{"id":"d57292bf-21c0-43c4-9616-be8843dc0cf6","arxiv_id":"2502.01939","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":1,"one_line_summary":"In a non-randomized pilot, EEG power spectral density fell when a hand was held near the scalp, but the effect is confounded and the 'human field' explanation is unsupported.","lead":"This pilot study reports that holding an experimenter's hand near a person's scalp decreases the power of their EEG signal. The authors interpret this as evidence that 'human fields' can modulate brain activity, but the design lacks a sham control and measured no field.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim that a human field caused the observed PSD decrease is unsupported because the design provides no sham control and no between-condition comparison; hand proximity is confounded with time, order, and sensory cues.","rationale":"The reader's verdict is appropriate: the paper reports an interesting pilot observation, but the causal inference to a 'human field' is not supported by the experimental design. The reader's weakest_assumption focuses on sensory cues; I agree that sensory cues are a serious confound, but the structural problem is broader. The control condition is not a valid counterfactual because it was run at a different time, on a partially different sample, and without any analogue of the hand-approach procedure. This makes the intervention-versus-control contrast confounded by order, practice, and non-specific attention, not only by unblinded sensory cues. The paper is honest about several limitations, including the inability to control subjects' thoughts and the post hoc outlier exclusions, and the authors do not claim to have measured any field directly. However, the load-bearing step—that the hand's proximity, rather than the many other differences between conditions, caused the PSD change—is not tested. A sham-controlled replication would directly address this gap, and the deposited data at least permit reanalysis of the original recordings. Given the high correctness risk in the causal claim, rejecting the manuscript as a demonstration of a human-field effect is the right call.","tokens_in":4609,"tokens_out":3187,"duration_ms":38649,"concrete_test":"Run a blinded, randomized, counterbalanced replication in which each participant experiences three conditions in random order while wearing an opaque mask and with white noise masking any sound: (1) the experimenter's hand held 5 cm above Cz, (2) an inert object of similar size and temperature moved and held in the same position, and (3) no proximal object. Compare PSD_pre to PSD_last using a pre-specified mixed-effects model with fixed effects for condition and order. If the hand condition does not differ significantly from the inert-object or no-object conditions, the original result is explained by sensory cues or time effects, not a human field.","verdict_should_be":"REJECT","load_bearing_attack":"The paper's central inference from Section 3 is that the intervention condition shows a statistically significant within-subject decrease in EEG PSD when the experimenter's hand is held near Cz. To support the 'human field' claim, the authors must show that this decrease is attributable specifically to the hand's proximity rather than to any other aspect of the intervention. That step fails. The control protocol described in Section 2 involved no analogous hand movement or object near the head, the intervention and control runs were not randomized in order (intervention first, control two weeks later), and some participants were in both protocols while others were in only one. Thus the reported p-value of 0.004353 is a within-group time trend, not a comparison against a matched sham condition. No between-group test of PSD_pre minus PSD_last is reported, so the apparent contrast between intervention and control cannot be evaluated statistically. In addition, Section 4 reports that several records were excluded post hoc as outliers without pre-specified criteria, which can inflate significance if exclusion is correlated with outcome. The absence of any sham condition means that air movement, sound, thermal cues, or the participant's awareness of a person approaching could account for the EEG change just as plausibly as an unmeasured 'field.' Because the causal attribution rests on this uncontrolled contrast, the central claim exceeds what the data can support.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This pilot study tests whether holding an experimenter's hand near the vertex of a participant's head without physical contact changes EEG power spectral density. Thirty participants were assigned to an intervention protocol and twenty to a control protocol; PSD was computed from the first 30 seconds of a baseline period and the last 30 seconds of a 60-second experimental period. The authors report a significant within-group decrease in the intervention group (exact Wilcoxon signed-rank test, p = 0.004353) and no significant decrease in the control group (p = 0.7667), which they interpret as evidence that a 'human field' associated with the hand modulates brain waves. They also discuss post hoc exclusion of outliers and acknowledge limitations including the narrow student sample and uncontrollable thoughts and feelings of participants.","tokens_in":4888,"tokens_out":6588,"duration_ms":69421,"significance":"A genuine causal effect of hand proximity on EEG would be surprising and potentially relevant to bioelectromagnetic interactions with the brain, so the question is worth asking. The manuscript has some virtues: it uses an exact nonparametric test, makes raw data available in a repository, and explicitly acknowledges several limitations. However, the central causal claim is not identifiable from the reported design: there is no sham control, no between-group statistical comparison, the intervention and control conditions are confounded with time and order, and outliers were excluded without pre-specified criteria. The paper is therefore best read as an exploratory pilot, not as evidence for a human-field effect on brain waves.","major_comments":[{"comment":"The design does not identify the intervention as the cause of the observed EEG change. The intervention condition includes the entire event of the experimenter moving a hand toward the participant's head and holding it there, while the control condition includes no analogous motor act, object, or sensory event. Air movement, faint sound, thermal cues, or peripheral awareness of a person could all plausibly change EEG, and no sham condition was used to control for these sensory and attentional features. The causal attribution to a 'human field' is therefore unsupported.","section":"Section 2, Protocol Description"},{"comment":"The reported statistics are two separate within-group Wilcoxon signed-rank tests: one comparing PSD_pre and PSD_last in the intervention group (p = 0.004353) and one in the control group (p = 0.7667). No direct between-group test is reported, such as a comparison of the within-subject changes (PSD_pre minus PSD_last) between the intervention and control groups, or an interaction test in a mixed model. Consequently, the conclusion that the intervention differs from control is not statistically established.","section":"Section 3, Results"},{"comment":"Condition is confounded with time and order: the intervention protocol was always performed first and the control protocol two weeks later, and only some participants took part in both protocols. Any time-dependent trend, increased familiarity with the laboratory, or change in alertness or mood could produce the reported pattern. In addition, because some participants contributed data to both groups, the two group-level tests are not independent, further complicating any informal comparison between them.","section":"Section 2, Protocol Description; Section 4, Discussion"},{"comment":"Several records were excluded after data inspection without a pre-specified criterion, as stated in the sentence 'several records had to be excluded because they were classified as outliers, with values far beyond the expected range.' This is outcome-dependent exclusion and can inflate the reported significance. The manuscript should state the exact number of excluded records, the quantitative rule used to define an outlier, and the results of an analysis that includes all data or a sensitivity analysis under different exclusion rules.","section":"Section 4, Discussion"},{"comment":"The sample-size description is internally inconsistent: the text says 'Thirty healthy students ... were recruited' but then gives 'twenty for the control group and thirty for the experimental group,' which implies fifty participants. The number actually analyzed in each group after exclusions is also not reported. The final N per group must be stated for the statistical results to be interpretable.","section":"Section 2, Methodology"}],"minor_comments":[{"comment":"The Wilcoxon signed-rank test does not directly compare medians; it tests a null hypothesis about the distribution of paired differences, often summarized by the pseudomedian. Please either report the Hodges-Lehmann estimate of the median difference or rephrase the claim as 'the distribution of PSD_pre is stochastically greater than PSD_last' rather than 'the median of PSD_pre is greater than the median of PSD_last.'","section":"Abstract and Results"},{"comment":"The manuscript does not report the EEG sampling rate, filter settings, FFT window parameters, or the frequency range over which the raw-signal PSD was computed. These details are needed for reproducibility.","section":"Data Acquisition and Processing"},{"comment":"Only p-values are reported; please add effect sizes and confidence intervals for the paired differences, and clarify whether the one-sided alternative was pre-specified or selected after observing the direction of the change.","section":"Results and Discussion"},{"comment":"There are several typographical errors: 'unknow laws' should be 'unknown laws,' 'Aproveed' should be 'Approved,' and the phrase 'inclusive of all genders' is awkward and should be replaced by a precise demographic description.","section":"Introduction and Section 2"},{"comment":"The caption describes PSD curves but calls them 'EEG readings'; please clarify that these are power spectral density curves, not raw EEG traces, and label the axes with units and a description of what the plotted quantities represent.","section":"Figure 2 caption"},{"comment":"The speculation about drowsiness in the control group is clearly labeled as speculation, which is appropriate, but it also illustrates that the control condition is not a matched comparison for the intervention; this should be acknowledged more explicitly when interpreting the control group's PSD increase.","section":"Section 4, Discussion"}],"recommendation":"reject","confidential_remarks":"The authors have made their data available and used an appropriate nonparametric test for the within-group comparison, which is commendable. However, the gap between the causal language in the abstract and what the design can actually support is too large to close with a reanalysis of the existing data: the missing sham condition, the lack of a between-group test, and the outcome-dependent outlier exclusion require new data collection under a different protocol. I would not invite a revision of the current manuscript, although a future properly controlled, randomized, and blinded study could build on this pilot's raw data and procedural notes."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe one thing you should know: this is a small pilot with a suggestive but confounded result, and the authors overclaim what it shows. The new bit is the specific observation that, when an experimenter held a palm near Cz without contact, EEG power spectral density dropped relative to baseline in that same session (p=0.0043). That is an empirical data point, and they've posted the data online, which is good practice.\n\nWhat the paper does well: it is transparent about its own limitations, considers drowsiness as an alternative explanation for the control group's PSD increase, and uses an exact test rather than an asymptotic approximation. The writing is clear and the protocol description is detailed enough to see the flaws.\n\nThe soft spots are the load-bearing kind. The control condition involved no hand movement, no sham object, and no matched sensory cue, so the intervention is confounded with time, order, and the participant's awareness that someone is doing something near their head. The intervention was always first and the control came two weeks later, so any time trend or order effect could produce the same pattern. The authors only report within-group Wilcoxon tests; they never test whether the change in PSD differs between intervention and control. A p-value of 0.0043 for a within-group decrease does not establish that the hand caused the decrease. In addition, records were excluded as outliers after the fact without pre-specified criteria, which can inflate significance if exclusion tracks outcome. And no 'human field' is measured, so the causal attribution is assumed rather than tested.\n\nI don't think the paper is incoherent or dishonest. The authors acknowledge several limitations and speculate cautiously. But the central claim—that a human field caused the PSD change—is not entailed by the data. It could just as easily be sensory cueing, relaxation, or order effects. The result is a pilot observation, not a demonstrated effect.\n\nWho is this for? Someone compiling examples of study-design pitfalls in biofield research, or a methods class on confounding. It is not a source to cite for the reality of human field effects. If this were submitted to a journal, I'd want it sent back for a redesign: sham hand, counterbalanced order, a direct between-condition comparison, and a pre-registered analysis. In its current form, I wouldn't send it to peer review; the design flaw is too fundamental to salvage by revision. But the authors' candor and data sharing make it a useful case to keep in mind.","headline":"A suggestive but confounded pilot: the EEG change is real, but the 'human field' interpretation is unsupported by a design that lacks any sham or between-group comparison.","tokens_in":5353,"tokens_out":3023,"would_cite":false,"duration_ms":30720,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["87.18.Sn","87.85.D-","87.85.Ng"],"model":"deepseek-v4-flash","headline":"Holding a palm near the top of the head, without contact or the participant's knowledge, was followed by a significant drop in EEG power; the no-intervention control showed no drop.","keywords":["EEG","power spectral density","human field","Wilcoxon signed-rank test","Cz electrode","non-contact intervention","brain waves","pilot study"],"falsifier":"Run the identical protocol with a sham object—say, an insulated plastic rod moved to the same spot above the head and held there for 60 seconds—with participants unable to tell whether it is a hand or a rod. If the same drop in brain-wave power appears with the rod, the effect is not specific to a human field; if it does not, the field interpretation is supported. A distance test would also settle it: if the effect fades smoothly as the hand is moved further away, that pattern fits a field; if it disappears once the hand cannot be seen or heard, a sensory cue is implicated.","tokens_in":4416,"feed_emoji":"🧠","tokens_out":8366,"duration_ms":78148,"temperature":0.7,"pith_summary":"This pilot study asks whether a field emitted by the human body can alter the brain's electrical activity, the way transcranial magnetic stimulation does. Thirty healthy participants sat with eyes closed while an experimenter held a palm motionless just above the top of the head, at the scalp location called Cz, for 60 seconds without touching the participant. The paper compares EEG power spectral density in the 30 seconds before the hand approached with the final 30 seconds while the hand was held there. In the intervention group the before value was significantly higher than the during value (exact Wilcoxon signed-rank test, p = 0.004353), while a no-intervention control group showed no such decrease. The authors take this as preliminary evidence that close, non-contact proximity of a human hand is associated with a drop in overall EEG power.","feed_headline":"Palm held near head lowers EEG power in pilot test","feed_subtitle":"In 30 subjects, EEG power fell when a hand hovered above the scalp; controls showed no drop.","key_machinery":"The argument runs on paired comparisons of EEG power spectral density (PSD), a summary of how much electrical power the raw EEG signal carries across frequencies, computed at the single scalp location Cz. PSD is calculated for the first 30 seconds of baseline and the final 30 seconds of the 60-second condition, deliberately from the raw signal rather than from separate frequency bands so that a change in any band registers. The statistical machinery is the exact Wilcoxon signed-rank test, chosen because the Shapiro-Wilk test showed the PSD values are not normally distributed; it tests whether the median of the paired differences favors the baseline over the final segment. The proposed mechanism—the 'human field'—is the paper's name for an unmeasured field with amplitude, frequency, and phase that could affect neurons through resonance or an unknown interaction.","core_discovery":"The central claim is that a human hand held near the scalp changes the brain's measured electrical state. In the intervention group, the median power spectral density of the raw EEG at Cz over the final 30 seconds of the hand-held-near-head condition was lower than the median PSD over the 30 seconds before the hand approached, with a p-value of 0.004353 under the exact Wilcoxon signed-rank test. In the control group, the before/after comparison gave p = 0.7667, meaning the median PSD did not decrease without the intervention; if anything it increased, which the authors attribute to drowsiness during long eyes-closed rest. The authors therefore conclude that the hand's proximity, not the passage of time or relaxation alone, accompanied the PSD reduction, and they frame the result as consistent with an unknown human field interacting with neuronal activity.","pith_inferences":["The paper does not randomize the order of intervention and control, and some participants took part in both; a cross-over design with a sham hand-shaped object would separate a genuine field effect from expectation, fatigue, or order effects.","A distance-dependence study (hand at a few centimeters, 10 cm, 30 cm) would discriminate a field-like effect, which should fade with distance, from a cue-based effect, which would disappear once the participant cannot see or hear the hand.","The authors' interpretation invokes human magnetic fields, but no magnetic measurement was made during the protocol; co-recording with a magnetometer would test whether any field fluctuation correlates with the EEG change."],"forward_implications":["If the effect is real, EEG experiments that involve a researcher leaning near the subject should control for or report hand position, since proximity alone may shift measured power.","A non-contact, subject-blinded method for reducing EEG power could be developed into a relaxation or sleep-induction aid, pending replication.","Because the paper collapses all frequency bands into one PSD value, a replication that decomposes the change by band would show whether the drop is a broadband suppression or a specific rhythm such as alpha being reduced.","The result would strengthen the case for short-range, non-electromagnetic interactions from living tissue, and would motivate instruments that directly measure fields near the hand during the EEG change."],"supporting_citations":[{"why":"Supplies the physiological basis of EEG: synchronized postsynaptic potentials of cortical pyramidal neurons generate the scalp signal whose power is analyzed.","marker":"[1]"},{"why":"Shows that externally applied electrical or magnetic stimulation near the brain can modify neural activity, the analogy that motivates testing a human field.","marker":"[2, 3]"},{"why":"Establishes that the human body generates measurable magnetic fields outside the body, grounding the claim that a human field could exist.","marker":"[4]"},{"why":"Shows that brain alpha-rhythm currents produce magnetic fields detectable outside the head, further grounding the human-field premise.","marker":"[5]"},{"why":"Defines the 10-20 electrode placement system and the Cz central scalp location where the EEG was recorded.","marker":"[9]"},{"why":"Provides evidence that increased PSD in the delta band accompanies excessive daytime sleepiness, used to explain why the control group's PSD rose during eyes-closed rest.","marker":"[10]"}],"fun_headline_variants":["Pilot: hand near head lowers EEG power","EEG power drops when palm nears skull: pilot","Human hand proximity reduces brain wave power","Small trial: palm over scalp cuts EEG power"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim that a 'human field' caused the measured decrease in brain-wave power assumes that the hand's proximity was the only meaningful difference between the intervention and control conditions. If subjects sensed the hand through air movement, sound, or peripheral vision, that sensory cue, not any field, could explain the change.","fun_headline_variants_meta":{"raw":{"variants":["Pilot: hand near head lowers EEG power","EEG power drops when palm nears skull: pilot","Human hand proximity reduces brain wave power","Small trial: palm over scalp cuts EEG power"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000239,"raw_usage":{"total_tokens":1545,"prompt_tokens":1003,"completion_tokens":542,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":619,"completion_tokens_details":{"reasoning_tokens":483}},"tokens_in":619,"tokens_out":542,"duration_ms":5034,"temperature":1.0,"reasoning_tokens":483,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T13:54:58.028908+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the identical protocol with a sham object—say, an insulated plastic rod moved to the same spot above the head and held there for 60 seconds—with participants unable to tell whether it is a hand or a rod. If the same drop in brain-wave power appears with the rod, the effect is not specific to a human field; if it does not, the field interpretation is supported. A distance test would also settle it: if the effect fades smoothly as the hand is moved further away, that pattern fits a field; if it disappears once the hand cannot be seen or heard, a sensory cue is implicated.","supporting_citations":[{"cited_title":"Bioelectrical Signal Processing in Cardiac and Neurological Applications","cited_arxiv_id":null,"evidence_quote":"Supplies the physiological basis of EEG: synchronized postsynaptic potentials of cortical pyramidal neurons generate the scalp signal whose power is analyzed."},{"cited_title":"Magnetic fields around the torso: Production by electrical activity of the human heart","cited_arxiv_id":null,"evidence_quote":"Establishes that the human body generates measurable magnetic fields outside the body, grounding the claim that a human field could exist."},{"cited_title":"Magnetoencephalography: Evidence of magnetic fields produced by alpha- rhythm currents","cited_arxiv_id":null,"evidence_quote":"Shows that brain alpha-rhythm currents produce magnetic fields detectable outside the head, further grounding the human-field premise."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the 10-20 electrode placement system and the Cz central scalp location where the EEG was recorded."},{"cited_title":"Excessive daytime sleepiness is associated with relative delta frequency power among patients with mild osa","cited_arxiv_id":null,"evidence_quote":"Provides evidence that increased PSD in the delta band accompanies excessive daytime sleepiness, used to explain why the control group's PSD rose during eyes-closed rest."}],"review_version":1}