Does Inference Feel Like Perception? Metacognitive Evaluation in Event Completion

Humans often perceive events as continuous and complete even when some visual information is missing. This process, known as event completion, fills gaps in perception using contextual information, which can make it difficult to distinguish between what was actually perceived and what was inferred. Prior research has demonstrated that inferred content can be mistaken for real perception, yet it remains unclear whether individuals' confidence in their judgments reflect their reliability. We first re-analyzed the data of Papenmeier et al. (2019) by applying Bayesian methods to estimate metacognitive efficiency in the event completion paradigm. Following this, we conducted a pre-registered replication study in which participants viewed soccer video clips containing either causal or non-causal continuations of a critical moment. Participants then indicated whether they had seen the moment and provided confidence ratings for their judgments. Our findings replicated previous results showing that events which are causally implied but not directly seen are often experienced as if perceived. Crucially, in both the secondary analysis and the replication experiment, metacognitive efficiency did not differ significantly between the causal and non-causal condition. These results indicate that while perceptual experience and bridging inference merge during event perception, their ability to evaluate the correctness of their judgments remains unaffected. Overall, this research advances understanding of how people monitor and regulate their cognitive processes when perceiving dynamic events. This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – 381713393 (Research Unit 2718: Modal and Amodal Cognition). notReviewed other