The Caffeine Paradox: Anxiety Trigger or Stress Protector?

How a substance that increases anxiety may also protect the brain from stress

By Jess Wight | 11 September 2026

“Caffeine is the most widely used mood-altering drug in the world… people often see coffee, tea, and soft drinks simply as beverages rather than vehicles for a psychoactive drug”

Roland Griffiths, PhD, professor in psychiatry and neuroscience at the Johns Hopkins University School of Medicine, describing how caffeine is culturally disguised

“Caffeine makes you anxious”. It’s a familiar message, and one that is supported by a substantial body of research. Caffeine is a central nervous system stimulant, and controlled studies have found that consumption – particularly in large amounts – can increase physiological arousal. At more moderate doses, caffeine can also improve alertness and attention: a 2025 systematic review and meta-analysis involving more than 1,400 participants indicated that acute caffeine consumption improved both reaction time and accuracy on measures of attention.

For some people, these effects can be experienced as nervousness, restlessness or anxiety. A 2024 meta-analysis of studies in healthy adults found that caffeine intake was associated with increased anxiety, with substantially larger effects at doses above 400 mg. this aligns with Health Canada’s recommended maximum daily intake of 400 mg for a healthy adult. The Coffee Association of Canada reports that coffee is Canada’s most popular beverage, with daily consumption even surpassing that of tap water as revealed by a 2020 survey.

However, many people are unaware of the amount of caffeine they consume. A standard single espresso is around 100 mg, and most standard coffee shops use a minimum of one double espresso in a small drink, meaning your large Starbucks order can contain upwards of 300 mg of caffeine in just one coffee. This means that some people think they are only having two coffees a day but could, in reality, be consuming closer to 700 mg a day.

More recently, a 2026 systematic review of 27 studies concluded that caffeine was generally associated with increased anxiety symptoms, also suggesting a dose-dependent relationship, finding that just 200 mg to be consistently associated with heightened anxiety.

This evidence has contributed to a broader narrative around coffee and mental health: that while a cup of coffee may improve alertness and concentration in the short term, consuming too much could ultimately be detrimental to psychological wellbeing.

There is some truth behind this, but new research suggests this isn’t the complete picture. A new study published this week suggests that the relationship between caffeine and mental health may be considerably more complex, particularly when we consider the effects of chronic stress. In a study of mice, regular consumption of caffeinated coffee appeared to protect against several behavioral, cognitive and neurobiological changes associated with prolonged stress. Importantly, the same effects were not observed in mice consuming decaffeinated coffee.

The findings do not mean that caffeine is an anxiety treatment, but they do challenge the assumption that caffeine’s effects on mental health are predominantly negative. In the context of chronic stress, the findings suggest that caffeine may offer previously underappreciated benefits, potentially protecting mood, cognitive function and aspects of brain health from some of the effects of prolonged stress. This shifts the focus from viewing caffeine simply as a potential risk factor for anxiety towards considering whether, under certain circumstances, its effects may be more beneficial than harmful.

Concerns surrounding caffeine and anxiety are not unwarranted. Caffeine works primarily by blocking adenosine receptors in the brain. Adenosine is involved in promoting sleepiness and regulating neural activity, so blocking its effects increases alertness and arousal. The importance of this mechanism was described in detail by Daly and colleagues in a paper published in 1994, which identifies the A1 and A2A receptors as the main targets responsible for caffeine’s effects.

At higher doses, this stimulation can produce sensations that overlap with the physical experience of anxiety. For some people, increased arousal, restlessness and other physiological effects of caffeine can become uncomfortable. It seems logical that this appears to be particularly relevant for people with existing anxiety-related conditions: a systematic review found that high-dose caffeine provoked panic attacks in more than half of participants with panic disorder, compared with just 1.7% of healthy individuals. This clearly demonstrates that some individuals may be particularly sensitive to the effects of caffeine.

Extending beyond the ‘coffee jitters’, there is another well-known route through which caffeine could affect psychological wellbeing: sleep disruption. Caffeine’s effects can persist for several hours after ingestion, meaning that when consumed later in the day, it can interfere with subsequent sleep. A 2023 systematic review and meta-analysis of 24 studies found that caffeine reduced total sleep time by approximately 45 minutes and reduced sleep efficiency by around 7%. It also increased the time required to fall asleep and the amount of time spent awake during the night, indicating that caffeine can disrupt every aspect of an individual’s sleep health.

The importance of this surpasses simply feeling tired the following day. People often underestimate the impact of poor sleep; however, a recent umbrella review indicates that sleep deprivation has implications far beyond fatigue: a review of 29 studies highlights the wide-ranging effects, from increasing the risk of physical health conditions like cardiovascular disease and diabetes to impaired emotional regulation, increased anxiety, stress, and depressive symptoms.

Taken together, these mechanisms help explain why caffeine has traditionally been regarded with caution in discussions of mental health. Its stimulant effects can increase physiological arousal, and consuming caffeine later in the day can meaningfully disrupt sleep.

However, there is an important limitation to this evidence – much of what we know about caffeine and anxiety concerns its immediate or relatively short-term effects. It tells us considerably less about what happens when caffeine is regularly consumed in the context of prolonged stress. This is a gap which the new study is beginning to fill.

Stress is far from simply an unpleasant psychological experience. Prolonged exposure to stress can actually produce measurable changes in the brain and subsequent behaviors. A major review from 2009 examined evidence from both human and animal research and concluded that chronic exposure to stress hormones can affect brain structures involved in cognition and mental health.

More recently, research has begun to shed light on what these changes look like at a cellular level. A 2023 review found that stressful experiences can alter hippocampal synaptic plasticity, neuronal activity, dendritic structure and neurogenesis – processes that are important for learning, memory and the brain’s ability to adapt. Not only are these integral functions of the brain impaired, but stress also appears to affect motivation.

Another review from 2015 describes evidence that stress can alter the neural circuits involved in motivated behavior and decision-making. Researchers noted that these effects were particularly relevant to psychiatric conditions such as depression, in which changes in motivation and reward processing can be prominent features.

It is clear from this research that prolonged stress is not simply something we feel. It can influence the biological systems that regulate mood, motivation, memory, and the brain’s capacity to adapt. Understanding how the brain responds to sustained stress is therefore an important part of understanding mental health. The new study approaches this question experimentally.

Researchers at the University of Coimbra exposed adult mice to a three-week protocol of chronic unpredictable stress. The animals were divided into groups receiving either water, caffeinated coffee or decaffeinated coffee. They then assessed a range of behaviors relating to anxiety, motivation, anhedonia and memory, alongside measures of brain function, generating striking results.

Stress produced several behavioral changes in the mice: they showed increased anxiety-like behavior, reduced motivation, greater anhedonia and impaired spatial memory. These changes were substantially attenuated in the mice consuming caffeinated coffee. Decaffeinated coffee, however, did not produce the same protective effect.

Furthermore, the significantly contrasting behaviors observed in the caffeinated and decaffeinated groups of mice were not the only stark differences the researchers found. Two biological markers relevant to brain function were measured: long-term potentiation (LTP) and brain-derived neurotrophic factor (BDNF). LTP is a process through which communication between neurons is strengthened, a fundamental mechanism underlying learning and memory. BDNF is a protein involved in neuronal health and synaptic plasticity, as well as mood regulation. These biological results mirrored the behavioral effects too – both measures were significantly reduced in the stressed mice; however, caffeine consumption prevented this reduction.

Together, the combination of biological and behavioral results suggests that caffeine may influence how the brain responds and adapts to prolonged stress – not simply how alert or stimulated an individual feels immediately after consumption.

At first glance, these findings may appear contradictory: a substance that increases feelings of anxiety could also potentially protect the brain against the consequences of chronic stress. This suggests that caffeine’s effects may depend, at least in part, on the context in which they are studied. The physiological arousal associated with acute caffeine consumption and the brain’s longer-term response to repeated stress involve different biological processes, meaning that an effect that increases anxiety in the short term does not necessarily preclude a protective effect under chronic stress.

The researchers suggest that adenosine A2A receptors may be involved. Previous work has implicated these receptors in the behavioral and synaptic effects of chronic stress, while earlier studies from the same research group found that caffeine and selective A2A receptor blockade could prevent some stress-induced changes in mood and memory in mice.

The new findings therefore add to an emerging line of research suggesting that caffeine may influence the brain’s response to chronic stress. Crucially, the protective effects were not seen in mice consuming decaffeinated coffee, suggesting that caffeine itself – not simply coffee consumption – may be responsible for the observed effects. This is an important methodological strength of the study: by directly comparing caffeinated and decaffeinated coffee, researchers were able to more specifically isolate caffeine as the variable responsible for the differences between groups, rather than simply comparing coffee consumption with a fundamentally different control such as water.

This makes the findings particularly interesting: the study is not simply showing that mice given coffee responded differently to chronic stress. It suggests that the caffeine component of coffee may be responsible for the effects.

But this certainly does not mean we should all start drinking four coffees a day.

The study was conducted in mice, not humans. This is an important limitation to highlight, but it does not make the findings irrelevant. Animal research has played an integral role in the development of psychology and neuroscience, particularly in helping researchers understand the biological mechanisms underlying learning, memory, motivation, stress and behavior. Many of the processes investigated in this study – including synaptic plasticity, neurogenesis and the effects of chronic stress on the brain – are precisely the kind of mechanisms that can be studied experimentally in animals in ways that would be difficult or impossible to study directly in humans.

Animal studies are particularly valuable because researchers can control factors such as stress exposure, caffeine intake and timing, while also examining changes in the brain at a cellular and molecular level. In this study, for example, the researchers were able to examine behavioral changes alongside measures such as LTP and BDNF, providing evidence regarding potential biological mechanisms rather than simply observing whether participants reported feeling more or less anxious.

The limitation is not that mice cannot tell us anything about human psychology. It is that findings in mice do not automatically translate into effects in humans. Mice and humans share many fundamental biological processes, but their brains, behavior, and responses to stress are not identical. A laboratory model of chronic unpredictable stress also cannot reproduce the complexity of human psychological stress, which is shaped by factors such as social relationships, environment, cognition and individual experience.

The study should therefore be viewed as an important piece of preclinical evidence, and we cannot conclude from these findings that drinking caffeinated coffee will prevent depression, reduce anxiety, or protect cognitive function during periods of chronic stress. Nor should we interpret the study as evidence that people experiencing anxiety should increase their caffeine intake.

What this study does provide is something more subtle – and arguably more scientifically interesting. It suggests that the relationship between caffeine and mental health is far more complex than “caffeine causes anxiety”, and it could be argued that it challenges this idea altogether.

Nutrition research is often communicated in deceptively simple terms. A food is described as “good” or “bad”. A nutrient is labelled as beneficial or harmful. A particular dietary habit is presented as either something we should adopt or something we should avoid. We have all been influenced at some point by this black and white food narrative, but biology rarely follows such rigidity.

The effects of a substance depend on dose, timing, frequency, individual physiology and the biological environment in which it is consumed. Caffeine is a particularly good example. For one person, a large amount of caffeine late in the day may interfere with sleep and amplify feelings of anxiety. For another, a moderate morning intake may improve alertness without producing noticeable adverse effects.

The broader evidence on coffee points in a similar direction. Coffee is not simply a vehicle for caffeine, but a more complex mixture of hundreds of compounds, and its effects appear to vary depending on the outcome being studied. Some research, like that of Harvard’s Chair of the Department of Nutrition, Frank Hu, links moderate coffee consumption with lower risks of several chronic diseases, while also recognizing that caffeine can cause anxiety, restlessness and sleep disruption in susceptible individuals. This is why the question of whether coffee is “good” or “bad” is probably the wrong one.

“For people who are already drinking coffee, I think the data is very reassuring that two to five cups per day is not just safe, it may be healthful in terms of lower risk of certain chronic diseases.”

Frank Hu, Chair of the Department of Nutrition at Harvard University, discussing his research surrounding the effects of coffee and caffeine

And, as the new research suggests, there may be longer-term effects that cannot be captured simply by asking whether someone feels more anxious immediately after their morning coffee fix. The study therefore does not overturn the existing evidence on caffeine and anxiety; instead, it adds another layer to it.

Caffeine may increase anxiety under some circumstances while also influencing brain’s ability to adapt to chronic stress. Both observations can be scientifically plausible at the same time. The important question is no longer simply whether caffeine is “good” or “bad” for mental health. It is how caffeine interacts with the brain, stress, and individual physiology – and whether those interactions differ between short-term stimulation and long-term adaptation.

There is a much more complicated story than “caffeine makes you anxious”. It is also a more interesting one.