The main findings of this study were an increase in psychotomimetic symptoms following administration of both THC alone and the combination of THC + CBD. Administration of both THC and the combination (THC + CBD) increased negative symptoms on the BPRS, along with perceptual distortions and cognitive disorganisation on the PSI. Lower frequency cannabis users experienced a reduction in psychotomimetic symptoms following CBD alone compared with placebo. Both THC and THC + CBD impaired performance on episodic and working memory tasks. Contrary to hypotheses, CBD did not offset the psychotomimetic effects of THC.

In line with previous studies, we found that THC in the laboratory at a dose of 8 mg inhaled in a Volcano vaporiser increased some psychotic-like symptoms^7,40. Contrary to predictions, CBD when given concurrently with THC in a 2:1 ratio had no impact on psychotomimetic symptoms, which were still higher than placebo or CBD alone. It has previously been suggested that CBD may ameliorate the harmful effects of THC on psychotic-like symptoms⁵¹ which has been supported by some empirical evidence^27,30,32 including two experimental studies (n = 6 in a crossover²⁷ and n = 48 in parallel groups³²). The latter study³² found a reduction in positive psychotic symptoms following intravenous THC, which did not reach significance following pre-treatment of oral CBD. However, a significant protective effects of CBD was found when the authors compared the number of people who met clinically significant psychosis (an increase from baseline of ≥3 points¹⁵). It may be the case therefore, that CBD is only protective when THC induces a strong psychotic reaction, which might be achieved by using higher doses of THC. In the current study, although THC increased scores on the PSI (perceptual distortion and cognitive disorganisation) and BPRS negative symptoms, it did not increase BPRS positive or PSI paranoia scores at a group level. This in itself is interesting as the same dose and administration route used here has been reliably shown to increase these symptoms in other studies (e.g., ref. ⁴⁰), possible reasons maybe repeating each of these measures in this four-way crossover design. Additionally both our groups were heavier cannabis users than previous studies which have administered similar doses in the lab and tolerance to psychotomimetic effects has been previously shown (D’Souza et al., 2005). Another important consideration is that it is unclear which CBD:THC ratio is most effective for reducing harm¹⁶. In this study doses of 8 mg THC and 16 mg CBD were chosen to create a CBD:THC ratio of 2:1, reflecting the upper limit (mean + 3 SD) found in high CBD/low THC cannabis preparations⁴². In addition, Englund et al. pre-treated volunteers with oral CBD (600 mg) and before intravenous THC (1.5 mg/kg). These very different doses and routes of administration (which influence absorption and metabolite profiles⁵²) may account for our divergent findings. In street cannabis, findings from our own lab did not suggest an acute protective effect of CBD against the psychotomimetic effects of THC³¹ but rather a chronic effect, related to levels in hair^30,51, which is in accordance with the findings of this current study.

Similarly with cognition, impairments to working and episodic memory were observed following both administration THC and the combination of THC + CBD. Contrary to our predictions, CBD did not offset the effects of THC. This differs from our naturalistic findings where users smoking cannabis with higher levels of CBD resulted in less impairment than those with lower levels³¹. One reason for this difference from naturalistic findings could be other chemicals present in whole plant cannabis material, including terpenes⁵³, which do not occur in our synthetically produced cannabinoids. However, interactive effects of CBD and THC on memory function were also reported by Englund et al. in a controlled study of synthetic oral CBD and intravenous THC³². Moreover, in the current data we also found that CBD improved facial affect recognition when administered alone, and offset impairments on the same task when co-administered with THC³⁸. It is worth noting that the protective effects in CBD that study were of a small effect size, and limited to a single intensity of facial stimuli (40%)³⁸. Moreover, CBD was not protective of immediate verbal recall in Englund et al.³², and although it showed evidence for a protective effect on delayed recall, these were not supported by a significant condition by group interaction. Thus, taken together, the ability of CBD to protect against THC-induced cognitive impairment may be of a small effect size, and/or influenced by vulnerability factors which have not been considered in research studies to date.

One interesting and unexpected finding was of the greater number of correct exemplars generated following THC + CBD than placebo, but no differences between placebo and THC alone and CBD alone. This is partly in line with previous findings with cannabis users⁵⁴ of increases in divergent thinking following acute cannabis use. This increase occurred only in the group given CBD alongside THC, one possible tentative explanation is that CBD’s pro-cognitive effects combine with the ability of THC to stimulate novel thinking to result in successful task performance.

In this study, no tolerance to the cognitively impairing or psychotomimetic effects of THC was observed for high compared to low frequency users which contrasts with previous studies^8–10. Differences between these studies could be attributable to dose and/or route of administration (e.g. smoked or intravenous versus vaporised).

Additionally, previous studies suggesting tolerance develops to the acute cognitive impairing effects of THC have generally been small-scale. However, our findings are consistent with a recent study which like ours, used inhalation of THC via a volcano vaporiser⁶⁰. This was a large-scale, cross-over study with 122 participants ranging from daily cannabis users to very infrequent users (as low as once in the last 3 months). Like our present data, they found no evidence of tolerance to the impairing effects of acute THC on neurocognitive function.

Our rationale for including users high and low in schizotypy was that those high in schizotypy may be more vulnerable to the pro-psychotic effects of THC^11–14. however this hypothesis was not borne out in the data. Additionally we saw no evidence of tolerance to the psychotomimetic effects when THC was administered in this manner in the lab to heavy cannabis users, when compared to lower frequency users, in contrast to previous findings⁸. There may be a variety of reasons for this, including the fact that the current study used a relatively low dose of THC (8 mg) compared to doses estimated in naturalistic studies (~35 mg in the UK⁴²; ~32 mg in the Netherlands⁶⁰). The inhalation procedure was standardised, with the aim of controling for dose titration that may occur with cannabis smokers in a naturalistic setting^42,60. Previous studies investigating associations between schizotypy and acute psychotomimetic effects of cananbis/THC have used naturalistic designs^11,12 or retrosepctive reports of drug effects^13,14. It is therefore possible that differences in dose, smoking behaviour and expectancy/recall may have influenced these findings to some extent. A controlled study¹⁵ reported similar psychotomimetic effects of THC in people diagnosed with schizophrenia and controls.

An interesting finding to emerge from this research was that CBD alone reduced baseline levels of psychotomimetic symptoms in light but not heavy users. Moreover, we found a trend level correlation in heavy users, suggesting that the antipsychotic effects of CBD were increasingly blunted as years of cannabis use increased. This finding is broadly consistent with our previous finding that CBD in hair was associated with fewer psychotic-like symptoms in light but not heavy users³⁰. This relates to recent work suggesting that CBD may be a potential treatment in schizophrenia through boosting brain levels of anandamide³³. Crucially, that study excluded patients with a positive urine screen for cannanabinoids, which might have influenced their antipsychotic response to CBD. The effect of CBD observed in this study was confined only to low frequency users, suggesting a potential tolerance to these effects in higher frequency users, which may have implications for the future use of this compound in the treatment of schizophrenia, which is highly co-morbid with cannabis use disorders⁶⁰. However, further research with repeated dosing of CBD in patients would be needed to support or refute this possibility. Neurobiologically, that these effects were confined to low frequency users fits with research suggesting that high frequency cannabis users have a reduction in anandamide⁶⁰, potentially as a result of chronic cannabis use, which may mean their endocannabinoid systems are less sensitive to exogenous cannabinoids. In addition, a subtle pro-cognitive effect of CBD was observed on the trailmaking task, but only as an increase in psychomotor speed, without any impact on cognitive flexibility, hence this finding should be treated with caution.

This study has several strengths; it used a large sample size in a four-way cross-over in a highly controlled laboratory setting and the use of well-validated tasks. The Volcano

Vaporiser method of administering cannabinoids produces similar plasma and pulmonal THC levels in comparison to smoked cannabis cigarettes^59,60 and delivers between 80% of the loaded THC^35,36. Limitations of the current study include not having plasma measures of THC and CBD, therefore we were not able to accurately verify our controlled inhalation procedure, although it has been verified in previous research⁴⁰.

In summary, our study replicated previous findings of a pro-psychotic effect of THC for some psychotic symptoms e.g. negative symptoms, cognitive disorganisation and perceptual distortions and extended these to include a similar effect of a combination when given in a 2:1 ratio (CBD: THC). Cognitive impairments were evident following THC and the combination but no cognitive impairments were observed following CBD alone. CBD was able reduce sub-clinical psychotic-like symptoms in low frequency cannabis users, but not in those using the drug more heavily. This highlights a potentially important area of further research, given the therepeutic potential of CBD for psychosis and the high incidence of cannabis use disorders in this population.