The Neural mechanisms of perceptual Stability in magnitude perception (NeSt)

How do we perceive the world around us? An essential piece of information that we need in order to understand the external environment is the magnitude dimension of the sensory experience: how many objects are around us? And how big are they? How long and how frequent are the important events occurring in a visual scene? To answer these questions, the brain must process magnitude information in a fast and reliable way. An important characteristic of sensory information allowing to make decisions in a reliable way is its stability: objects do not suddenly change in their appearance or position, and we usually perceive the world as a continuous, seamless, stream of information. This is however not trivial, as the machinery of our brain and sensory organs is not exact. Neurons are noisy, and the information they process is made discontinuous by the intrinsic instability of biological sensors like the eye, constantly moving and shifting to scan the environment. The overarching goal of NeSt (“Neural mechanisms of perceptual Stability in magnitude perception”) was thus to understand how the brain ensures a stable and continuous processing of magnitude information. Understanding the mechanisms of stability is indeed fundamental to further understand how the brain works on a broader level, and to understand how our subjective conscious experience is constructed based on the available sensory information. With NeSt, we have extensively explored the neural mechanisms and functional properties of stability in magnitude perception, characterizing the rules governing stability across different perceptual dimensions and sensory modalities, the consequences of stability on perception and decision-making, and the dynamics of brain processes involved in stability.

Across two main work packages, we used psychophysical, electroencephalography (EEG), and brain stimulation techniques to address the encoding, processing, and stabilisation of magnitude information. The first research work package (WP2) was based on psychophysics and EEG. First, we investigated how the brain exploits information over time across multiple magnitude dimensions (size, duration, and numerosity) in the service of perceptual stability and continuity, using psychophysics and EEG (Exp. 1). To this aim, we measured the psychophysical “serial dependence” effect, a perceptual bias whereby a stimulus that we are currently seeing appears more similar to what we saw in the recent past. Concurrently with the testing of serial dependence, we measured the EEG responses to the visual stimuli. Additionally, we further measured the neural signature of stability across magnitude dimensions using a passive-viewing approach during EEG recording (Exp. 2), which allows to capture the purely perceptual aspects of stability in the absence of decision-making. In three subsequent studies (Exp. 3 Exp. 4, and Exp. 5), we also addressed the specificity of the serial dependence effect for the task-relevance of different magnitude dimensions (i.e., whether they are actively judged or not), the interaction between the processes of perceptual stability and magnitude integration, and the nature of the magnitude integration effect itself. Magnitude integration is indeed an important feature of magnitude perception, and defines how we perceive a given dimension when multiple magnitudes are concurrently modulated. However, it is unknown what kind of brain mechanism might support magnitude integration, and whether this process might share similarities or even overlapping processes with perceptual stability. Finally, in Exp. 6 and Exp. 7, we addressed whether stability in magnitude perception involves a common, generalized mechanism encompassing different perceptual dimensions and sensory modalities, or a series of distinct and potentially independent mechanisms. Overall, our results show that the neural (EEG) and behavioural (serial dependence) signatures of stability can be dissociated, as the first seems to more genuinely capture perceptual processing, while the second also involves cognitive decision-making processes. Our results also provided important evidence concerning the nature of magnitude integration: it is mediated by an active mechanism “binding” the different dimensions of an object, and not by a trivial contextual interference. Magnitude integration however does not seem to interact with stability, although past information and the history of attention to different magnitude dimensions seem to influence the pattern of magnitude integration. Finally, our results also suggest that the brain possesses different and potentially independent mechanisms mediating stability in different perceptual domains and sensory modalities, following different rules and showing different patterns of brain activity. The work in this WP has been disseminated by presenting the results at the European Society of Cognitive Psychology (ESCoP) 2019 meeting, at the CyPi Neuroscience workshop in 2019, at the European Workshop on Cognitive Neuropsychology (EWCN) in 2020, and will be presented at the European Conference on Visual Perception (ECVP) in 2021.

In the second main research work package (WP3), we used the transcranial magnetic stimulation (TMS) technique to further address the mechanisms of perceptual stability in magnitude perception. TMS involves delivering a magnetic pulse disrupting brain activity in a specific cortical area at a specific time, with a high degree of spatial and temporal precision. In this study, we focused on the serial dependence effect in numerosity and time perception, considering it as a correlate of perceptual stability (i.e., the consequence of stability on perception), and delivered TMS to the primary visual cortex (V1) at different times while participants judged the numerosity or duration of visual stimuli. Our results show that TMS to area V1 can successfully disrupt serial dependence, abolishing the effect, and crucially it does so in a time-sensitive fashion, depending on which brain processing stage is disrupted. Namely, the crucial processing stages linked to establishing stability (and the serial dependence effect) correspond to the processing of the past stimulus at late latencies, and the processing of the current stimulus at both early and late latencies. With this study, we have provided for the first time causal evidence for the involvement of the primary visual cortex in perceptual stability, and demonstrated the timing of its involvement.

Overall, the work carried out during NeSt has pushed the boundaries of our understanding of magnitude processing and perceptual stability, going significantly beyond the state of the art. Work in WP2 indeed provided new evidence suggesting a possible dissociation between the mechanisms purely dedicated to perceptual stability, and the processes resulting in the serial dependence effect. This is particularly important, and suggests that a shift of paradigm is necessary in the serial dependence and perceptual stability research fields, which could benefit from a more extensive use of techniques such as the EEG to reach more robust results. This observation is thus very promising and has the potential to spur research in new directions to further advance our understanding of perceptual stability and serial dependence. Additionally, the finding of different stability mechanisms in different perceptual domains and sensory modalities is important as well, as it might explain the inconsistency of results obtained by different studies when testing serial dependence with different paradigms. Second, evidence gathered in WP3 provided for the first time robust cause-effect evidence for the involvement of the primary visual cortex in establishing serial dependence, showing that although potentially modulated by decision-making, this effect is deeply rooted into the sensory processing stream. This in turn might inform the development of new models and theoretical frameworks of stability and serial dependence, which should take into account the physiological properties of primary sensory areas. Taken together, these results are expected to have an important impact in the perceptual science and cognitive neuroscience research field, opening new research frontiers and deepening our understanding not only of perceptual stability and magnitude perception, but also of how our brain works on a broader level. On a broader level, the findings achieved during NeSt represent an important step forward towards reaching the ultimate goal of how our conscious experience of the external world is generated by the brain.

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 838823