Mediumship can be defined as a communication coming from a source that is considered to exist at another level or dimension, beyond the known physical reality, and not stemming from the medium's mind (Klimo, 1998, as cited in Almeida et al., 2004). This experience presents a change in the person`s global qualitative pattern of consciousness in which he/she is able to hear, see or feel the presence of supposed spiritual entities or deceased people (Dalgalarrondo, 2008). The reports of this type of experience are vast. Barros and coworkers (2022) investigated the prevalence of Spiritual and Religious Experiences in the Brazilian general population, sampling people from all regions of the country. The study included 1,053 Brazilian citizens, with an average age of 40.8 years, 52.1 % female. They found that circa 92 % of the participants reported having had at least one of the Spiritual and Religious Experiences at some point in their lives. Mediumistic experiences were reported by 58.3 %, for instance, “feeling the presence of someone who has died” (50.4 %), “seeing apparitions or spirits” (41.3 %), and “hearing voices of someone who has died” (30.8 %). Moreover, 11 % of participants claimed to frequently feel the presence of someone who has died and see spirits. Women are more likely to have mediumistic experiences (80 %) (Barros et al., 2022), which corroborates other studies (Almeida, 2004).

Currently, it is possible to think about an approach that studies mediumship from a neuroscientific perspective. This line of research is still in its initial phase, but we have already been able to find consistent studies in the academic literature that use rigorous neuroimaging methods to evaluate mediumship (Bastos et al., 2016; Delorme et al., 2013; Hughes; Melville, 1990; Kawai et al., 2017; Mainieri et al., 2017; Oohashi et al., 2002; Peres et al., 2012; Wahbeh et al., 2019; Rosario-Gilabert; Vigué-Guix, 2025).

So far, EEG has been the most widely used equipment in mediumship research but functional magnetic resonance imaging (fRMI) (Mainieri et al., 2017) and single-photon emission computed tomography (SPECT) (Peres et al., 2012) have also been used in some studies. In fact, these studies represent a major advance in the investigation of brain function during the mediumistic phenomenon. Currently, the understanding that the brain functions in an interconnected and integrated way has gained centrality for neuroscientific approaches.

This theoretical framework proposes that different brain areas interact in a synchronized manner in order to optimize information processing. Mapping the functional brain networks (FCN) allows the understanding of how spatially separated regions may be functionally connected. Dynamic functional connectivity assessment detects statistical connections of anatomically different brain regions based on the temporal dependence of neuronal activation patterns (Lang et al., 2012), using mainly statistical dependencies of correlation and covariance. Thus, dynamic brain functional networks (BFN) are capable of capturing the temporal dynamics of brain relationships, as these types of networks can display changes in small periods of time (Lang et al., 2012). To measure the oscillation of these dynamics, the most suitable equipment is the EEG, due to its high temporal resolution, non-invasiveness (Lang et al., 2012) and low-cost (Azevedo, 2005).

In general, based on studies on connectivity, dynamic BFN are complex and characterized by vertices that are densely connected through functional interactions. Since this method allows for a structural and dynamic topological description of graphs, it has been used in several studies to characterize patterns of neuronal activity both at rest and during the performance of tasks. According to the graph theory, the cortical electrical activity of a given electrode is the vertex and the edges are the functional connections that the different electrodes establish (Lang et al., 2012).

Hence, the present work aims to describe the topological indices of the brain's functional networks during alleged mediumistic trances, for which the dynamic BFN method will be used. The null hypothesis was that there is no difference in the topological indices of the brain's functional networks between the mediumistic trance and the baseline state in the Mediums Group and between the prayer activity and the baseline state in the Control Group. The alternative hypothesis was that the mediumistic trance alters the topological patterns of the brain's functional network compared to the baseline state in the Mediums Group, while the prayer activity does not alter the same patterns compared to the baseline state in the Control Group.