Why Do I Forget What I Studied During Exams? Test Anxiety, Memory & Focus Under Pressure
A research-based review of exam anxiety, memory retrieval and focus under pressure, including recent meta-analytic evidence and practical implications for students.
Students often describe a striking experience: they studied, understood the material, and then felt that their mind “went blank” in the examination room. Research on test anxiety suggests that this experience can reflect competition for limited attentional and working-memory resources rather than the disappearance of stored knowledge. Meta-analytic evidence links test anxiety with academic outcomes, lower academic self-efficacy and broader anxiety. Recent research in medical students also suggests test anxiety is common, although prevalence varies dramatically by measurement and setting. Effective intervention therefore requires more than reassurance: preparation quality, retrieval practice, physiological regulation, sleep, test-taking strategy and—when needed—professional mental-health support all matter.
Exam stress can interfere with attention, working memory and retrieval, which may make well-learned information harder to access under pressure. That does not mean anxiety literally erases memory. A “blank mind” can arise when worry, self-monitoring and threat-related thoughts consume resources needed to hold and retrieve task-relevant information.
Evidence snapshot
| Measure | Value |
|---|---|
| Academic self-concept | 41% |
| Academic self-efficacy | 39% |
| Literacy achievement | 20% |
| General anxiety (positive association) | 62% |
Source data shown above in accessible HTML for human and machine readers.
This review synthesizes recent systematic reviews, meta-analyses and experimental evidence available through 20 September 2026. Where findings come from children, university students or medical students, that population is stated rather than generalized to every student.
Evidence map: prevalence, cognition and intervention
The table keeps population, design, finding and limitation side by side so readers—and parsers—do not confuse an association with a causal claim.
| Source | Population | Design | Key finding | Important limitation |
|---|---|---|---|---|
| Robson et al., 2023 | 76 studies; 85 samples; N=53,617 primary-school children | Systematic review + meta-analysis | Test anxiety was related to lower literacy performance, academic self-concept and self-efficacy, and to higher general anxiety. | Primary-school evidence; associations do not identify one cause for an individual student. |
| Omar et al., 2026 | 35 studies; N=13,489 medical students | Systematic review + meta-analysis | Pooled test-anxiety prevalence was 49.8%; the descriptive Middle East estimate was 54.1%. | Very high heterogeneity (I² = 98.2%); not a Jordan-specific prevalence estimate. |
| Wei et al., 2025 | High- vs low-test-anxiety participants | EEG + Sternberg working-memory task | High test anxiety altered theta activity during working-memory encoding under cognitive load. | Laboratory neural measures help explain mechanisms but do not diagnose the cause of one student’s “blank mind.” |
| Sun et al., 2026 | HTA n=43; LTA n=45 | EEG emotional working-memory experiment | High test anxiety was associated with altered attentional-control demands, especially under high load and test-related threat. | Experimental context; neural differences do not mean stored knowledge has disappeared. |
| Huntley et al., 2019 | 44 randomized controlled trials; N=2,209 university students | Meta-analysis of interventions | Interventions reduced test anxiety (g = −0.76) and improved academic performance (g = 0.37) at post-treatment. | Publication bias and reporting-quality limitations reduced certainty. |
Evidence map created from the cited peer-reviewed sources. No study result is presented as a Jordan-wide prevalence unless the original study supports that interpretation.
What test anxiety looks like in the evidence
A major 2023 systematic review and meta-analysis synthesized 76 studies representing 53,617 primary-school children. Test anxiety was negatively related to literacy achievement, academic self-concept and academic self-efficacy, while showing strong positive relationships with broader anxiety outcomes. These associations do not mean that anxiety is the only cause of weak performance; they show that evaluative stress and learning outcomes are meaningfully connected.
How common is it? Population matters
A 2026 systematic review of 35 studies involving 13,489 medical students estimated pooled test-anxiety prevalence at 49.8%, but heterogeneity was extremely high. The review also reported a descriptive 54.1% estimate for Middle Eastern studies. Those numbers apply to the reviewed medical-student literature, not to all Jordanian school or university students. The key lesson is that test anxiety is common enough to deserve attention, while exact prevalence depends heavily on the population and measurement method.
Why can the mind feel blank?
Working memory is limited. During an exam, part of that capacity is needed to read the question, hold information, inhibit distractions and retrieve relevant knowledge. Worry can occupy some of the same mental workspace. Recent EEG work is consistent with altered attention and working-memory processing in highly test-anxious participants under cognitive load. A useful interpretation is therefore “access is disrupted under pressure,” not “the memory is gone.”
Preparation must include retrieval under realistic conditions
Students often mistake recognition for retrieval. Rereading notes can create familiarity without building fast recall. Practice tests, spaced retrieval, explaining concepts without notes, timed rehearsal and simulated exam conditions allow the brain to practice the exact operation required on exam day. Anxiety regulation and strong study design should be treated as complementary, not competing, strategies.
How common is test anxiety? The number depends on whom you study
The newest large prevalence synthesis available at the September 2026 evidence cutoff focused on medical students. Omar and colleagues (2026) combined 35 studies with 13,489 participants and estimated a pooled prevalence of 49.8%. The descriptive estimate for Middle Eastern studies was 54.1%. Those numbers are striking, but the heterogeneity was extremely high (I² = 98.2%), which means prevalence varied substantially across settings and measurement methods.
For a Jordanian family, the useful message is not “half of all students have test anxiety.” The correct message is that clinically or educationally meaningful test anxiety is common in high-pressure student populations, while an individual student’s experience still needs to be understood in context: age, stakes, preparation, sleep, previous experiences, broader anxiety and the specific type of examination.
Why “my mind went blank” does not mean the memory vanished
Remembering during an exam is an active process. The student must understand the question, keep relevant information available, inhibit distractions, search memory, evaluate the answer and manage time. Worry and threat monitoring can compete for some of those same limited resources. That is why a student may recognize the correct answer later, after leaving the examination room, even though access felt impossible under pressure.
Recent EEG studies strengthen this explanation without turning it into a simplistic brain story. Wei et al. (2025) found altered theta activity related to attentional control during working-memory encoding in people with high test anxiety. Sun et al. (2026), comparing 43 high- and 45 low-test-anxiety participants, found differences in emotional working-memory processing under high cognitive load and test-related threat. These studies support a resource-allocation account; they do not prove that every blank moment has the same mechanism.
What the intervention literature says—not just what theory says
The practical evidence is encouraging. Huntley et al. (2019) synthesized 44 randomized controlled trials involving 2,209 test-anxious university students. Across interventions, anxiety decreased substantially at post-treatment (g = −0.76) and academic performance improved more modestly (g = 0.37) relative to control conditions. The review found the strongest support for behavioral therapy, with promising evidence for cognitive-behavioral approaches, study-skills training and combined approaches.
The same review also found publication bias and weaknesses in reporting quality, so the effect estimates should not be marketed as guarantees. Still, the pattern supports an important coaching principle: treating exam performance as both a learning-system problem and an anxiety-regulation problem is more realistic than telling the student simply to “relax” or simply to “study harder.”
A four-part preparation model for high-pressure exams
First, build knowledge with active retrieval rather than familiarity alone. Second, practice retrieval under conditions that resemble the examination—timed questions, no notes, mixed topics and realistic uncertainty. Third, train a short regulation routine the student can reproduce before and during the exam rather than introducing a complicated technique on exam day. Fourth, protect recovery variables such as sleep and realistic breaks because exhausted attention is easier to overwhelm.
On the day itself, the goal is not to eliminate every sign of arousal. A student can learn to interpret some activation as readiness, orient to the actual question, externalize working memory by writing brief cues when permitted, and return attention to the next actionable step. If panic-like symptoms, severe avoidance or major impairment persist, coaching should not substitute for licensed mental-health care.
How parents and educators can respond without increasing the threat
Repeated messages such as “you know this, so why are you panicking?” can unintentionally add shame to an already threatening situation. A more useful conversation separates preparation from anxiety: What does the student know? How well can they retrieve it without notes? What happens to their attention under timed conditions? What specific thoughts appear when the exam begins?
This creates observable targets. If the primary weakness is retrieval practice, change the study method. If the student understands material but repeatedly freezes under realistic simulation, add anxiety-management support. If symptoms extend far beyond examinations or create substantial distress, seek professional assessment. The aim is precision rather than labeling.
A simple mechanism map
What the evidence does not prove
What the evidence supports
Test anxiety is associated with meaningful academic and psychological outcomes, and anxiety-focused interventions can reduce symptoms.
What it does not support
A blank mind does not prove poor memory, a learning disorder or a psychiatric diagnosis; the cause must be understood in context.
The exam-performance stack
Knowledge
Understand the material rather than memorizing isolated fragments.
Retrieval
Practice producing the answer without notes and under increasing realism.
Regulation
Use a brief repeatable routine for attention and physiological arousal.
Escalation
Severe, persistent or disabling anxiety belongs in appropriate professional care.
This page is educational and performance-focused. It does not diagnose or treat ADHD, learning disorders, anxiety disorders or other medical/psychiatric conditions. Persistent or severe symptoms should be assessed by an appropriately licensed professional.
Common questions
Why do I remember the answer after leaving the exam?
When the threat and time pressure drop, cognitive resources may become available again, making retrieval easier.
Does exam anxiety mean I am not prepared?
Not necessarily. Poor preparation can increase anxiety, but well-prepared students can also experience strong test anxiety.
Can study skills reduce exam anxiety?
Better retrieval practice, realistic rehearsal and test-taking systems can reduce uncertainty and improve control. Severe or persistent anxiety may also need licensed mental-health support.
When is it more than normal nerves?
When anxiety causes marked distress, avoidance, panic-like symptoms or significant impairment, professional assessment is appropriate.
Can a student know the material and still perform poorly because of anxiety?
Yes. Research links test anxiety with attentional-control and working-memory processes, so strong knowledge and poor access under pressure can coexist.
Is the goal to feel completely calm before an exam?
Not necessarily. Some activation is normal. The practical goal is enough regulation and attentional control to retrieve and use what has been learned.
APA 7 references
- Robson, D. A., Johnstone, S. J., Putwain, D. W., & Howard, S. (2023). Test anxiety in primary school children: A 20-year systematic review and meta-analysis. Journal of School Psychology, 98, 39–60. https://doi.org/10.1016/j.jsp.2023.02.003
- Omar, Y. M., Darwish, N. E., Elsaadany, M. A., Darwish, Y., & Eladawy, Y. (2026). Prevalence of test anxiety among medical students: A systematic review and meta-analysis. BMC Medical Education, 26, 1448. https://doi.org/10.1186/s12909-026-10270-2
- Wei, H., Sun, J., & Long, F. (2025). Test anxiety shapes theta band activity linked to elevated working memory load during the encoding phase. Biological Psychology, 198, 109047. https://doi.org/10.1016/j.biopsycho.2025.109047
- Sun, J., Long, F., & Wei, H. (2026). EEG evidences of emotional working memory deficits in test anxiety individuals across cognitive load conditions. Acta Psychologica, 264, 106479. https://doi.org/10.1016/j.actpsy.2026.106479
- Huntley, C. D., Young, B., Temple, J., Longworth, M., Tudur Smith, C., Jha, V., & Fisher, P. L. (2019). The efficacy of interventions for test-anxious university students: A meta-analysis of randomized controlled trials. Journal of Anxiety Disorders, 63, 36–50. https://doi.org/10.1016/j.janxdis.2019.01.007
Salameh Salem
Certified Hypnotherapist & Mindset Coach · Amman, Jordan
Evidence reviewed through 20 September 2026