ABSTRACT
Detecting changes in the environment is fundamental for survival, as these may indicate potential rewards or threats. According to predictive coding theory, detecting these irregularities relies on both incoming sensory information and our prior beliefs; with incongruity between the two manifesting as a prediction error (PE) response. Many changes occurring in our environment do not pose any threat and may go unnoticed. Such subtle changes can nevertheless be detected by the brain without ever emerging into consciousness. What remains unclear is how such sensory changes are processed in the brain at the network-level. Here, we developed a visual oddball paradigm, where participants engaged in a central letter task during electroencephalographic (EEG) recordings while presented with task-irrelevant (and unreported) high- or low-coherence background random dot motion. Critically, once in a while, the direction of the dots changed. After the EEG session, we behaviourally confirmed that changes in motion direction at high- and low-coherence were visible and invisible, respectively. ERP analyses revealed that changes in motion direction elicited PE regardless of visibility of such changes but with distinct spatiotemporal patterns. To better understand these responses at the network-level, we applied Dynamic Causal Modelling (DCM) to the EEG data. Bayesian Model Averaging analysis of the DCMs showed that both visible and invisible PE relied on a release from adaptation (or repetition suppression) within sensory areas of V1 and MT. Furthermore, while feedforward upregulation was present for invisible PE, results for the visible change PE also included downregulation of feedback connections between right MT to V1. Overall, our findings reveal a complex interplay of modulation in causal network underlying visible and invisible motion changes.
Footnotes
Conflict of Interest: The authors declare no competing financial interests.








