Neural Mechanisms of Visual Mental Imagery in the Healthy and Damaged Brain

In the absence of visual input from the external world, humans are able to internally generate vivid mental images of external stimuli. This cognitive process is known as visual mental imagery, and it is involved in many forms of complex reasoning and problem-solving. Using functional neuroimaging, several studies revealed that visual mental imagery recruits a network of prefrontal, parietal and inferotemporal regions. Moreover, under certain conditions, imagining external entities have been shown to induce recruitment of retinotopically-organized visual areas, traditionally thought to be dedicated to the perception of external stimuli. The recruitment of low-level visual areas following visual mental imagery of different stimuli could have important implications for the implementation of new rehabilitative techniques directed to patients suffering from visual field defects. In fact, following lesions affecting retrochiasmatic visual pathways, one of the most common deficits is homonymous hemianopia. This visual impairment is characterized by the loss of sight in one half of the visual field and has a profound impact on patients’ emotional and social wellbeing. Several studies indicated preserved visual imagery abilities in hemianopic patients, both in the sighted and damaged hemifield. This led us to hypothesize the possibility to recruit functionally preserved portions of early visual cortices in the affected hemisphere of hemianopic patients by means of visual mental imagery. If this revealed to be true, the recruitment of early visual areas would potentially induce plastic mechanisms of change that could reinstate perceptual awareness, increasing the size of the perceived visual field. In the present thesis, we explored neural substrates of visual mental imagery both in the healthy and in the damaged brain using fMRI. In Study 1, by means of a delayed spatial judgment task, we investigated in healthy participants the degree of complexity of the information encoded in primary visual cortex, its similarities and differences with representations of perceived stimuli, and how this information is encoded in areas outside early visual cortex. We found significant encoding of complex stimulus categories in early visual areas, as well as in inferotemporal and parietal cortices. Additionally, in agreement with previous studies, we found that a subset of these regions showed a certain degree of shared representations with perception. Moreover, in Study 2, we explored whether it is possible to selectively recruit individual quadrants within the visual field using visual mental imagery. To this aim, we tested a group of normal-sighted individuals and patients suffering from homonymous hemianopia in a visual imagery paradigm. Results indicated that normal-sighted individuals are able to recruit early visual cortex by means of top-down mechanisms. In the group of patients, we observed a large amount of interindividual variability that allowed reliable recruitment limited to the healthy hemisphere. Together, the results of this thesis provide evidence for distinct roles of parietal and premotor areas, involved in processing the spatial layout of imagined stimuli, and temporal regions, representing the content of internally generated representations. Moreover, the results are in line with the view that, in the absence of bottom-up visual stimulation, early visual cortex is able to access information about both content and spatial layout of imagined stimuli via feedback connections. In addition, we demonstrated that the top-down modulation of low-level visual areas occurring during visual mental imagery is feasible to recruit retinotopically-organized early visual cortex, both in normal-sighted participants and in the healthy hemisphere of hemianopic patients. Albeit preliminary, these results open new perspectives on the potential use of visual mental imagery as a rehabilitation tool in the clinical treatment of visual field defects.