| dc.description.abstract |
Multistable perception describes the spontaneous fluctuation between two or more perceptual states when sensory input is ambiguous. An example hereof is bistability, which occurs when a stimulus has two competing interpretations that perceptually alternate over time. For instance, in structure- from-motion (SFM) bistable perception, the coherent movement of dots creates the illusion of a rotating sphere, where the direction of movement is uncertain. Another example is binocular rivalry (BR), which occurs when the two eyes are presented with dissimilar visual stimuli in the same retinal space, leading to an alternation of conscious awareness between the two stimuli. Multistable perception has been used to investigate the neural correlates of conscious experience, since an unchanging stimulus leads to a change in awareness, hence dissociating consciousness from sensory processing. Functional magnetic resonance imaging (fMRI) has consistently shown activity of the right intraparietal sulcus (IPS) and right superior parietal lobule (SPL) during perceptual transitions in multistable perception. Previously, transcranial magnetic stimulation (TMS) and in particular inhibitory theta burst stimulation (cTBS) has been used on the IPS to probe its causal role in multistable perception. That endeavour has produced inconsistent results on whether IPS inhibition shortens or lengthens multistable dominance durations. Problematically, the neural effects of cTBS over IPS during multistable perception are unknown, as is indeed the causal role of IPS in mediating perceptual reversals.
Chapter 1
cTBS was applied over IPS or over vertex control site, between two sessions of fMRI, to illuminate the changes in neural activity accompanied by IPS cTBS. During the fMRI sessions, participants viewed alternating blocks of a bistable SFM stimulus or a replay condition using depth-cue disambiguated SFM. Behaviourally, it was found that IPS cTBS lengthened dominance durations when comparing pre vs post cTBS as well as when comparing IPS with vertex stimulation. Neurally, IPS cTBS led to a decrease in blood-oxygen-level dependent (BOLD) response in thalamus, foveal V1, right superior parietal lobule and middle frontal gyrus compared to vertex cTBS. Moreover, a decrease of functional connectivity between activity in IPS and ipsilateral hippocampus was observed. The present results suggest that the combined effects of a reduction of sensory processing as well as decoupling between IPS and the memory site hippocampus allows inhibitory TMS over parietal cortex to stabilise the current perceptual content. Together, these results provide a hitherto unreported insight into the brain networks that subserve the resolution of bistable perception and how IPS stimulation modulates them to bring about a behavioural effect.
Chapter 2
Next to the IPS, also the more posterior SPL has been indicated as serving a causal role in multistable perception. TMS has been used to modulate bistable dominance durations for both sites, but in opposite directions. This led to the proposal that parietal cortex is fractionated, such that IPS and SPL serve opposing functions. However, neuroimaging evidence also suggests that higher cortical activity, including parietal cortex, is diminished when BR percept switches are either unreported or unreportable. This suggests that parietal regions have no causal role in multistable perception, but are active only as consequence thereof. To resolve this conflict, chapter 2 investigates whether cTBS to the IPS as well as the SPL affects the temporal dynamics of BR using regular button press rivalry as well as no-report and invisible rivalry paradigms. Specifically, it was hypothesised that cTBS would lead to a change in BR dominance when it was visible or unreported, but not when invisible. However, contrary to expectation, not only was it not possible to replicate the previously observed functional fractionation of parietal cortex, but also no difference was found between any cTBS condition. To verify if cTBS had its desired inhibitory effect, also motor-evoked potentials (MEP) were recording prior and following cTBS to primary motor cortex. It was found that cTBS to M1 decreases MEP amplitude. However, this effect did not correlate with the main findings over parietal cortex, leading to the conclusion that cTBS is not an apt neurostimulation technique to answer the present research question.
Chapter 3
Relative intensities of steady-state responses (SSRs) over early visual cortex have been reported to correlate with conscious perception in paradigms like BR and have even be used to predict the content of consciousness. However, their causal role in perception remains uninvestigated despite their common use. Are modulations of SSRs mere epiphenomena of perception or do they aid in determining its content? To test this, it was enquired if interference with the SSR by means of transcranial alternating current stimulation (tACS) would affect conscious perception. Sham or real tACS across left and right parieto-occipital cortex was applied at either the same or a different frequency or in and out of phase with an SSR eliciting flicker stimulus, while participants viewed either BR or tried to detect stimuli masked by continuous flash suppression (CFS). It was found that tACS did not differentially affect conscious perception in the forms of BR predominance, CFS detection accuracy, reaction time, or metacognitive sensitivity. Importantly, the present null-findings are supported by Bayesian statistics. In conclusion, the application of tACS at frequencies and phases of stimulus-induced SSRs does not have perceptual effects. The relationship of tACS with SSRs and the possibility that SSRs are epiphenomenal to conscious perception is discussed.
Chapter 4
One reason for the difference between findings of studies, which attempted to modulate multistable dominance durations thought cTBS to the IPS, may be that different stimuli were used, dissimilar properties of which modulated the TMS effect direction. To test this, cTBS was applied to the IPS between two sessions of SFM bistable perception (chapter 1), random dot motion BR (chapter 2), as well as checkerboard BR (chapter 3). It was foremost hypothesised that the findings of the first two chapters would be replicated, and moreover that the TMS effect would correlate between stimuli. Contrary to this hypothesis, cTBS neither consistently affected dominance durations in any of the stimuli, nor were effect sizes correlated across participants. This is supported by Bayesian statistics. Baseline dominance durations prior to TMS correlated across the three stimuli, suggesting a common mechanism to resolve multistability. However, the lack of correlation pertaining to the cTBS effect points towards the absence of any cTBS effect. Considering the present results, the small samples and effect sizes of previous studies, as well as recent literature of variable cTBS effects on motor cortex, this chapter concludes that there is good reason to cast general doubt over the ability of parietal cTBS to modulate dominance durations in multistable perception.
Chapter 1 pointed towards the importance of multiple brain networks including the IPS in the resolution of multistability. Chapters 2 & 4 by contrast presented with null results that do not allow inference as to the causal role of IPS. Similarly, the use of tACS to modulate SSRs in chapter 3 was not able to demonstrate conclusively whether SSRs have a causal role in multistability. The search for the contribution of IPS by use of cTBS or tACS has been hindered by methodological concerns over whether these methods have an interpretable or even any effect on IPS activity. In summary, the causal role of IPS activity in multistable perception remains elusive. |
en |
