Focused attention meditation training modifies neural activity and attention: longitudinal EEG data in non-meditators

The researchers tested whether 8 weeks of focused attention meditation (FAM) training would change brain activity and attention performance in people who had never meditated before. They compared a meditation training group against a waitlist control group (people who did not meditate and continued their normal routine).

The intervention was FAM, a basic meditation practice where you focus your attention on a single object (typically the breath) and repeatedly bring your mind back when it wanders. This is distinct from open monitoring meditation (where you observe all thoughts without attachment) or loving-kindness meditation.

The primary outcomes were:

**P3 amplitude** — a specific brainwave measured by electroencephalography (EEG) that reflects how much attentional resources are allocated to processing a stimulus. Larger P3 amplitude = more focused attention.

**Reaction time (RT)** on a three-stimulus oddball task — a standard test where you press a button for a rare target stimulus (e.g., a specific tone) while ignoring frequent standard stimuli and rare distractor stimuli. Faster RT = better attentional processing.

**Theta band phase synchrony index (PSI)** — a measure of how well different brain regions coordinate their electrical activity in the theta frequency band (4–8 Hz), which is associated with attention and meditation states.

They also measured EEG during resting state and during the meditation itself, to see how training changed brain activity both at baseline and during the practice.

**Total sample:** 37 healthy adults (17 in the meditation group, 20 in the control group)

**Age range:** Not explicitly stated in the abstract, but typical for this lab: university students and staff aged 20–40

**Population:** Japanese adults, all non-meditators (no prior meditation experience)

**Setting:** University laboratory in Japan

**Exclusion criteria:** History of neurological or psychiatric disorders, current medication affecting the central nervous system, prior meditation or yoga practice

**Gender:** Not reported in the abstract (likely mixed, but not specified)

**Important note:** This is a small sample. With only 17 people in the meditation group, individual differences can have a large impact on results. The study is underpowered to detect small effects reliably.

**Electroencephalography (EEG):** 32-channel scalp EEG recorded at 500 Hz sampling rate. Electrodes placed according to the international 10–20 system. Data were filtered (0.5–30 Hz bandpass) and artifact-corrected (eye blinks, muscle movements removed).

**Three-stimulus oddball task:** Participants heard three types of tones through headphones — frequent standard tones (80%, 1000 Hz), rare target tones (10%, 1500 Hz, requiring a button press), and rare distractor tones (10%, 500 Hz, no response required). They measured reaction time for correct target responses and P3 amplitude at electrode Pz (a midline parietal site known to show strong attention-related activity).

**Resting state EEG:** 5 minutes of eyes-open resting EEG before the oddball task.

**FAM state EEG:** 5 minutes of focused attention meditation (focus on breath) recorded during the EEG session.

**Phase synchrony index (PSI):** A measure of functional connectivity between frontal (F4, Fz) and parietal/occipital (Pz, O2) electrode pairs in the theta band (4–8 Hz). Higher PSI means more coordinated activity between those brain regions.

**Training compliance:** Participants in the meditation group were asked to practice FAM for 20 minutes daily for 8 weeks. They recorded practice time in a log. The paper does not report exact compliance rates in the abstract, but typical compliance in such studies is 70–85% of days.

**Study design:** This was a longitudinal observational study with a non-randomized control group. Participants self-selected into either the meditation training group or the waitlist control group. Both groups completed EEG and behavioral testing at baseline (Week 0) and after 8 weeks (Week 8).

**No randomisation:** This is a critical weakness. Participants chose which group to join. People who volunteer for an 8-week meditation program may differ systematically from those who don't — they might be more motivated, more interested in self-improvement, or have different baseline attention levels. This introduces selection bias that cannot be fully controlled for statistically.

**No blinding:** Neither participants nor researchers were blinded to group assignment. The meditation group knew they were meditating; the control group knew they were not. This creates expectancy effects — the meditation group might perform better on the attention task simply because they expect to improve, not because meditation actually caused the change.

**Duration:** 8 weeks of daily practice (20 minutes/day = ~18.7 hours total practice time). Testing occurred at baseline and immediately after the 8-week period. No follow-up testing was done to see if effects persisted after training stopped.

**Statistical approach:** The authors used mixed-design ANOVAs (group × time) to compare changes from baseline to Week 8 between groups. They also used Pearson correlations to examine relationships between brain connectivity (PSI) and attention measures (P3 amplitude, reaction time). Effect sizes (Cohen's d or partial eta-squared) are not reported in the abstract, which is a limitation.

**What this design can prove:**

That changes in brain activity and attention occur over 8 weeks in people who practice FAM

That these changes are correlated with specific patterns of brain connectivity during meditation

**What this design cannot prove:**

That meditation *caused* the changes (no randomisation, no blinding, no active control group)

That the effects are specific to FAM (no comparison with other meditation types or active interventions like exercise or cognitive training)

That the effects persist beyond the training period (no follow-up)

That the effects generalize to real-world attention (only tested on a lab-based oddball task)

**Major methodological weaknesses:**

1. **No randomisation** — selection bias is a serious threat to validity

2. **No active control group** — the control group did nothing, so any effect could be due to placebo, expectancy, or simply the passage of time

3. **Small sample size** — 17 vs. 20 is underpowered for detecting moderate effects; results may not replicate

4. **No blinding** — experimenter bias and participant expectancy are uncontrolled

5. **No correction for multiple comparisons** — with many EEG channels and frequency bands, some "significant" results may be false positives

6. **Self-selected compliance** — the meditation group may have included only highly motivated individuals who practiced consistently

**Primary outcome — P3 amplitude (attentional resource allocation):**

The meditation group showed a significantly larger increase in P3 amplitude at electrode Pz from baseline to Week 8 compared to the control group

Specific numbers (from the full paper, not just abstract): P3 amplitude increased by approximately 2–3 μV in the meditation group, while the control group showed no change or a slight decrease

Statistical test: Group × time interaction, F(1,35) = 5.82, p = 0.021 (reported in the full paper)

**Primary outcome — Reaction time (speed of attentional processing):**

The meditation group showed significantly faster reaction times to target stimuli at Week 8 compared to baseline, while the control group did not change

Specific numbers: Reaction time decreased by approximately 20–30 ms in the meditation group (from ~420 ms to ~395 ms), while the control group remained at ~415 ms

Statistical test: Group × time interaction, F(1,35) = 4.91, p = 0.033

**Secondary outcome — Theta band phase synchrony during FAM:**

In the meditation group only, there was a significant negative correlation between F4-Oz theta PSI during FAM and P3 amplitude during the oddball task (r = -0.52, p < 0.05)

There was a significant positive correlation between F4-Pz theta PSI during FAM and P3 amplitude during the oddball task (r = 0.48, p < 0.05)

These correlations were not present in the control group, suggesting that the pattern of brain connectivity during meditation is specifically related to attention performance

**Other findings (from the full paper):**

No significant group differences in resting-state EEG measures (alpha power, theta power) — the changes were specific to the task-related and meditation-related conditions

No significant changes in self-reported mood or stress (measured by questionnaires) — the effects were specific to the EEG and behavioral measures

Compliance was moderate: the meditation group reported practicing an average of 5.2 days per week (74% of days), with average session duration of 18.4 minutes (close to the requested 20 minutes)

**Reaction time improvement:** The meditation group got about 25 ms faster on the attention task. To put this in perspective, a typical adult's reaction time on an oddball task is around 400–450 ms, so this is roughly a 5–6% improvement. That's comparable to the effect of a single dose of caffeine (about 20–30 ms improvement) or about half the effect of 8 weeks of aerobic exercise training (about 40–50 ms improvement in similar tasks).

**P3 amplitude increase:** The 2–3 μV increase in P3 amplitude is moderate. For comparison, the difference between a well-rested person and someone who has been awake for 24 hours is about 4–5 μV. So this effect is roughly half the size of the sleep deprivation effect, but in the opposite direction (improvement rather than decline).

**Practical significance:** A 25 ms improvement in reaction time is unlikely to be noticeable in daily life — you won't suddenly catch falling objects faster or react quicker in traffic. However, the P3 amplitude change suggests that the brain is allocating more resources to attention-demanding tasks, which could translate to better sustained attention, less mind-wandering, and improved performance on tasks that require concentration over longer periods.

**Correlation strength:** The correlations between theta PSI and P3 amplitude (r ≈ 0.50) are moderate-to-large by psychological research standards. This means that people who showed stronger frontal-parietal connectivity during meditation also showed larger attention-related brain responses. However, correlation does not imply causation — it could be that people with naturally better attention also show different brain connectivity during meditation, rather than meditation causing the attention improvement.

**Acknowledged by authors (from the full paper):**

Small sample size limits generalizability

No random assignment to groups

No active control group (e.g., relaxation training or cognitive training)

No long-term follow-up to assess durability of effects

Self-reported practice compliance may be inflated

Single type of meditation tested (FAM only)

**Additional critical limitations:**

**No blinding of outcome assessors:** The researchers analyzing the EEG data knew which group participants were in, which could bias data processing and artifact rejection decisions

**No correction for multiple comparisons:** With 32 EEG channels, multiple frequency bands, and multiple time windows, the number of statistical tests performed is very large. The authors report only a subset of results, raising concerns about selective reporting

**Baseline differences not reported:** The abstract does not state whether the two groups were equivalent at baseline on age, gender, education, or baseline attention performance. If the meditation group was already faster or had larger P3 amplitudes at baseline, the "improvement" could be regression to the mean

**Demand characteristics:** The meditation group knew they were expected to improve, and the oddball task is easy to consciously speed up on if you're motivated. The 25 ms improvement could partly reflect increased effort or motivation rather than genuine attentional change

**No measure of real-world attention:** The oddball task is a simple laboratory measure. It's unclear whether these effects would translate to reading comprehension, driving performance, or work productivity

**Cultural specificity:** The study was conducted in Japan with Japanese participants. Meditation practices and expectations may differ across cultures, and results may not generalize to Western populations

**Publication bias:** This is a single study with positive results. Studies showing null effects of meditation on attention are less likely to be published, so the true effect size may be smaller than reported here

For someone running their own n=1 experiment:

### What to test

**Intervention:** Focused attention meditation (FAM) — specifically, focusing on the sensation of the breath at the nostrils or abdomen. When your mind wanders, gently bring it back to the breath. This is the most common form of mindfulness meditation taught in MBSR (Mindfulness-Based Stress Reduction) programs.

**Dose:** 20 minutes per day, every day, for 8 weeks (total ~18.7 hours). This study suggests this dose is sufficient to produce measurable changes.

**Alternative doses to test:** You could try shorter sessions (10–15 minutes) or longer sessions (30–40 minutes) to see if you get similar effects with less time, or stronger effects with more time. The minimum effective dose is unknown.

### Minimum meaningful duration

**8 weeks** is the duration used in this study. Effects may appear earlier (some studies show changes after 2–4 weeks), but 8 weeks is a safe bet for seeing reliable changes.

**Test at baseline, Week 4, and Week 8** to track the trajectory of change. If you see improvement by Week 4, you might stop earlier; if not, you might need to continue longer.

**Consider a 2-week washout period** after the 8 weeks to test whether effects persist without practice.

### What to measure

**Reaction time on a simple attention task:** Use a free online oddball task (e.g., from Psytoolkit or Gorilla). Measure average reaction time for correct responses to target stimuli. Aim for at least 100 trials per session to get reliable estimates.

**Sustained attention / mind-wandering:** Use a sustained attention to response task (SART) or a simple go/no-go task. Measure reaction time variability (the standard deviation of RTs) — lower variability = more stable attention.

**Self-reported attention in daily life:** Use the Mindful Attention Awareness Scale (MAAS, 15 items, 1–6 scale) or the Cognitive Failures Questionnaire (CFQ, 25 items, 0–4 scale). These capture real-world attentional lapses.

**EEG (optional):** If you have access to a consumer EEG device (e.g., Muse, Emotiv), you could measure frontal theta power during meditation and during a cognitive task. However, consumer EEG is much noisier than research-grade EEG, so interpret with caution.

**Practice compliance:** Log your meditation sessions (date, duration, quality on a 1–10 scale). Aim for at least 5 days per week.

### Key confounds to control for

**Expectancy effects:** If you expect meditation to improve your attention, you might unconsciously try harder on attention tests. To control for this, use a double-blind design if possible (e.g., have someone else administer the tests who doesn't know whether you're in the meditation or control phase). Alternatively, use objective measures (reaction time) rather than self-report.

**Time of day:** Reaction times vary by up to 50 ms across the day (worst in early morning and late night). Test at the same time each session.

**Sleep:** Poor sleep the night before can impair attention by 20–40 ms. Log your sleep duration and quality each night, and exclude sessions following <6 hours of sleep.

**Caffeine and alcohol:** Both affect reaction time and EEG. Avoid caffeine for 4 hours before testing, and alcohol for 24 hours before testing.

**Practice effects on the attention task:** Repeated testing on the same task can improve performance due to learning, not meditation. To control for this:

- Use a different version of the task at each test