Our main objective is to design Dimensional Analysis (DA) algorithms for the multidimensional dialect PyLucid of Lucid, the equational data flow language. The significance is that the DA is indispensable for an efficient implementation of multidimensional Lucid and should aid the implementation of other data flow systems, such as Google’s TensorFlow. Data flow is a form of computation in which components of multidimensional datasets (MDDs) travel on communication lines in a network of processing stations. Each processing station incrementally transforms its input MDDs to its output, another (possibly very different) MDD. MDDs are very common in Health Information Systems and data science in general. An important concept is that of relevant dimension. A dimension is relevant if the coordinate of that dimension is required to extract a value. It is very important that in calculating with MDDs we avoid non-relevant dimensions, otherwise we duplicate entries (say, in a cache) and waste time and space. Suppose, for example, that we are measuring rainfall in a region. Each individual measurement (say, of an hour’s worth of rain) is determined by location (one dimension), day, (a second dimension) and time of day (a third dimension). All three dimensions are a priori relevant. Now suppose we want the total rainfall for each day. In this MDD (call it N) the relevant dimensions are location and day, but time of day is no longer relevant and must be removed. Normally this is done manually. However, can this process be automated? We answer this question affirmatively by devising and testing algorithms that produce useful and reliable approximations (specifically, upper bounds) for the dimensionalities of the variables in a program. By dimensionality we mean the set of relevant dimensions. For example, if M is the MDD of raw rain measurements, its dimensionality is {location, day, hour}, and that of N is {location, day}. Note that the dimensionality is more than just the rank, which is simply the number of dimensions. Previously, there’s extensive research on dataflow itself, which we summarize. However, an exhaustive literature search uncovered no relevant previous DA work other than that of the GLU (Granular Lucid) project in the 90s. Unfortunately the GLU project was funded privately and remains proprietary – not even the author has access to it. Our methodology is that we proceeded incrementally, solving increasingly difficult instances of DA corresponding to increasingly sophisticated language features. We solved the case of one dimension (time), two dimensions (time and space), and multiple dimensions. We also solved the difficult problem (which the GLU team never solved) of determining the dimensionality of programs that include user defined functions, including recursively defined functions. We do this by adapting the PyLucid interpreter (to produce the DAM interpreter) to evaluating the entire program over the (finite) domain of dimensionalities. As a result, the experimentally validated algorithms in our dissertation can produce useful upper bounds for the dimensionalities of the variables in multidimensional PyLucid programs. That also includes those with user-defined functions / Graduate

This thesis presents TransLucid, a low-level, purely declarative, intensional programming language. Built on a simple algebra and with just a small number of primitives, TransLucid programs define arbitrary dimensional infinite data structures, which are then queried to produce results. The formal foundations of TransLucid come from the work in intensional logic by Montague and Scott. The background chapters give a history of intensional logic and its predecessors in the Western world, as well as a history of intensional programming and Lucid, the first intensional programming language. The semantics of TransLucid are fully specified in the form of operational semantics. Three levels of semantics are given, in increasing order of efficiency, with the sequential warehouse semantics, the most efficient, being presented together with a proof that any expression will be evaluated by only examining relevant dimensions in the current context. The language is then extended in three important ways, by adding versioned identifiers, (declarative) side-effects and timestamped equations and demands. Adding versioned identifiers to TransLucid enriches the expressiveness of the language and allows the encoding of a variety of programming paradigms, ranging from manipulating large data-cubes to pattern-matching. Adding side-effects supports one of the main reasons for TransLucid: namely, to provide a target language, together with a methodology, for translating the main programming paradigms, thus creating a uniform end platform that can be the focus for optimisation and program verification. A translation of imperative programs into TransLucid is given. Timestamped equations and demands enable TransLucid to become a language for synchronous programming in real-time systems, as well as allowing runtime updates to a program's equations. The language TransLucid represents a decisive advance in declarative programming. It has applications in many fields of computer science and opens up exciting new avenues of research.

On étudie l'abstraction et la généricité dans les langages de programmation. On décrit les principaux outils de la programmation générique. On traite ensuite de la spécification des types et des fonctions et on étudie les methodes de vérification de programmes spécifiés algébriquement.

This thesis presents TransLucid, a low-level, purely declarative, intensional programming language. Built on a simple algebra and with just a small number of primitives, TransLucid programs define arbitrary dimensional infinite data structures, which are then queried to produce results. The formal foundations of TransLucid come from the work in intensional logic by Montague and Scott. The background chapters give a history of intensional logic and its predecessors in the Western world, as well as a history of intensional programming and Lucid, the first intensional programming language. The semantics of TransLucid are fully specified in the form of operational semantics. Three levels of semantics are given, in increasing order of efficiency, with the sequential warehouse semantics, the most efficient, being presented together with a proof that any expression will be evaluated by only examining relevant dimensions in the current context. The language is then extended in three important ways, by adding versioned identifiers, (declarative) side-effects and timestamped equations and demands. Adding versioned identifiers to TransLucid enriches the expressiveness of the language and allows the encoding of a variety of programming paradigms, ranging from manipulating large data-cubes to pattern-matching. Adding side-effects supports one of the main reasons for TransLucid: namely, to provide a target language, together with a methodology, for translating the main programming paradigms, thus creating a uniform end platform that can be the focus for optimisation and program verification. A translation of imperative programs into TransLucid is given. Timestamped equations and demands enable TransLucid to become a language for synchronous programming in real-time systems, as well as allowing runtime updates to a program's equations. The language TransLucid represents a decisive advance in declarative programming. It has applications in many fields of computer science and opens up exciting new avenues of research.

Made available in DSpace on 2014-12-17T15:36:38Z (GMT). No. of bitstreams: 1 SergioAMR_TESE_1-152.pdf: 4687573 bytes, checksum: d13791598d6a440077f9a5703901fe3b (MD5) Previous issue date: 2012-06-19 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / Lucid dreaming (LD) is a mental state in which the subject is aware of being dreaming while dreaming. The prevalence of LD among Europeans, North Americans and Asians is quite variable (between 26 and 92%) (Stepansky et al., 1998; Schredl & Erlacher, 2011; Yu, 2008); in Latin Americans it is yet to be investigated. Furthermore, the neural bases of LD remain controversial. Different studies have observed that LD presents power increases in the alpha frequency band (Tyson et al., 1984), in beta oscillations recorded from the parietal cortex (Holzinger et al., 2006) and in gamma rhythm recorded from the frontal cortex (Voss et al., 2009), in comparison with non-lucid dreaming. In this thesis we report epidemiological and neurophysiological investigations of LD. To investigate the epidemiology of LD (Study 1), we developed an online questionnaire about dreams that was answered by 3,427 volunteers. In this sample, 56% were women, 24% were men and 20% did not inform their gender (the median age was 25 years). A total of 76.5% of the subjects reported recalling dreams at least once a week, and about two-thirds of them reported dreaming always in the first person, i.e. when the dreamer observes the dream from within itself, not as another dream character. Dream reports typically depicted actions (93.3%), known people (92.9%), sounds/voices (78.5%), and colored images (76.3%). The oneiric content was related to plans for upcoming days (37.8%), and memories of the previous day (13.8%). Nightmares were characterized by general anxiety/fear (65.5%), feeling of being chased (48.5%), and non-painful unpleasant sensations (47.6%). With regard to LD, 77.2% of the subjects reported having experienced LD at least once in their lifetime (44.9% reported up to 10 episodes ever). LD frequency was weakly correlated with dream recall frequency (r = 0.20, p <0.001) and was higher in men (?2=10.2, p=0.001). The control of LD was rare (29.7%) and inversely correlated with LD duration (r=-0.38, p <0.001), which is usually short: to 48.5% of the subjects, LD takes less than 1 minute. LD occurrence is mainly associated with having sleep without a fixed time to wake up (38.3%), which increases the chance of having REM sleep (REMS). LD is also associated with stress (30.1%), which increases REMS transitions into wakefulness. Overall, the data suggest that dreams and nightmares can be evolutionarily understood as a simulation of the common situations that happen in life, and that are related to our social, psychological and biological integrity. The results also indicate that LD is a relatively common experience (but not recurrent), often elusive and difficult to control, suggesting that LD is an incomplete stationary stage (or phase transition) between REMS and wake state. Moreover, despite the variability of LD prevalence among North Americans, Europeans and Asians, our data from Latin Americans strengthens the notion that LD is a general phenomenon of the human species. To further investigate the neural bases of LD (Study 2), we performed sleep recordings of 32 non-frequent lucid dreamers (sample 1) and 6 frequent lucid dreamers (sample 2). In sample 1, we applied two cognitive-behavioral techniques to induce LD: presleep LD suggestion (n=8) and light pulses applied during REMS (n=8); in a control group we made no attempt to influence dreaming (n=16). The results indicate that it is quite difficult but still possible to induce LD, since we could induce LD in a single subject, using the suggestion technique. EEG signals from this one subject exhibited alpha (7-14 Hz) bursts prior to LD. These bursts were brief (about 3s), without significant change in muscle tone, and independent of the presence of rapid eye movements. No such bursts were observed in the remaining 31 subjects. In addition, LD exhibited significantly higher occipital alpha and right temporo-parietal gamma (30-50 Hz) power, in comparison with non-lucid REMS. In sample 2, LD presented increased frontal high-gamma (50-100 Hz) power on average, in comparison with non-lucid REMS; however, this was not consistent across all subjects, being a clear phenomenon in just one subject. We also observed that four of these volunteers showed an increase in alpha rhythm power over the occipital region, immediately before or during LD. Altogether, our preliminary results suggest that LD presents neurophysiological characteristics that make it different from both waking and the typical REMS. To the extent that the right temporo-parietal and frontal regions are related to the formation of selfconsciousness and body internal image, we suggest that an increased activity in these regions during sleep may be the neurobiological mechanism underlying LD. The alpha rhythm bursts, as well as the alpha power increase over the occipital region, may represent micro-arousals, which facilitate the contact of the brain during sleep with the external environment, favoring the occurrence of LD. This also strengthens the notion that LD is an intermediary state between sleep and wakefulness / O sonho l?cido (SL) ? um estado mental no qual o sujeito est? consciente de estar sonhando durante o sonho. A preval?ncia do SL em Europeus, Norte-Americanos e Asi?ticos ? bastante vari?vel (entre 26 e 92%) (Stepansky et al., 1998; Erlacher & Schredl, 2011; Yu, 2008) e em Latino-Americanos ainda n?o foi investigada. Al?m disso, as bases neurais do SL permanecem controversas. Diferentes estudos observaram um aumento da pot?ncia na frequ?ncia alfa (Tyson et al., 1984), na oscila??o beta na ?rea parietal (Holzinger et al., 2006) e no ritmo gama na regi?o frontal (Voss et al., 2009) durante o SL em rela??o ao n?o l?cido. Assim, para investigar a quest?o epidemiol?gica (Estudo 1), elaboramos um question?rio online sobre sonhos que foi respondido por 3427 volunt?rios. Em nossa amostra, 56% s?o mulheres, 24% s?o homens e 20% n?o responderam o g?nero; a mediana de idade foi de 25 anos. Um total de 76,5% dos indiv?duos refere que lembra dos sonhos pelo menos uma vez por semana. Cerca de dois ter?os dos sujeitos observam o sonho em primeira pessoa, ou seja, vendo o sonho da pr?pria perspectiva e n?o como mais um dos personagens do sonho. Os elementos mais comuns nos sonhos s?o movimentos/a??es (93,3%), pessoas conhecidas (92,9%), sons/vozes (78,5%) e imagens coloridas (76,3%). O conte?do on?rico se relaciona principalmente com planos para o dia seguinte (37,8%) e mem?rias do dia anterior (13,8%). Os pesadelos apresentam principalmente ansiedade/medo (65,5%), ser perseguido (48,5%) e sensa??es desagrad?veis que n?o envolvem dor (47,6%). Assim, sonhos e pesadelos podem ser evolutivamente entendidos como uma simula??o das situa??es frequentes que acontecem na vida e que se relacionam com a nossa integridade social, psicol?gica e biol?gica. Observamos tamb?m que a maioria dos indiv?duos (77,2%) relata ter tido pelo menos um SL, tendo experimentado na sua maior parte at? 10 epis?dios (44,9%). A frequ?ncia do SL foi fracamente correlacionada com a frequ?ncia de lembran?a dos sonhos (r=0,20, p<0,001) e foi tamb?m maior em homens (?2=10,2, p= 0,001). O controle do SL ? raro (29,7%) e inversamente correlacionado com o tempo de dura??o do SL (r=- 0,38, p<0,001), que normalmente ? curto: para 48,5% dos sujeitos o SL dura menos que 1 minuto. A ocorr?ncia do SL ? principalmente facilitada pela possibilidade de dormir sem hora para acordar (38,3%) que aumenta a chance de ter sono REM (SREM), e estresse (30,1%) que aumenta tamb?m as transi??es do SREM para a vig?lia. Como conclus?o, nossos resultados indicam que o SL ? uma experi?ncia relativamente comum (mas n?o recorrente), geralmente fugaz e dif?cil de controlar, o que sugere que o SL ? um est?gio intermedi?rio, incompleto e estacion?rio (ou fase de transi??o) entre o SREM e a vig?lia. Al?m disso, apesar das popula??es Europeias, Norte-Americanas e Asi?ticas terem uma preval?ncia de SL bastante vari?vel, nossos dados de uma amostra de Latino-Americanos fortalecem a no??o de que o SL ? um fen?meno universal da esp?cie humana. Para investigar as bases neurais do SL (Estudo 2), realizamos registros de sono em 32 sujeitos que n?o apresentam SL de forma frequente, e investigamos 6 sujeitos que apresentam SL recorrentemente. A primeira amostra foi submetida a duas t?cnicas cognitivo-comportamentais para induzir o SL: sugest?o pr?- sono (n = 8) e incuba??o de est?mulos do ambiente (pulsos de luz) no sonho durante o SREM (n = 8). Um grupo controle n?o foi submetido a nenhuma das duas t?cnicas (n = 16). Os resultados indicam que ? muito dif?cil induzir SL em laborat?rio, uma vez que conseguimos obter apenas um SL em um sujeito, que era do grupo em que aplicamos a t?cnica de sugest?o pr?-sono. O sinal eletroencefalogr?fico deste volunt?rio apresentou pulsos de ritmo alfa (7-14Hz) anteriores ao SL, de forma breve (aproximadamente 3s), sem altera??o significativa do t?nus muscular e independente da presen?a de movimentos oculares r?pidos. O SL desse sujeito apresentou tamb?m uma maior pot?ncia do ritmo alfa (7-14Hz) na regi?o occipital e um aumento de atividade gama (20- 50Hz) na regi?o temporo-parietal direita. Nos 6 sujeitos que frequentemente t?m SL, o mesmo apresentou em m?dia um aumento de pot?ncia em gama alto (50-100Hz) na regi?o frontal em compara??o com o SREM n?o-l?cido; no entanto, isso aconteceu de forma clara para apenas um dos indiv?duos. Observamos tamb?m que quatro desses volunt?rios apresentaram um aumento da pot?ncia do ritmo alfa na regi?o occipital, pouco antes do SL, ou durante o mesmo. Dessa forma, nossos resultados preliminares sugerem que o SL apresenta diferentes caracter?sticas neurofisiol?gicas dos estados t?picos de SREM e vig?lia: 1) Os pulsos de ritmo alfa, bem como o aumento da pot?ncia dessa oscila??o na regi?o occipital, podem ser micro-despertares. Estes facilitam o contato do c?rebro durante o sono com o meio externo, favorecendo a ocorr?ncia do SL e fortalecendo a ideia de que o SL seria um estado intermedi?rio entre o sono e a vig?lia. 2) Como as regi?es temporoparietal direita e frontal se relacionam com a forma??o da auto-consci?ncia e da imagem corporal, sugerimos que um aumento de atividade nessas regi?es durante o sono pode ser o mecanismo neurobiol?gico subjacente ao SL

Cette thèse vise à identifier et évaluer les impacts biogéophysiques des changements d'usage des sols depuis les 150 dernières années jusqu'à la fin du XXIe siècle sur le climat en Afrique de l’Ouest à partir des modèles LUCID et des scénarios CMIP5 utilisés dans le contexte LUCID-CMIP5. Les analyses menées dans cette thèse se sont d'abord basées dans le Sahel et dans le Golfe de Guinée où les changements passés de la couverture terrestre sont supérieurs à 5%. Les simulations LUCID ont été d'abord évaluées dans cette thèse en comparant les précipitations et la température de l'air simulées par les modèles aux données d'observation. Les analyses ont montré que la moyenne et la variabilité inter-annuelle observées des précipitations et de la température sont respectivement sous-estimées et surestimées par la plupart des modèles de climat LUCID même si la température semble mieux simulée que les précipitations. Dans cette étude, les deux simulations actuelles forcées respectivement par une distribution actuelle et pré-industrielle de la couverture terrestre ont été comparées. Les résultats montrent qu'il n'y a pas de différence évidente entre ces deux simulations par rapport aux valeurs moyennes climatiques des précipitations et de la température dans les modèles comme si les changements de la couverture terrestre n'ont pas vraiment d'importance sur la représentation de ces variables. Dans le Golfe de Guinée, les analyses montrent que l'expansion des surfaces cultivées et des pâturages s'est effectuée au détriment d'une déforestation entraînant une diminution du LAI, une augmentation d'albédo et une diminution de la rugosité de surface. Les analyses montrent que les impacts historiques des changements d'occupation des sols sur le climat dans ces régions restent très petits par rapport aux changements induits par l'augmentation des gaz à effet de serre dans l’atmosphère. Le LAI simulé par les modèles de surface LUCID et leur relation avec le climat en Afrique de l'Ouest ont été évalués, les résultats montrent que les précipitations sont fortement et positivement corrélées à la densité de feuillage avec des valeurs supérieures ou égales à 0.8 dans les deux régions. La plupart des modèles de climat montrent que la corrélation entre le LAI et la température de l'air est positive dans le Sahel et négative dans le Golfe de Guinée et suggèrent que plus de LAI dans le golfe de Guinée conduit plus d'évapotranspiration et donc une surface plus froide, alors que dans le Sahel l'effet d'albédo de l'augmentation du LAI peut dominer et augmenter la température de surface.Dans un second temps, l'impact biophysique des changements futurs de la couverture terrestre sur le climat de surface du XXIe siècle a été évalué à l'aide des simulations spécifiques similaires aux scénarios RCP8.5 mais avec une végétation fixe en 2006. Les analyses révèlent qu'à l’échelle régionale, les impacts biophysiques des changements d'occupation des sols dans les scénarios ont été globalement faibles mais statistiquement significatifs au Sahel et en Afrique centrale où la déforestation est prescrite dans le futur (>10%), mais avec une large dispersion sur la réponse du climat résultant aux différentes paramétrisations de la surface terrestre dans les modèles de climat. / By climate models developed in the LUCID project and CMIP5 models used in the LUCID-CMIP5 projet, this thesis aims to identify and evaluate biogeophysical impacts of LULCC of the past 150 years and the end of XXIst century on surface climate in West Africa. Focusing analysis in two contrasted regions of West Africa: Sahel and Guinea where land cover change is above 5% since pre-industrial times, results reveal expansion of crops and pasture and deforestation in Guinea in all LUCID models. In this work, simulations of present-day rainfall and surface air temperature have been compared with observed datasets. Results show that the observed mean and inter-annual variability of rainfall are respectively underestimated and overestimated by most of the seven climate models. Overall surface air temperature is better simulated than precipitation.Two simulations of rainfall and surface air temperature, forced respectively with present-day and pre-industrial land cover distribution are also compared. Results show that there is no obvious/visible difference between the two simulations with respect to mean climatic values of both rainfall and temperature as if the changes in land cover did not really matter for the good representation of those variables. Finally, this thesis evaluates leaf area index (LAI) in the LUCID models and its relationships with surface climate. Observations reveal that precipitation is highly and positively correlated to foliage density with values larger or equal to 0.8 in both the Sahel and Guinea. Five out of seven models show positive correlations, but not as large as in the observations. However none of the models is able to capture a larger correlation between precipitation and LAI in Guinea than in the Sahel. Most of climate models show that correlation between LAI and surface air temperature is positive in the Sahel and negative in Guinea. It suggests that more LAI in Guinea will lead to more evapotranspiration and therefore cooler surface, while in the Sahel the albedo effect of increased LAI may dominate and increase surface temperature. Finally, analysis reveals that historical effects of land-use changes are not regionally significant among the seven climate models due to a small land-cover change prescribed in these regions compared to the changes induced by large scale forcing such as sea surface temperatures changes and CO2 concentration increase.Furthermore, biogeophysical impact of land-use change in the XXIst Century climate were evaluated using specific simulations similar to RCP8.5 scenarios but with a prescribed fixed land cover map on 2006. The analysis reveals, that in contrast of last 150 years, deforestation continues in the coming years in tropical region in scenarios resulting from the extension of the cultivated area reaching 15 million km2 in 2100 over tropical Africa. Regionally, the biogeophysical impacts of projected changes in land cover in RCP8.5 scenarios were generally small but statistically significant in the Sahel and Central Africa regions where deforestation is more than 10% with a wide dispersion of climate response due to differents parameterizations of land surface in climate models.

Il y a une longue histoire derrière l’intérêt porté pour les rêves lucides –conscience de rêver pendant un rêve– datant de l’époque de la Grèce antique. La compréhension du rêve lucide a d’abord reposée sur des observations personnelles de certains individus ayant eux-mêmes expérimentés leurs propres rêves lucides. La découverte de l’inconscient, de même que celle du sommeil paradoxal (SP), a grandement contribué à l’essor du rêve lucide, qui a finalement constitué l’objet d’étude de plusieurs recherches empiriques, dont celles menées par le groupe de Stanford, qui est l’un des premiers à avoir évalué la capacité des rêveurs à signaler leurs rêves lucides à l’aide de différents mouvements du corps, en temps réel. Le rêve lucide est un phénomène prévalent, comprenant de nombreuses dimensions, allant de la conscience de soi jusqu’au contrôle sur le contenu du rêve. Certains auteurs ont également parlé du rêve lucide en tant qu’état mental hybride, combinant des caractéristiques des rêves survenant en SP et d’autres caractéristiques propres à l’éveil. De plus, il s’avère que les individus rapportant des rêves lucides fréquemment présentent des caractéristiques psychologiques particulières, sans compter les corrélats physiologiques mesurables qui ont également été associés aux rêves lucides, de même que des corrélats neurophysiologiques. Certains auteurs se sont questionnés sur les différentes façons d’induire des rêves lucides en sommeil. Les techniques proposées sont nombreuses et diversifiées, allant des techniques comportementales aux techniques de stimulation électrique. À cet effet, deux études ont utilisé la stimulation transcrânienne par courant direct (tDCS) (Stumbrys, Erlacher, & Schredl, 2013) et alternatif (tACS) (Voss et al., 2014) dans les régions frontales pendant le SP, montrant une association entre l’augmentation de la conscience de soi dans les rêves et l’activité gamma fronto-temporale. Toutefois, ces études renferment plusieurs faiblesses méthodologiques. Nous avons donc tenté de reproduire ces résultats considérables en appliquant la tACS en SP dans les régions frontales pendant 2.5 min à une fréquence de 40 Hz lors de siestes matinales –favorisant des périodes riches en SP– et en utilisant des tests statistiques plus conservateurs ainsi que des mouvements oculaires pour signaler la lucidité. Pour s’y faire, nous avons recruté 33 sujets dans une étude randomisée à simple aveugle dans laquelle la tACS a été appliquée (STIM) ou non (SHAM) en SP. Les sujets ont été réveillés 3 min après la fin de la STIM ou du SHAM pour répondre à un journal de rêves et une échelle évaluant le degré de lucidité contenu dans les rêves. Bien que plusieurs signaux de lucidité ont été observés, ces derniers se sont produits dans les deux conditions expérimentales, et ce, sans différence significative. L’auto-évaluation du contenu des rêves ne différait pas non plus entre les conditions. Par conséquent, les résultats de la présente étude ne permettent pas de conclure que la tACS appliquée à une fréquence de 40 Hz en SP engendre plus de rêves lucides signalés en temps réel, un plus grand nombre de rêves lucides tels que définis par la LuCiD Scale et des scores plus élevés aux facteurs de cette même échelle. Plusieurs considérations méthodologiques peuvent avoir jouer un rôle dans la production de rêves lucides en sommeil, tels que les critères de recrutement, la pratique des signaux de lucidité et les effets indirects de la tACS, et avoir masqué l’effet de la tACS, qui, dû à son effet sans doute minime, n’a peut-être pas été en mesure d’influencer significativement la production de rêves lucides. Somme toute, plusieurs sujets ont été en mesure de signaler ou de rapporter des rêves lucides dans le cadre de cette étude, ce pourquoi il serait intéressant d’étudier davantage les différentes applications des rêves lucides, autant dans le but d’accroitre nos connaissances sur les rêves et les théories de la conscience que pour développer de potentiels outils cliniques. / There is a long history of interest in lucid dreaming–the awareness of dreaming while dreaming–from the time of ancient Greece. The understanding of lucid dreaming was initially based on personal observations of certain individuals who had themselves experienced lucid dreams. Discovery of the unconscious, as well as rapid-eye-movement (REM) sleep, greatly contributed to the rise of interest in lucid dreaming, which finally became the object of several empirical studies. The Stanford group is one of the first to have conducted laboratory studies specifically on lucid dreams assessing, among other attributes, the ability of dreamers to signal their lucid dreams with different body movements in real-time. Lucid dreaming is a prevalent phenomenon, comprising many dimensions, ranging from self-awareness to control over dream content. Some authors have also described lucid dreaming as a hybrid mental state, mixing REM dreaming and wakefulness characteristics. Moreover, frequent lucid dreamers display certain psychological characteristics, such as high creativity, an internal locus of control, and increases in several cognitive functions. Lucid dreaming has measurable physiological and neurophysiological correlates. Some authors have speculated about the different ways to induce lucid dreams during sleep. The proposed techniques are many and varied, ranging from behavioral techniques to electrical stimulation techniques. Critically, two studies employing frontal transcranial direct (tDC) (Stumbrys et al., 2013) and alternating (tAC) current stimulation (Voss et al., 2014) during REM sleep, have shown an association between increased self-awareness in dreams and fronto-temporal gamma electroencephalographic activity. However, these studies suffer from several methodological weaknesses. We attempted to replicate these important findings using frontal tAC stimulation during morning, REM-rich naps, real-time signal verification and more appropriate conservative statistical tests. We recruited 33 subjects in a single-blind randomized group study in which tAC stimulation was applied (STIM) or not (SHAM) during the REM sleep of a morning nap. During the STIM condition, we applied frontal 40 Hz tAC stimulation during REM sleep for 2.5 min. Subjects were awakened 3 min after stimulation offset for dream reporting and administration of a scale to assess dream lucidity. Although several episodes of signal-verified dream consciousness were observed, these were equally frequent in SHAM and STIM conditions. Self-ratings of dream content also did not distinguish conditions. Therefore, the findings of the present study do not indicate that frontal gamma stimulation increases dreamed self-awareness as measured by real-time signal-verified lucid dreams and self-ratings of dreams. Several methodological considerations may have played a role in the negative findings, such as recruitment criteria, the eyes-closed practice signal and the indirect effects of tAC stimulation during the phosphene and sensation tests. The latter may have masked a real, albeit weak, tAC stimulation effect. Thus, many subjects were able to signal or report lucid dreams in this study, and this supports the notion that further study of lucid dreaming may lead to fruitful applications. Investigating lucid dreaming could be useful for increasing our knowledge of the physiological substrate of dreams, for clarifying theories of consciousness, and for developing potential clinical tools.