Cannabis is the second most widely used drug in the United States by adolescents and young adults with a lifetime prevalence rate of 15.4% for adolescents and 51.5% for young adults [1,2]. As adolescence is a time of critical brain development [3], specifically within the endocannabinoid (eCB) system [4,5,6,7], adolescents may be significantly impacted by early, repeated cannabis use. Despite these concerns, neurocognitive alterations linked with cannabis exposure following monitored abstinence in adolescents is understudied and findings to date are inconsistent.

The primary psychoactive component of cannabis is Δ⁹-tetrahydrocannabinol (THC) that activates the endogenous endocannabinoid (ECB) system. THC interacts and binds to the cannabinoid receptor 1 (CB1) and animal research has demonstrated that repeated cannabis results in downregulation or desensitization of the CB1 receptor [8]. Notably, the ECB system continues to develop throughout adolescence into young adulthood [3,9]. Repeated use during adolescence may result in alterations in regions rich in eCB signaling, including the parietal cortex [10,11,12], a region that continues to develop into adolescence and underlies sustained attention [13]. Consistent with this, several neuropsychological studies have shown evidence that young regular cannabis users, typically defined as at least weekly cannabis use, have demonstrated poorer sustained and selective attention, attentional control, and increased lapses in attention [14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36]. Thus, repeated exposure to cannabis during the neurodevelopmental phase of adolescence may result in alterations in the attentional network.

One criticism of the current literature is that cannabis effects may only be observed within a few days of use, and this may be related to either very acute effects or even cannabis withdrawal versus more chronic impact [37]. Indeed, some research has suggested, in humans, there is evidence of recovery in CB1 receptor density even only after a few days of cannabis abstinence [38]. In addition, other research demonstrates recovery of cognitive function, particularly on memory-based tasks [39]. For example, Schweinsburg et al., (2010), [40] found that among heavy adolescent cannabis users who had remained abstinent (between 27 and 60 days) showed improvements in a spatial working memory task relative to recent cannabis users (between 2 and 7 days). Some studies have found attentional abnormalities measured on neuropsychological tests in cannabis using adolescent and young adults compared to controls after two to three weeks of monitored abstinence [22,28,33,41]. Although [33] Hanson et al. (2010) found improved verbal working memory after a three-week period of abstinence among adolescent cannabis users (ages 15–19), however, attention accuracy on a sustained and selective attention task remained impaired. Notably, our group has found improvements in a sustained attention task after two weeks of abstinence among a slightly older sample of adolescent and young adult cannabis users (ages 16–26), but found no significant difference between controls and cannabis users on verbal working memory and verbal learning tasks [16]. However, at a later time point in this same sample, our group also found evidence of deficits on another sustained attention task (CPT) [41], as well as working memory and psychomotor speed [23]. Thus, some evidence suggests attentional differences remain even after sufficient time for THC metabolites to be eliminated from the body and withdrawal symptoms have generally recovered [42,43]. Taken together, previous evidence suggests that impairments in sustained attention among cannabis users may take longer to recover compared to deficits shown in verbal or working memory. Despite these findings, few studies to date have examined underlying neuronal signaling in the attentional networks following a period of monitored abstinence.

It is theorized that there are two anatomically and functionally segregated but complementary attention networks [44], the dorsal and ventral attention networks (DAN and VAN). The DAN employs dorsal fronto-parietal areas which involves top–down visual attentional control and voluntary allocation of attention. The visual attention system is a top–down system controlled by both cognitive factors (i.e., prior knowledge, expectation, and current goals), and bottom–up factors reflecting sensory stimulation [44]. The DAN consists of the bilateral inferior parietal sulcus (IPS) and bilateral frontal eye fields (FEF) [44]. The DAN is activated when there is prior knowledge or expectation of seeing an object in a particular location, for example, movement in a certain direction [45]. The DAN is also involved in associating relevant stimuli to responses and is activated during task preparation [46]. The DAN responds alongside the VAN, a secondary system including the ventral fronto-parietal areas. The VAN is activated during this bottom–up processing of new or unexpected sensory stimuli [44]. This processing of new and behaviorally significant stimuli can cause a shift in attention or goal directed processes [44]. The VAN is theorized to be involved when an important stimulus appears outside the context of the current task or focus of attention, and the VAN initiates a reorienting or “circuit-breaking” to interrupt the dorsal signaling, ultimately redirecting attention [47]. Structurally, the VAN is localized in the right hemisphere and consists of the right temporoparietal junction (TPJ) and the right ventral front cortex (VFC) [44,45].

The developmental trajectory of both attention networks continues through to young adulthood [48]. Resting state functional connectivity studies have found an inverted U-shaped trajectory of the DAN through development, as the DAN connections become more dominant throughout childhood, until the age of 30 in which synaptic pruning eliminates many short-range connections from development [49,50]. Specifically, there is stronger within-network connectivity in the DAN among adolescents aged 11–13 relative to 19–25-year-olds [51]. Particularly, for the DAN, relative to adolescents, adults show greater functional connectivity between the FEF and the posterior cingulate cortex [52]. As the DAN becomes more dominant through childhood and eventually pruned in adulthood, the VAN shows greater functional connectivity among adults relative to children [52,53]. For the VAN, relative to adults, adolescents show greatest functional connectivity among the VFC and nodes related to salience, including the anterior insula and the anterior cingulate cortex (ACC), which may demonstrate a reduced segregation of these networks in childhood relative to adulthood [52]. Given this ongoing development, exposure to exogenous cannabis may alter this development of attentional networks in regular cannabis using youth.

To better understand these attentional networks, resting state functional connectivity (RSFC) can be used to measure the synchrony of regional brain activation during rest, and are particularly useful in the absence of a task [54,55,56,57,58,59]. In previous studies, the DAN and VAN are shown to be consistently segregated systems using RSFC [60,61]. Previous research has found that the DAN and VAN are associated with performance on behavioral tasks including the Attention Network Test-Revised test, a task assessing ability of alerting, orienting, and other attentional functions [62] among healthy adults. Functional MRI studies have shown associations of the connectivity of the DAN and VAN associated with behavioral performance on the Attention Network Task, suggesting that the DAN and VAN are correlated with alerting, orienting, and executive control among healthy controls [63], indicating an intrinsic functional network and important neuronal mechanism in attention. Additionally, other work has shown associations with the DAN and VAN with other sustained attention tasks including the gradual Conners’ Continuous Performance Test (grad CPT) and the Test of Variables of Attention (TOVA) [64]. Other studies have shown intraindividual fluctuations in accuracy on attention tasks associated with the DAN [65,66,67] that may be explained by low DAN connectivity can precede an attention lapse. Additionally, a previous study demonstrated that functional connectivity in the DAN is associated with selective attention skills in early childhood [68]. A study assessing functional connectivity analyses among the DAN and VAN networks and the default mode network (DMN) found a strong negative relationship between the DMN and both attentional networks, and this negative relationship was found to have a robust relationship with behavioral performance on the flanker task [69]. Specifically, the strength of this negative correlation varied among individuals’ response times of the flanker task and were related to intraindividual variability, which can be a useful index in how efficient attention is regulated, particularly as an indicator of neurological dysfunction [69].

Although there is strong behavioral evidence to support alterations in attentional tasks and performance among cannabis users, there have been limited studies using resting state functional connectivity (RSFC) to measure the DAN and the VAN connectivity among adolescent and young adult regular cannabis users. In a study of, primarily, male young adults (n = 103, 18–38 years old) with and without both early psychosis and cannabis use history, controls with a history of cannabis use showed increased connectivity in the DAN compared to non-using participants, and DAN strength positively correlated with severity of cannabis use dependence [70]. Notably, no abstinence period was required, the sample included participants outside of the young adult developmental period and was primarily male; thus, findings may not generalize to younger samples. [71] Peeters and colleagues (2015) found that cannabis use did not moderate the relationship between psychosis risk and functional connectivity in frontoparietal regions [71]; they did not find any association or moderation of cannabis use. Notably, this study was limited to the frontoparietal region and did not specifically distinguish the DAN or VAN networks. Thus, due to the paucity of this research, further examination of RSFC among DAN and VAN and how differences in connectivity may correlate with cannabis use symptoms may shed light in determining mechanisms of cannabis use effects on attentional neural networks.

The primary goal of the present study was to investigate the effects of regular, or weekly, cannabis use on the dorsal and ventral attentional networks using seed-based resting-state functional connectivity in adolescents and young adults after a two-week period of monitored abstinence. The secondary goal explored whether the corresponding differences are associated with cannabis use patterns and behavior. We hypothesized that increases in connectivity within the dorsal (DAN) and ventral (VAN) attentional networks will be demonstrated for chronic cannabis users. Additionally, we hypothesized that the increase in intraconnectivity within either the DAN or VAN will be associated with increased cannabis use severity, decreased length of abstinence, earlier age of regular use onset, and increased severity of Cannabis Use Disorder (CUD) symptoms.