low-precision sensory input biases the system toward interpreting the state as wakefulness. In effect, the perceptual system becomes tuned to detect and prioritize wake-like features.

Neurophysiological findings support this account. Elevated high-frequency EEG activity during sleep has been observed in individuals with insomnia and is thought to reflect persistent cortical activation [10,11]. This “wake-like” neural signature may contribute to the salience of internal signals and reinforce the perception of being awake.

Interoceptive Inference and Internal Signal Processing

The concept of interoceptive inference extends predictive processing to the domain of internal bodily states. Rather than directly sensing internal conditions, the brain constructs interoceptive experience through predictive integration of signals and expectations [15,16].

This framework is particularly relevant to sleep, which lacks a single discrete sensory signal indicating its presence. Instead, sleep involves gradual changes in neural activity, responsiveness, and awareness. Residual cognitive activity, bodily sensations, or brief arousals may be interpreted as evidence of wakefulness, particularly when these signals are assigned high precision.

Heightened interoceptive attention may amplify this process. Individuals who are highly attuned to internal states may detect subtle fluctuations that would otherwise go unnoticed. When combined with strong priors about poor sleep, these signals are more likely to be interpreted as confirming wakefulness. This creates a feedback loop in which increased attention leads to increased detection, reinforcing the prior expectation.

This account also helps explain individual differences in susceptibility to SSM. Not all individuals with insomnia develop persistent misperception, suggesting that vulnerability may depend on how internal signals are processed and interpreted rather than on the presence of those signals alone.

Emerging research suggests that individuals with insomnia may exhibit altered interoceptive processing, further shaping subjective experience [21]. In SSM, these processes may contribute to the persistent perception of wakefulness despite underlying sleep.

Monitoring behaviors represent a behavioral manifestation of altered precision weighting. Individuals with SSM frequently attend to their internal state, checking whether they are asleep, tracking time, or evaluating cognitive activity [7,17]. This monitoring increases the frequency with which internal signals are sampled.

In predictive processing terms, increased sampling enhances the availability and salience of wake-like signals. Because these signals are experienced clearly and encoded in memory, they exert disproportionate influence on perceptual inference. Over time, this leads to a data set dominated by wake-like experiences, reinforcing the perception of wakefulness.

This process also interacts with attentional bias. Individuals with insomnia have been shown to preferentially attend to sleep-related cues and internal sensations [18,19]. In SSM, this bias may be directed specifically toward signals interpreted as evidence of wakefulness, further amplifying perceptual distortion.

Perception involves classification under uncertainty. In the case of sleep, the brain must determine whether a given state represents wakefulness or sleep based on incomplete and ambiguous information. This classification process depends on implicit thresholds that define what counts as sleep.

Subjective reports of sleep depend on retrospective reconstruction. Sleep is associated with reduced encoding, whereas wakefulness is associated with continuous awareness and memory formation [9]. As a result, remembered experiences disproportionately reflect wake like states.

Predictive processing models emphasize that perception and memory are intertwined inferential processes. In SSM, remembered episodes of awareness are treated as representative of the entire night, while unremembered periods of sleep are treated as absent. This reconstruction reinforces priors and contributes to the stability of misperception.

From a predictive processing perspective, this reflects a shift in decision boundaries influenced by priors and precision weighting. Ambiguous states are resolved in favor of wakefulness because they do not meet the internal criteria required for classification as sleep.

Subjective reports of sleep are inherently retrospective and depend on memory reconstruction. Sleep, particularly deeper stages, is associated with reduced encoding, whereas wakefulness is associated with continuous awareness and memory formation [9].

Predictive processing models emphasize that perception and memory are intertwined through inferential processes. Retrospective judgments are constructed from available evidence and shaped by prior expectations. In SSM, remembered episodes of awareness are interpreted as representative of the entire night, while unremembered periods of sleep are effectively treated as absent.

This reconstruction process contributes to the stability of misperception. Because the resulting narrative is coherent and consistent with prior beliefs, it reinforces those beliefs and resists updating. The individual experiences a continuous and convincing account of wakefulness, even when objective evidence suggests otherwise.

Taken together, these components form a self-reinforcing system. Strong priors bias interpretation toward wakefulness. Hyperarousal increases the precision of internal signals. Monitoring amplifies sampling of wake like experiences. Memory processes reinforce the resulting interpretation.

Within predictive processing, such systems are expected when priors dominate and prediction error is minimized through reinterpretation rather than updating [12,14]. Sleep state misperception is therefore not simply an error in estimation, but a predictable outcome of an inferential system operating under conditions of uncertainty.

Conceptualizing SSM as a disorder of perceptual inference generates several testable hypotheses.

First, individuals with SSM should demonstrate altered precision weighting of interoceptive signals, reflected in increased sensitivity to internal cognitive or physiological activity during sleep.

Second, thresholds for classifying sleep onset may be elevated, requiring deeper or less ambiguous states before labeling a state as sleep.

Third, individuals with SSM may exhibit reduced updating of priors in response to disconfirmatory evidence, such as polysomnographic feedback.

Fourth, interventions that reduce monitoring, modify expectations, or alter attention to internal signals should improve subjective sleep perception independent of changes in objective sleep architecture.

These hypotheses provide a framework for future research and suggest that SSM may serve as a model for studying perception under conditions of reduced sensory precision.

Conceptualizing sleep state misperception (SSM) as a disturbance of perceptual inference reframes the primary clinical target. Traditional insomnia treatments emphasize improving sleep generation—reducing sleep latency, increasing total sleep time, and consolidating sleep. While these goals remain important, they may be insufficient in SSM, where the central difficulty lies in how sleep is perceived, interpreted, and remembered.

Within a predictive processing framework, treatment can be understood as perceptual recalibration. The aim is not only to change sleep physiology, but to modify the inferential system that constructs subjective experience. This includes altering priors (expectations about sleep), reducing the precision of misleading internal signals, decreasing maladaptive monitoring, and increasing tolerance for ambiguity in internal states.

This reframing helps explain a common clinical observation: objective improvements in sleep do not always lead to immediate changes in subjective experience. From a predictive processing perspective, this reflects the persistence of entrenched priors and biased interpretation rather than treatment failure. Clinically, this underscores the importance of explicitly targeting perceptual processes alongside sleep behavior.

Cognitive behavioral therapy for insomnia (CBT-I) is the first-line treatment for chronic insomnia [5,23]. Although typically conceptualized in behavioral and cognitive terms, many of its components can be reinterpreted as mechanisms that influence perceptual inference.

Stimulus Control and Reduction of Ambiguous Sampling

Stimulus control procedures reduce time spent awake in bed and weaken conditioned associations between the bed and wakefulness. From a predictive processing perspective, this intervention also reduces exposure to ambiguous states that are prone to misclassification. By limiting time in bed to periods of high sleep probability, stimulus control decreases the frequency of wake-like signals entering the perceptual system, thereby reducing biased sampling.

Sleep Restriction and Reweighting of Evidence

Sleep restriction (or sleep compression) increases homeostatic sleep drive and reduces time spent in bed awake. In doing so, it increases the density of sleep-related experiences relative to wake-like experiences. This shift in the experiential dataset may facilitate updating of priors by providing more consistent evidence of sleep. Over time, this can weaken entrenched expectations such as “I do not sleep.”

Cognitive Interventions as Priors Modification

Cognitive restructuring targets maladaptive beliefs about sleep, which function as priors within a predictive processing framework. Interventions that challenge rigid expectations such as “If I’m aware, I’m awake” or “I never sleep,” introduce alternative models that allow for reinterpretation of ambiguous states.

Importantly, cognitive interventions are most effective when paired with experiential learning. Behavioral experiments that test predictions (e.g., reducing clock-checking or delaying sleep estimation) create opportunities for prediction error. When outcomes differ from expectations, priors may be updated, particularly when the discrepancy is experienced directly rather than explained abstractly.

A defining feature of SSM is heightened monitoring of internal states. Patients often track thoughts, bodily sensations, and time in an effort to determine whether they are asleep. Within predictive processing, this monitoring increases the precision and salience of internal signals, making them more influential in shaping perception.

Interventions that reduce monitoring can therefore be understood as reducing precision weighting. Standard CBT-I strategies such as eliminating clock-checking, discouraging active sleep evaluation, and promoting disengagement from sleep-related performance directly target this mechanism.

Mindfulness- and acceptance-based approaches may further enhance this effect. By encouraging nonjudgmental awareness without evaluation, these approaches reduce the tendency to assign meaning to transient internal states. Thoughts and sensations are experienced as events rather than evidence, thereby reducing their impact on perceptual inference [24].

Given the role of interoceptive inference in SSM, treatment may also benefit from explicitly addressing how internal signals are interpreted. Patients can be educated that sleep does not require the absence of awareness and that residual cognitive activity can occur within physiological sleep. This reframing expands the range of states that can be classified as sleep.

At the same time, interventions may aim to increase tolerance for ambiguity. Rather than attempting to definitively determine whether one is asleep or awake, patients are encouraged to allow uncertainty. This shift reduces the need for continuous evaluation and weakens the link between ambiguous states and categorical judgments.

Paradoxical intention is particularly relevant in this context. By instructing patients to relinquish the goal of sleep or to remain awake, this technique reduces performance pressure and monitoring. Within a predictive processing framework, it may also weaken priors related to the necessity of achieving sleep, thereby reducing bias toward wakefulness.

Consider a patient who reports sleeping “zero hours” despite PSG evidence of approximately six hours of sleep. The patient describes lying awake all night, monitoring thoughts and bodily sensations, and frequently checking the clock. When shown objective data, the patient acknowledges its validity but insists, “That’s not what it felt like.”

Within a predictive processing framework, this presentation can be understood as the result of strong priors (“I do not sleep”), high-precision internal signals (thoughts and awareness), and frequent monitoring that amplifies wake-like experiences. Treatment focuses not only on improving sleep continuity, but on reducing monitoring, modifying expectations, and reframing the meaning of awareness during sleep.

Over time, the patient may begin to report uncertainty (“Maybe I slept some”) before reporting improvement (“I think I slept more than I realized”). This shift reflects not only changes in sleep, but changes in the inferential processes that generate subjective experience.

Conceptualizing SSM as a disorder of perceptual inference also broadens the definition of treatment success. In addition to improvements in sleep continuity and duration, meaningful outcomes include changes in how sleep is experienced and interpreted.

Patients may continue to experience occasional awareness during the night, but no longer interpret this as evidence of failure. Subjective sleep satisfaction may improve even when objective sleep parameters change modestly. This distinction is clinically important and highlights the need to assess both physiological and perceptual outcomes in treatment.

For clinicians, this framework suggests several practical considerations:

First, assessment should include not only sleep parameters, but also how patients define and evaluate sleep. Questions about what “counts” as sleep and how judgments are made can reveal underlying priors and interpretive processes.

Second, treatment should explicitly address misperception when present. Standard CBT-I protocols may require emphasis on perceptual processes, particularly in patients reporting extreme discrepancies.

Third, clinicians should anticipate that changes in perception may lag behind changes in sleep. Preparing patients for this trajectory can reduce frustration and improve adherence.

Finally, this framework supports a collaborative therapeutic stance in which the clinician helps the patient understand how their experience is constructed, rather than attempting to correct it through reassurance alone. This approach may enhance engagement and reduce resistance.

The present paper advances a reframing of sleep state misperception (SSM) as a disturbance of perceptual inference rather than solely a disorder of sleep generation, hyperarousal, or cognitive bias. While existing models have identified important contributors—including heightened arousal, selective attention, monitoring behaviors, and memory asymmetry—these processes are typically considered in isolation [7,10,19]. A predictive processing framework integrates these elements within a unified account of how subjective sleep experience is constructed under conditions of uncertainty [12-14].

From this perspective, the discrepancy between subjective and objective sleep is not simply a measurement problem or a failure of estimation. Rather, it reflects the operation of general perceptual mechanisms applied to internally generated and often ambiguous signals. Strong priors bias interpretation toward wakefulness, hyperarousal increases the precision of internal signals, monitoring amplifies sampling of wake-like experiences, and memory processes reinforce wake-based reconstructions [11,9,6]. The resulting percept is coherent, stable, and resistant to change because it emerges from a system designed to generate consistent interpretations rather than veridical ones [14].

This framework helps explain several longstanding clinical observations. First, objective reassurance alone is often insufficient. Presenting polysomnographic data does not directly alter the inferential processes that generate subjective experience [5]. Second, improvements in objective sleep do not necessarily produce immediate changes in perception. Priors and interpretive biases may persist even as sleep becomes more consolidated [6]. Third, interventions that reduce monitoring, modify expectations, and increase tolerance for ambiguity can improve subjective outcomes even when physiological changes are modest [23,24].

More broadly, sleep state misperception can be understood as a model case of perception under conditions of reduced sensory precision. Unlike exteroceptive modalities such as vision or audition, which are anchored by relatively stable external inputs, the perception of sleep depends largely on internally generated signals that are inherently ambiguous [15,16]. This makes sleep an especially informative domain for examining how priors, attention, and inference interact to shape subjective experience [13].

This perspective also situates SSM within a broader class of phenomena involving altered perceptual inference. Conditions characterized by heightened interoceptive sensitivity, such as anxiety-related hypervigilance or somatic symptom amplification, similarly involve strong priors interacting with ambiguous internal signals [25,15]. While these conditions differ in content and clinical presentation, they share common underlying mechanisms related to attention, expectation, and interpretation. Understanding SSM within this broader framework may therefore contribute to a more general account of how the brain constructs internal experience [26-30].

The present model also generates testable predictions that extend beyond descriptive accounts. For example, individuals with SSM should exhibit altered precision weighting of interoceptive signals, elevated thresholds for classifying sleep, and reduced updating of priors in response to disconfirmatory evidence [12,14]. These hypotheses can be examined using multimodal approaches, including PSG combined with subjective reporting, interoceptive assessment, and experimental manipulation of expectation [31-41].

In addition, neurophysiological investigations may further clarify the relationship between cortical activation and subjective experience. Elevated high-frequency EEG activity during sleep has been observed in insomnia and is thought to reflect persistent cortical activation [10,11]. Integrating these findings with predictive processing models may help identify neural correlates of altered perceptual inference.

Finally, this framework suggests that subjective objective discrepancy should not be viewed solely as an error to be corrected, but as a phenomenon to be understood. Rather than attempting to eliminate discrepancy, treatment may be more effective when it targets the processes that generate and maintain it. This shift in perspective aligns with broader developments in psychiatry that emphasize the role of perception and interpretation in shaping clinical experience [13,15].

Several limitations should be acknowledged. First, this paper presents a conceptual synthesis rather than a systematic empirical analysis. Although the model integrates findings from multiple domains, the proposed mechanisms have not been tested together within a single experimental framework. As such, the framework should be viewed as hypothesis-generating rather than definitive.

Second, direct empirical research specifically examining predictive processing mechanisms in SSM remains limited. While components of the model, such as hyperarousal, attentional bias, and memory asymmetry, are well supported, their integration within a predictive processing framework is largely theoretical. Future studies are needed to examine how priors, precision weighting, and perceptual thresholds operate in individuals with SSM.

Third, SSM likely represents a heterogeneous phenomenon. Some individuals may exhibit greater contributions from physiological hyperarousal, whereas others may be more influenced by cognitive or attentional factors. The present framework is intended to provide an organizing model rather than a singular causal explanation, and it may require refinement to account for individual differences.

Fourth, clinical implications derived from this model require empirical validation. Although the framework aligns with observed treatment effects and provides a coherent rationale for intervention, controlled studies are needed to determine whether explicitly targeting perceptual processes improves outcomes beyond standard CBT-I approaches.

Sleep state misperception represents a clinically significant and conceptually informative form of insomnia in which subjective experience diverges from objective measurement. Traditional models have identified key contributing processes but have not fully accounted for how these processes combine to produce a stable and compelling perception of wakefulness.

Conceptualizing SSM as a disorder of perceptual inference provides a unifying framework that integrates hyperarousal, attentional monitoring, boundary ambiguity, and memory reconstruction within a coherent model of how subjective experience is constructed. Within this framework, the perception of sleep is understood as an inferential process shaped by prior expectations and ambiguous internal signals rather than a direct reflection of physiological state.

This perspective has important implications for both research and clinical practice. It suggests that effective treatment may require not only improving sleep continuity, but also modifying how sleep is perceived and interpreted. Interventions that reduce monitoring, weaken maladaptive priors, and increase tolerance for ambiguity may help recalibrate perceptual processes and improve subjective outcomes.

More broadly, positioning SSM within a predictive processing framework bridges sleep medicine and the science of perception. It highlights the role of inferential processes in shaping internal experience and suggests that sleep misperception may serve as a model for understanding perception under conditions of uncertainty. Future research integrating behavioral, neurophysiological, and computational approaches may further clarify these mechanisms and inform more targeted and effective interventions.

Ethics Approval and Consent to Participate

Not applicable. This study is a conceptual review and does not involve human participants or identifiable data.

Author Contributions

Bruce D. Forman conceptualized the manuscript, conducted the literature review, and wrote and revised the manuscript in its entirety.

Use of Generative AI

Generative AI tools were used to support language refinement and organization. The author takes full responsibility for all content.

1. Morin CM, Benca R. Chronic insomnia. Lancet. 2012; 379: 1129-1141.
2. Riemann D, Baglioni C, Bassetti C. European guideline for the diagnosis and treatment of insomnia. J Sleep Res. 2017; 26: 675-700.
3. Perlis ML, Giles DE, Mendelson WB, Bootzin RR, Wyatt JK. Psychophysiological insomnia. Psychological Bulletin. 1997; 121: 97-111.
4. Harvey AG, Tang NKY. (Mis)perception of sleep in insomnia. Sleep Medicine Reviews. 2012; 16: 567-580.
5. Edinger JD, Arnedt JT, Bertisch SM, Carney CE, Harrington JJ, Lichstein KL, et al. Behavioral and psychological treatments for chronic insomnia disorder in adults: An American Academy of Sleep Medicine clinical practice guideline. J Clin Sleep Med. 2021; 17: 255-262.
6. Harvey AG, Tang NKY. (Mis)perception of sleep in insomnia: A puzzle and a resolution. Psychol Bull. 2012; 138: 77-101.
7. Harvey AG. A cognitive model of insomnia. Behav Res Ther. 2002; 40: 869-893.
8. Buysse DJ. Sleep health: Can we define it? Does it matter? Sleep. 2014;37: 9-17.
9. Nir Y, Tononi G. Dreaming and the brain. Trends Cogn Sci. 2010; 14: 88-100.
10. Riemann D, Spiegelhalder K, Feige B. The hyperarousal model of insomnia. Sleep Med Rev. 2010; 14: 19-31.
11. Perlis ML, Smith MT, Andrews PJ, Orff H, Giles DE. Beta EEG activity in insomnia. Sleep. 2001; 24: 110-117.
12. Friston K. The free-energy principle: A unified brain theory? Nat Rev Neurosci. 2010; 11: 127-138.
13. Clark A. Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behav Brain Sci. 2013; 36: 181-204.
14. Hohwy J. The predictive mind. Oxford University Press. 2013.
15. Irwin MR. Sleep and inflammation. Nat Rev Immunol. 2019; 19: 702-715.
16. Seth AK. Interoceptive inference. Trends Cogn Sci. 2013; 17: 565-573.
17. Critchley HD, Garfinkel SN. Interoception and emotion. Curr Opin Psychol. 2017; 17: 7-14.
18. Kyle SD, Beattie L, Spiegelhalder K, Rogers Z, Espie CA. Altered emotion perception in insomnia. Sleep. 2014; 37: 775-783.
19. Spiegelhalder K, Regen W, Feige B. Increased EEG sigma and beta power in insomnia. Sleep. 2015; 35: 247-255.
20. Levenson JC, Kay DB, Buysse DJ. The pathophysiology of insomnia. Chest. 2015; 147: 1179-1192
21. Blume C, Lechinger J, Santhi N, Del Giudice R, Gnjezda MT, Pichler G, et al. Signatures of sleep depth in the brain: The role of interoception. Neurosci Biobehav Rev. 2019; 99: 1-12.
22. Wei Y, Van Someren EJW, Van der Werf YD. Interoceptive processing in insomnia. Sleep Med Rev. 2022; 64: 101645.
23. Means MK, Edinger JD, Glenn DM, Fins AI. Accuracy of sleep perceptions among insomnia sufferers. Sleep Med. 2023; 4: 285-296.
24. Trauer JM, Qian MY, Doyle JS. CBT-I meta-analysis. Ann Intern Med. 2015; 163: 191-204.
25. Ong JC, Crawford MR, Dawson SC, Wyatt JK. Insomnia and daily affect. Sleep Med Rev. 2021; 57: 101471.
26. Paulus MP, Stein MB. Interoception in anxiety and depression. Brain Struct Funct. 2010; 214: 451-463.
27. Altena E, Micoulaud-Franchi JA, Geoffroy PA, Bioulac S, Philip P. Mechanisms of cognitive behavioral therapy for insomnia. J Sleep Res. 2023; 32: e13860.
28. Baglioni C, Spiegelhalder K, Lombardo C, Riemann D. Sleep and emotions: A focus on insomnia. Sleep Med Rev. 2010; 14: 227-238.
29. Ben Simon E, Vallat R, Barnes CM, Walker MP. Sleep loss and the socio-emotional brain. Trends Cogn Sci. 2020; 24: 435-450.
30. Clark A. Surfing uncertainty: Prediction, action, and the embodied mind. Oxford University Press. 2016.
31. Espie CA, Emsley R, Kyle SD, Gordon C, Drake CL, Siriwardena AN, et al. Effect of digital cognitive behavioral therapy for insomnia on health outcomes: A randomized clinical trial. JAMA Psychiatr. 2019; 76: 21-30.
32. Fortier-Brochu E, Beaulieu-Bonneau S, Ivers H, Morin CM. Insomnia and daytime cognitive performance: A meta-analysis. Sleep Med Rev. 2012; 16: 83-94.
33. Garfinkel SN, Seth AK, Barrett AB, Suzuki K, Critchley HD. Knowing your own heart. Biol Psychol. 2015; 104: 65-74.
34. Goldstein AN, Walker MP. The role of sleep in emotional brain function. Ann Rev Clin Psychol. 2014; 10: 679-708.
35. Hertenstein E, Feige B, Gmeiner T. Insomnia as a predictor of mental disorders. Sleep Med Rev. 2019; 43: 96-105.
36. Kalmbach DA, Anderson JR, Drake CL. Sleep reactivity. J Sleep Res. 2018; 27: e12710.
37. Kyle SD, Espie CA, Morgan K. Insomnia and health-related quality of life. Sleep Med Rev. 2010; 14: 69-82.
38. Seth AK, Friston KJ. Active interoceptive inference. Biological Psychiatr. 2016; 81: 737-738.
39. Simon EB, Walker MP. Sleep loss causes social withdrawal. Nature Communications. 2018; 9: 3146.
40. Tempesta D, Socci V, De Gennaro L, Ferrara M. Sleep and emotional processing. Front Psychol. 2018; 9: 113.
41. Troxel WM. Relational sleep. Psychosomatic Med. 2010; 72: 578-586.
42. Walker MP, Stickgold R. Sleep and memory. Annual Rev Psychol. 2008; 57: 139-166.