According to the National Institute of Alcohol Abuse and Alcoholism, a binge is a pattern of drinking alcohol that brings blood alcohol concentration (BAC) to 0.08 g percent or higher, which corresponds to consuming five or more drinks for men and four or more for women within a 2-h interval (National Institute of Alcohol, Abuse and Alcoholism, 2004). This type of excessive alcohol use is a highly prevalent pattern, mostly among high school and college students (Courtney and Polich, 2009). Recent reports show that around one third of young Europeans and North Americans are binge drinkers (BDs; Kraus et al., 2016; SAMHSA, 2016).

This prevalent pattern has been related to an array of negative consequences, including traffic collisions, poor academic performance, risky sexual behavior and neurocognitive deficits (White and Hingson, 2013; Carbia and López-Caneda, 2016). Concerning the latter, studies using neuropsychological batteries have demonstrated that BDs exhibit poor performance on tasks involving a number of cognitive processes such as, verbal and prospective memory (Heffernan et al., 2010; Carbia et al., 2017), inhibitory control (Czapla et al., 2015) or decision making (Moreno et al., 2012).

Electrophysiological measures of brain activity have also proved to be efficient methods for the comprehension of neurocognitive function as well as for the understanding of dysfunctional processes linked to various psychiatric disorders, including alcoholism (Parvaz et al., 2011; Kamarajan and Porjesz, 2015). Impairments in a variety of neurophysiological parameters have been consistently observed in alcoholic patients both during task execution and during resting state conditions (Campanella et al., 2009). In the resting electroencephalogram (EEG)—i.e., the electrophysiological recording of oscillatory brain activity while the person is relaxed- increased power in the beta and theta frequency bands has been reported extensively. In this sense, findings from several laboratories have repeatedly showed that alcoholics manifest higher resting beta power than non-dependent subjects (Costa and Bauer, 1997; Rangaswamy et al., 2002, 2003; Coutin-Churchman and Moreno, 2008; Ehlers et al., 2010; Mumtaz et al., 2016; Meyers et al., 2017). Furthermore, increases in the beta band in relapsers in comparison with alcoholics who remained abstinent (Bauer, 2001; Saletu-Zyhlarz et al., 2004) and in non-alcoholic individuals compared to alcoholic relatives (Pollock et al., 1995; Rangaswamy et al., 2004), appears to strength the relationship between the alcoholic spectrum and the beta power.

As with beta rhythms (although perhaps less consistently), elevated theta activity at rest has been reported from resting-state EEG studies in alcohol abusers (Pollock et al., 1992; Rangaswamy et al., 2003; Mumtaz et al., 2016; but see Saletu-Zyhlarz et al., 2004; Coutin-Churchman et al., 2006 for different findings).

Despite the functional meaning of these higher beta and theta rhythms in alcoholics is still under discussion, the increased power in this frequency bands has often been interpreted as reflecting cortical hyperexcitability—for beta rhythms- and reduced-cognitive processing capacity—for theta frequencies (Rangaswamy et al., 2002, 2003; Porjesz and Begleiter, 2003; Campanella et al., 2009).

Although comparatively much less studied, electrophysiological measures of brain activity also appear to be sensitive to the BD pattern (López-Caneda et al., 2014a; Petit et al., 2014b). As such, anomalous brain responses have been documented in BD youths during cognitive tasks using event-related potentials (ERPs) and event-related oscillations (EROs; Crego et al., 2009; Maurage et al., 2012; Smith and Mattick, 2013; Petit et al., 2014a; Watson et al., 2014; López-Caneda et al., 2017). Despite its well documented potential for detecting neurofunctional anomalies in chronic alcoholics, brain activity during task-free resting states has been virtually unaddressed in the BD population. As such, to the best of our knowledge only two studies have directly assessed the effects of BD on the frequency spectrum of the whole brain at rest (Courtney and Polich, 2010; Correas et al., 2015). Specifically, the EEG study conducted by Courtney and Polich (2009) showed increased delta and beta power in high-BDs during passive viewing, which was considered a potential biomarker for future alcoholism. Correas et al. (2015), using magnetoencephalography (MEG), reported increased theta and decreased alpha power in the occipital region of young BDs during eyes-closed resting state, which was considered an initial sign of anomalous neural activity caused by BD in youth. Although both studies point to anomalies in neural activity at rest in young BDs, the neurofunctional effects of this pattern of binge alcohol intake on the resting brain are far from well-studied.

The current study is the first to use source localization analysis to examine multichannel EEG data to detect the active brain regions during eyes-open and eyes-closed resting state in young BDs. To this end, we used the exact Low Resolution Electromagnetic Tomography Analysis (eLORETA), a well-validated source reconstruction model for mapping source activations of scalp-recorded EEG (Pascual-Marqui, 2007; Pascual-Marqui et al., 2011). We aimed to identify potentially abnormal intracortical EEG patterns in young BDs as compared to age-matched controls in specific brain regions to provide new neurophysiological evidence that might prove useful for early detection of brain damage associated with the BD pattern. We hypothesized that young BDs will display anomalous patterns of brain activity compared to control subjects, specifically in those frequency spectra typically impaired in alcohol abusers, namely, theta, and beta frequencies.

The major finding of the present study is that young university students reporting a BD pattern displayed, in comparison with age-matched control subjects, a significantly higher intracranial current density in the beta frequency band in the right temporal lobe (parahippocampal and fusiform gyri) during eyes-open resting state as well as higher intracranial current density in the theta band in the bilateral occipital cortex (cuneus and lingual gyrus) during eyes-closed resting condition (see Figure 3).

As noted in the introduction, increased beta frequency rhythms in the EEG of alcoholics, mostly during rest, has been widely documented (Rangaswamy and Porjesz, 2014). Although it is difficult to associate oscillatory activity of a single frequency rhythm with a unique brain function, it has been proposed that tonic beta rhythms indicate states of enhanced arousal (Steriade et al., 1993; Engel and Fries, 2010). Thus, an imbalance in these oscillations may affect the overall level of neural excitation, so that an increment in the beta band may reflect a state of cortical hyperexcitability (Edenberg et al., 2004). In this sense, the increased beta power commonly observed in chronic alcoholics has been associated with a neural disinhibition resulting from an imbalance between excitatory and inhibitory neurons (Begleiter and Porjesz, 1999; Porjesz and Begleiter, 2003; Coutin-Churchman et al., 2006).

Our data extend the notion of abnormally increased cortical excitability seen in alcoholics to youths with a BD pattern, who also show higher beta power at rest than age-matched controls. In the same vein, the only previous study that, to our knowledge, has assessed the resting EEG of young BDs also observed increased beta- and delta-power in high-BDs during passive viewing (Courtney and Polich, 2010). According to these authors and to our results, this increment in beta power would be suggestive of brain overactivity in young BDs.

The results of the present study also revealed that the BD group exhibited higher theta power compared to the control group during eyes-closed resting state. As stated early, increased resting theta activity has been reported from several studies in alcohol abusers (Pollock et al., 1992; Rasgaswamy et al., 2003; Mumtaz et al., 2016).

In addition to alcoholics, enhanced tonic theta power has been observed in a variety of neurological disorders including Alzheimer's disease (Dauwels et al., 2010), vascular dementia (Babiloni et al., 2004), or attention deficit hyperactivity disorder (Burke and Edge, 2013). Likewise, increases in the tonic theta band have been remarked in altered neurophysiological states of the brain such as, drowsy and sleep states (Makeig et al., 2000; Tanaka et al., 2000) and after consumption of anesthetic drugs (Voss and Sleigh, 2007) or depressant substances such as, alcohol (Stenberg et al., 1994; Ehlers et al., 1998; Ilan and Gevins, 2001), which appears to suggest that when the ability to respond to external stimuli decreases, spontaneous theta activity increases (Klimesch, 1999).

It is important to note here that tonic activity differs from phasic or event-related changes in their origin and in their functional implications. While phasic changes are more or less under volitional control and take place at a rapid rate, tonic changes are (usually) not deliberate and occur over the lifetime in response to circadian rhythms, fatigue, distress, etc. (Klimesch, 1999). Regarding theta frequency band, tonic and phasic theta oscillations seem to be dissociated with respect to cognitive performance. In this sense, it has been reported that phasic theta power is enhanced as a function of task difficulty, memory load or attentional and cognitive control demands, and that increments in this phasic oscillation are positively associated with increasing cognitive performance (Klimesch et al., 2008; Nigbur et al., 2011; Cavanagh and Shackman, 2015; Clayton et al., 2015). Meanwhile, the opposite holds true for tonic theta activity, i.e., a pronounced suppression of tonic theta power is observed with increasing cognitive activity, whereas increments are associated with impaired global cognitive function (Jeong, 2004; Niedermeyer, 2005; Dubbelink et al., 2014). Consequently, large power in the range of tonic theta frequency has been considered an index of reduced cognitive processing capacity (Klimesch, 1999), and this is the reason why the higher resting theta activity present in the EEG of alcoholics has been suggested to reflect a deficit in information processing capacity (Rangaswamy et al., 2003; Campanella et al., 2009). Accordingly, the increased tonic theta power observed in the BD group of the present study might indicate a poorer information processing capacity compared to the control group.

Supporting the notion that BD may lead to similar spontaneous oscillatory activity as that identified in alcoholics, the only study examining whole brain activity measured by MEG in young BDs and conducted from our research group also reported higher theta power in the occipital region in BDs in comparison with controls during eyes-closed resting state (Correas et al., 2015). According to other BD studies assessing potential neural anomalies at the functional and structural level (Squeglia et al., 2012; Howell et al., 2013; Doallo et al., 2014; López-Caneda et al., 2014b; Correas et al., 2016), studies suggest that abnormal increases in the theta frequency band might correspond with a neuromaturational delay induced by a binge pattern of alcohol consumption during the adolescent period.

Adolescence is a key developmental stage with profound changes in brain morphology, generally reflected in an increase in global white matter volumes and a decrease in global gray matter structures (Lenroot and Giedd, 2006; Fuhrmann et al., 2015). Mirroring these neuromaturational changes, the oscillatory activity in the brain at rest is reduced in a broad frequency range from low-delta to high-beta between late childhood and early adulthood (Dustman et al., 1999; Boord et al., 2007; Lüchinger et al., 2012; Rodríguez-Martínez et al., 2017). Specifically, Whitford et al. (2007) and Buchmann et al. (2011) reported a reduction in gray matter volumes with age (ranging from 10 to 30 years and from 8 to 19 years, respectively) that was accompanied by a significant decrease in spectral power—most prominent in low frequencies—during awake and sleep EEG recordings. This juxtaposed decline in EEG power and gray matter has led to the suggestion that the synaptic pruning that typically takes place during the transition from childhood to adulthood is responsible for the observed reduction in EEG frequencies (Boord et al., 2007; Whitford et al., 2007; Feinberg and Campbell, 2010; Buchmann et al., 2011; Gómez et al., 2017).

Within the heterogeneous pattern of neurodevelopmental trajectories, the temporal cortex is the region where gray matter matures last, peaking generally around 17 years of age (Giedd, 2004; Gogtay et al., 2004), although some studies have extended this peak to around 30 years of age (Sowell et al., 2003). Taking into account the relationship between EEG power and gray matter changes, the increased beta power recorded over the right temporal cortex (parahippocampal and fusiform gyri) in the young BDs—aged 18 years—of our study might be related to neuroanatomical immaturity in these subjects compared to controls. As mentioned above, this would be in accordance with several neurostructural studies indicating enlarged gray matter (in some cases gender- and region-specific) in the BD population (Squeglia et al., 2012; Howell et al., 2013; Doallo et al., 2014; Kvamme et al., 2016; Morris et al., 2017). In particular, Kvamme et al. (2016) found greater gray matter volumes in female BDs in the bilateral fusiform gyrus and the right parahippocampal gyrus, showing that these regions may be affected—i.e., may undergo delayed synaptic pruning—due to BD. However, the lack of studies directly comparing electrophysiological and structural data in young BDs coupled with the fact that other studies have reported reduced instead increased cortical volumes in this population (Mashhoon et al., 2014; Squeglia et al., 2015) exhort us to be cautious in this interpretation.

Besides that, a key feature of EEG maturation is the progressive replacement of tonic theta power in the occipital region in favor of faster rhythms from childhood to adolescence and early adulthood, continuing until the third decade of life (Niedermeyer, 2005; Segalowitz et al., 2010). Indeed, despite resting theta activity having its maximum in occipital regions, this is not prevalent in normal adult waking EEG (Puligheddu et al., 2005; Kamarajan and Porjesz, 2015; Gómez et al., 2017). Thus, the higher occipital theta power observed in young BDs relative to age-matched controls may be linked to a slowing down in the reduction of theta rhythms that naturally occurs during development. However, further studies examining the development of theta activity in young BDs are warranted to confirm this hypothesis.

In summary, the present study used EEG-source localization analysis in order to identify potentially anomalous resting EEG frequency patterns in young BDs. Results showed a significantly higher intracranial current density in the beta frequency band over the right medial temporal cortex (parahippocampal and fusiform gyri) during eyes-open rest, and in the theta band over the bilateral occipital cortex (cuneus and lingual gyrus) during eyes-closed, in 18 year-old BDs compared to age-matched controls. These findings seem to be in line with previous results in alcohol-dependent subjects and BDs which observed increased beta and/or theta power following chronic or heavy alcohol drinking. As in these studies, the augmented tonic beta and theta oscillations obtained in our study suggest cortical hyperexcitability and potential difficulties in the information processing capacity of young BDs. The presence of this anomalous EEG pattern in young non-dependent BDs with a relatively short history of consumption could be due to more deleterious effects of alcohol on still-in-development brains, perhaps by means of a delay on neuromaturational processes. Finally, although additional research involving the study of spontaneous EEG rhythms in BDs is required, this abnormally elevated tonic neural activity might act as a potential marker of early brain damage associated with BD.

EL and AnC collected the data. EL and AlC analyzed the data. SR, FC, and FM designed the study. EL wrote the first draft of the manuscript and all authors read, revised, and approved the final manuscript.