The search yielded 2394 studies across three databases (EBSCOhost: n = 885; SCOPUS: n = 911 and PubMed: n = 598). Inclusion criteria were applied to initial searches resulting in 1916 eligible records with 478 records excluded for not meeting inclusion criteria. Initial title and abstract searches were conducted, excluding 1884 records. The remaining studies (n = 32) included 22 duplicates, which were removed, leaving ten studies suitable for reference list searches. One additional record was identified that met all inclusion criteria, resulting in 11 studies included in the final scoping review, as shown in Fig. 1.

Flow chart of the preferred reporting items for systematic reviews

An overview of the characteristics of these studies are shown in Table 2 (experimental studies) and Table 3 (self-report survey designs), which outlines the citation information, country of data collection, type of study design, experimental conditions or survey information, measures of cognition, sample size and participant details, and study findings. More than half of the studies (n = 6) were published in the USA (Hobkirk et al. 2018; MacLean et al. 2021; Palmer and Brandon 2019; Wade et al. 2022; Xie et al. 2020a, b), with two studies each being published in England (Dawkins et al. 2013, 2012), Korea (Kim et al. 2022a, b) and one in Italy (Caponnetto et al. 2017).

Seven studies (sample sizes ranging from n = 9 to n = 128) were experimental with participants randomly allocated to experimental conditions, such as different types of e-cigarette devices or different concentrations of nicotine, to determine effect on a measure of cognition (Caponnetto et al. 2017; Dawkins et al. 2013; Dawkins et al. 2012; Hobkirk et al. 2018; Kim et al. 2022a; MacLean et al. 2021; Palmer and Brandon 2019). Four studies adopted cross-sectional survey designs (Kim et al. 2022b, b; Wade et al. 2022; Xie et al. 2020a, b). Among these, three utilised large-scale nationally representative self-report surveys, with sample sizes ranging from n = 18,535 to n = 886,603 (Kim et al. 2022b; Xie et al. 2020a, b). However, within these surveys, cognition was only measured by a single question (e.g., “Do you have serious difficulty concentrating, remembering, or making decisions?”; Xie et al., 2022). In another smaller survey, participants (N = 203) were divided into three groups based on their self-reported history of e-cigarette or cigarette use (e.g., ‘never users’, ‘e-cigarette-only users’, and ‘cigarette-only users’), with group differences in self-reported cognitive function compared (Wade et al. 2022).

Of the studies included in this review, two studies included adolescent and young adult participants (Wade et al. 2022; Xie et al. 2020a), while the rest recruited adult participants only (Caponnetto et al. 2017; Hobkirk et al. 2018; MacLean et al. 2021; Dawkins et al. 2013; Kim et al. 2022a, b; Xie et al. 2020b). The majority of studies (n = 6) included participants who were either current smokers only (Caponnetto et al. 2017; MacLean et al. 2021; Dawkins et al. 2012; Dawkins et al. 2013; Kim et al. 2022a), or e-cigarette users with a history of cigarette smoking (Palmer and Brandon 2019). Another study included only participants with a history of e-cigarettes and vaping e-liquid containing nicotine (Hobkirk et al. 2018). The remaining four studies grouped participants according to user status, such as ‘non-cigarette-smokers’, ‘past smokers’ and ‘current smokers’ (Kim et al. 2022b), or ‘exclusive e-cigarette users’, ‘exclusive cigarette smokers’ and ‘never users’ (Wade et al. 2022; Xie et al. 2020a, b).

Cognition is a very broad construct and incorporates many domains. Experimental studies included in this review primarily focused on domains of memory and attention. Memory was assessed using either computerised cognitive tests, such as the N-BACK Working Memory test [8], the 3-BACK Alphabet/Digit recognition task (Kim et al. 2022a), and the Continuous Performance Task (MacLean et al. 2021), or pen-and-paper tests, such as the Cambridge Prospective Memory Test (Dawkins et al. 2013) or the Brown-Peterson Memory Task (Dawkins et al. 2012). Attention was assessed by either computerised measures, such as the Continuous Performance Test – AX version [8], the Automated Neuropsychological Assessment Metrics (MacLean et al. 2021), and the Rapid Visual Information Processing Task (Palmer and Brandon 2019), or a self-reported rating scale, e.g. difficulty in concentration visual analogue scale from 0/not at all to 100/very much (Hobkirk et al. 2018).

For the survey design studies, a self-reported response to a single question (as part of a larger survey) related to cognitive problems was used in three studies (Kim et al. 2022b; Xie et al. 2020a, b). The remining one utilised the National Institute of Health Neurocognitive Toolbox to measure memory and attention (Wade et al. 2022).

When examining the cognitive effects of e-cigarettes across 11 studies, results were inconsistent: three studies each reported either cognitive improvements (Dawkins et al. 2012, 2013; Palmer and Brandon 2019) or cognitive impairments (Kim et al. 2022a; Xie et al. 2020a, b), while five studies reported no significant effect of e-cigarettes on cognition (Caponnetto et al. 2017; Hobkirk et al. 2018; Kim et al. 2022b; MacLean et al. 2021; Wade et al. 2022).

Dawkins et al. (2012) examined the cognitive effects of e-cigarettes with a group of current cigarette smokers who had never used e-cigarettes. They randomly allocated participants to either 18 mg nicotine e-cigarette (nicotine condition), 0 mg nicotine e-cigarettes (placebo condition), or a’just hold’ condition, where participants simply held the e-cigarette but did not inhale (Dawkins et al. 2012). The results showed e-cigarette use led to improved performance in memory recall but had no effect on attention-related visual-spatial information processing (Dawkins et al. 2012). Similarly, Dawkins et al. (2013) examined the effect of e-cigarettes on prospective memory with a sample of regular cigarette smokers who were randomly allocated a test order for two e-cigarette conditions: 18 mg nicotine e-cigarettes and a control of 0 mg nicotine e-cigarettes. They reported that e-cigarettes improved time-based prospective memory but had no effect on event-based prospective memory. Furthermore, Palmer and Brandon (2019) examined the immediate cognitive effects of e-cigarettes by randomly allocating current e-cigarette users with a history of cigarette smoking to one of four experimental conditions: 12 mg nicotine e-cigarette and were told the e-cigarette contained nicotine, 0 mg nicotine e-cigarette and were told the e-cigarette contained nicotine, 12 mg nicotine e-cigarette and were told the e-cigarette did not contain nicotine, and 0 mg nicotine e-cigarette and were told the e-cigarette did not contain nicotine. It was shown that nicotine e-cigarette improved sustained attention, and the improvement was more salient among females (Palmer and Brandon 2019). However, these studies only tested individuals with a history of nicotine use and lacked control groups (i.e., participants who had never used e-cigarettes or cigarettes), which makes it difficult to isolate the actual effect of e-cigarettes from nicotine, and improved cognition might be related to reduced withdrawal effect associated with e-cigarettes use (Dawkins et al. 2013; Palmer and Brandon 2019). Nevertheless, a study compared the self-reported difficulty in concentration in regular nicotine e-cigarette users pre- and post-overnight abstinence, they found no significant effect of e-cigarettes on cognition (Hobkirk et al. 2018).

In contrast to the reported cognitive improvements, three studies reported cognitive impairments related to e-cigarette use. A repeated-measures study examining the cognitive effects of vaping e-cigarettes (containing 16 mg/ml nicotine) on current cigarette smokers (N = 22) in relation to regular cigarette use following overnight cessation (about 12 h of smoking abstinence) had opposite findings (Kim et al. 2022a). It was reported that participants performed more poorly in memory task following e-cigarette use compared to using their regular brand cigarette (Kim et al. 2022a). It was argued that regular cigarette smokers were not fully satiated by vaping e-cigarettes, and this led to differences in behavioural measures (Kim et al. 2022a). In line with this, studies that adopted large-scale nationally representative survey reported increased risks of subjective cognitive impairments (Xie et al. 2020a, b). This risk amongst exclusive e-cigarette users appears higher than amongst ‘never users’ or exclusive cigarette users (Xie et al. 2020b). Furthermore, Xie et al. (2020a) reported that adolescent e-cigarette users were at a significantly higher risk of difficulties in concentration, remembering, and making decisions compared to ‘never users’, and that the risk was even higher in adolescents who initiated e-cigarettes at a younger age. However, it should be noted that survey question as a measure of cognition does not identify unknown causes of cognitive difficulties and there are also survey studies that failed to identify significant association between e-cigarette use and cognitive decline (Kim et al. 2022b; Wade et al. 2022).

Overall, it appears that experimental studies that reported acute effects of e-cigarette use suggest cognitive improvements, which may have more to do with the cognitive effects of nicotine, rather than e-cigarettes specifically. Studies reporting significant results from surveys found cognitive impairments of e-cigarette use among participants who have never smoked cigarettes as well as amongst those with a history of cigarette smoking in both adult and adolescent samples.