Introduction

Aligning neuronal activity with the content of perception is a fundamental concern in consciousness studies. This not only contributes to addressing the ”what-it-is-like” problem (Nagel, 1974) but also establishes the technical foundation for developing virtual perceptions, such as bionic eyes and auditory prostheses. While the brain itself does not receive direct light stimulation, understanding the ”visual code” connecting the brain and perception remains crucial. One way to investigate this process more directly is via the use of phosphene—the phenomenon of experiencing visual light without external light stimulation. Studies have demonstrated that phosphene experiences can be induced through direct electrical stimulation of the visual cortex (Brindley & Lewin, 1968; Foerster, 1929; Fox et al., 2020) or magnetic pulses applied over the visual cortex (Barker et al., 1985; Meyer & Allen, 1991). Particularly, transcranial alternating current stimulation (tACS), involving a current with a constant polarity exchange, has been shown to elicit phosphenes (Kanai et al., 2008). Phosphenes induced by tACS are commonly perceived as repetitive flickering in the visual field. Previous research has indicated that the perception of tACS-induced phosphenes can be dependent on the stimulation frequency (Evans et al., 2019, 2021; Hsu et al., 2023; Kanai et al., 2008; Turi et al., 2013). In contrast to conventional tACS, amplitude-modulated tACS (AM-tACS) has garnered attention, driven in part by efforts to minimize phosphene occurrence during stimulation while examining the impact of tACS on visual cognition performance (Thiele et al., 2021). A strategy employed to achieve this goal involves embedding the target frequency into a rapidly oscillating tACS in AM form (Thiele et al., 2021). This approach aims to prevent phosphene occurrence (Thiele et al., 2021) and contamination of neighboring ongoing neural activity (Witkowski et al., 2016) during neural entrainment. Recent studies have emphasized the crucial role of AM information in sensory encoding (Clarke et al., 2015; Juan et al., 2021; Nguyen et al., 2019; Ryu et al., 2017). Specifically, one study by Hsu et al.(2023) provided direct insights into the effects of AM-tACS on visual awareness, shedding light on the role of AM in tACS-induced phosphene perception. The study applied AM-tACS with carrier frequencies of 10, 14, 18, and 22 Hz, and AM frequencies of 0, 2, and 4 Hz. The results indicated that AM-tACS-induced phosphene perception exhibited higher thresholds and a slower flash rate, suggesting that the AM frequency exerts a more substantial influence on phosphene perception than the carrier frequency. Nevertheless, our understanding of how AM frequency influences visual perception is still in its early stages (Juan et al., 2021; Nguyen et al., 2019). In the study, we build upon our previous work and investigate the neural mechanisms underlying phosphene perception induced by transcranial electric brain stimulation. To comprehend phosphene perception, it is imperative to deconstruct the biophysiological effects induced by tACS. This technique entails the application of an alternating current, where the polarities—referred to as the ”source” and ”sink” of the electricity—rhythmically switch between the two electrodes. Unlike phosphene elicitation through transcranial magnetic stimulation, where magnetic pulses induce neural firing in the visual cortex, scalp electric stimulation is not potent enough to directly provoke action potentials in the cortex (Asamoah et al., 2019; Y.-Z. Huang, 2017; Vöröslakos et al., 2018). Nevertheless, the oscillations of the electric field can modulate the rhythms of spontaneous neural firing, resulting in synchronous excitation and inhibition during specific oscillatory phases (Wischnewski et al., 2023). Additionally, tACS may induce ephaptic coupling (Anastassiou et al., 2011; Han et al., 2018), a phenomenon wherein non-synaptic interneural communication occurs via extracellular potential oscillation. This is considered one of the electromagnetic foundations for the occurrence of consciousness (Hunt, 2020; Hunt et al., 2022; Hunt & Jones, 2023; McFadden, 2021). The application of tACS, therefore, brings about both rhythmic changes in the electric field and the alternating polarity (excitatory or inhibitory phase). Considering that no phosphene has been reported during monotonous transcranial direct current stimulation (tDCS), it is plausible that electric oscillation plays a pivotal role in inducing phosphene perception. One potential method to test this hypothesis is to dissociate the effects of rhythmic changes in electric field vs. alternating polarity. To do this, the present study employs oscillatory transcranial direct current stimulation (otDCS), where the current oscillation is confined to one polarity—either positive or negative—to eliminate the influence of polarity switching. Previous studies utilizing otDCS for modulating cognitive function have demonstrated that otDCS can induce changes in excitability (Antal et al., 2008; Bergmann et al., 2009; Groppa et al., 2010) and neural entrainment (Vulić et al., 2021; Živanović et al., 2022), combining characteristics of both tDCS and tACS. Evans et al. (2021) used sinusoidal direct current on the scalp, eliciting phosphenes without polarity differences, focusing only on threshold measurement. In the study, we employed various measurements, including response times (RT), ratings of perceived brightness, flash rate, and confidence, along with indices of threshold intensity and pattern drawing. This comprehensive approach allows us to thoroughly investigate the features of phosphenes induced by both otDCS and tACS, facilitating direct comparison of potential differences between them. We hypothesize that if current oscillation plays a crucial role in inducing phosphene perception, phosphenes should be observed in both anodal and cathodal otDCS. While the primary neural effect of tACS is synchronization rather than a general firing rate increase (Wischnewski et al., 2023), the polarity of otDCS may not affect the threshold intensity or interact with the AM frequency. However, ratings on other perceptual reports may be influenced by polarity if it is involved in the general firing rate. Hsu and his colleagues (2023) delved into the characterization of phosphene content by instructing participants to report on the pattern and flash rate of tACS-induced phosphenes. In terms of flash rate scoring, they demonstrated a linear effect of carrier frequency in sinusoidal tACS. Interestingly, they found that the amplitude modulation (AM) frequency could override the carrier frequency, disrupting the linear relationship and resulting in a slower flash rate. These findings suggested the intriguing possibility that AM may take precedence over carrier frequency in influencing the perception system. Numerous previous studies have indicated the ability of the visual neural system to capture envelope information at both retinal (Ryu et al., 2017) and visual cortical (Nguyen et al., 2019; Shapley, 1998) levels. Additionally, a simulation study (Negahbani et al., 2018) demonstrated that the local field potential could synchronize with the AM frequency when applying AM-tACS to simulated pyramidal cortical cells. Collectively, these studies strongly suggest that AM information could play a pivotal role in perception by exerting an overriding influence on carrier frequency. However, an alternative hypothesis could potentially explain the observed slower flash ratings. It is conceivable that the slower flash perception may result from the waveform of the AM-tACS. Specifically, the amplitude of the carrier oscillation might be modified by the AM frequency, causing it to change over time. It is plausible that the amplitude of the carrier oscillation reaches the perception threshold only when the amplitude of the AM frequency peaks at phase angles of 90 or 270 degrees. Consequently, AM-tACS may trigger a phosphene percept at the AM frequency, leading to a slower flash that is less sensitive to the carrier frequency. This concern is addressed by comparing ratings from trials with intensities above the mean threshold. The hypothesis posits that if the phosphene flash follows this pattern, participants should report a higher flash rate during suprathreshold trials, as more oscillations could exceed the phosphene threshold. A null result from the test could imply that the slower ratings support the perceptual dominance of the AM frequency or are simply a result of missing values and data variation due to not every participant having trials with the same intensity. Therefore, to better elucidate the source of the slow flash rate in AM-tACS, we introduced an additional suprathreshold condition that stimulates at 120% intensity of the threshold. We anticipate that a higher stimulation intensity will enhance neural alignments to the oscillation that determines our flash percept. If the AM frequency holds an advantage in phosphene perception, the flash rate would align with the AM frequency and remain unaffected by the stimulation intensity. In contrast, if the flash rate reflects the frequency of carrier oscillations surpassing the perception threshold, ratings for the suprathreshold condition should be faster than those for the threshold condition. The current study delves into the neural mechanisms underlying transcranial electric stimulation (tES) induced phosphene perception by investigating quantitative indicators of phosphene experiences. The specific objectives of this study are to explore (1) the essential role of neural electric field oscillation by isolating oscillation from polarity, (2) the impact of polarity in oscillatory tDCS on phosphene perception, and (3) the influence of neural alignment with AM frequency on phosphene flash rate through the implementation of suprathreshold stimulation. By incorporating multiple behavioral measurements beyond the threshold, we anticipate gaining a more comprehensive understanding of the neural mechanisms that connect neural oscillations with visual perception.