Intellectual disabilities and power spectra analysis during sleep: a new perspective on borderline intellectual functioning

Intellectual disabilities and power spectra analysisduring sleep: a new perspective on borderlineintellectual functioning

M. Esposito & M. Carotenuto

Sleep Clinic for Developmental Age, Clinic of Child and Adolescent Neuropsychiatry, Second University of Naples, Naples, Italy

Abstract

Background The role of sleep in cognitive proc-esses has been confirmed by a growing number ofreports for all ages of life. Analysing sleep electroen-cephalogram (EEG) spectra may be useful to studycortical organisation in individuals with BorderlineIntellectual Functioning (BIF), as seen in other dis-turbances even if it is not considered a disease. Theaim of this study was to determine if the sleep EEGpower spectra in children with BIF could be differ-ent from typically developing children.Methods Eighteen BIF (12 males) (mean age 11.04;SD � 1.07) and 24 typical developing children (14

men) (mean age 10.98; SD � 1.76; P = 0.899)underwent an overnight polysomnography (PSG)recording in the Sleep Laboratory of the Clinic ofChild and Adolescent Neuropsychiatry, after oneadaptation night. Sleep was subdivided into 30-sepochs and sleep stages were scored according tothe standard criteria and the power spectra werecalculated for the Cz-A2 channel using the sleepanalysis software Hypnolab 1.2 (SWS Soft, Italy) bymeans of the Fast Fourier Transform and thepower spectrum was calculated for frequenciesbetween 0.5 and 60 Hz with a frequency step of

1 Hz and then averaged across the following bandsdelta (0.5–4 Hz), theta (5–7 Hz), alpha (8–11 Hz),sigma (11–15 Hz), and beta (16–30 Hz), gamma(30–60 Hz) for S2, SWS and REM (Rapid EyeMovement) sleep stages.Results BIF have a reduced sleep duration (totalsleep time; P < 0.001), and an increased rate ofstage shifts (P < 0.001) and awakenings (P < 0.001)and WASO (wakefulness after sleep onset) %(P < 0.001); the stage 2% (P < 0.001), and REM%(P < 0.001) were lower and slow-wave sleep per-centage was slightly higher (P < 0.001). All chil-dren with BIF had an AHI (apnoea–hypopneaindex) lessthan 1 (mean AHI = 0.691 � 0.236) with a meanoxygen saturation of 97.6% and a periodic legmovement index (PLMI) less than 5 (meanPLMI = 2.94 � 1.56). All sleep stages had a signifi-cant reduction in gamma frequency (30–60 Hz)(P < 0.001) and an increased delta frequency(0.5–4.0 Hz) (P < 0.001) power in BIF subjectscompared with typically developing children.Conclusion Our findings shed light on the impor-tance of sleep for cognition processes particularly incognitive borderline dysfunction and the role ofEEG spectral power analysis to recognize sleepcharacteristics in BIF children.

Keywords borderline intellectual functioning, FastFourier Transform, intellectual disability, polysom-nography, power spectra analysis, sleep

Correspondence: Prof. Marco Carotenuto, Sleep Clinic for Devel-opmental Age, Clinic of Child and Adolescent Neuropsychiatry,Via Sergio Pansini 5 PAD XI, 80131 Naples, Italy (e-mail:[email protected]).

Journal of Intellectual Disability Research doi: 10.1111/jir.120361

bs_bs_banner

© 2013 The Authors. Journal of Intellectual Disability Research © 2013 Blackwell Publishing Ltd

mailto:[email protected]

Introduction

Borderline intellectual functioning (BIF) is a neu-ropsychological entity identified as an intelligencequotient (IQ) ranging between 71 and 84 (i.e.between -2 and -1 standard deviations), with mala-daptive functioning in academic, social, or voca-tional areas (American Psychiatric Association2000; Ninivaggi 2001).

Although there is a 7% prevalence of BIF in theschool-aged population (Karande et al. 2008), thereare few reports about this interesting condition.Although idiopathic and poorly studied, BIF couldrepresent a model to study other factors in thedevelopment of cognitive function.

On the other hand, the role of sleep in cognitiveprocesses has been confirmed by a growing numberof reports for all ages (Curcio et al. 2006; Aricòet al. 2010; Ahrberg et al. 2012; Bruni et al. 2012)and in intellectual disabilities (IDs) (Miano et al.2008; Verrillo et al. 2009; Aricò et al. 2010; Masonet al. 2011; van Dijk et al. 2012).

Our previous study showed the neurophysiologi-cal characteristics of sleep and non-REM instabil-ity (Cyclic Alternating Pattern – CAP – analysis)in BIF children, suggesting the correlation betweensleep organisation and lower IQ of these subjectsthan comparisons (Esposito & Carotenuto2010).

To highlight the relationship between cognitionand healthy sleep, McDermott showed howprolonged continuous wakefulness could impairhippocampal long-term synaptic plasticity andhippocampus-dependent memory formation, whichmay underlie sleep deprivation-induced impair-ments in synaptic plasticity and cognitive function(McDermott et al. 2003).

Alternatively, this effect may also be linked toaltered oscillatory rhythms >30 Hz (gamma band),which are involved in many processes such asmemory and learning (Tallon-Baudry et al. 1998;Taylor et al. 2005; Ursino et al. 2009).

The peculiar sleep organisation of children withBIF, such as the reduced slow wave activity (SWA)and the prolonged CAP A1 pattern phases (Esposito& Carotenuto 2010) may suggest a sort of putativeintrinsic inefficiency in oscillations of frontal genera-tor of A1 identified as linked with high cognitivefunctioning (Aricò et al. 2010).

In this light, the analysis of sleep electroencepha-logram (EEG) spectra could be an another usefultool to study the cortical organisation in individualswith BIF, as shown in other IDs (Della Marca et al.2011; Gombos et al. 2011; Yang et al. 2011), even ifBIF is not considered a disease.

Therefore, the aim of study was to determinewhether the sleep EEG power spectra of childrenwith BIF could be different from typical developingchildren.

Materials and methods

Study population

For this study, 18 BIF (12 males) (mean age 11.04;SD � 1.07) referred for generic academic difficultiesand 24 typical developing children (14 men) (meanage 10.98; SD � 1.76; P = 0.899) underwent anovernight PSG recording in the Sleep Laboratory atthe Clinic of Child and Adolescent Neuropsychia-try, after one adaptation night, in order to avoidthe first-night effect. The subjects of both groupswere recruited from the same urban area, were ofCaucasian origin, and had middle socio-economicbackgrounds.

Following the Diagnostic and Statistical Manualof Mental Disorder (DSM)-IV, we included thechildren with a total intelligence quotient (TIQ)between 71 and 84 in BIF group.

Exclusion criteria were: mental retardation (IQbelow 70), genetic syndromes with IDs (i.e. Downsyndrome, Prader–Willi syndrome, fragile-X syn-drome, Williams syndrome, deletions on the chro-mosome 22, etc), hypothyroidism, psychiatricdiseases [schizophrenia, mood disorders, attentiondeficit hyperactivity disorder (ADHD)], epilepsy,obesity, sleep breathing disorders and periodic limbmovement disorder (PLMd).

All evaluations were performed after informedparental consent was obtained for all the childrenenrolled, according to the World Medical Associa-tion (2008). The study was approved by theDepartmental Ethics Committee at the SecondUniversity of Naples.

Cognitive assessment

Intellective functioning was assessed by the Italianversion of the Wechsler Intelligence Scale for Chil-

2Journal of Intellectual Disability Research

M. Esposito & M. Carotenuto • BIF and power spectra sleep analysis


dren – III edition (WISC-III) (Wechsler 1991;Orsini & Picone 2006), applicable for childrenranging from 6 to 16 years. The WISC-III is com-posed of 13 distinct subtests divided into two scales,a verbal scale and a performance scale. The sixverbal scale tests use language-based items, whereasthe seven performance scales use visual-motor itemsthat are less dependent on language. Five of thesubtests in each scale produce scale-specific IQs asverbal IQ (VIQ) and performance IQ (PIQ), andthe 10 subtest scores produce a total scale IQ(TIQ).

Polysomnographic sleep recordings

The EEG recordings and electrode placement wereperformed according to the 10–20 system (Jasper1958) and the PSG montage included at least 19

EEG channels (Fp2, Fp1, F3, F4, F7, F8, C3, C4,T3, T4, P3, P4, T5, T6, O1, O2, Fz, Cz, Pz) refer-enced to the contralateral mastoid, left and rightelectrooculogram (EOG), chin electromyogram(EMG), left and right tibialis EMG, electrocardio-gram (ECG) (1 derivation), nasal cannula, thoraxand abdominal effort, peripheral oxygen saturation,pulse and position sensors.

Recordings were carried out using a Brain QuickMicromed System 98 recording machine, andsignals were sampled at 256 Hz and stored onhard disk for further analysis. EEG signals weredigitally band-pass filtered at 0.1–120 Hz, 12-bitA/D precision. Sleep signals were sampled at 200

or 256 Hz and stored on hard disk in Europeandata format (EDF) (Kemp et al. 1992) for furtheranalysis. EEG signals, in particular, were firstacquired with a wide band analogue filter(0.001–70 Hz) and then digitally band-passfiltered at 0.1–50 Hz.

All recordings started at the subjects’ usualbedtime and continued until spontaneous morningawakening.

Sleep stage scoring

Sleep was subdivided into 30-s epochs and sleepstages were scored according to the standard criteria(Rechtschaffen & Kales 1968).

The following conventional sleep parameters wereevaluated:• Time in bed (TIB);• Sleep period time (SPT): time from sleep onset tosleep end;• Total sleep time (TST): the time from sleep onsetto the end of the final sleep epoch minus timeawake;• Sleep latency (SL): time from lights out to sleeponset, defined as the first of two consecutive epochsof sleep stage 1 or one epoch of any other stage, inminutes;• REM latency (RL): time from sleep onset to thefirst REM sleep epoch;• Number of stage shifts/hour (SS/h);• Number of awakenings/hour (AWN/h);• Sleep efficiency (SE%): the percentage ratiobetween total sleep time and time in bed (TST/TIB ¥ 100);• Percentage of SPT spent in wakefulness aftersleep onset (WASO%), i.e. the time spent awakebetween sleep onset and end of sleep; and• Percentage of SPT spent in sleep stages 1 (S1%),2 (S2%), slow-wave sleep (SWS%), and REM sleep(REM%).All these variables were analysed by means of theHypnolab 1.2 sleep software analysis (SWS Soft,Italy) and all the recordings were visually scored byone of the investigators (M.C.), and the sleepparameters derived were tabulated for statisticalanalysis.

To exclude sleep-related breathing disorders, thenocturnal respiratory parameters (i.e. central,obstructive, and mixed apnoea events) werecounted according to the standard criteria(American Thoracic Society 1996). The apnoea–hypopnea index (AHI) was defined as the numberof apnoeas and hypopneas per hour of total sleeptime; an obstructive apnoea index >1 was selectedas the cut-off for normality (Marcus et al. 1992;Traeger et al. 2005).

To identify episodes of periodic limb movements(PLMs) the standard criteria were used. The fre-quency of leg movements was represented as theperiodic leg movement index (PLMI; number/h oftotal sleep time). Episodes of PLMS were definedas leg movements with an amplitude increase of8 mV above the baseline value, a duration of 0.5–10 s, a period length between two consecutive




movements of 5–90 s, and a minimum of fourconsecutive movements (Zucconi et al. 2006). APLMI �5 was considered abnormal.

Power spectra analysis

According to previous studies on paediatric samples(Bruni et al. 2009a,b), the power spectra were cal-culated for the Cz-A2 channel using the sleep analy-sis software Hypnolab 1.2 (SWS Soft, Italy), afterWelch windowing, (wn = 1 - ((n1/2(N - 1))/1/2(N + 1))2) in order to minimise the truncation errorand reduce spectral leakage by suppressingsidelobes (Press et al. 1989), by means of the FastFourier Transform (FFT) (Cooley & Tukey 1965),on 2-s EEG epochs. Fifteen 2-s epoch FFTs wereperformed for each 30-s artefact-free sleep epochand their results averaged. Subsequently, the powerspectrum was calculated for frequencies between0.5 and 60 Hz with a frequency step of 1 Hz andthen averaged across the following bands: delta(0.5–4 Hz), theta (5–7 Hz), alpha (8–11 Hz),sigma (11–15 Hz), and beta (16–30 Hz), gamma(30–60 Hz) for S2, SWS and REM sleep stages.

Statistical analysis

Comparisons between age, sex, intellectual abilities,z-BMI, sleep architecture, respiratory parametersand power spectra analysis, obtained in BIF chil-dren and typical developing individuals, werecarried out by the t-test followed by the Mann–Whitney U-test. To verify the matching sex ratio(M/F) the chi-squared test was performed.

Bonferroni correction was applied. P values <0.01

were considered statistically significant. The com-mercially available software STATISTICA (dataanalysis software system), version 6, StatSoft, Inc.(Tulsa, OK, USA) was used for all statistical tests.

Results

Table 1 shows the characteristics of two groups(BIF vs. Controls), the matching for age(P = 0.899), sex (P = 0.819), and z-BMI (P = 0.559)and the obvious differences in the means of intellec-tive parameters (VIQ, P < 0.001; PIQ, P < 0.001,and TIQ, P < 0.001).

Table 2 shows the comparisons of macrostruc-tural sleep parameters between children with BIFand normal controls. Children with BIF have areduced sleep duration (TST; P < 0.001), and anincreased rate of stage shifts (P < 0.001), awaken-ings (P < 0.001) and WASO% (P < 0.001); the stage2% (P < 0.001), and REM% (P < 0.001) were lowerand slow-wave sleep percentage was slightly higher(P < 0.001). All children with BIF had a less than 1

AHI (mean AHI = 0.691 � 0.236) with a meanoxygen saturation of 97.6% and a PLM Index(PLMI) less than 5 (mean PLMI = 2.94 � 1.56).

Table 3 shows differences in power spectral dataobtained in all sleep stages evaluated (2non-REM,SWS and REM) (Table 3). Particularly, all stageshad a significantly lower gamma frequency (30

-60 Hz) (P < 0.001) and higher delta frequency(0.5–4.0 Hz) (P < 0.001) power in BIF subjectscompared with typically developing children.

Table 1 Population characteristics

BIF (n = 18) Control (n = 24) P

Age 11.04 � 1.07 10.98 � 1.76 0.899Sex ratio (M/F) 12/6 14/10 0.819z-BMI 0.67 � 0.31 0.72 � 0.24 0.559VIQ 81.692 � 7.903 112.078 � 8.941 <0.001PIQ 83.286 � 5.418 102.029 � 11.984 <0.001TIQ 80.017 � 2.863 110.436 � 7.241 <0.001

Table 1 shows the mean of age, zeta score Body Mass Index (z-BMI), Verbal IntelligentQuotient (VIQ), Performance Intelligent Quotient (PIQ) and Total Intelligent Quotient(TIQ) of individuals with borderline intellectual functioning (BIF) and typical developingsubjects (Control).In order to verify the matching about sex ratio (M/F) the chi-squared test was performed.P < 0.05 was considered significant.




Discussion

Approaching IDs can be difficult, especially indevelopmental age. Therefore, neurophysiologicalsleep evaluation can be useful to broaden thecurrent knowledge and understanding of thecomplex interrelationships between the cerebralareas.

Decreasing sleep efficiency REM ratio have beenreported as characteristic neurophysiological signsin several developmental disabilities such as Downsyndrome (Diomedi et al. 1999; Miano et al. 2008),autism (Diomedi et al. 1999), Angelman syndrome(Miano et al. 2004) and in ADHD (Sobanski et al.2008).

Moreover, lower sleep efficiency, higher WASO,increases in non-REM sleep EEG (relative) deltaand region-dependent drops in sigma/high fre-quency activities have been reported in subjectswith Asperger syndrome (Lázár et al. 2010).

Finally, reduced total sleep time, lower sleep effi-ciency percentage, higher WASO, increases in fron-tally measured non-REM sleep EEG delta power

and SWS time, as well as region-dependentdecreases in sigma power and reduced REM sleeppercent have been reported in Williams syndrome(Gombos et al. 2011). Thus, several papers havebeen reporting similar sleep-EEG alterations in dif-ferent conditions affecting intellectual functioning.

Conversely, the findings of the present study tendto confirm the physiological sleep fragmentationassociated with a higher SWS% (P < 0.001) andlower S2% (P < 0.001) and REM% (P < 0.001)than comparisons, as previously showed in 2010

(Esposito & Carotenuto 2010).In addition, reduced REM percentage has been

identified in autism, particularly associated withdevelopmental delay (Buckley et al. 2010), and inRett syndrome (Carotenuto et al. 2012). Alterna-tively, REM sleep has most recently been implicatedin the process of human memory consolidation andseveral studies suggest that it is crucial to normalcognitive function and in the processing of emotionin memory systems (Maquet 2001; Stickgold 2005).

The role of REM sleep for cognition was alreadyproposed in 1966 by Roffwarg and is still being

Table 2 Macrostructural sleeparchitecture in subjects with borderlineintellectual functioning (BIF) and controls

BIF (n = 18) Controls (n = 24) Mann–Whitney test

Mean SD Mean SD P value*

TIB-min 473.600 31.183 498.800 41.824 NSSPT-min 453.780 36.814 487.163 51.952 NSTST-min 369.438 82.417 472.397 49.718 <0.001SOL-min 14.8933 12.736 22.744 13.018 NSFRL-min 148.923 74.130 108.314 42.917 NSSS-h 8.925 3.142 4.562 1.948 <0.001AWN-h 3.519 1.946 1.028 0.913 <0.001SE% 87.935 7.418 90.017 5.614 NSWASO % 15.418 12.729 3.047 2.113 <0.001S1 % 3.428 1.739 4.018 2.126 NSS2 % 31.724 11.020 44.731 7.046 <0.001SWS % 33.947 10.769 21.431 6.918 <0.001REM % 12.463 8.024 24.149 6.062 <0.001AHI 0.691 0.236 0.575 0.341 NSODI 0.594 0.496 0.519 0.372 NSPLM-I 2.944 1.563 2.719 1.623 NS

* Bonferroni-corrected value.TIB, time in bed; SPT, sleep period time; TST, total sleep time; SOL, sleep onset latency;FRL, first REM sleep latency; SS, stage shifts; AWN, awakenings; SE, sleep efficiency;WASO, wake time after sleep onset; S1, sleep stage 1; S2, sleep stage 2; SWS, slow-wavesleep; REM, rapid eye movement sleep; Apnoea/hypopnea index (AHI); Oxygen Desatura-tion Index (ODI); Periodic Limb Movement Index (PLM-I).




explored today, and it has been hypothesised thatthe increased ponto-thalamo-cortical activity seenduring REM sleep could provide the endogenousstimulation needed to form and stabilise durablesynaptic connections in the developing brain(Roffwarg et al. 1966; Marks et al. 1995; Hobson2009).

Therefore, our findings tend support the impor-tance of sleep for cognition processes as highlightedby the FFT analysis.

It is well known that gamma frequencies areinvolved in many cognitive functions such asmemory and learning (Tallon-Baudry et al. 1998;Singer 1999; Fries et al. 2001; Simos et al. 2002;Fitzgibbon et al. 2004; Bichot et al. 2005;Taylor et al. 2005; Pipa et al. 2009; Ursino et al.2009).

According to the neuropsychological perspective,the gamma activity could be considered as a sort oflink between many cognitive functions allocated incortical and subcortical areas (i.e. cortex, hippocam-

pus, diencephalon, cerebellum) anatomically andfunctionally distant (Ribary et al. 1991; Traub et al.1996; Fisahn et al. 1998; Pesaran et al. 2002;Lakatos et al. 2008; Popescu et al. 2009).

In this light, BIF subjects have less gamma activ-ity in all sleep stages compared with normal indi-viduals, which could be linked to the typicallearning and/or memory difficulties of this condi-tion (Table 3).

Interpretation of the spectral analysis findingscould be fairly difficult because of the substantiallack of quantitative sleep EEG studies in BIFsubjects.

In our sample, the delta activity seems to bedominant in SWS, and its increased presence inthe power spectra might be related to the higherSWS ratio in BIF compared with controls(33.947 � 10.769 vs. 21.431 � 6.918; P < 0.001)(Table 2).

The increased delta (in both non-REM and REMsleep) as well as the decreased non-REM sigma

Table 3 Power spectra sleep analysis insubjects with borderline intellectualfunctioning (BIF) and controls

BIF (n = 18) Controls (n = 24)

Mean SD Mean SD P*

S2 0.5–4.0 Hz 263.427 119.081 112.476 53.027 <0.001S2 5.0–7.0 Hz 41.141 25.197 35.614 28.418 NSS2 8.0–11.0 Hz 26.564 19.647 16.179 13.553 NSS2 11.0–15.0 Hz 19.916 13.389 12.202 10.182 NSS2 16.0–30.0 Hz 33.361 27.542 46.791 39.440 NSS2 30.0–60.0 Hz 17.175 10.756 143.120 116.528 <0.001SWS 0.5–4.0 Hz 306.104 183.418 127.913 79.361 <0.001SWS 5.0–7.0 Hz 58.404 34.261 28.868 24.259 NSSWS 8.0–11.0 Hz 32.361 24.063 20.889 17.739 NSSWS 11.0–15.0 Hz 23.662 18.261 12.355 10.465 NSSWS 16.0–30.0 Hz 30.837 29.840 36.485 31.435 NSSWS 30.0–60.0 Hz 14.373 9.519 104.674 87.289 <0.001REM 0.5–4.0 Hz 199.433 98.306 102.690 71.389 <0.001REM 5.0–7.0 Hz 33.545 19.729 25.482 17.065 NSREM 8.0–11.0 Hz 22.374 14.870 13.540 11.424 NSREM 11.0–15.0 Hz 14.580 9.551 9.780 8.291 NSREM 16.0–30.0 Hz 27.398 17.982 36.795 31.223 NSREM 30.0–60.0 Hz 13.038 9.214 107.867 88.087 <0.001

* Bonferroni-corrected value.Table 3 shows the means and SD between individuals with borderline intellectual function-ing (BIF) and typical developing subjects (Control) in power spectra analysis of sleep stages(stage 2 – S2; slow wave sleep – SWS, Rapid Eye Movement – REM) according withFast Fourier Transform (FFT): delta (0.5–4 Hz), theta (5–7 Hz), alpha (8–11 Hz), sigma(11–15 Hz), and beta (16–30 Hz), gamma (30–60 Hz).




activity in individuals with BIF could be consideredas reminiscent of sleep deprivation effects on theEEG and may suggest an elevated sleep propensityin BIF (Dijk et al. 1993).

It is also known that SWS percentage and thedelta EEG power in non-REM sleep tend todecrease in typical brain maturation (Feinberg &Campbell 2010) and the persistence of a slightlyhigher rate of delta in BIF may be interpreted as asort of alteration/delay in maturation of cerebralareas and/or in their interconnections.

The increased SWS percentage and frontal deltamight be due to either fatigue or delayed brainmaturation in BIF. The frontopolar localisation ofthe increased delta power spectra may also berelated to the increases of grey matter in the frontallobes (Campbell et al. 2009) and white matter (vonStein & Sarnthein 2000; Kamarajan et al. 2004).

Moreover, our BIF sample had a significantreduction in the gamma band representation(P < 0.001) at all sleep stages, based on findingsthat memory and learning are often impaired inthese subjects (Karande et al. 2008; Fernell & Ek2010).

In conclusion, our findings focus on the impor-tance of sleep for cognition processes particularly incognitive borderline dysfunction and the role ofEEG spectral power analysis to recognise sleepcharacteristics in BIF children.

Acknowledgements

The authors thank Joseph Sepe, MD, for revisingthe text.

References

Ahrberg K., Dresler M., Niedermaier S., Steiger A. &Genzel L. (2012) The interaction between sleep qualityand academic performance. Journal of PsychiatricResearch 46, 1618–22.

American Psychiatric Association (2000) Diagnostic andStatistical Manual of Mental Disorders, DSM-IV-TR. APA,Washington, DC.

American Thoracic Society (1996) Standards and indica-tions for cardiopulmonary sleep studies in children.American Journal of Respiratory and Critical Care Medi-cine 153, 866–78.

Aricò D., Drago V., Foster P. S., Heilman K. M., William-son J. & Ferri R. (2010) Effects of NREM sleep instabil-ity on cognitive processing. Sleep Medicine 11, 791–8.

Bichot N. P., Rossi A. F. & Desimone R. (2005) Paralleland serial neural mechanisms for visual search inmacaque area V4. Science 308, 529–34.

Bruni O., Ferri R., Novelli L., Finotti E., Terribili M.,Troianiello M. et al. (2009a) Slow EEG amplitude oscil-lations during NREM sleep and reading disabilities inchildren with dyslexia. Developmental Neuropsychology 34,539–51.

Bruni O., Novelli L., Finotti E., Luchetti A., Uggeri G.,Aricò D. et al. (2009b) All-night EEG power spectralanalysis of the cyclic alternating pattern at differentages. Clinical Neurophysiology 120, 248–56.

Bruni O., Kohler M., Novelli L., Kennedy D., LushingtonK., Martin J. et al. (2012) The role of NREM sleepinstability in child cognitive performance. Sleep 35, 649–56.

Buckley A. W., Rodriguez A. J., Jennison K., Buckley J.,Thurm A., Sato S., Swedo S. (2010) Rapid eye move-ment sleep percentage in children with autism comparedwith children with developmental delay and typicaldevelopment. Archives of Pediatrics & Adolescent Medicine164, 1032–7.

Campbell L. E., Daly E., Toal F., Stevens A., Azumaa R.,Karmiloff-Smith A. et al. (2009) Brain structural differ-ences associated with the behavioural phenotype in chil-dren with Williams syndrome. Brain Research 1258,96–107.

Carotenuto M., Esposito M., D’Aniello A., Rippa C. D.,Precenzano F., Pascotto A. et al. (2012) Polysomno-graphic findings in Rett syndrome: a case-control study.Sleep and Breathing 17, 93–8.

Cooley J. W. & Tukey O. W. (1965) An algorithm for themachine calculation of complex Fourier series. Math-ematics of Computation 19, 297–301.

Curcio G., Ferrara M. & De Gennaro L. (2006) Sleeploss, learning capacity and academic performance. SleepMedicine Reviews 10, 323–37.

Della Marca G., Leoni C., Dittoni S., Battaglia D.,Losurdo A., Testani E. et al. (2011) Increased sleepspindle activity in patients with Costello syndrome(HRAS gene mutation). Journal of Clinical Neurophysiol-ogy 28, 314–18.

Dijk D. J., Hayes B. & Czeisler C. A. (1993) Dynamics ofelectroencephalographic sleep spindles and slow waveactivity in men: effect of sleep deprivation. BrainResearch 626, 190–9.

van Dijk E., Hilgenkamp T. I., Evenhuis H. M. & EchteldM. A. (2012) Exploring the use of actigraphy to investi-gate sleep problems in older people with intellectual dis-ability. Journal of Intellectual Disability Research 56, 204–11.




Diomedi M., Curatolo P., Scalise A., Placidi F., Caretto F.& Gigli G. L. (1999) Sleep abnormalities in mentallyretarded autistic subjects: Down’s syndrome with mentalretardation and normal subjects. Brain and Development21, 548–53.

Esposito M. & Carotenuto M. (2010) Borderline intellec-tual functioning and sleep: the role of cyclic alternatingpattern. Neuroscience Letters 485, 89–93.

Feinberg I. & Campbell I. G. (2010) Sleep EEG changesduring adolescence: an index of a fundamental brainreorganization. Brain and Cognition 72, 56–65.

Fernell E. & Ek U. (2010) Borderline intellectual function-ing in children and adolescents – insufficiently recog-nized difficulties. Acta Paediatrica 99, 748–53.

Fisahn A., Pike F. G., Buhl E. H. & Paulsen O. (1998)Cholinergic induction of network oscillations at 40 Hzin the hippocampus in vitro. Nature 394, 186–9.

Fitzgibbon S. P., Pope K. J., Mackenzie L., Clark C. R. &Willoughby J. O. (2004) Cognitive tasks augmentgamma EEG power. Clinical Neurophysiology 115,1802–9.

Fries P., Reynolds J. H., Rorie A. E. & Desimone R.(2001) Modulation of oscillatory neuronal synchroniza-tion by selective visual attention. Science 291, 1560–3.

Gombos F., Bódizs R. & Kovács I. (2011) Atypical sleeparchitecture and altered EEG spectra in Williams syn-drome. Journal of Intellectual Disability Research 55, 255–62.

Hobson A. (2009) REM sleep and dreaming: towards atheory of protoconsciousness. Nature Reviews. Neuro-science 10, 803–13.

Jasper H. H. (1958) The 10–20 electrode system of theinternational federation. Electroencephalography and Clini-cal Neurophysiology 10, 370–5.

Kamarajan C., Porjesz B., Jones K. A., Choi K., ChorlianD. B., Padmanabhapillai A. et al. (2004) The role ofbrain oscillations as functional correlates of cognitivesystems: a study of frontal inhibitory control in alcohol-ism. International Journal of Psychophysiology 51, 155–80.

Karande S., Kanchan S. & Kulkarni M. (2008) Clinicaland psychoeducational profile of children with border-line intellectual functioning. Indian Journal of Pediatrics75, 795–800.

Kemp B., Värri A., Rosa A. C., Nielsen K. D. & Gade J.(1992) A simple format for exchange of digitized poly-graphic recordings. Electroencephalography and ClinicalNeurophysiology 82, 391–3.

Lakatos P., Karmos G., Mehta A. D., Ulbert I. &Schroeder C. E. (2008) Entrainment of neuronal oscilla-tions as a mechanism of attentional selection. Science320, 110–13.

Lázár A. S., Lázár Z. I., Bíró A., Gyori M., Tárnok Z.,Prekop C. et al. (2010) Reduced fronto-cortical brainconnectivity during NREM sleep in Asperger syndrome:

an EEG spectral and phase coherence study. ClinicalNeurophysiology 121, 1844–54.

McDermott C. M., LaHoste G. J., Chen C., Musto A.,Bazan N. G. & Magee J. C. (2003) Sleep deprivationcauses behavioral, synaptic, and membrane excitabilityalterations in hippocampal neurons. Journal of Neuro-science 23, 9687–95.

Maquet P. (2001) The role of sleep in learning andmemory. Science 294, 1048–52.

Marcus C. L., Omlin K. J., Basinki D. J., Bailey S. L.,Rachal A. B., Von Pechmann W. S. et al. (1992) Normalpolysomnographic values for children and adolescents.The American Review of Respiratory Disease 146, 1235–9.

Marks G. A., Shaffery J. P., Oksenberg A., Speciale S. G.& Roffwarg H. (1995) A functional role for REM sleepin brain maturation. Behavioural Brain Research 69, 1–11.

Mason T. B., Arens R., Sharman J., Bintliff-Janisak B.,Schultz B., Walters A. S. et al. (2011) Sleep in childrenwith Williams Syndrome. Sleep Medicine 12, 892–7.

Miano S., Bruni O., Leuzzi V., Elia M., Verrillo E. & FerriR. (2004) Sleep polygraphy in Angelman syndrome.Clinical Neurophysiology 115, 938–45.

Miano S., Bruni O., Elia M., Scifo L., Smerieri A.,Trovato A. et al. (2008) Sleep phenotypes of intellectualdisability: a polysomnographic evaluation in subjectswith Down syndrome and Fragile-X syndrome. ClinicalNeurophysiology 119, 1242–7.

Ninivaggi F. J. (2001) Borderline intellectual functioningin children and adolescents: reexamining an underrecog-nized yet prevalent clinical comorbidity. ConnecticutMedicine 65, 7–11.

Orsini A. & Picone L. (2006) WISC III. Contributo allataratura italiana Giunti O.S. Organizzazioni Speciali,Firenze.

Pesaran B., Pezaris J. S., Sahani M., Mitra P. & AndersenR. A. (2002) Temporal structure in neuronal activityduring working memory in macaque parietal cortex.Nature Neuroscience 5, 805–11.

Pipa G., Städtler E. S., Rodriguez E. F., Waltz J. A.,Muckli L. F., Singer W. et al. (2009) Performance- andstimulus-dependent oscillations in monkey prefrontalcortex during short-term memory. Frontiers in IntegrativeNeuroscience 3, 1–13.

Popescu A. T., Popa D. & Paré D. (2009) Coherentgamma oscillations couple the amygdale and striatumduring learning. Nature Neuroscience 12, 801–7.

Press W. H., Flannery B. P., Teukolsky S. A. & VetterlingW. T. (1989) Numerical Recipes. The Art of ScientificComputing. Press Syndicate of the University of Cam-bridge, Cambridge.

Rechtschaffen A. & Kales A. (1968) A Manual of Stand-ardized Terminology,Techniques, and Scoring System forSleep Stages of Human Subjects. Washington PublicHealth Service, US Government Printing Office, Wash-ington, DC.




Ribary U., Ioannides A. A., Singh K. D., Hasson R.,Bolton J. P., Lado F. et al. (1991) Magnetic field tomog-raphy of coherent thalamocortical 40-Hz oscillations inhumans. Proceedings of the National Academy of Sciencesof the United States of America 88, 11037–41.

Roffwarg H. P., Muzio J. N. & Dement W. C. (1966)Ontogenetic development of the human sleep-dreamcycle. Science 152, 602–19.

Simos P. G., Papanikolaou E., Sakkalis E. & Micheloyan-nis S. (2002) Modulation of gamma-band spectralpower by cognitive task complexity. Brain Topography 14,191–6.

Singer W. (1999) Neuronal synchrony: a versatile code forthe definition of relations? Neuron 24, 49–65.

Sobanski E., Schredl M., Kettler N. & Alm B. (2008)Sleep in adults with attention deficit hyperactivity disor-der (ADHD) before and during treatment with methyl-phenidate: a controlled polysomnographic study. Sleep31, 375–81.

von Stein A. & Sarnthein J. (2000) Different frequenciesfor different scales of cortical integration: from localgamma to long range alpha/theta synchronization. Inter-national Journal of Psychophysiology 38, 301–13.

Stickgold R. (2005) Sleep-dependent memory consolida-tion. Nature 437, 1272–8.

Tallon-Baudry C., Bertrand O., Peronnet F. & Pernier J.(1998) Induced gamma-band activity during the delay ofa visual short-term memory task in humans. The Journalof Neuroscience 18, 4244–54.

Taylor K., Mandon S., Freiwald W. A. & Kreiter A. K.(2005) Coherent oscillatory activity in monkey area V4

predicts successful allocation of attention. CerebralCortex (NewYork, N.Y.: 1991) 15, 1424–37.

Traeger N., Schultz B., Pollock A. N., Mason T., MarcusC. L. & Arens R. (2005) Polysomnographic values in

children 2–9 years old: additional data and review of theliterature. Pediatric Pulmonology 40, 22–30.

Traub R. D., Whittington M. A., Stanford I. M. & JefferysJ. G. (1996) A mechanism for generation of long-rangesynchronous fast oscillations in the cortex. Nature 383,621–4.

Ursino M., Magosso E. & Cuppini C. (2009) Recognitionof abstract objects via neural oscillators: interactionamong topological organization, associative memory andgamma band synchronization. IEEE Transactions onNeural Networks 20, 316–35.

Verrillo E., Bruni O., Franco P., Ferri R., Thiriez G.,Pavone M. et al. (2009) Analysis of NREM sleep in chil-dren with Prader-Willi syndrome and the effect ofgrowth hormone treatment. Sleep Medicine 10, 646–50.

Wechsler D. (1991) Wechsler Intelligence Scale for Children,3rd edn. The Psychological Corporation, San Antonio,TX.

World Medical Association (2008) World Medical Associa-tion Declaration of Helsinki: Ethical Principles for MedicalResearch Involving Human Subjects, Available at: http://www.wma.net/en/30publications/10policies/b3/ (retrieved06 December 2012).

Yang H. H., Savostyanov A. N., Tsai A. C. & Liou M.(2011) Face recognition in Asperger syndrome: a studyon EEG spectral power changes. Neuroscience Letters492, 84–8.

Zucconi M., Ferri R., Allen R., Baier P. C., Bruni O.,Chokroverty S. et al. (2006) The official World Associa-tion of Sleep Medicine (WASM) standards for recordingand scoring periodic leg movements in sleep (PLMS)and wakefulness (PLMW) developed in collaborationwith a task force from the International Restless LegsSyndrome Study Group (IRLSSG). Sleep Medicine 7,175–83.

Accepted 18 February 2013




http://www.wma.net/en/30publications/10policies/b3/
















Documents

Intellectual disabilities and power spectra analysis during sleep: a new perspective on borderline intellectual functioning