Wormwood turns up in a surprising number of brain-health product descriptions. The pitch is usually assembled from two true statements: the plant is high in antioxidant polyphenols, and antioxidant activity is relevant to neurodegeneration. What sits between those statements and a benefit in an actual person is the part worth examining.
This article covers what has and has not been tested. Nothing here is medical advice, and wormwood has not been approved by the FDA to diagnose, treat, cure, or prevent Alzheimer’s disease, Parkinson’s disease, or any other condition affecting memory or cognition.
Key Takeaways
- The entire neurological evidence base for wormwood is preclinical. There is no human trial of wormwood for memory, cognition, or any neurodegenerative disease.
- The most quotable result, a 2021 rat study reporting improved spatial memory and reduced amyloid plaques, was retracted. The journal published a formal withdrawal notice in 2022 [2] [3].
- A separate rat study using scopolamine to induce memory impairment did find improved maze and avoidance performance at 50 to 200 mg/kg, alongside reduced markers of oxidative damage in cortex and hippocampus [1]. It has not been replicated.
- Two cell-culture studies report that wormwood extract protects neural cell lines from chemically induced injury [4] [5]. Cell-line protection is a screening result, not a treatment.
- The doses used in these rodent studies are not translatable to a supplement dose, and none of the papers addressed whether a human taking that much would exceed thujone exposure limits [6].
- A 2010 review in this exact area warned that promising neuroprotective findings in rats needed a careful risk-benefit analysis before any human application, because wormwood has a well-defined profile of adverse properties [6].
The Retracted Study, and Why It Matters
Start with the paper that gets cited most, because the citation is no longer valid.
In 2021, Neurobiology of Learning and Memory published a study in which 48 male Wistar rats received an injection of amyloid-beta into the brain, a standard way of modelling the plaque pathology seen in Alzheimer’s disease. Treated animals were then given hydroalcoholic wormwood extract by gavage at 10, 50, or 100 mg/kg per day. The reported result was striking: dose-dependent improvement in spatial memory on the Morris water maze, plus reduced tissue degeneration, fewer amyloid plaques on Congo red staining, and less apoptosis on TUNEL staining [2].
That is the single most favourable finding in the wormwood neuroscience literature, and it is the reason wormwood appears in Alzheimer’s-adjacent marketing at all.
The following year the same journal published a withdrawal notice for it [3]. PubMed now carries the article with the publication type Retracted Publication.
A retraction does not prove the underlying claim is false. It does mean the paper can no longer be used as evidence for anything, and anyone still citing it either has not checked or is not telling you. This is worth knowing precisely because retracted papers keep circulating: the abstract stays indexed, the summary stays quotable, and the withdrawal notice is a separate record that almost nobody clicks through to.
What the Surviving Rodent Study Actually Found
Removing the retracted paper leaves one animal study of reasonable size, and it is a genuinely different experiment.
Fifty rats were divided into five groups. One control group received solvent only. A second received scopolamine at 2 mg/kg, a drug that blocks acetylcholine signalling and reliably produces memory impairment, which is the standard pharmacological model of amnesia. The remaining three groups received scopolamine plus hydroalcoholic wormwood extract at 50, 100, or 200 mg/kg before the scopolamine dose [1].
Two behavioural tests were used. In the Morris water maze, extract-treated animals reached the hidden platform faster and swam a shorter distance to find it. In passive avoidance testing, they took longer to enter the compartment associated with a foot shock and spent less time there overall, which is read as better retention of the aversive memory [1].
The biochemistry moved in the same direction. In cortex and hippocampus, the extract lowered malondialdehyde and nitric oxide metabolites, both markers of oxidative stress, and raised total thiol content along with catalase and superoxide dismutase activity, all antioxidant defences [1].
Read carefully, that is a coherent result: an antioxidant-rich plant extract blunted a chemically induced memory deficit in rats and moved oxidative markers the way you would expect. Read carelessly, it becomes a claim that wormwood improves memory. The gap between the two is the whole point. Scopolamine-induced amnesia in a rat is not age-related forgetfulness in a person, and blunting a drug effect is not the same as improving baseline cognition.
The Cell-Culture Work
Two further studies used cell lines rather than animals, and they belong to a different tier of evidence again.
In the first, OLN-93 cells, a rat oligodendrocyte line, were pretreated with wormwood extract for two hours and then exposed to quinolinic acid, a metabolite that accumulates in several neurodegenerative conditions and is toxic to neural tissue. Pretreatment preserved cell viability and reduced reactive oxygen species, lipid peroxidation, and apoptosis [4].
In the second, SH-SY5Y human neuroblastoma cells were treated with ethanolic wormwood extract for 24 hours and then exposed to 6-hydroxydopamine, a toxin used to model the dopaminergic cell loss of Parkinson’s disease. At 6.25 to 25 micrograms per millilitre, the extract reduced reactive oxygen species, apoptosis, and malondialdehyde, and raised glutathione and superoxide dismutase [5].
These are competent experiments and they are also screening assays. A protective effect in a dish tells you a compound is worth investigating; it says nothing about whether an oral dose survives digestion, reaches the brain in a meaningful concentration, or does anything useful once there. The literature is full of botanicals that clear this bar and then fail every subsequent one.
A Third Line: Withdrawal Symptoms in Rats
One 2025 study sits outside the memory question but is often lumped in with it. Rats were given MDMA for seven days, then challenged with naloxone to precipitate withdrawal, having received ethanolic wormwood extract at three concentrations alongside [7].
The authors reported that the two higher extract concentrations reduced the severity of withdrawal behaviours such as writhing and body dragging, while the lowest concentration was described as possibly not well tolerated and needing re-evaluation to minimise adverse effects [7].
It is a small animal study in a niche model, and the honest summary is that it does not support any human use. It is included here because it is part of the neuro literature people find when they search, and because a study whose own authors flag a tolerability concern in one arm should not be presented as a safety signal in the other direction.
The Dose Problem Nobody Resolves
Every rodent result above shares one unaddressed weakness, and a 2010 review in the Journal of Ethnopharmacology named it directly.
Reviewing an earlier round of neuroprotective findings in rats, the author noted that it remained unclear in those studies whether the threshold dose for thujone, as set by the European Medicines Agency, would be exceeded during therapeutic use. Thujone exhibits neurotoxic properties above threshold concentrations, producing dose-dependent tonic-clonic seizures in animals, likely through modulation of GABA type A receptors. The review’s conclusion was that application in humans should be preceded by a thorough and careful risk-benefit analysis [6].
That warning was published in 2010 and has not been answered. A neuroprotection claim for a plant whose signature compound lowers the seizure threshold is not automatically wrong, but it does need the dose arithmetic done, and none of the studies here do it. Our article on wormwood and thujone covers the safety side in detail, and wormwood drug interactions covers why anticonvulsants and centrally acting medications are the specific concern.
How to Read Brain-Health Marketing for This Plant
A few practical filters, given the state of the evidence.
Check whether a cited study is still live. The amyloid paper is the clearest case on this site of a retracted result that continues to circulate. PubMed labels retractions plainly; the label is on the record page.
Watch for species substitution. Reviews of the Artemisia genus discuss neuroprotective potential across more than 500 species [8], and a claim sourced from a genus-level review is not a claim about Artemisia absinthium. Our comparison of wormwood and sweet wormwood covers why that distinction keeps mattering.
Treat cell and rodent results as questions, not answers. Language such as shows promise, supports brain health, or has been studied for memory is compatible with everything on this page, including the retracted paper. It is compatible with almost nothing a person would actually want to know.
A Note on the Evidence
The neuro evidence for wormwood is entirely preclinical, and the single most citable result was formally withdrawn. No human trial has tested wormwood for memory, cognition, or any neurodegenerative condition.
Frequently Asked Questions
Does wormwood improve memory in humans?
No human trial has tested it. Every result described here comes from rats or from cells in a dish, and effects in those models routinely fail to reproduce in people [1].
What happened to the wormwood Alzheimer’s study?
A 2021 paper in Neurobiology of Learning and Memory reported that wormwood extract improved spatial memory and reduced amyloid plaques in rats. The journal published a withdrawal notice for it the following year, so it should not be cited as evidence [2] [3].
Is there anything on wormwood and Parkinson’s disease?
Only a cell-culture experiment. Wormwood extract reduced markers of damage in a neuroblastoma cell line exposed to 6-hydroxydopamine, a chemical used to model dopaminergic injury. That is several steps removed from a human disease [5].
Does the thujone problem apply to brain-health use?
It applies more, not less. Thujone reduces inhibitory signalling at GABA type A receptors and produces dose-dependent seizures in animals, so a compound proposed for neurological benefit carries a documented neurological risk at higher doses [6].
Should I take wormwood for brain fog or cognitive decline?
There is no evidence base to support that use. Cognitive symptoms have many causes that deserve a real diagnosis, and this is a conversation for a clinician rather than a supplement decision.
References
- Hosseini M et al. The effects of Artemisia absinthium L. on scopolamine-induced learning and memory impairment and brain tissue oxidative damage in adult rats. Avicenna Journal of Phytomedicine (2023). PMID 36698740
- Rahmani F et al. Artemisia absinthium improves spatial performance and neuronal injury induced by amyloid-beta in the CA1 hippocampal area of male Wistar rats. RETRACTED. Neurobiology of Learning and Memory (2021). PMID 34478861
- Withdrawal notice to Artemisia absinthium improves spatial performance and neuronal injury induced by amyloid-beta in the CA1 hippocampal area of male Wistar rats. Neurobiology of Learning and Memory (2022). PMID 34998673
- Artemisia Absinthium Extract Attenuates the Quinolinic Acid-Induced Cell Injury in OLN-93 Cells. Current Drug Discovery Technologies (2023). PMID 36998142
- Protective effect of Artemisia absinthium on 6-hydroxydopamine-induced toxicity in SH-SY5Y cell line. Avicenna Journal of Phytomedicine (2021). PMID 34046320
- Lachenmeier DW. Wormwood (Artemisia absinthium L.), a curious plant with both neurotoxic and neuroprotective properties? Journal of Ethnopharmacology (2010). PMID 20542104
- Artemisia absinthium L. (Wormwood) Extract Effect on 3,4-Methylenedioxymethamphetamine Withdrawal in Rats: An Animal Study. Addiction and Health (2025). PMID 41431632
- Selim S et al. Unveiling the multifaceted potential of Artemisia: insights into nutritional benefits and emerging therapeutic applications. Journal of Pharmaceutical Analysis (2026). PMID 42199532
These statements have not been evaluated by the Food and Drug Administration. This information is not intended to diagnose, treat, cure, or prevent any disease. Content is for informational purposes only and is not medical advice; consult a qualified healthcare provider before starting any supplement. As an Amazon Associate we earn from qualifying purchases.