New Study: Recreational Cannabis Users Test Like Non-Users

A study published earlier this year in the peer-reviewed journal Comprehensive Psychiatry took a careful look at a question that often goes muddled in the courtroom: does using cannabis make you cognitively impaired in some lasting, measurable way or does that depend entirely on how a person uses it?

The answer the researchers reached, simplified: it depends. People who use cannabis recreationally performed essentially like non-users on a comprehensive battery of cognitive tasks. People whose use rose to the level of problematic showed one selective deficit in complex working memory. Across the broader range of executive functions and learning tasks, all three groups looked largely the same. NORML’s May 28 summary described the finding as recreational consumers performing similarly to abstainers.

For a Texas drugged driving defense practice, the study is useful but only if it is read precisely. It does not say marijuana cannot impair driving. It does not say it is safe to drive high. It does say something more specific (and more useful in the courtroom) about the assumptions prosecutors, juries, and even some experts bring to a marijuana DUID case.

What the Study Actually Did

The study (Pesthy et al., “Dissociating the cognitive underpinnings of recreational cannabis use from problematic use,” Comprehensive Psychiatry, Vol. 147 (2026)) compared three groups of adults: non-users, recreational users, and a group at risk for cannabis use disorder, classified using the Cannabis Use Disorder Identification Test–Revised (CUDIT-R), a validated screening tool. The at-risk group cleared a cutoff score of 13 on the CUDIT-R, which past research has tied to high sensitivity and specificity for clinically diagnosed cannabis use disorder.

All three groups were given a comprehensive neuropsychological battery measuring working memory (both simple and complex), inhibitory control, cognitive flexibility, and implicit learning. The result that drew attention:

• Recreational users did not differ from non-users on any major cognitive domain measured by the study.
• The at-risk (problematic-use) group showed a selective deficit, but only in complex working memory (the ability to hold and manipulate multiple pieces of information at once).
• Inhibitory control, cognitive flexibility, and implicit learning were preserved across all three groups.

In other words, the cognitive cost of cannabis use (to the extent the study could detect one) was not a feature of cannabis use generally. It tracked addiction risk. That is a meaningful distinction the courts and the general public regularly fail to make.

The study’s scale is worth stating plainly. One hundred twenty-two participants were included in the analysis: 43 non-users, 36 recreational users, and 43 in the problematic-use group. Group assignment came from a self-report screening questionnaire rather than a clinical diagnosis. The only abstinence requirement was that participants not use cannabis on the day of testing, and that was self-reported rather than confirmed by any toxicology. One participant who arrived intoxicated was excluded from the data.

The working memory finding is also narrower than a headline conveys. It appeared on a single task, at p = .044, with an effect size the authors’ own stated benchmarks place below the medium range. The study’s multiple-comparison correction was applied to its moderation analyses rather than across the full battery of outcome measures. Reporting those numbers is not a concession. It is what keeps the rest of this usable when somebody checks.

What the Study Does Not Say

Before this kind of finding is used in any courtroom, the limits have to be honest:

• It is not a driving study. Participants were not asked to operate vehicles or perform divided-attention tasks under acute intoxication. The study measured baseline neuropsychological performance in the non-intoxicated state.
• It is not a study of acute impairment. Nothing in it suggests that smoking a joint and getting behind the wheel an hour later is safe. Cannabis acutely affects reaction time, divided attention, and motor coordination, and that is well established in the literature.
• It is a snapshot, not a longitudinal trial. The design compared groups at a single point. It is not a long-term prospective study of what happens to a person’s cognition over years.
• “Recreational” in the study has a specific meaning. It refers to users below the CUDIT-R threshold for problematic use so not anyone who has ever consumed cannabis in any quantity and not heavy daily users.

Used correctly, the study is a careful refutation of a single, broad claim: that any cannabis use produces measurable, lasting cognitive damage. That claim, repeated in cultural shorthand for decades, is not what the science shows.

Why This Matters for Texas Drugged Driving Cases

Texas, like every other state, prohibits operating a motor vehicle while impaired by a controlled substance. Under Texas Penal Code § 49.04, a person commits DWI by operating a motor vehicle while intoxicated. “Intoxicated” is defined in § 49.01(2) as not having the normal use of mental or physical faculties by reason of the introduction of alcohol, a controlled substance, a drug, a dangerous drug, a combination of two or more of those substances, or any other substance into the body or having an alcohol concentration of 0.08 or more. There is no per se THC concentration in Texas. The State must prove actual impairment.

That last point matters more than people realize, because in practice DUID cases are often tried on a different theory than the statute requires. A typical marijuana DUID case looks like this: an officer pulls a driver over for a traffic violation, smells marijuana, asks questions, conducts field sobriety tests, makes an arrest, and obtains a blood draw. The lab reports the presence of THC or its metabolites in the driver’s blood. The State then asks the jury to combine those facts into a conclusion of impairment.

The Pesthy study is one more piece of peer-reviewed support for the position the defense has been taking on those cases for years:

• The mere fact that a person uses cannabis recreationally does not establish that they are cognitively impaired in any baseline, lasting way. A juror who believes “marijuana users have impaired brains” (a belief the State will not correct) is starting from a premise the science does not support.
• THC in the blood is not impairment. THC and its metabolites can be detected in blood and urine well beyond any period of acute psychoactive effect, particularly in regular users. A positive blood test alone tells the jury that a person used cannabis at some point; it does not tell them whether the person was impaired at the time of driving.

That proposition does not rest on inference. In its 2017 report to Congress on marijuana-impaired driving, the National Highway Traffic Safety Administration concluded that THC level in blood does not appear to be an accurate and reliable predictor of impairment from THC, and that peak impairment does not occur when THC concentration in the blood is at or near peak levels. A 2022 meta-regression in Neuroscience and Biobehavioral Reviews examined 28 publications and more than 800 driving-related outcomes and found correlations between blood THC and driving performance of roughly negative 0.08 to negative 0.10. Among regular users, meaning weekly or more often, the authors found no significant relationship at all.

The counterweight belongs here too. A 2025 meta-regression in Drug Science, Policy and Law found a real dose-response relationship between blood THC and crash culpability at the population level, with culpability roughly doubling near five nanograms. Both things are true. Population-level risk is a different question from whether a particular concentration proves that a particular driver had lost the normal use of his faculties at the time of operation, which is the question the statute actually asks.

• The State’s burden is impairment at the time of operation. Not impairment generally. Not impairment yesterday. Not impairment as a lifestyle inference. The statutory text is specific, and the science around residual versus acute effects supports holding the State to that specific burden.

How a Defense Lawyer Can Use This

Studies like Pesthy are not magic bullets. They are pieces of context that change how juries hear the State’s case. A few practical uses:

• Cross-examination of the State’s drug recognition expert. A drug recognition evaluator who suggests, expressly or by implication, that habitual cannabis users are cognitively compromised can be pressed on the current peer-reviewed literature. The Pesthy study, the broader meta-analytic record on adult recreational use, and existing reviews from the NIH databases are fair territory.
• Voir dire and opening. Jurors come into the box with cultural beliefs about cannabis that often do not match what the science currently shows. Identifying those assumptions in voir dire, and previewing in opening that the case will turn on impairment at the time of driving (not on lifestyle inferences) reframes how the evidence is received.
• Closing argument. “The State has shown you that my client uses cannabis. The State has not shown you that my client was impaired at the time he was driving” is the structurally honest closing in a residual-THC DUID case, and the Pesthy line of research helps make it credible.
• Pretrial motions. Where the State intends to offer expert opinion that conflates THC presence with impairment, the current scientific record is a basis for limiting the opinion or, in appropriate cases, a Kelly challenge to the methodology underlying the opinion.

The Bigger Picture

Marijuana cases are going to remain a significant part of the Texas criminal docket. The Texas Legislature has not legalized recreational cannabis, but the cultural reality of cannabis use in Texas, the rise of hemp-derived THC products, and the growth of medical cannabis under the Texas Compassionate Use Program have made cannabis far more visible in everyday life than the criminal code reflects. DUID cases involving cannabis are not going to slow down.

The science around cannabis and cognition is, in turn, becoming more careful. The decades-old caricature of the burnout user, slow and dimmed by his habit, has not held up well under scrutiny. The 2026 Pesthy study is part of a growing literature that says cognitive harm from cannabis is real but specific. It’s concentrated in problematic, addiction-level use, and largely absent in recreational use measured outside the window of acute intoxication.

One more paper belongs in this picture, and it cuts in both directions. Pilon and colleagues published a meta-analysis in Addictive Behaviors in 2025 finding small to moderate impairments across 10 of 13 cognitive domains in cannabis use disorder. Pesthy and colleagues cite it themselves. Set beside their finding of a single selective deficit, it suggests the 2026 study may sit at the optimistic end of the current literature rather than in the middle of it.

It also supports the distinction this post is drawing. The deficits Pilon found are in disordered use, not in recreational use. That is the same line Pesthy draws, and it is the line that matters in a courtroom.

For a Texas defense lawyer, that nuance is not just intellectually interesting. It is the structure of an honest, evidence-based DUID defense.

Deandra Grant holds a Master of Science in Pharmaceutical Science, a Graduate Certificate in Forensic Toxicology, and the ACS-CHAL Forensic Lawyer-Scientist designation. Deandra Grant Law handles DWI and DUID cases throughout North and Central Texas. Call (214) 225-7117 for a free, confidential case review.