insomnia: neurophysiological and neuropsychological approaches

REVIEW

Insomnia: Neurophysiological and NeuropsychologicalApproaches

Célyne H. Bastien

Received: 4 November 2010 /Accepted: 6 January 2011 /Published online: 20 January 2011# Springer Science+Business Media, LLC 2011

Abstract Insomnia is a symptom, a syndrome and acomorbid disorder. Its diagnosis relies on subjective reportsfrom the afflicted individual and is defined as difficulties ininitiating sleep, maintaining sleep, waking up too early ornon-restorative sleep. However, insomnia and especially,primary insomnia, has received much attention in insomniaresearch with the use of objective measures. Insomnia, itspeculiarities, most frequent subtypes and two most prom-inent models will first be briefly introduced. Then,insomnia will be reviewed according to results obtainedwith the use of neurophysiological measures as basic/traditional as polysomnography to more sophisticated onessuch as power spectral analysis, neuroimaging, cyclicalternating patterns and event-related potentials. In addition,a review of the discrepancies between subjective andobjective reports of cognitive alterations through neuropsy-chological testing is offered. The need to combine measuresis then highlighted in conclusion.

Keywords Insomnia . Subtypes . PSG . Neurophysiologicalmeasures

Acronyms

APA American Psychiatric AssociationARAS Ascending Reticular Activating System

BZD (BZDs) Benzodiazepine (Benzodiazepines)CAP Cyclic Alternating PatternCBT-I Cognitive-Behavioral Therapy for

InsomniaCNS Central Nervous SystemEEG ElectroencephalogramEKG ElectrocardiogramEMG ElectromyogramEOG ElectrooculogramERPs Event-Related PotentialsFFT Fast Fourier TransformsfMRI Functional Magnetic Resonance ImagingGABA Gamma-aminobutyric acidLORETA Low Resolution Electromagnetic

Tomography AnalysisN100 (N1) Negative wave peaking at about

100 ms after stimulus onsetN350 Negative wave peaking at about

350 ms after stimulus onsetNCAP Non-Cyclic Alternating PatternNREM Non Rapid Eye MovementP50 Positive wave peaking at about

50 ms after stimulus onsetP200 (P2) Positive wave peaking at about

200 ms after stimulus onsetP300 (P3) Positive wave peaking at about

300 ms after stimulus onsetPET Positron Emission TomographyPSA Power Spectral AnalysisPSG PolysomnographyREM Rapid Eye MovementSPECT Single-Photon-EmissionComputed

TomographySWA Slow Wave ActivitySWS Slow Wave SleepTMS Transcranial Magnetic Stimulation

C. H. Bastien (*)École de Psychologie, FAS-Local 1012, 2325,rue des Bibliothèques, Université Laval,Québec, QC, Canada G1V 0A6e-mail: [email protected]

C. H. BastienLaboratoire de Neurosciences Comportementales Humaines,Institut de Recherche en Santé Mentale,Centre de Recherche Université Laval Robert-Giffard,Québec, QC, Canada

Neuropsychol Rev (2011) 21:22–40DOI 10.1007/s11065-011-9160-3

Introduction

Insomnia is among the most common health complaints inmedical practice and the most prevalent of all sleep disorders.Approximately 30% of the general population experiencessome insomnia symptoms occasionally and, 10% suffer fromchronic and persistent insomnia (Morin et al. 2006; Ohayon2002). Frequently reported consequences related to insomniainclude fatigue, sleepiness, mood disruption (Zammit 1988),impaired attention, and memory deficits (Hauri 1997). It hasbeen observed that often, it is not the sleep difficulties per sethat will be an incentive to seek medical help or treatment,but the daily consequences associated with the sleepdifficulties (Morin 1993). Sleep aids, increased medicalvisits, and absenteeism from work imply important costsfor the patient as well as society (Hillman et al. 2006;St-Jean and Bastien 2008a). Insomnia is multi-dimensionaland multifaceted, yet its evaluation and diagnosis are basedprimarily on clinical history (Sateia et al. 2000). This review,in addition of stating some peculiarities of the disorder, willprovide an overview of research devoted at circumscribingthe objective deficits in insomnia sufferers while exploringthe most recent research avenues offered by neurophysio-logical and neuropsychological approaches.

Definition

Insomnia is defined as a complaint of prolonged sleep latency(labeled “sleep-onset insomnia”), difficulties in maintainingsleep (labeled “sleep maintenance insomnia”), waking up tooearly in the morning (labeled “terminal insomnia”), a mix ofdifferent sleep complaints (labeled “mixed insomnia”) and/orthe experience of non-restorative sleep. In addition, the DSM-IV-TR (APA 2000) specifies that to be considered a disorder,“primary” insomnia or its perceived consequences causeclinically marked distress or significant impairment ofoccupational or social functioning. Furthermore, it is notcaused by a mental or physical disorder, a primary sleepdisorder, or the effects of substance abuse or a medication.Although insomnia can be acute (less than 1 month), thecurrent definition of primary insomnia specifies that it mustbe present at least three times a week, for at least 1 month(thus “chronic”). Increasingly, the epidemiological literatureon insomnia classifies individuals in two categories: “syn-drome” and “symptom”. To be included in the syndromecategory, an individual must meet the definition for insomniadisorder as specified by the DSM-IV-TR. On the other hand,an individual presenting one or more symptoms withoutcorresponding to the full spectrum of criteria of the DSM-IV-TR (for example, having a sleep difficulty only once a weekor not presenting any marked distress related to his/her sleepdifficulties), would be classified in the symptom category.

Peculiarities of the Disorder

Insomnia can be a problem from childhood through old age. Itcan stand alone as a disorder (primary), can be comorbid withanother active sleep or mental disorder or be present as asymptom of another disorder. In this regard, insomnia is by farthe most frequent sleep complaint amongst psychologicaldisorders. While the essential clinical features of insomnia aresimilar for the symptom and the syndrome, in comorbidinsomnia, the sleep disturbance evolves in the context ofanother disorder. Comorbid insomnia (National Institutes ofHealth 2005) replaces “secondary insomnia” as the directionof causality is not always easy to identify when insomnia ispresent in the context of other conditions. Comorbidinsomnia can be caused by other conditions, can becompletely independent of them or can also cause thedisorders with which it coexists. Disorders may be psychi-atric (depression and anxiety), medical (typically involvingpain), circadian (phase-delay syndrome), or other sleepdisorders (periodic limb movements, sleep-related breathingdisorders). Like many other sleep disorders, insomniacomprises a spectrum of complaints reflecting dissatisfactionwith the quality, duration, or efficiency of sleep. Unlikemany other sleep disorders that are first noticed by othersand require objective polysomnographic laboratory valida-tion (for example, apnea, REM sleep behavior disorder,nocturnal myoclonous), insomnia is first reported by theindividual and does not require an objective confirmation.Furthermore, the severity of the complaint is not classifiedinto categories such as light, medium or severe such is thecase for obstructive sleep apnea syndrome, but is ratherdistributed along a severity continuum.

Subtypes of Insomnia

The ICSD (ICSD-2, 2005) distinguishes among 11 differentsubtypes of insomnia. However, three of them are the mostfrequent: psychophysiological insomnia, sleep-state misper-ception (or paradoxical insomnia), and idiopathic insomnia.In some cases, complaints of sleep difficulties are objec-tively observable, in other cases, they are not. In otherwords, patient may either complain of sleep difficultieswhile objective polysomnographic recordings appear nor-mal, or there may be some form of constant and importantgap between objective and subjective measures of sleep(Edinger et al. 2004). Depending on the diagnostic criteriaused, as much as 50% of individuals suffering frominsomnia could be poor estimators and, consequently,classified as suffering from paradoxical insomnia (Edingerand Krystal 2003; St-Jean and Bastien 2009). Whileparadoxical insomniacs and good sleepers seem to beequivalent on measures of sleep macroarchitecture (for

Neuropsychol Rev (2011) 21:22–40 23

example the observed percentage of different sleepstages), there may be subtle and finer differences in themicroarchitecture characterizing both groups (Krystal etal. 2002). Among insomniacs estimating their sleep quiteaccurately, and presenting “real” sleep-onset or sleepmaintenance difficulties are patients suffering from psy-chophysiological insomnia. These individuals are assumedto be conditioned by sleep-related stimuli in the environ-ment (e.g., bedroom) or bedtime routine, which are thenassociated in the sufferer’s mind with the anxiety, fear, orfrustration of having trouble getting to sleep (Hauri andFisher 1986; Harvey 2002; Espie 2002). Consequently, thelevel of arousal is increased, preventing a good nightsleep. A third form of primary insomnia, less common butstill acknowledged, is idiopathic or childhood-onsetinsomnia. In this case, insomnia is reported as starting inearly childhood, typically before the age of 10. A higherrate of difficult and premature deliveries is found inidiopathic insomnia compared to the general population(Regestein and Reich 1983; Hauri and Olmstead 1980).Despite evidence that their sleep is often more disturbedthan in psychophysiological insomnia, individuals withidiopathic insomnia tend to show less emotional distress,perhaps due to resignation or to coping mechanisms theyhave developed over their lifetime. Finally, differentsubtypes of insomnia are not all mutually exclusive. Anindividual can show objective sleep-onset, sleep-maintenanceor mixed objective sleep difficulties once his or her sleep hasbeen recorded in the sleep laboratory but also presentimportant discrepancies between his or her reports andlaboratory sleep observations. As such, an individual mightbe complaining of mixed sleep difficulties over a long periodof time, without showing any signs of depression or anxietyand at the same time, be subtyped as having paradoxicalinsomnia.

Daily Consequences and NeuropsychologicalAssessment

In clinical practice, individuals suffering from insomniaoften report fatigue as well as attention/concentration andmemory impairments (Zammit 1988). These impairmentsare attributed to the sleep difficulties, create great concernsand are often the reason that will prompt an individual toseek treatment (Morin 1993). Deterioration in occupational,social and cognitive functioning is a diagnostic criterion forchronic insomnia (APA 1994, 2000).

Ancillary Measures

A number of studies have confirmed that insomniacs comparedto good sleepers, score higher on questionnaires measuring

fatigue and depressive/anxious symptoms (Lichstein et al.2001; Moul et al. 2002) as well as on quality of life indexes(Hatoum et al. 1998; Zammit et al. 1999). Some epidemio-logical studies also report that insomniacs complain moreoften than good sleepers of concentration and memorydeficits and of mistakes related to work demands (Légeret al. 2001; Roth and Ancoli-Israel 1999). Individualssuffering from insomnia also score higher than self-defined good sleepers on the Cognitive Failures Question-naire (Broadbent et al. 1982).

Neuropsychological Testing

Some studies have observed that in insomniacs, compared togood sleepers, deficits were present in tasks measuringequilibrium, attention, reaction time (Hauri 1997) and accessto information stored in semantic memory (Mendelson et al.1984). However, no between groups differences are detectedon psychomotor measures (Broman et al. 1992; Mendelsonet al. 1984), episodic memory (Mendelson et al. 1984),divided attention (Broman et al. 1992), verbal learning andface and word recognition (Broman et al. 1992). On theother hand, contrary to what Hauri (1997) reported, Bromanand colleagues (1992) had not previously observed anydeficits in reaction time in insomniacs.

While many studies do report at least one impairment ininsomniacs relative to good sleepers (for example: Altena etal. 2008a; Bonnet and Arand 1995; Edinger et al. 2008;Hauri 1997; Mendelson et al. 1984; Randazzo et al. 2000;Schmutte et al. 2007; Schneider-Helmert 1987; Schneideret al. 2004; Sugerman et al. 1985; Varkevisser and Kerkhof2005; Vignola et al. 2000), other studies do not observe anydifference whatsoever between the two groups of sleepersin similar studied domains (Broman et al. 1992; Dorsey andBootzin 1997; Haimov et al. 2008; Orff et al. 2007).

Neuropsychological studies aimed at evaluating vigi-lance (mainly attention and motor speed) and sleepinessoften yield ambiguous reports. While some studies haveobserved similar vigilance levels between insomniacs andgood sleepers (Hauri 1997; Sugerman et al. 1985; Mendelsonet al. 1984; Stepanski et al. 1988) others find better vigilancein insomniacs than good sleepers (Gu et al. 2005; Seidelet al. 1984). Our own group has also recently observed thatinsomniacs reported more sleepiness at pre-test, before acognitive task (event-related potentials, ERPs), than goodsleepers did. The between groups difference disappeared atpost-testing (after the ERPs tasks). Thus, once insomniacsare engaged in cognitive tasks their sleepiness level becomessimilar to those of good sleepers (Ropars et al. 2010).According to the cortical activation theory (see below,neurocognitive model, Perlis et al. 1997), Bonnet and Arand(1997) suggested that enhanced vigilance levels in insom-niacs would result from greater cortical activation, this one

24 Neuropsychol Rev (2011) 21:22–40

already interfering with difficulties in initiating or maintainingsleep. A compensation phenomenon could also explain thatinsomniacs’ expectations towards their performance increasewhen they perceived that have not slept well (Broman et al.1992; Schneider-Helmert 1987). As such, insomniacs aremore likely to report having to provide more effort toperform and continue to perform following a bad night’ssleep than good sleepers (Broman et al. 1992). It also seemsthat performance in insomniacs, compared to good sleepers,varies during the day being lower late at night and early inthe morning (Varkevisser and Kerkhof 2005).

A meta-analysis is often indicated when ambiguousresults are reported. To date, four meta-analyses have beenconducted to better understand the impairments or extent ofcognitive deficits observed in insomnia (Riedel andLichstein 2000; Fulda and Schulz 2001; Shekleton et al.2010; Fortier-Brochu et al. 2008).

Riedel and Lichstein (2000) observed that only 24% of54 compared studies revealed cognitive deficits and similarconclusions were reached by Fulda and Schultz a year later(2001). Fortier-Brochu and colleagues (2008) observed thatgenerally deficits in insomniacs were more common insustained attention and vigilance, episodic and workingmemory, verbal fluidity and problem solving abilities.Finally, Shekleton et al. (2010) compiled studies comparinginsomniacs and good sleepers on measures of attention(selective, sustained and alternate), psychomotor speed,working memory, learning and executive functioning.Again, ambiguous and sometimes contradictory resultswere observed between studies included in this meta-analysis. Nonetheless, these authors pointed out thatindividuals suffering from insomnia, compared to goodsleepers, appear to present more deficits in tasks involvingcomplex processes in attention (alternate and sustained) aswell as tasks involving working memory.

Relating subjective/objective sleep parameters to neuro-psychological testing scores, a study done in our own labsuggests that certain objective sleep parameters are associ-ated with different aspects of cognitive performance(Bastien et al. 2003a). In fact, in older insomniacs andgood sleepers, a good night of sleep is usually associatedwith a better daily cognitive performance. In addition, inboth groups of sleepers, perceived sleep quality is moreimportant than subjective sleep duration in terms of itsimpact on next day performance.

Short- and long-term memory consolidation are also partof the neurocognitive model and seemingly affected byhyperarousal. Since sleep has been associated with theconsolidation of memory (Stickgold 2005), both declarativeand procedural memory consolidation have been investi-gated in insomniacs and good sleepers. Nissen et al. (2006)reported impairments in procedural memory in insomniacscompared to good sleepers, and Backhaus et al. (2006)

observed impairments in declarative memory in insomniacscompared to good sleepers. Although these results areinteresting and appear to support the neurocognitive model,more research in this area needs to be conducted to assesshow these findings might parallel reported daily conse-quences in insomnia.

Altogether, neuropsychological studies conducted so farsuggest that only subtle cognitive deficits are observed ininsomniacs, despite their subjective reports of dailycognitive alterations. It is possible that neuropsychologicaltesting used so far is not appropriate to detect objectivedeficits (ecological testing or testing reflecting moreadequately tasks encountered each and every day mightbe more appropriate and informative) or on the other hand,subtle deficits are enough to interfere with daily activities ininsomniacs. Studies using more precise measures of activecognitive processes (such as ERPs) or combining evalua-tion measures (for example, fMRI and cognitive evokedpotentials) might be the next step in trying to shed somelight on cognitive deficits in insomniacs. In addition, asmentioned earlier, the quality of the preceding night of theevaluation needs to be accounted for since it appears to bareimportant weight, albeit be only for the perception ofperformance.

Theoretical Models

Many theoretical models of insomnia have been developed.The two most influential are the “neurocognitive” (Perlis et al.1997) and “psychobiological” (Espie 2002) models. Both ofthese models are currently being tested using objectivelaboratory measures as a means to provide empiricalvalidation. The different sections on power spectral analysis(PSA), neuroimaging and event-related potentials (ERPs) ofthis paper will expose different experimental designs thathave been used to validate these models.

The Neurocognitive Model

Many predisposing (Spielman and Glovinski 1991), pre-cipitating (Bastien et al. 2004), and maintaining factors(Spielman and Glovinski 1991) have been suggested asbeing linked to the appearance or perpetuation of insomnia.Hyperarousal has been suggested as being a core feature ininsomnia. This hypothesis posits that insomniacs are overlyaroused (thus “hyperaroused”) at the somatic and cognitivelevel compared to good sleepers; in turn, this state ofhyperarousal would interfere with sleep initiation andmaintenance and would also interfere with daily activities(Bonnet and Arand 1997; Harvey 2002). The neurocogni-tive model of insomnia (Perlis et al. 1997; see Fig. 1adapted from M. Perlis et al., Principles and Practice in


Sleep Medicine, 2010) proposes that insomniacs developconditioned cortical arousal from the association of sleeprelated stimuli and encountered sleep difficulties. Thisconditioned hyperarousal may lead to “enhanced sensoryand information processing” around sleep onset, as well asin sleep—wake transitions throughout the night. In addi-tion, other cognitive alterations such as short- and long-term memory formation would also be jeopardized.Furthermore, hyperarousal can also be translated in theamount, or degree of discrepancies, between subjective andobjective ratings of sleep (Perlis et al. 1997).

So far, many studies have provided empirical validationfor the neurocognitive model, mainly through the use ofPower Spectral Analysis (PSA) showing elevated betapower (Freedman 1986; Merica and Gaillard 1992; Jacobset al. 1993; Lamarche and Ogilvie 1997; Merica et al. 1998;Perlis et al. 2001; Krystal et al. 2002; Buysse et al. 2008) orneuroimaging techniques (Nofzinger et al. 1999). Both PSAand neuroimaging studies are described later. For a morecomprehensive review on hyperarousal, its interaction withthe sleep-wake regulation system and contribution to thepathophysiology of insomnia, see Riemann et al. 2009.

The Psychobiological Model

The psychobiological model, put forward by Espie (2002),suggests that high levels of cortical arousal in insomnia, as

hypothesized in the neurocognitive model, may not besufficient to produce or maintain sleep difficulties. In thistheory, “normal sleep processes” are deregulated and theautomaticity of normal unattended sleep initiation isbreached. In other words, it would be an inability tode-arouse or disengage from active wake processes thatinterferes with the normal initiation of sleep processes ininsomnia. Thus higher cortical arousal might be present atany time before sleep onset and during the night, but anotherprocess, inhibition of arousal, might be absent or deficient asinsomniacs attempt to fall asleep or return to sleep during thenight. Some studies have provided empirical evidence forthis theory (Bastien et al. 2008a, b; Smith et al. 2002).

Although the literature so far appears to provide muchempirical support for the “hyperarousal” put forward by theneurocognitive model of insomnia, the inability to “de-arouse” proposed by the psychobiological model is gaininginteresting intuitive and scientific support from ongoinginsomnia research.

General Insomnia Measures

Subjective Measures: The Basis of All

Although essential to make the diagnosis, a detailed clinicalhistory/assessment of the patient’s subjective complaint will

Fig. 1 Neurocognitive Model ofInsomnia (Model is reproducedwith the permission of Dr. MichaelL. Perlis and adapted from Perlis etal. (2011). In Principles and Prac-tice in Sleep Medicine, Kryger,Roth and Dement (Eds), 2011).After insomnia has becomechronic (predisposing and precip-itating factors have usually beenidentified), this model suggeststhat neurocognitive factors suchas conditioned arousal and neuro-cognitive alterations interfereswith sleep initiation and/or sleepmaintenance. Neurocognitive fac-tors (somatic, cortical and cogni-tive) interplay one with the other.“Cortical conditioned arousal” canbe expressed through enhancedsensory and information process-ing as well as through enhancedshort-term and long-term memoryformation. These ones can bemeasured through experimentalprotocols (for example, event-related potentials for informationprocessing)


greatly benefit or even depend on complementary data frommore systematic sources such as subjective, neurophysio-logical and neuropsychological assessments. Sleep diarymonitoring is an irreplaceable tool to document theperceived severity of insomnia. A sleep diary commonlyrequires self-recording of bedtime and arising time, alongwith morning estimates of sleep-onset latency, number andduration of awakenings, total sleep time, and severalindices of sleep quality for the previous night. Even thoughthey do not reflect absolute values obtained from poly-somnography (see below), daily morning estimates of sleepparameters such as sleep-onset latency or wake after sleeponset yield a reliable and valid index of insomnia (Coates etal. 1982). By tracking sleep over several consecutivenights, a sleep diary is more likely to capture the night-to-night variability that often characterizes the sleep of chronicinsomnia (Vallières et al. 2005) compared to one or twonights of polysomnography. Based on the diary measures,sleep-onset insomnia is typically defined by a latency tosleep onset greater than 30 min, and sleep-maintenanceinsomnia is defined by time awake after sleep onset greaterthan 30 min. A sleep efficiency (total sleep time divided bytime in bed) lower than 85% also characterizes insomnia.On the other hand, early morning awakening can bedescribed as a complaint of waking up earlier (more than30 min) than desired, with an inability to go back to sleep,and before total sleep time reaches 6.5 h. These criteria,while seemingly arbitrary, are useful to operationalize thedefinition of insomnia.

The Gold Standard for Sleep: Polysomnography

A standard polysomnographic (PSG) montage providescomprehensive data on sleep continuity (sleep latency, timeawake after sleep onset, sleep time and derived sleepefficiency) and sleep architecture (% of time spent in differentstages of sleep such as non-rapid eye movement (NREM)stages 1, 2, 3 and 4 and rapid eye movement (REM) sleep,stage shifts, number and duration of cycles, and awakenings).Standard montages for sleep include horizontal and verticaleye movements recordings (electrooculogram, EOG), fourelectroencephalographic leads (EEG; C3, C4, O1, O2) andsubmental muscle tone recordings (electromyogram, EMG).These three basic recordings (EOG, EEG and EMG) are usedto score sleep recordings (thus achieve sleep staging).Additional measures (e.g. airflow, respiratory effort, oxygensaturation, and anterior tibialis—leg muscle—EMG) areincreasingly used by laboratories in order to detect othersleep disorders such as sleep apnea or periodic limb move-ments. In healthy adults, stage 1 occupies about 5% of thetotal night of sleep and stage 2 about 50%, stages 3–4(combined) and REM sleep take up respectively about 15%–20% and 25%. Even though EEG leads are fixed to the scalp

and then amplified and read through computerized systems,this measure of cerebral activity still remains limited. Itstemporal resolution is low and it does not permit theidentification of peculiar brain structures involved in normalor disturbed sleep.

On a day to day basis, PSG is not indicated or practicalfor the routine clinical evaluation of insomnia (Reite et al.1995). Still, it is most useful to validate a subjective sleepcomplaint as in the case of psychophysiological insomniaand to rule out the presence of other sleep disorders thatmight contribute to the insomnia complaint (e.g., periodicleg movement). PSG is becoming a standard practice ininsomnia research and is essential in mechanism/evaluationstudies as well as in treatment efficacy studies (Buysse et al.2006). Insomnia protocols are typically conducted on twoor three consecutive nights of PSG recordings for pre andpost-treatment comparisons. Another important limitationof PSG is that it does not provide a valid sample of apatient’s typical sleep at home. Indeed, the sleep ofotherwise good sleepers is more disrupted during their firstnight of recording in the sleep laboratory (i.e., first nighteffect: worst sleep efficiency on the first night compared tosubsequent ones), whereas the sleep of insomniacs mayactually improve in the laboratory (i.e., reverse first-night-effect: better sleep efficiency on the first night than onsubsequent ones) (Pennestri et al. 2003). Still, PSG isparticularly helpful when subtypes of insomnia are understudy and especially in the case of paradoxical insomnia.Our own research group has suggested that this subtype isbest identified after large subjective-objective discrepancieshave been observed in at least two out of four consecutiverecording nights in the laboratory (Bastien et al. 2008a). Inany case, modest PSG differences are usually observedbetween insomniacs and good sleepers. “Objective” meas-ures of insomniacs reveal they tend to spend more time instage 1, less time in stages 3–4, and display more frequentstage shifts through the night (Coates et al. 1982; Reynoldset al. 1984; Hauri and Fisher 1986).

Differences may also be apparent in phasic EEG eventsduring sleep such as K-complexes and sleep spindles. K-complexes are single occurrences of high amplitude lowfrequency waveforms that occur spontaneously in stage 2and slow wave sleep and can also be evoked by externalstimuli (Colrain 2005; Oswald et al. 1960). Sleep spindlesare half second bursts of synchronized 12–15 cycles persecond activity (De Gennaro and Ferrara 2003).

Both K-complexes and sleep spindles are thought toreflect sleep protective mechanisms (De Gennaro andFerrara 2003; Jankel and Niedermeyer 1985; Hess 1965;Walter 1953) and both are used to mark the onset of stage 2sleep (Rechtschaffen and Kales 1968). One insomniahypothesis might be that sleep protective mechanisms aredeficient or inefficient in this disorder. To test this


hypothesis, some researchers have systematically studiedthe behaviors of these phasic events in insomnia sufferers.While Wauquier et al. (1995) observed a lower occurrenceof spontaneous K-complexes in sleep disordered individualscompared to controls, our group has shown that theoccurrence and density of both spontaneous K-complexesand sleep spindles were similar in psychophysiologicalinsomniacs and good sleepers (Bastien et al. 2009a, b).Altogether, results from phasic event studies generallysuggest that sleep protection mechanisms appear intact inpsychophysiological insomnia.

In line with the perception of sleep which is oftenerroneous in insomnia, Coates et al. (1983) reported that agreater percentage of insomniacs claimed to be awake afterthe appearance of the 1st spindle while a greater percentageof good sleepers reported being drowsy at that time. On theother hand, the best concordance between subjective reportsand objective measures of sleep onset latency is obtained whensleep onset is defined as the first epoch of stage 2 and followedby 15 min of uninterrupted sleep (Hauri and Olmstead 1983).Some studies have compared arousal thresholds in insomniacsand controls, usually by presenting tones of increasingintensity or frequency (Mendelson et al. 1986; Haynes et al.1985). Results suggest that arousal thresholds are similar ininsomniacs and good sleepers and thus insomnia cannot beexplained by differences in arousal threshold.

Finer Measures of the EEG

Power Spectral Analysis (PSA)

EEG brain waves have different amplitudes (measured inμV) and frequencies (measured in Hz). Usually, the sloweris a brainwave, the higher its amplitude. PSA is performedby computing Fast Fourier Transforms (FFT) revealing thefrequency and amplitude of the sine waves constituents ofan analyzed portion of the EEG. Results are combined byaveraging the data in different frequency bands usuallydefined as: slow waves (0–1 Hz), delta (1–4 Hz), theta (4–7 Hz), alpha (7–11 Hz), sigma (11–14 Hz), beta1 (14–20 Hz), beta2 (20–35 Hz) and gamma (35–60 Hz). Theresults of the FFT are presented as EEG power (μV2),which corresponds to the square of the amplitude in each ofthe specified frequency bands (see Fig. 2 for a SpectralGraph example). Data may be presented as such (absolutepower) or as the proportion of the power in a frequency bandrelative to the sum of the power in all frequency bands(relative power). Usually, elevated powers in fast frequencybands (e.g. beta bands) reflect increased activity (increasedcortical activation) whereas an increase in a slower band (e.g.slow waves and delta) presupposes less activity (decreasedcortical activation).

The hypothesis of a conditioned aroused state ininsomnia has thus been the impetus for quantitative studyusing PSA at the sleep onset period, during the night aswell as during wakefulness. Freedman (1986) first com-pared sleep-onset insomniacs to good sleepers from lightsout to the end of the first sleep cycle. Insomniacs hadsignificantly more beta activity and less alpha activity thannormal sleepers. Merica and Gaillard (1992) and Merica etal. (1998) also observed more beta and less delta activity atsleep onset and as sleep is initiated in insomniacs compared tocontrols. In another study, Lamarche and Ogilvie (1997)observed higher relative beta activity during wakefulness inpsychophysiological insomniacs compared to controls. Duringsleep, similar observations were also made. Freedman (1986)has also observed more beta activity in sleep-onset insom-niacs than controls during stage 1 and REM sleep althoughbetween groups differences disappeared during stages 2, 3, or4. Looking at the first four sleep cycles, Merica et al. (1998)found that insomniacs showed increased beta activitycompared to good sleepers in both REM and NREM sleep.In REM sleep, insomniacs also presented lower powers in thedelta and theta bands.

When insomniacs are compared with individuals suffer-ing from comorbid insomnia and depression and goodsleepers, insomniacs show elevated powers in beta (bothbeta1 and beta2) as well as in the gamma range (Nofzingeret al. 1999; Perlis et al. 2001) (See Fig. 3). As such,increased in beta activity was observed in insomniacscompared to good sleeper controls. Recently, it was alsoreported that in addition to an increase in beta power,women displayed more low frequency activity (delta/theta)than men, whether experiencing insomnia or being goodsleepers (Buysse et al. 2008). Altogether, these resultssuggest higher cortical arousal, especially in the beta band

Fig. 2 Spectral graph. The spectral graph represents the result ofPSA. EEG has been recorded at C3 (A1 being the reference) and PSAwas performed on selected epochs of stage 2 sleep. Frequencies (Hz)are presented on the x-axis and power (μV2) is found on the y-axis


in insomniacs than in controls, at sleep-onset, during sleepas well as during wake, strongly supporting the neuro-cognitive model of insomnia, with hyperarousal reflectedthrough increased beta activity.

Regarding the different subtypes of insomnia, a report byKrystal et al. (2002) suggested that paradoxical insomniacsappeared to display greater high frequency activity thaneither psychophysiological insomnia sufferers or goodsleepers. In a preliminary report from our own researchteam in Quebec (St-Jean and Bastien 2008b), usingdifferent sub typing criteria than Krystal et al. (2002), nosignificant differences in absolute power of the differentEEG bands was observed between psychophysiological andparadoxical insomniacs. Recently, while accounting for ageand gender, Ouellet et al. (2010) observed that paradoxicalinsomniacs displayed greater powers in lower frequencybands than those with psychophysiological insomnia.Results from these studies are difficult to reconcile,

possibly because criteria for subtypes are not yet welldefined, and vary between studies. On the other hand, it isalso possible that PSA does not provide the informationthat could be related to the severity of sleep complaints. Inthat regard, it is possible that an individuals’ perceivedsleep severity is only partially explained by his or her EEGactivity. As such, when subjective-objective discrepanciesare too important (or at the far end of the severitycontinuum), they may not be reflected through elevatedEEG activities. Furthermore, it is also possible thataveraging multiple nights together instead of targetingnights during which “bad sleep” is reported, obscuresdifferences between groups.

Using PSA to test the hypothesis that sleep mechanismsmight be deficient in insomniacs compared to the ones ofgood sleepers, our group (Forget et al. 2007) has combinedPSA with spontaneous and evoked K-complexes to deter-mine if these events served the same sleep protective role in

Fig. 3 (from Perlis et al. 2001).Delta and beta relative powerplotted as a function of elapsedtime (30 s epochs) from sleeponset to lights on. The data fromthe good sleepers shows theinverse cyclic nature of therelationship between delta andbeta activity. More beta activityis seen in the insomnia groupthan in either the good sleepersof depressed patients


insomniacs and good sleepers. PSA was measured beforeand after the presence or absence of spontaneous or evokedK-complexes in early and late stages 2 of sleep. Resultsshowed significant similar changes following evoked andspontaneous K-complexes (increased activity in the deltafrequency band and decreased activity in theta, sigma, andbeta frequency bands) for both groups of sleepers, whilethere were no such marked changes in EEG activity when K-complexes were not evoked by stimuli. These results suggestthat both spontaneous and evoked K-Complexes promotedeeper sleep, further supporting a sleep protection role ofthese events and that this role is unaffected in insomnia.

Cyclic Alternating Pattern (CAP)

Most of the arousal-related phasic events (spontaneous orevoked during NREM sleep) follow a peculiar timeorganization described as cyclic alternating pattern (CAP)(Terzano et al. 1985). CAP displays a 20–40 s cyclicrhythm and is characterized by periods of cerebralactivation (phase A) followed by periods of deactivation(phase B; see Fig. 4 for an example). Phase A allows ahierarchic classification of different subtypes: A1, A2 andA3 (Ferri et al. 2005a, b; Terzano and Parrino 2000). Whilesubtype A1 appears to be linked to the synchronization ofthe EEG during sleep (Ferri et al. 2005a, b), subtypes A2and A3 appear to be linked to the EEG desynchronizingprocesses (Ferri et al. 2005a, b; Terzano and Parrino 2000).When the interval between two consecutive A phasesexceeds 60 s, the CAP sequence ends and sleep enters intoa non-CAP (NCAP) mode. While CAP sequences corre-

spond to segments of unstable sleep, NCAP sequences areassociated with stable sleep (Terzano et al. 2000, 2005).

Previous studies have shown that the A phase (activationphase) of the CAP was longer in insomniacs than in goodsleepers while the B phase (deactivation phase) was shorter(Halasz et al. 2004; Parrino et al. 2004). Furthermore, CAPrate (defined as the percentage ratio of total CAP time tototal NREM sleep time) was greater in insomniacs than ingood sleepers (Halasz et al. 2004). Phase B occupied about65% of the CAP sequence in good sleepers while only 50%in insomniacs. Higher CAP rates also correlated with poorersubjective sleep quality (Terzano et al. 1995, 2003).Recently, Parrino et al. (2009) compared 20 paradoxicalinsomniacs and 20 good sleepers on PSG and CAPparameters. Total CAP rate (58.1% vs. 35.5%) wassignificantly higher in paradoxical insomniacs than in goodsleepers. Paradoxical insomniacs also showed significantlyhigher amounts of CAP rate in stages 1 (62.7% vs. 37.5%)and 2 (53.3% vs. 33.1%), but not in slow wave sleep (SWS)compared to good sleepers. In addition, A2 subtypes weresignificantly more frequent in paradoxical insomniacs thanin controls (31% vs. 24%). The results from these studiestend again to suggest greater arousal (or “hyperactivation”)in the EEG of insomniacs than in the EEG of good sleepers.

Event-related Potentials (ERPs)

An external physical stimulus or internal psychologicalevent elicits small amplitude changes in the EEG, thereforeproviding a means to “probe” the extent of informationprocessing within the nervous system during wake and

Fig. 4 Cyclic Alternating Pattern(CAP). Reproduced with thepermission of M.G. Terzano. Thetop portion of the figure shows theCyclic Alternating Pattern in Stage2 sleep, with the two differentphases A (EEG activation) and B(EEG deactivation). Phase A andB composes a cycle and 60 sequals to a sequence. The bottomportion of the figure display aNon Cyclic Alternating Pattern(NCAP) period of sleep,corresponding to sleep stability.The recordings are shown fromfrontal, central and occipitalleads. In addition, the figuredisplays EOG and EMG, EKG(heart rate), abdominal (O-NPNG) and thoracic (THOR)respiratory efforts as well as bothright and left tibialis (respectivelyTIB ANT R and TIB ANT L)


sleep. Event-related potentials (ERPs) have a high temporalresolution (about one tenth of a millisecond) and are thus auseful tool to provide a precise timely measure of aspects ofinformation processing. ERP components are classifiedaccording to their latencies: early components(<80 msapproximately) reflect sensory processing, those around100 ms are sensitive to arousal and attention, and latercomponents usually reflect higher order central nervoussystem processing related to cognitive functions such asattention, vigilance, memory and inhibition of informationprocessing in wake and sleep (Colrain and Campbell 2007).The valence and latency of peaks are used to label thedifferent components: “N” refers to a Negative wave and“P” to a Positive wave. Thus, for example, “N100” wouldrefer to a negative wave appearing 100 ms after stimulusonset (note that most ERPs are usually termed as N1, P2,P3 even though their respective latency is 100, 200 and300 ms).

Although reduced relative to wakefulness, informationprocessing still remains present during sleep. Usually, as anindividual falls asleep, attention/vigilance processes de-crease (thus N1 amplitude is lower from wake to sleep) andthe inhibition of external stimuli becomes important topromote sleep (thus the amplitude of P2 is greater andN350 appears at sleep onset as a sign of an individualfalling asleep). Because it is impossible for subjects torespond verbally or behaviorally to a stimulus while theyare sleeping, the study of information processing duringsleep is limited. According to Colrain and Campbell (2007),ERPs can be very useful in the assessment of daytimeconsequences following sleep disruption and the determi-nation of the central nervous system function during sleep.As mentioned earlier, EEG studies have suggested thatcentral nervous system hyperarousal is associated withinsomnia. ERPs thus offer a means for direct evaluation ofcortical activation to an experimental stimulus during wakeand sleep. The oddball paradigm (a train of auditory stimulicontaining a frequent standard tone and a deviant tonewhich occurs rarely) is usually used to measure corticalexcitability. Participants can be instructed to pay attentionto the auditory stimulation (target stimuli generating a“P3”) or they can be asked to simply ignore or not to payattention to the stimulation (recordings of N1 and P2 duringwake and N1, P2 and N350 during sleep-onset and sleep).

For ERPs, general hypotheses linked to the neuro-cognitive model and hyperarousal theory of insomnia mightbe as follows, since these individuals should show signs ofenhanced sensory and information processing during wakeand sleep: 1) if sensory impairments are present ininsomniacs compared to good sleepers, P50 (reflectingprimary auditory cortex activation) amplitude should belower; 2) if insomniacs show signs of hyperarousal(enhanced information/cognitive processing), they should

display a larger N1 and a smaller P2 to standard and deviantstimuli as well as larger P3 to target stimuli relative to goodsleepers; 3) if insomniacs have difficulties inhibitingcortical arousal, a smaller N350 will be observed at sleeponset in these individuals relative to good sleepers.

In 1993, Hull conducted a study using an oddballparadigm during wakefulness and sleep onset. Resultsshowed that N350 was smaller in poor sleepers than innormal controls during the initial part of stage 2 sleep.Greater P300 amplitude immediately before sleep onsetand shorter response latency in insomniacs relative togood sleepers during wakefulness was also observed inthis study, suggesting a hyperarousal state in poorsleepers. That same year, Regestein and his colleagues(1993) reported a significantly larger P1-N1 in insomniacscompared to good sleepers during wakefulness. Loewyand Bootzin (1998) and Loewy et al. (1999) also recordeda larger N1 and a smaller P2 in insomniacs relative togood sleepers during wakefulness. Bastien et al. (2008b)compared primary chronic psychophysiological insom-niacs and good sleepers on N1, P2 and N350 componentsin a multi-assessment protocol. They reported hyperarous-al upon awakening in the morning (greater N1 amplitude)in psychophysiological insomniacs compared to goodsleepers. In addition, these authors recorded a smallerN350 in insomniacs than in good sleepers at sleep-onset,just as Hull had reported in 1993. Altogether, these studiesshowed that insomniacs are more vigilant (or “aroused”)during wakefulness than good sleepers controls (SeeFig. 5). However, studies conducted at sleep-onset alsotended to show that inhibition deficits are present ininsomniacs. Finally, while comparing different subtypes ofinsomnia, Bastien (2008) observed that paradoxicalinsomniacs, compared with psychophysiological insom-niacs and good sleepers, presented larger N1 and P2 aswell as lower N350 in the evening as well as during sleep-onset. Bastien et al. (2008b) thus suggested that bothprocesses, cortical arousal and inhibition deficits, areunderlying neurophysiological mechanisms of chronicinsomnia.

Does hyperarousal persists during sleep? First, Yang andLo (2007) reported that insomniacs show larger N1 andsmaller P2 to rare tones as well as smaller N350 to standardtones than controls during the first 5 min of continuousstage 2 sleep. However, Turcotte et al. (2009) observed thatduring the different stages of sleep, and contrary to whatYang and Lo (2007) had reported, insomniacs and goodsleepers did not differ on measures of N1 and P2 during thenight. Thus the issue of hyperarousal during the night, asmeasured with ERPs, requires more investigations ascontradictory results are reported to this point.

So far, studies using ERPs have mainly concentratedtheir efforts at providing empirical support for enhanced


cognitive information processing. An innovative study leadby Cote’s group (Milner et al. 2009), however, exploredwhether sensory information processing was deficient inpoor sleepers compared to normal sleepers. On a singlenight, paired-click stimuli were delivered to groups ofsleepers at pre-sleep (wake), REM and stage 2 sleep. P50amplitude to stimuli provided an index of sensory gating, asthe P50 amplitude observed to the second stimulus in eachpair is typically “gated” with substantially reduced ampli-tude relative to that seen to the first stimulus. Lower P50amplitudes were observed in poor sleepers compared withgood sleepers during wake, similar amplitudes in REMwhile no P50 was present in stage 2. Thus gating wasimpaired in poor sleepers during wake, was similar in thetwo groups in REM sleep and absent in stage 2 in bothgroups. Milner et al. (2009) thus provide very interestingempirical data, however they remain somewhat ambiguous.These authors interpreted the data as supporting thehypothesis that poor sleepers experience enhanced sensoryprocessing (i.e., hyperarousal) since insomniacs appeared topresent deficits at “gating” the stimulation. However, sincegating was less efficient in insomniacs relative to goodsleepers, a failure in inhibitory mechanisms could alsoexplain the results of their study.

Because variability in sleep quality is challenging ininsomnia, it is possible that information processing variesaccording to the quality of the night (preceding orfollowing ERPs recordings). Some studies have thus triedto associate sleep quality in insomnia (subjective, measuredwith a diary and/or objective, measured with PSG) todaytime measures of ERPs. Devoto and colleagues (2003)compared insomniacs and good sleepers after a subjective“good night” and after a subjective “bad night”. Theyobserved that P3 amplitude was lower after a good night ininsomniacs compared to good sleepers. On the other hand,

after a bad night, P3 amplitude was significantly higheramong insomniacs than among good sleepers. In a similardesign but this time using actigraphy for measuring sleep,the same group of researchers (Devoto et al. 2005)observed larger P3 in insomniacs on the worst night ofsleep compared to good sleepers. Both sets of results thusappeared to support the neurocognitive theory of hyper-arousal. Sforza and Haba-Rubio (2006) examined therelationship between evening-morning changes in ERPsamplitudes (N1, P2 or P3) and general PSG measures ininsomniacs and controls. ERPs in the two groups weresimilar in the evening; however, N1 amplitude was smallerin the morning relative to the pre-sleep N1 and wassignificantly negatively correlated with the amount of stage1 sleep, the number of stage changes and the number ofawakenings during the night in the insomniacs. No suchrelations were observed for good sleepers. Turcotte andBastien (2009) investigated the relationship between objec-tive sleep parameters and the amplitudes and latencies ofERPs components in a multi-assessment protocol. Theseauthors found that as the amplitude of N1 and P2 increasedbefore going to sleep among insomniacs, the sleep qualityof the following night decreased. In addition, the sleepquality of the previous night also appeared to be linked tothe ERPs’ amplitudes recorded on the following morning.Turcotte and Bastien (2009) concluded that the existinghyperarousal and inhibition deficits in insomniacs aredirectly associated with a poorer sleep quality. Consideringthese results, it is most probable that daily consequences ofinsomnia such as reported memory and attention deficitsalso vary with the quality of the night preceding theevaluation. Thus, it would be advisable to document thequality of the preceding night whichever cognitive evalu-ation is to be taken place in the morning.

Together, these studies suggest the presence of height-ened cortical/cognitive arousal levels in insomnia, as washypothesized at the beginning of this section. Thisincreased arousal appears to be present both whileinsomniacs are awake as well as during sleep. However,the recordings of some ERPs (N350 and P50) suggest thatinhibitory deficits appear also present in insomnia. It is thuspossible that both mechanisms are at play and depending ontask and demands from a cognitive point of view, resultsdiffer. Moreover, since limited data are available underconditions requiring both a motor and cognitive response,that ERPs protocols have not yet specifically targetedinhibition in a cognitive task, and because sleep quality(albeit be subjective or objective) is not always docu-mented, the issue of hyperarousal and/or inhibition deficitsin insomnia remains one of debate. Nonetheless, the utilityof ERPs as a useful tool to investigate cognitive functionsor information processing during wake and/or sleep ininsomniacs is undisputable. In this regard, contrary to PSA

Fig. 5 (from Bastien et al. 2008b). ERPs for good sleepers (GS) andpsychophysiological insomnia sufferers (INS) at sleep onset (SO).Between groups significant differences are illustrated with the blackcircles. N1 to the standard tones and N350 to the deviant tones aresmaller for insomniacs than good sleepers


and although unlike imaging they do not identify the activebrain structures at one moment in time, ERPs provide adirect measure and precise timing of cortical arousal anddynamic neural mechanisms.

Neuroimaging

Neuroimaging is relatively new in insomnia research.Although many methodological endpoints still need to beaddressed (e.g. small sample sizes, small window ofanalysis, sleeping in the machine, etc.), neuroimagingstudies have enhanced our understanding of the underlyingneurophysiology of this disorder. Detailed information onstudied samples, methods and recordings of the work citedon neuroimaging is depicted in Table 1.

Single-photon-emission computed tomography (SPECT)is a method allowing the visualization of cerebral metab-olism by analyzing the signal emitted by a short-livedradio-active tracer injected intravenously before scanning.Using SPECT, Smith and collaborators (2002), observedthat insomniacs, relative to self-defined good sleepers,showed a decrease in cerebral blood flow in the medialfrontal, occipital and parietal cortices, with a profounddeactivation in the basal ganglia, suggesting that patientswith primary insomnia were hypometabolic rather thanhypermetabolic in a 2 min window of perfusion during thefirst NREM sleep period when compared to good sleepers.These results thus appeared to contradict the hyperarousaltheory. It is however possible that insomniacs are morearoused over the totality of NREM sleep. While this studyproves the feasibility of a neuroimaging technique incombination with traditional PSG, the results remaindifficult to interpret in light of methodological restrictions(small sample size and limited time window for analyses).

Positron emission tomography (PET) is a more accuratemeasure of whole brain activity than SPECT. It can be usedduring wakefulness as well as during sleep and involves theinjection (usually into blood circulation) of a short-livedradioactive tracer isotope which decays by emitting apositron, chemically incorporated into a metabolicallyactive molecule. Using PET, Nofzinger et al. (2004)showed an increased global cerebral glucose metabolismin insomniacs compared with good sleeper controls duringNREM sleep (See Fig. 6). In addition, a smaller decrease incerebral activity from wakefulness to NREM sleep states inthe ascending reticular activating system (ARAS), thehypothalamus, thalamus, insular cortex, the amygdala, thehippocampus, the anterior cingulate and medial prefrontalcortices was observed in insomniacs compared with goodsleepers. During wakefulness, insomniacs showed relativehypometabolism in bilateral prefrontal, left hemisphericsuperior temporal, parietal and occipital cortices, thethalamus, hypothalamus and brainstem reticular formation

compared to good sleepers, the sleep of insomniacs beingthus associated with increased brain metabolism. Thisrelative hypometabolism in similar brain structures hadpreviously been reported in healthy good sleepers followingsleep deprivation (Thomas et al. 2000). These findingsmight thus suggest that insomniacs are chronically sleep-deprived and their sleep is associated with increased brainmetabolism. The ‘inability’ to fall asleep may thus beexpressed by arousal mechanisms failing to be less active.

Recently, Altena et al. (2008a, b) replicated theprefrontal hypoactivation during wakefulness in insom-niacs using functional magnetic resonance imaging (fMRI)during a test performance of a category and a letter fluencytask. The fMRI measures the haemodynamic responserelated to neural activity in the brain. Altena andcolleagues (2008a, b) observed less activation in insom-niacs than in controls in the left medial prefrontal cortexand left inferior frontal gyrus while individuals wereperforming both tasks.

Szelenberger and Niemcewicz (2001) used low resolu-tion electromagnetic tomography analysis (LORETA) ofmulti-channel EEG (a non-invasive method that allows themapping of current density measures according to astandard brain atlas, in order to localize electrical activityin the brain). They found that insomniacs showed lesscurrent density in cerebral regions involved in affect andcognitive performance (i.e., orbitofrontal, medial prefrontaland anterior cingulated cortices) relative to normal controlsduring wake.

In summary, these studies reveal both a hypermetabolismin certain brain regions indicating an increase in arousal ora difficulty in inhibiting arousal and a hypometabolism inspecific areas that have been related to sleep deprivation.Moreover, these findings also highlight the presence ofinteracting neural networks including a general arousalsystem (ascending reticular formation and hypothalamus),an emotion regulating system (hippocampus, amygdala andanterior cingulate cortex), and a cognitive system (prefron-tal cortex) in the neurobiology of insomnia. Neuroimagingstudies performed on larger and heterogeneous samples ofinsomniacs could help classify the different subtypes (psy-chophysiological or paradoxical) and shed some light onpossible shared neurobiological bases of different clinicalrepresentations of the disorder (stand-alone syndrome orcomorbid with another psychopathology).

Treatment

Pharmacotherapy

Pharmacotherapy, using benzodiazepines (BZDs) and non-benzodiazepine hypnotic agents (like Zolpidem), is the most


Tab

le1

Neuroim

agingstud

ies

Autho

rsObjectiv

eParticipants

Metho

dsSleep

stages

Results

Interpretatio

n

Studies

comparing

insomniasufferersto

controls

Smith

etal.20

02Provide

prelim

inarydata

onregion

alcerebral

activ

ityin

INSvs

GS.

−5IN

S(W

)Mage

37.8

(12.1)

years

3PSG

nigh

ts1st NREM

sleep

cycle

INSshow

edapattern

ofdecreasedrCBF

over

allregion

s.Com

paredto

GS,the

largestandsignificantreductions

inIN

Sarelocalized

inthebasalgang

lia,

themedialfron

talcortex,theoccipital

cortex

andtheparietal

cortex.

Con

tradictsthehy

perarousal

theory

ofinsomnia.

Insomniamight

beassociated

with

abno

rmal

CNS

activ

itydu

ring

NREM

sleep.

−4GS(W

)Mage

34.5

(11.9)

Night

1=adaptatio

n

Night

2=evaluatio

n

Night

3=PSG

+SPECT

scan

Nofzing

eretal.2

004

Investigatethe

neurob

iologicalbasisof

poor

sleepanddaytim

efatig

uein

insomnia

−7IN

S(4W,3M

)Mage34

.2(8.9)years.

3PSG

nigh

tsWakeand

NREM

sleep.

INSshow

edincreasedglob

alcerebral

glucosemetabolism

comparedwith

GS

during

both

wakingandNREM

sleep,

asm

allerdecrease

inrelativ

emetabolism

from

wakefulness

toNREM

sleepandredu

cedrelativ

emetabolism

intheprefrontal

cortex

while

awake.

Insomniaisassociated

with

greater

brainmetabolism.Daytim

efatig

uemay

reflectdecreasedactiv

ityin

the

prefrontal

cortex

resulting

from

inefficientsleep.

−20GS(13W

,7M)Mage

32.6

(8.4)years.

Night

1=adaptatio

nNight

2=evaluatio

n+

PETscan

during

the

morning

Night

3=PSG

+PETscan

Studies

implying

treatm

entforinsomnia

Smith

etal.20

05Evaluatewhether

behaviou

rtherapymight

beassociated

with

norm

alizationin

brain

functio

n.

−4IN

S(W

)Mage34

.5(12)

years.

SPECTscan

during

NREM

sleepbefore

andafterbehavior

therapy.

NREM

sleep

INSshow

edasign

ificantincrease

incerebralbloo

dflow

inthebasalg

anglia

aftertreatm

ent.

InIN

S,sleep

therapymay

beassociated

with

areversal

incerebral

deactiv

ation.

(See

Smith

etal.20

02)

Altena

etal.

2008

a,b

Investigatefunctio

nal

brain

activ

ationin

INSand

mod

ifications

aftera

mutli-mod

alsleep

therapy.

−21IN

S(17W

,4M

),Mage61

(6.2)years.

fMRIscanning

during

theperformance

testing

INSshow

edless

activ

ationthan

controls

incertainregion

sdu

ring

the

performance

ofbo

thtasks.

Insomniainterferes

inareversible

way

with

activ

ationof

theprefrontal

cortex

during

daytim

eperformance.

12GS(9W:3M

)Mage60

(8.2)years.

Sleep

therapyfor6weeks

After

treatm

ent,theactiv

ationwas

restored

inbo

thtasks.

INSInsomniasufferers;GSGoo

dsleepers,W

Wom

en,M

Men;Mag

eMeanage;SDin

parentheses,PSG

Polysom

nography,P

ETPositron

Emission

Tom

ograph

y,SP

ECTSingle-ph

oton

-emission

compu

tedtomog

raph

y,fM

RIFun

ctionalmagnetic

resonanceim

aging,

NREM

nonrapideyemov

ement


common treatment for insomnia (Hohagen et al. 1994; Morinand Wooten 1996; Kupfer and Reynolds 1997). Thesepsychotropic drugs bind with gamma-amino butyric acid(GABA) receptors, thus increasing CNS inhibitory effects.Although short-term use of hypnotic medications (less than4 weeks) may be useful and indicated for acute insomnia,(NIH 1991), there is still little information on long-termefficacy. On a short-term basis, pharmacotherapy can behelpful at decreasing sleep onset and sleep maintenancedifficulties as well as improving sleep continuity. On theother hand, these compounds, especially BZDs, are associat-ed with potential adverse effects (e.g., memory impairments)and altered sleep structure (e.g., reduces stages 3 and 4). Inaddition, the efficacy of long-term use is questioned and ithas been documented that long-term use causes tolerance andphysical dependence in addition to a withdrawal syndromeupon cessation of use (Lader 2008; Lader et al. 2009).

Bastien et al. (2003) performed PSA on the EEG of threegroups of older adults: medication free or long-termbenzodiazepine users insomniacs and good sleepers. Benzo-diazepines did not mask the effect of insomnia and less delta,less theta and more beta1 activity were observed in thedifferent stages of sleep. Some increases in power in thedelta activity band have also been reported in insomniacsand women subjectively complaining of insomnia with theuse of Zolpidem, a short-acting non benzodiazepine hypnotic(Monti et al. 2000; Declerck et al. 1992). However, resultswere not significantly related to objective and subjectivesleep improvements.

Unfortunately, many insomniacs use medications to allevi-ate their sleep difficulties (Holbrook et al. 2000). Knowing thatusage of BZDs produce adverse effects and its long-term usedependence, and because the adverse long-term effects ofnon-benzodiazepines agents are still not well known, more

research is needed in this area. In addition, since theGABAergic system is the most important inhibitory mecha-nism in the central nervous system (CNS) (Gottesmann2002), and that it has recently been shown that a globalreduction in GABA neurotransmitters is present in insomniacscompared to good sleepers (Winkelmann et al. 2008), one canagain question if insomniacs do not suffer from a cortical/cerebral unbalance in inhibition and excitation mechanisms.

Non-pharmacological Treatments

Cognitive-Behavioral Treatment for insomnia (CBT-I) hasbeen recognized as an effective non-pharmacological treat-ment for insomnia (Morin 2004). In most studies, CBT-Iincludes behavioral components (stimulus control and sleeprestriction), cognitive therapy, and sleep hygiene. Othercomponents may be added to treatment: relaxation (Lichstein2000), biofeedback (Bootzin and Rider 1997), bright lightexposure (Lack and Wright 2007), body temperaturemanipulations (Van Someren 2006) and physical activity(Martin et al. 2007). Altogether, non-pharmacological treat-ments for insomnia have shown short and long-termsustainable gains such as decreased sleep onset latency andwake after sleep-onset, increased total sleep time and sleepefficiency.

The stimulus control instructions (Bootzin et al. 1991)are aimed at re associating the bed, bedroom, and bedtimestimuli with sleep rather than with the frustration andanxiety associated with sleeplessness. The sleep restrictionprocedure (Spielman et al. 1987) consists of limiting thetime spent in bed to the actual time spent sleeping. Therationale is that individuals with insomnia often spendexcessive amounts of time in bed in a misguided attemptto get more sleep. Participants are instructed to determine

Fig. 6 (from Nofzinger et al. 2004). Brain structures that did notshow decreased metabolic rate from waking to sleep states in patientswith insomnia (a) and brain structures where relative metabolism

while awake was higher in healthy subjects than in patients withinsomnia (b) a Differences in all regions shown reached statisticalsignificance at the p<0.05, corrected, level


their allowable time in bed according to their totalsubjective sleep time. Cognitive therapy of insomniaconsists of identifying, challenging, and altering a set ofdysfunctional beliefs and attitudes about sleep (Morin1993). The objective of cognitive therapy is to break thevicious circle of insomnia, dysfunctional thoughts, andemotional distress that lead to further sleep disturbance.Sleep-hygiene education consists of teaching individualsabout the impact of certain lifestyle habits (e.g., diet, druguse, and exercise) and the influence of some environmentalfactors (e.g., light, noise, and temperature) on sleep (Hauri1991).

Non-pharmacological treatments for insomnia appear tohave a positive impact on EEG frequency. For example,Jacobs et al. (1993), using a multifactor behavioralintervention for insomnia, observed a significant decreasein beta activity in insomniacs at post-treatment compared topre-treatment. The intervention thus appeared to havereduced pre-sleep CNS arousal. Later, with CBT-I, Cervenaet al. (2004) observed EEG changes at post-treatment.Some of these changes were in NREM sleep, during whichslow wave activity (SWA) powers increased and betadecreased. These results led the authors to suggest thatCBT-I might decrease CNS hyperarousal during sleep andreinforce the cortical synchronization and sleep pressure tomaintain sleep longer.

Smith et al. (2005) examined the effects of behaviortherapy on cerebral blood flow (SPECT) during NREMsleep. When comparing SPECT measures of insomniacspost- to pre-treatment, a significant increase in cerebralblood flow occurred in the basal ganglia. Using fMRI,Altena and colleagues (2008a, b) observed that after CBT-Ithe activation of the different brain regions was restored ininsomniacs. Therefore, it was suggested that insomniainterferes in a reversible way with activation of theprefrontal cortex during daytime performance. Recently,Van der Werf and colleagues (2010) used transcranialmagnetic stimulation (“TMS”) to compare the degree ofintracortical excitability in insomniacs and good sleepers.These authors delivered short-lived pulses of a strongmagnetic field over the scalp of 16 insomniacs who wereassigned either to CBT-I (plus bright light exposure, bodytemperature manipulations, and structured physical activity;for details see Altena et al. 2008a, b) or a waitlist controlgroup. Pre and post-treatment results showed that even ifsubjective sleep latency decreased and sleep efficiencyincreased, an increased absolute excitability remained aftertreatment in insomniacs. The authors suggested that thisincreased intracortical excitability might be a sustainabletrait in insomniacs.

It thus appears that non pharmacological treatment ofinsomnia can decrease ‘hyperarousal’ as observed withfindings measuring the combination of PSA or neuro-

imaging with subjective sleep measures (pre- and post-treatment comparisons). This combination might afterwardbe helpful at disentangling the different subtypes ofinsomnia. Furthermore, measuring brain activity levelsthrough EEG or neuroimaging after treatment targetingbehavior or both behavior and cognitive processes (CBT-I)not only reflect therapeutic gains but validate the usefulnessand genuine benefits of these treatments. In addition, on amore theoretical note, the comparison of pre and post EEGmeasures enhances our understanding of other activeneurophysiological processes in insomnia, and one canquestion if CBT-I for example, unlike long-term use ofBZDs, does not re-establish (at least partly) arousal levelsin insomniacs.

Conclusion

Whether true insomnia is purely subjective in nature orrequires presentation of objectively measured sleep diffi-culties is a matter of genuine debate. Insomnia, bydefinition, is at least primarily a subjective disorder, as itis the subjective report of the individual that will lead to thediagnosis of insomnia per se. Diagnostic criteria from theAmerican Psychiatric Association (APA) are currentlybeing updated and will most likely change not only ourresearch approach to insomnia but also the whole languageof it (for more information on the next set of criteria,readers are invited to visit the APAweb page at http://www.dsm5.org/Pages/Default.aspx). For example, the proposednew criteria now specify that to be chronic, sleepdifficulties must now last of at least 3 months.

Protocols that incorporate both subjective and objectiveratings of sleep quality and cognitive processing are gainingin importance and might be more informative than anysingle measure used alone, albeit be PSG or more refinedEEG measures. For example, by combining PSG andsubjective sleep ratings with fMRI or ERPs, we can notonly more thoroughly explore subjective and objectivediscrepancies, we can also provide invaluable informationon the neurophysiological bases of the disorder. Although itappears that “hyperarousal” as has been put forward in theneurocognitive model of insomnia can be measured throughEEG (e.g. PSA), the interplay between arousal andde-arousal (i.e., inhibitory) mechanisms cannot be ignoredany longer. For example, ERPs studies, both measuringsensory and information processing, have shown that gatingdeficits and an inability to initiate normal sleep processesare indeed present in individuals with insomnia comparedto good sleepers. Individuals with insomnia might thus beas afflicted with inadequate “dis-inhibition” mechanisms(expressed in an inability to disengage) in cognitive sensoryand emotional processes as these individuals are inappro-


http://www.dsm5.org/Pages/Default.aspx

http://www.dsm5.org/Pages/Default.aspx

priately aroused by these same processes. Thus on a strictlytheoretical aspect, our understanding of “arousal” and“de-arousal” processes at play in insomnia is still limited,but recent findings in neuroimaging, combined or not withnon-pharmacological treatment, are promising and maylead to uncovering other aspects of this multidimensionaldisorder.

In everyday life, and general clinical practice, objectivereports (from PSG to neuroimaging) are not available, norpractical. However, one must never forget that there are dailyconsequences to insomnia and increasingly, insomnia isrecognized as a 24 h problem, even with a waxing and waningintensity/severity in sleep difficulties during defined periods oftime. As such, not only are sleep difficulties important toconsider, the consequences linked to these sleep difficultiesmust be thoroughly explored and documented. The currentreview has focused on the state of research on neurophysio-logical and neuropsychological measures while suggestingthat combining measures might be the best avenue offering amore genuine picture of insomnia. In addition, promoting andincreasing the use of treatments known to be well effective forinsomnia, such as CBT-I, will alleviate greatly the personalburden of this disorder for individuals with insomnia whilealso decreasing the economic cost for our society.

References

Altena, E., Van Der Merf, Y. D., Strijers, R. J., & Van Someren, E. J. W.(2008). Sleep loss affects vigilance: effects of chronic insomnia andsleep therapy. Journal of Sleep Research, 17(3), 335–343.

Altena, E., Van Der Merf, Y. D., Sanz-Arigita, E. J., Voorn, T. A.,Rombouts, S. A., Kuijer, J. P., et al. (2008). Prefrontal hypoactiva-tion and recovery in insomnia. Sleep, 31(9), 1271–1276.

AmericanAcademy of SleepMedicine. (2005). International classificationof sleep disorders: Diagnostic and coding manual (2nd ed.).Westchester: American Academy of Sleep Medicine.

American Psychiatric Association. (1994). Diagnostic and statisticalmanual of mental disorders (4th ed.). Washington: AmericanPsychiatric Association.

American Psychiatric Association. (2000). Diagnostic and statisticalmanual of mental disorders. Fourth Edition, Text Revision (DSM-IV-TR). Washington: American Psychiatric Association.

Backhaus, J., Junghanns, K., Born, J., Hohaus, K., Faasch, F., &Hohagen, F. (2006). Impaired declarative memory consolidationduring sleep in patients with primary insomnia: influence of sleeparchitecture and nocturnal cortisol release. Biological Psychiatry,60, 1324–1330.

Bastien, C. H. (2008). ERP measures during wakefulness and sleep-onset in psychophysiological and paradoxical insomnia sufferers.Journal of Sleep Research, 17(Suppl. 1), 235.

Bastien, C. H., Fortier-Brochu, E., Rioux, I., LeBlanc, M., Daley, M.,& Morin, C. M. (2003). Cognitive performance and sleep qualityin elderly suffering of chronic insomnia: relations betweenobjective and subjective measures. Journal of PsychosomaticResearch, 54(1), 39–49.

Bastien, C. H., Leblanc, M., Carrier, J., & Morin, C. M. (2003). SleepEEG power spectra insomnia and chronic use of benzodiazepines.Sleep, 26(3), 313–317.

Bastien, C. H., Vallières, A., & Morin, C. M. (2004). Precipitatingfactors of insomnia. Behavioral Sleep Medicine, 2(1), 50–62.

Bastien, C. H., Guimond, S., Lemelin, S., St-Jean, G., & Turcotte, I.(2008a). Signs of insomnia in borderline personality individuals.Journal of Clinical Sleep Medicine, 4(5), 462–470.

Bastien, C. H., St-Jean, G., Turcotte, I., Morin, C. M., & Carrier, J.(2008b). Chronic psychophysiological insomnia: hyperarousaland/or inhibition deficits? An ERPs investigation. Sleep, 31(6),887–898.

Bastien, C. H., St-Jean, G., Turcotte, I., Morin, C. M., Lavallée, M., &Carrier, J. (2009a). Sleep spindles in chronic psychophysiologicalinsomnia. Journal of Psychosomatic Research, 66, 59–65.

Bastien, C. H., St-Jean, G., Turcotte, I., Morin, C. M., Lavallée, M.,Carrier, J., et al. (2009b). Spontaneous K-complexes in chronicpsychophysiological insomnia. Journal of Psychosomatic Re-search, 67, 117–125.

Bonnet, M. H., & Arand, D. L. (1995). 24-Hour metabolic rate ininsomniacs and matched normal sleepers. Sleep, 18, 581–588.

Bonnet, M. H., & Arand, D. L. (1997). Hyperarousal and insomnia.Sleep Medicine Reviews, 1(2), 97–108.

Bootzin, R. R., & Rider, S. P. (1997). Behavioral techniques andbiofeedback for insomnia. In M. R. Pressman & W. C. Orr (Eds.),Understanding sleep: The evaluation and treatment of sleepdisorders (pp 315–338). APA.

Bootzin, R. R., Epstein, D., & Wood, J. M. (1991). Stimulus controlinstructions. In P. Hauri (Ed.), Case studies in insomnia (pp. 19–28). New York: Plenum Press.

Broadbent, D. E., Cooper, P. F., Fitzgerald, P., & Parkes, K. R. (1982).The Cognitive Failures Questionnaire (CFQ) and its correlates.The British Journal of Clinical Psychology, 21, 16–21.

Broman, J. E., Lundh, L. G., Aleman, K., & Hetta, J. (1992). Subjectiveand objective performance in patients with persistent insomnia.Scandinavian Journal of Behavior Therapy, 21, 115–126.

Buysse, D. J., Ancoli-Israel, S., Edinger, J. D., Lichstein, K. L., &Morin, C. M. (2006). Recommendations for a standard researchassessment of insomnia. Sleep, 29(9), 11551173.

Buysse, D. J., Germain, A., Hall, M. L., Moul, D. E., Nofzinger, E.A., Begley, A., et al. (2008). EEG spectral analysis in primaryinsomnia: NREM period effects and sex differences. Sleep, 31(12), 1673–1682.

Cervena, K., Dauvilliers, Y., Espa, F., Touchon, J., Matousek, M.,Billiard, M., et al. (2004). Effect of cognitive behavioral therapyfor insomnia on sleep architecture and sleep EEG power spectrain psychophysiological insomnia. Journal of Sleep Research, 13(4), 385–393.

Coates, T. J., Killen, J. D., George, J., Marchini, E., Silverman, S., &Thoresen, C. (1982). Estimating sleep parameters: a multitrait—multimethod analysis. Journal of Consulting and ClinicalPsychology, 50, 345–352.

Coates, T. J., Killen, J. D., Silverman, S., George, J., Marchini, E.,Hamilton, S., et al. (1983). Cognitive activity, sleep disturbance,and stage specific differences between recorded and reportedsleep. Psychophysiology, 20, 243–250.

Colrain, I. M. (2005). The K-complex: a 7-decade history. Sleep, 28(2), 255–273.

Colrain, I. M., & Campbell, K. B. (2007). The use of evoked potentials insleep research. Sleep Medicine Reviews, 11, 277–293.

De Gennaro, L., & Ferrara, M. (2003). Sleep spindles, an overview.Sleep Medicine Review, 7, 423–440.

Declerck, A. C., Ruwe, F., OHanlon, J. F., & Wauquier, A. (1992).Effects of Zolpidem and Flunitrazepam on nocturnal sleep ofwomen subjectively complaining of insomnia. Psychopharma-cology, 106(4), 497–501.

Devoto, A., Violani, C., Lucidi, F., & Lombardo, C. (2003). P300amplitude in subjects with primary insomnia is modulated bytheir sleep quality. Journal of Psychosomatic Research, 54, 3–10.


Devoto, A., Manganelli, S., Lucidi, F., Lombardo, C., Russo, P. M., &Violani, C. (2005). Quality of sleep and P300 amplitude inprimary insomnia: a preliminary study. Sleep, 28(7), 859–863.

Dorsey, C. M., & Bootzin, R. R. (1997). Subjective and psychophys-iologic insomnia: an examination of sleep tendency andpersonality. Biological Psychiatry, 41, 209–216.

Edinger, J. D., & Krystal, A. D. (2003). Subtyping primary insomnia:is sleep state misperception a distinct clinical entity? SleepMedicine Reviews, 7(3), 203–214.

Edinger, J. D., Bonnet, M. H., Bootzin, R. R., et al. (2004). Derivationof research diagnostic criteria for insomnia: report of anAmerican Academy of Sleep Medicine Work Group. Sleep, 27(8), 1567–1596.

Edinger, J., Means, M. K., Carney, C. E., & Krystal, A. D. (2008).Psychomotor deficits and their relation to prior’snight sleepamong insomnia individuals with primary insomnia. Sleep, 31(5),599–607.

Espie, C. A. (2002). Insomnia: conceptual issues in the development,persistence, and treatment of sleep disorders in adults. AnnualReview of Psychology, 53, 215–243.

Ferri, R., Bruni, O., Miano, S., Plazzi, G., & Terzano, M. G. (2005).All-night EEG power spectral analysis of the cyclic alternatingpattern components in young adult subjects. Clinical Neurophys-iology, 116, 2429–2940.

Ferri, R., Bruni, O., Miano, S., & Terzano, M. G. (2005). Topographicmapping of the spectral components of the cyclic alternatingpattern (CAP). Sleep Medicine, 6(1), 29–36.

Forget, D., Morin, C., & Bastien, C. (2007). Spectral characteristicssurrounding the spontaneous K-Complex in good sleepers. Sleep,30, A34.

Fortier-Brochu, E., Beaulieu-Bonneau, S., Ivers, H., & Morin, C. M.(2008). Insomnia and neuropsychological performance. Sleep, 31(Suppl. 1), A226.

Freedman, R. R. (1986). EEG power spectra in sleep-onset insomnia.Electroencephalography and Clinical Neurophysiology, 63, 408–413.

Fulda, S., & Schulz, H. (2001). Cognitive dysfunction in sleepdisorders. Sleep Medicine Reviews, 5(6), 423–445.

Gottesmann, C. (2002). GABA mechanisms and sleep. Neuroscience,111, 231–239.

Gu, Y., Wahlund, B., Liljenström, H., von Rosen, D., & Liang, H.(2005). Analysis of phase shifts in clinical EEG evoked by ECT.Neurocomputing, 65–66, 475–483.

Haimov, I., Hanuka, E., & Horowitz, Y. (2008). Chronic insomnia andcognitive functioning among older adults. Behavioral SleepMedicine, 6(1), 32–54.

Halasz, P., Terzano, M. G., Parrino, L., & Bodizs, R. (2004). Thenature of arousal in sleep. Journal of Sleep Research, 13, 1–23.

Harvey, A. G. (2002). A cognitive model of insomnia. Behavior andResearch Therapy, 40, 869–893.

Hatoum, H. T., Kong, S. X., Kania, C. M., Wong, J. M., &Mendelson, W. B. (1998). Insomnia, health-related quality oflife and healthcare resource consumption. A study of managed-care organisation enrolees. Pharmacoeconomics, 14, 629–637.

Hauri, P. J. (1991). Case studies in insomnia. New York: PlenumPress.

Hauri, P. J. (1997). Cognitive deficits in insomnia patients. ActaNeurologica Belgia, 97, 113–117.

Hauri, P. J., & Fisher, J. (1986). Persistent psychophysiologic(learned) insomnia. Sleep, 9(1), 38–53.

Hauri, P., & Olmstead, E. (1980). Childhood onset insomnia. Sleep, 3(1), 59–65.

Hauri, P. J., & Olmstead, E. (1983). What is the moment of sleep onsetfor insomniacs? Sleep, 6(1), 10–15.

Haynes, S. N., Fitzgerald, G., Shute, G., & O'Meary, M. (1985).Responses of psychophysiologic and subjective insomniacs to

auditory stimuli during sleep: a replication and extension.Journal of Abnormal Psychology, 94(3), 338–345.

Hess, R. (1965). Sleep and sleep disturbances in the electroenceph-alogram. In K. Akert, C. Belly, & J. Schade (Eds.), Sleepmechanisms (pp. 127–139). Amsterdam: Elsevier.

Hillman, D. R., Murphy, A. S., Antic, R., & Pezzullo, L. (2006). Theeconomic cost of sleep disorders. Sleep, 29(3), 299–305.

Hohagen, F., Kappler, C., Schramm, E., Rink, K., Weyerer, S.,Riemann, D., et al. (1994). Prevalence of insomnia in elderlygeneral practice attenders and the current treatment modalities.Acta Psychiatrica Scandinavica, 90, 102–108.

Holbrook, A. M., Crowther, R., Lotter, A., Cheng, C., & King, D.(2000). Meta-analysis of benzodiazepine use in the treatmentof insomnia. Canadian Medical Association Journal, 162, 225–233.

Hull, J. S. (1993). Event-related potentials during the wake/sleeptransition in adults with and without primary insomnia [disser-tation]. Hattiesburg: Univ. of Southern Mississippi.

Jacobs, G. D., Benson, H., & Friedman, R. (1993). Home-basedcentral nervous system assessment of a multifactor behavioralintervention for chronic sleep onset insomnia. Behavior Therapy,24(1), 159–174.

Jankel, W. R., & Niedermeyer, E. (1985). Sleep spindles. Journal ofClinical Neurophysiology, 2, 1–35.

Krystal, A. D., Edinger, J. D., Wohlgemuth, W. K., & Marsh, G. R.(2002). NREM sleep EEG frequency spectral correlates of sleepcomplaints in primary insomnia subtypes. Sleep, 25(6), 630–640.

Kupfer, D. J., & Reynolds, C. F. (1997). Management of insomnia.The New England Journal of Medicine, 336, 341–346.

Lack, L. C., & Wright, H. R. (2007). Treating chronobiologicalcomponents of chronic insomnia. Sleep Medicine, 8, 637–644.

Lader, M. (2008). Effectiveness of benzodiazepines: do they work ornot? Expert Review of Neurotherapeutics, 8(8), 1189–1191.

Lader, M., Tylee, A., & Donoghue, J. (2009). Withdrawing benzo-diazepines in primary care. CNS Drugs, 23(1), 19–34.

Lamarche, C. H., & Ogilvie, R. D. (1997). Electrophysiologicalchanges during the sleep onset period of psychophysiologicalinsomniacs, psychiatric insomniacs and normal sleepers.Journal of Sleep Research and Sleep Medicine, 20(9), 724–733.

Léger, D., Scheuermaier, K., Philip, P., Paillard, M., & Guilleminault,C. (2001). SF36: evaluation of quality of life in severe and mildinsomniacs compared with good sleepers. Psychosomatic Medi-cine, 63, 49–55.

Lichstein, K. L. (2000). Relaxation. In K. L. Lichstein & C. M. Morin(Eds.), Treatment of late-life insomnia (pp. 185–206). ThousandOaks: Sage.

Lichstein, K. L., Durrence, H. H., Bayen, U. J., & Riedel, B. W.(2001). Primary versus secondary insomnia in older adults:subjective sleep and daytime functioning. Psychology and Aging,16(2), 264–271.

Loewy, D. H., & Bootzin, R. (1998). Event-related potentialsmeasures ofinformation processing in insomniacs at bedtimeand during sleep. Sleep, 21(Suppl. 1), 98.

Loewy, D. H., Burdik, R. S., Al-Shajlawi, A., Franzen, P., & Bootzin,R. (1999). Enhanced information processing at the peri-sleeponset period in insomniacs as measured by event-relatedpotentials. Sleep, 22(Suppl. 1), S152.

Martin, J. L., Marler, M. R., Harker, J. O., Josephson, K. R., & Alessi, C.A. (2007). A multicomponent nonpharmacological interventionimproves activity rhythms among nursing home residents withdisrupted sleep/wake patterns. The Journals of Gerontology. SeriesA: Biological Sciences and Medical Sciences, 62(1), 67–72.

Mendelson, W. B., Garnett, D., Gillin, J. C., & Weingartner, H.(1984). The experience of insomnia and daytime and nighttimefunctioning. Psychiatry Research, 12(3), 235–250.


Mendelson, W. B., James, S. P., Garnett, D., Sack, D. A., &Rosenthal, N. E. (1986). A psychophysiological study ofinsomnia. Psychiatry Research, 19, 267–284.

Merica, H., & Gaillard, J. (1992). The EEG of the sleep onset periodin insomnia: a discriminant analysis. Physiology & Behavior, 52,199–204.

Merica, H., Blois, R., & Gaillard, J.-M. (1998). Spectral character-istics of sleep EEG in chronic insomnia. The European Journalof Neuroscience, 10, 1826–1834.

Milner, C. E., Cuthbert, B. P., Kertesz, R. S., & Cote, K. A. (2009).Sensory gating impairments in poor sleepers during presleepwakefulness. NeuroReport, 20(3), 331–336.

Monti, J. M., Alvariño, F., & Monti, D. (2000). Conventional andpower spectrum analysis of the effects of Zolpidem on sleep EEGin patients with chronic primary insomnia. Sleep, 23(8), 1075–1084.

Morin, C. M. (1993). Insomnia: Psychological assessment andmanagement. New York: Guilford Press.

Morin, C. M. (2004). Cognitive-behavioral approaches to thetreatment of insomnia. The Journal of Clinical Psychiatry, 65(Suppl. 16), 33–40.

Morin, C. M., & Wooten, V. (1996). Psychological and pharmaco-logical approaches to treating insomnia. Clinical PsychologyReview, 6, 521–542.

Morin, C. M., LeBlanc, M., Daley, M., Grégoire, J. P., & Mérette, C.(2006). Epidemiology of insomnia: prevalence, self-help treat-ments, consultations, and determinant of help-seeking behaviors.Sleep Medicine, 7, 123–130.

Moul, D. E., Nofzinger, E. A., Pikonis, P. A., Houck, P. R., Miewald,J. M., & Buysse, D. J. (2002). Symptom reports in severe chronicinsomnia. Sleep, 25(5), 553–563.

National Institutes of Health. (1991). Consensus development confer-ence statement: the treatment of sleep disorders of older people.Sleep, 14, 169–177.

National Institutes of Health. (2005). National Institutes of Health State ofthe Science Conference statement on manifestations and manage-ment of chronic insomnia in adults. Sleep, 28, 1049–1057.

Nissen, C., Kloepfer, C., Nofzinger, E. A., Feige, B., Voderholzer, U.,& Riemann, D. (2006). Impaired sleep-related memory consol-idation in primary insomnia. Sleep, 29, 1068–1073.

Nofzinger, E. A., Nowell, P. D., Buysee, D. J., Vasko, R. C., Thase,M. E., Frank, E., et al. (1999). Towards a neurobiology of sleepdisturbance in insomnia and depression: a comparison ofsubjective, visually-scored, period amplitude, and power spectraldensity sleep measures. Sleep, 22(Suppl. 1), S99.

Nofzinger, E. A., Buysse, D. J., Germain, A., Price, J. C., Miewald, J.M., & Kupfer, D. J. (2004). Functional neuroimaging evidencefor hyperarousal in insomnia. American Journal of Psychatry,161(11), 2126–2128.

Ohayon, M. M. (2002). Epidemiology of insomnia: what we knowand we still need to learn. Sleep Medicine Reviews, 6, 97–111.

Orff, H. J., Drummond, S. P., Nowakowski, S., & Perlis, M. L. (2007).Discrepancy between subjective symptomatology and objectiveneuropsychological performance in insomnia. Sleep, 30(9),1205–1211.

Oswald, I., Taylor, A. M., & Treisman, M. (1960). Discriminativeresponses to stimulation during human sleep. Brain, 83, 440–452.

Ouellet, D. R., Turcotte, I., St-Jean, G., Carrier, J., & Bastien, C. H.(2010). Contribution of age, gender and insomnia subtypes to themicrostructure of sleep. Journal of Sleep Research, 19(Suppl. 2),162.

Parrino, L., Ferrillo, F., Smerieri, A., Spaggiari, M. C., Palomba, V.,Rossi, M., et al. (2004). Is insomnia a neurophysiological

disorder? The role of sleep microstructure. Brain ResearchBulletin, 63(5), 377–383.

Parrino, L., Milioli, G., De Paolis, F., Grassi, A., & Terzano, M. G.(2009). Paradoxical insomnia: the role of CAP and arousals insleep misperception. Sleep Medicine, 10(10), 1139–1145.

Pennestri, M., Adam, B., Paquet, J., Denesle, R., & Montplaisir, J.(2003). First-night effect in severe psychophysiological insom-nia. Sleep, 26(Suppl. 1), A311–A312.

Perlis, M. L., Giles, D. E., Mendelson, W. B., Bootzin, R. R., &Wyatt, J. K. (1997). Psychophysiological insomnia: the behav-ioral model and a neurocognitive perspective. Journal of SleepResearch, 5(3), 179–188.

Perlis, M. J., Kehr, E. L., Smith, M. T., Andrews, P. J., Orff, H., &Giles, D. E. (2001). Temporal and stagewise distribution of highfrequency EEG activity in patients with primary and secondaryinsomnia and good sleeper controls. Journal of Sleep Research,10, 93–104.

Perlis, M., Shaw, P. J., Cano, G., & Espie, C. A. (2011). Models ofinsomnia. In M. H. Kryger, T. Roth, & W. C. Dement (Eds.),Principles and practice of sleep medicine (5th ed., pp. 850–865).St Louis: Elsevier Saunders.

Randazzo, A. C., Schwitzer, P. K., Stone, K. L., Compton, J. D., &Walsh, J. K. (2000). Impaired cognitive function in insomniacsvs normals. Sleep, 23(Suppl. 2), 4.

Rechtschaffen, A., & Kales, A. A. (1968). Manual of standardizedterminology, techniques and scoring system for sleep stages ofhuman subjects. Los Angeles: Brain Information/Brain ResearchInstitute, UCLA.

Regestein, Q. R., & Reich, P. (1983). Incapacitating childhood-onsetinsomnia. Comprehensive Psychiatry, 24(3), 244–248.

Regestein, Q. R., Dambrosia, J., Hallett, M., Murawski, B., & Paine,M. (1993). Daytime alertness in patients with primary insomnia.The American Journal of Psychiatry, 150, 1529–1534.

Reite, M., Buysse, D., Reynolds, C., & Mendelson, W. (1995). Theuse of polysomnography in the evaluation of insomnia. Sleep, 18(1), 58–70.

Reynolds, C. F., III, Taska, L. S., Sewitch, D. E., Restifo, K.,Coble, P. A., & Kupfer, D. J. (1984). Persistent psychophys-iologic insomnia: preliminary research diagnostic criteria andEEG sleep data. The American Journal of Psychiatry, 141,804–805.

Riedel, B. W., & Lichstein, K. L. (2000). Insomnia and daytimefunctioning. Sleep Medicine Reviews, 4(3), 277–298.

Riemann, D., Kloepfer, C., & Berger, M. (2009). Functional andstructural brain alterations in insomnia: implications for patho-physiology. The European Journal of Neuroscience, 29, 1754–1760.

Ropars, S., Turcotte, I., St-Jean, G., & Bastien, C. H. (2010). Variationof sleepiness level after a cognitive task in the morning betweeninsomnia sufferers and good sleeper controls. Journal of SleepResearch, 19(Suppl. 2), P840, 287–288.

Roth, T., & Ancoli-Israel, S. (1999). Daytime consequences andcorrelates of insomnia in the United States: results of the1991 National Sleep Foundation Survey. II. Sleep, 22, S354–S358.

Sateia, M. J., Doghramji, K., Hauri, P. J., & Morin, C. M. (2000).Evaluation of chronic insomnia. An American Academy of SleepMedicine Review. Sleep, 23(2), 243–308.

Schmutte, T., Harris, S., Levin, R., Zweig, R., Katz, M., & Lipton, R.(2007). The relation between cognitive functioning and self-reported sleep complaints in nondemented older adults: resultsfrom the Bronx aging study. Behavioral Sleep Medicine, 5(1),39–56.

Schneider, C., Fulda, S., & Schulz, H. (2004). Daytime variation inperformance and tiredness/sleepiness ratings in patients with


insomnia, narcolepsy, sleep apnea and normal controls. Journalof Sleep Research, 13(4), 373–383.

Schneider-Helmert, D. (1987). Twenty-four-hour sleep-wake functionand personality patterns in chronic insomniacs and healthycontrols. Sleep, 10(5), 452–462.

Seidel, W. F., Ball, S., Cohen, S., Patterson, N., Yost, D., & Dement,W. C. (1984). Daytime alertness in relation to mood, perfor-mance, and nocturnal sleep in chronic insomniacs and non-complaining sleepers. Sleep, 7, 230–238.

Sforza, E., & Haba-Rubio, J. (2006). Event-related potentials inpatients with insomnia and sleep-related breathing disorders:evening-to-morning changes. Sleep, 29(6), 805–813.

Shekleton, J. A., Rogers, N. L., & Rajaratnam, S. M. (2010).Searching for the daytime impairments of primary insomnia.Sleep Medicine Reviews, 14(1), 47–60.

Smith, M. T., Perlis, M. L., Chengazi, V. U., Pennington, J., Soeffing,J., Ryan, J. M., et al. (2002). Neuroimaging of NREM sleep inprimary insomnia: a Tc-99-HMPAO single photon emissioncomputed tomography study. Sleep, 25(3), 325–335.

Smith, M. T., Perlis, M. L., Chengazi, V. U., Soeffing, J., & McCann,U. (2005). NREM sleep cerebral blood flow before and afterbehavior therapy for chronic primary insomnia: preliminarysingle photon emission computed tomography (SPECT) data.Sleep Medicine, 6(1), 93–94.

Spielman, A. J., & Glovinski, P. B. (1991). The varied nature ofinsomnia. In P. Hauri (Ed.), Case studies in insomnia (pp. 1–15).New York: Plenum.

Spielman, A. J., Saskin, P., & Thorpy, M. J. (1987). Treatment ofchronic insomnia by restriction of time in bed. Sleep, 10, 45–56.

Stepanski, E., Zorick, F., Roehrs, T., Young, D., & Roth, T. (1988).Daytime alertness in patients with chronic insomnia comparedwith asymptomatic control subjects. Sleep, 11, 54–60.

Stickgold, R. (2005). Sleep-dependent memory consolidation. Nature,437, 1272–1278.

St-Jean, G., & Bastien, C. H. (2008a). Sleep disorders and quality oflife. In J. C. Verster, S. R. Pandi-Perumal, & D. Streiner (Eds.),Sleep and quality of life in medical illness (pp. 37–46). HumanaPress: Springer.

St-Jean, G., & Bastien, C. H. (2008b). Rem and N-Rem powerspectral analysis on two consecutive nights in relation to sleepquality in psychophysiological and paradoxical insomnia suffer-ers. Sleep, 31, A247.

St-Jean, G., & Bastien, C. H. (2009). Classification of insomniasufferers based on laboratory PSG recordings and subjectivesleep reports. Sleep, 32, A284.

Sugerman, J. L., Stern, J. A., & Walsh, J. K. (1985). Daytime alertnessin subjective and objective insomnia: some preliminary findings.Biological Psychiatry, 20, 741–750.

Szelenberger, W., & Niemcewicz, S. (2001). Event-related currentdensity in primary insomnia. Acta Neurobiologiae Experimenta-lis (Wars), 61, 299–308.

Terzano, M. G., & Parrino, L. (2000). Origin and significance of thecyclic alternating pattern (CAP). Sleep Medicine Reviews, 4,101–123.

Terzano, M. G., Mancia, D., Salati, M. R., Costani, G., Decembrino, A.,& Parrino, L. (1985). The cyclic alternating pattern as a physiologiccomponent of normal NREM sleep. Sleep, 8, 137–145.

Terzano, M. G., Parrino, L., Boselli, M., Dell’Orso, S., Moroni, M., &Spaggiari, M. C. (1995). Changes of cyclic alternating pattern(CAP) parameters in situational insomnia under brotizolam andtriazolam. Psychopharmacology, 120, 237–243.

Terzano, M. G., Parrino, L., Boselli, M., Smerieri, A., & Spaggiari, M.C. (2000). CAP components and EEG synchronization in the first3 sleep cycles. Clinical Neurophysiology, 111, 283–290.

Terzano, M. G., Parrino, L., Spaggiari, M. C., Palomba, V., Rossi, M.,& Smerieri, A. (2003). CAP variables and arousals as sleepelectroencephalogram markers for primary insomnia. ClinicalNeurophysiology, 114, 1715–1723.

Terzano, M. G., Parrino, L., Smerieri, A., Carli, F., Nobili, L.,Donadio, S., et al. (2005). CAP and arousals are involved in thehomeostatic and ultradian sleep processes. Journal of SleepResearch, 14, 359–368.

Thomas, M., Sing, H., Belenky, G., Holcomb, H., Mayberg, H.,Dannals, R., et al. (2000). Neural basis of alertness and cognitiveperformance impairments during sleepiness, I: effects oh 24 h ofsleep deprivation on waking human regional brain activity.Journal of Sleep Research, 9, 335–352.

Turcotte, I., & Bastien, C. H. (2009). Is quality of sleep related tothe N1 and P2ERPs in chronic psychophysiological insomniasufferers? International Journal of Psychophysiology, 72, 314–322.

Turcotte, I., Adam, A., Lecarpentier, M., & Bastien, C. H. (2009). ERPsmeasures during sleep in psychophysiological and paradoxicalinsomnia sufferers. Sleep, 32, A271.

Vallières, A., Ivers, H., Bastien, C. H., Beaulieu-Bonneau, S., &Morin, C. M. (2005). Variability and predictability in sleeppatterns of chronic insomniacs. Journal of Sleep Research, 14,447–453.

Van der Werf, Y. D., Altena, E., Van Dijk, K. D., Strijers, R. L. M., DeRijke, W., Stam, C. J., et al. (2010). Is disturbed intracorticalexcitability a stable trait of chronic insomnia? A study usingtranscranial magnetic stimulation before and after multimodalsleep therapy. Biological Psychiatry, 68, 950–955.

Van Someren, E. J. (2006). Mechanisms and functions of couplingbetween sleep and temperature rhythms. Progress in BrainResearch, 153, 309–324.

Varkevisser, M., & Kerkhof, G. A. (2005). Chronic insomnia andperformance in a 24-h constant routine study. Journal of SleepResearch, 14(1), 49–59.

Vignola, A., Lamoureux, C., Bastien, C. H., & Morin, C. M. (2000).Late-life insomnia, chronic use of benzodiazepines, and daytimeperformance. Journal of Gerontology: Psychological Sciences,55, 54–62.

Walter, W. G. (1953). The living brain. New York: Norton andCompany.

Wauquier, A., Aloe, L., & Declerck, A. K. (1995). K-Complexes: arethey signs of arousal or sleep protective? Journal of SleepResearch, 4, 138–143.

Winkelmann, J. W., Buxton, O. M., Jensen, E., Benson, K. L.,O’Connor, S. P., Wang, W., et al. (2008). Reduced brain GABAin primary insomnia: preliminary data from 4T proton magneticresonance spectroscopy (1H-MRS). Sleep, 31, 1499–1506.

Yang, C. M., & Lo, H. S. (2007). ERP evidence of enhancedexcitatory and reduced inhibitory processes of auditory stimuliduring sleep in patients with primary insomnia. Sleep, 30, 585–592.

Zammit, G. K. (1988). Subjective ratings of the characteristics andsequelae of good and poor sleep in normals. Journal of ClinicalPsychology, 44, 123–130.

Zammit, G. K., Weiner, J., Damato, N., Sillup, G. P., & McMillan, C.A. (1999). Quality of life in people with insomnia. Sleep, 22,S379–S385.


insomnia: neurophysiological and neuropsychological approaches

Documents