In keeping with previous studies, we demonstrate a higher presence of auditory, haptic, olfactory and gustatory sensation in dreams of congenitally blind, by comparison to normally sighted individuals11,12. Our report of oneiric visual imagery in congenitally blind subjects (Figure 1,2; Table 3), however, challenges the negative findings in the majority of the previous studies11,12. On the other hand, our results appear to be in keeping with more recent studies that have demonstrated (oneiric) visual imagery in congenitally and totally blind subjects lacking any previous visual perception or experience12,38,40.
We also report, for the first time, an excerpt from the interview with the congenitally blind woman (Table 3; DreamBank45). Her elaboration of oneiric visual experiences is in broad agreement with other anecdotal reports where subjects refute common understanding that their visual imagery may reflect merely metaphoric5,44or mental representations with preserved spatial and metric properties40.
Historically, it has been recognised that the major experimental conundrum in delineating processes that may underlie any such visual imagery predominantly resides in the limited objectivity of otherwise highly personal and subjective dream reports. Similarly, the overwhelming neurophysiologic complexity of the visual system presents further hindrance49. For example, the visual system is comprised by multiple parallel and interacting processing pathways in the brain that relate and process neural information on form, motion, and colour40,49. However, it is uncertain whether there is anatomical separation between the visual cortical areas serving visual imagery and those serving visual perception40. Over the years some neuroscientists have advanced that the regions used for visual imagery present a subset of those engaged in perception, whilst others have maintained that the regions subserving visual perception and imagery are the same (please see40). In summary, to date the ambiguity remains how these separate pathways are brought together into a single image, and whether the reevoking of images inevitably activate all of them on all the hierarchic levels40,49,50.
Another interesting possibility could be that, arguably, in a theoretical parallel to Jungian’s notion of archetypal symbols (e.g. protoconsciousness and oneiric primordial images)51, the eccentric genetic wiring of our early visual cortex22 supports a possibility of elementary (primordial) ‘visual-like’ or homoiōma ("likeness", in Ancient Greek) neural representations that are inbuilt a priori and onto which other sensory modalities feedback non-visual and potentially predictive information. If this indeed was the case, this, in turn, would then enable for a typical spatio-temporal organization of early visual areas by eccentricity49 to develop even in the life-long absence of vision.32 Moreover, such notion would arguably also explain striking ability of congenitally and totally blind subjects to draw symbolic representations of various visual images 38 in eerie likeness to those drawn by normally sighted subjects. Somewhat analogous hypotheses have been in past advanced to explain the protracted language acquisition in the autistic individuals by comparison to those with neurotypical development, and are in line with notion of Hebbian correlation learning in neuroanatomically structured networks which yield distributed circuits binding action and perception information50.
At the neurophysiologic level, the existence of homoiōma could be argued by the demonstrations of cross-modal neuroplasticity, as evidenced by the neuroimaging 15,21,52 and sensory substitution 53-55 studies. For instance, parts of the occipital cortex, such as V1 region, have been shown to undergo cross-modal plastic adaptation in the congenitally blind, and to contribute to non-visual processing13,18,20. However, other occipital areas - such as the extrastriate body area14, the lateral-occipital tactile-visual area 56, and the fusiform and inferior temporal gyri 21 maintain the higher-order, multisensory integration functions that they have in the sighted, and therefore, presumably at least in part, may contribute to formation of our reported oneiric homoiōma (Figure 2).
Finally, despite some obvious limitations of our cross-sectional study that compared individuals from different time eras and ages, hence, restricting claims of any causality, we advance that our findings are supportive of the presence of homoiōma, or oneiric visual imagery in congenitally blind. Our findings also support the growing calls for multi-centre and multimodal imaging studies of dreaming and sleep physiology in congenitally blind people. Deciphering the mechanistic nature and the genesis of homoiōma may open new possibility in utilisation of neuroplasticity and its potential role for treatment of neurodisability.