In 2012 a neuroimaging study appeared in PLoS ONE that has been discussed in consciousness research ever since: Brazilian psychiatrist Julio Peres, medical professor Alexander Moreira-Almeida and US neuroscientist Andrew Newberg studied ten Brazilian psychography mediums using SPECT imaging — and encountered a finding that defies intuitive expectation: the more experienced the medium, the quieter the frontal lobe during trance; and the more complex the produced text.
What Is Psychography?
Psychography (from Greek psyche = soul, graphein = to write) is the spiritist form of automatic writing: a medium holds a pen and writes texts in a trance-like state, understood as communications from deceased spirits. In Brazil the practice is closely associated with Brazilian Spiritism after Allan Kardec and is best known through Chico Xavier (over 400 books) and Divaldo Franco (over 250 books). The precise nature of the phenomenon — whether unconscious dissociation, creative imagination, or an external source — remains scientifically open.
The Experiment
Peres and colleagues recruited ten Brazilian Spiritists with long-standing psychography practice and divided them into two groups:
- Experienced group: five mediums averaging 37 years of regular psychography practice
- Less experienced group: five mediums averaging 14 years of practice
All participants completed two tasks in randomised order:
- Normal condition: conscious, free writing of a self-composed text
- Trance condition: psychography — automatic writing in a self-induced trance state
Brain activity was measured via SPECT during both tasks. The produced texts were subsequently rated by independent assessors for linguistic complexity (vocabulary, sentence structure, information density) — without knowledge of which condition had generated them.
The SPECT Method
SPECT (Single Photon Emission Computed Tomography) measures regional cerebral blood flow: a radiolabelled tracer (technetium-99m-HMPAO) is injected intravenously and distributes through the brain in proportion to local neural activity. The result is a three-dimensional map of activity distribution. Compared with fMRI, SPECT has lower temporal resolution but is more robust against motion artefacts — an important advantage when participants are actively writing during measurement.
The Frontal Lobe Paradox
The surprise lay in the combination of two findings:
Brain activity: In the experienced mediums, activity in frontal lobe areas — particularly the left inferior frontal gyrus, the anterior cingulate cortex, and parts of the prefrontal cortex — was reduced during trance compared with normal writing. In the less experienced mediums, by contrast, activity in those same regions was increased during trance.
Text complexity: The trance-state texts of the experienced mediums received significantly higher complexity ratings from the independent assessors than their normal texts. In the less experienced group the pattern was reversed: their normal texts were more linguistically elaborated than their trance texts.
The pattern can be summarised as follows:
- Experienced mediums: less frontal lobe activity → more complex texts
- Less experienced mediums: more frontal lobe activity → less complex texts
This relationship contradicts the naive assumption that more cognitive output always demands more brain activity. The authors interpreted it as evidence of possible dissociation: the conscious, planning frontal system of experienced mediums was less engaged during trance — and yet, or perhaps therefore, linguistically richer texts emerged.
"The reduced activity in frontal cortical areas in experienced psychography mediums during trance, combined with higher text complexity, suggests a dissociative process that differs from normal creative writing." — Peres et al., PLoS ONE 2012
Comparison with Other Dissociation Research
This finding does not stand alone. Other research fields have documented that highly practised skills can lead to the conscious monitoring system becoming less active while output quality remains high or even improves:
- Jazz improvisation: Limb and Braun (2008) showed via fMRI that professional jazz musicians deactivate the dorsolateral prefrontal cortex — a centre for conscious self-monitoring — during free improvisation, while creative areas become more active.
- Deep meditation: Studies on long-term meditators show similar patterns: experienced practitioners require less frontal control activity to reach states that novices can only achieve with considerable cognitive effort.
- Hypnosis: Highly hypnotisable individuals show reduced anterior cingulate cortex activity during trance — precisely the region that was also less active in the experienced mediums.
The findings of Peres et al. thus fit an established picture: expertise can cause conscious monitoring to recede — which raises an interesting question about the relationship between consciousness and brain activity.
The Researchers
Julio Fernando Prieto Peres is a Brazilian psychiatrist and neuroscientist at Hospital Israelita Albert Einstein in São Paulo. He has published several studies on psychography, religious experiences and trauma, and is one of the few researchers to have applied neuroimaging to spiritist phenomena.
Alexander Moreira-Almeida is Professor of Psychiatry at the Universidade Federal de Juiz de Fora (UFJF) and founder of NUPES (Núcleo de Pesquisas em Espiritualidade e Saúde). Internationally one of the leading researchers in spirituality and mental health, he has published over 200 peer-reviewed articles and sits on advisory boards of the World Psychiatric Association.
Andrew Newberg is a neuroscientist at Thomas Jefferson University in Philadelphia and one of the founders of modern neurotheology — the imaging-based study of religious and spiritual experience. He is known for SPECT studies of meditating Tibetan Buddhists and praying Franciscan nuns, as well as research on glossolalia. His 2009 book How God Changes Your Brain summarises this work for a general audience.
What the Study Shows — and What It Does Not
The study does not prove that mediums actually communicate with the deceased. It shows something more specific and methodologically tangible: that the state in which experienced mediums practise psychography differs measurably in neurophysiological terms from ordinary writing — and that this difference correlates systematically with years of experience.
Limitations of the study: the sample of ten participants is small; SPECT has lower spatial resolution than fMRI; text complexity rating involves a subjective element. Replications with larger samples have yet to appear.
Nevertheless it remains the most methodologically rigorous neuroimaging investigation specifically of psychography to date — and it has prompted scientific discussion about whether automatic writing deserves to be taken seriously as a dissociative state, independently of the question of its ultimate cause.
Sources
- Peres JF, Moreira-Almeida A, Caixeta L, Leuchsenring F, Newberg A. Neuroimaging During Trance State: A Contribution to the Study of Dissociation. PLoS ONE. 2012;7(11):e49360. doi:10.1371/journal.pone.0049360
- Limb CJ, Braun AR. Neural Substrates of Spontaneous Musical Performance: An fMRI Study of Jazz Improvisation. PLoS ONE. 2008;3(2):e1679.
- Newberg A, Waldman MR. How God Changes Your Brain. Ballantine Books, 2009.
- Moreira-Almeida A, Lotufo Neto F, Cardeña E. Comparison of Brazilian spiritist mediumship and dissociative identity disorder. Journal of Nervous and Mental Disease. 2008;196(5):420–424.
- For psychography practice: Chico Xavier and Psychography; for context: Mediumship in Brazil.