Table 2. Exclusion criteria for tDCS experiments.

Exclusion	Explanation
Mock Priming (Higher than 66%)	Mock displays are fake binocular rivalry displays – priming on these trials indicates that participants are showing a response/demand characteristic and as such we cannot trust their priming scores, as they may either be responding in a way that they think we want them too, or they are not attending to the task correctly. A score of more than 66% indicates that the participant has primed on these mock trials more than once.
Low Priming (lower than 40%)	Participants whose imagery scores were lower than 40% were excluded, as the score becomes difficult to interpret: The measure of imagery strength is predicated on how the energy of a stimulus impacts on binocular rivalry. Very weak perceptual stimuli prime binocular rivalry up unto a certain point. At this tipping point, as the image becomes stronger, it begins to suppress binocular rivalry (Brascamp et al., 2007). For this reason, very low priming can either mean that the participant’s visual imagery is so strong that it leads to suppression, or that the opposite is the case, and imagery is in fact very weak. Alternatively, it may also be that a participant is not completing the task correctly. 40% was chosen as the cut off as this is 10% lower than chance values (50%).
Mixed Percepts (higher than 1/3 or 33%)	We analyse our priming data as percent primed,that is the percentage of trials a participant imagined an image, then saw that image in the following display. Mixed trials cannot be labelled as either ‘primed’ or ‘not primed’, and as such these trials are not included in the analysis. A high number of mixed trials reduces the number of usable trials for analysis. This can lead to large changes that may be spurious (much larger percentage changes due to a single trial) and not due to the stimulation parameters. We excluded an individual’s data set if more than a third of the trials were mixed percept’s.
Attrition	Attrition occurred when a participant did not turn up to or cancelled the second and/or third day of testing.
Impedance (Exceeding voltage: impedance greater than 55 kΩ)	For safety reasons, our tDCS machine was programmed to shut off once the impedance exceeded 55 kΩ (this is pre-programmed into the tDCS machine). As the participants completed the experiment in a blackened room by themselves we could not monitor the impedance of the machine in real-time and as such the machine could switch off during a block of the experiment. As we cannot know how much stimulation this participant received we are unable to use them reliably in our analysis – as different stimulation durations will lead to different excitability changes. The second (1.5mA Occip + cheek) and fourth (1.5mA pFC + cheek) tDCS experiments were run at the same time. There were a large number of cases of impedance being exceeded in these studies – it was discovered that this was due to a faulty wire, which was replaced halfway through the experimental data collection.
Phosphene Determination	If participants reported phosphenes in the wrong visual hemifield (e.g. left visual hemisphere was stimulated and participants reported phosphenes in the left visual hemifield) their data was excluded. Participants who had very large eye-blinks in the REPT procedure were excluded from the experiment, due to this potentially resulting in a missed phosphene leading to incorrect phosphene threshold estimation.