Web search heavily relies on click-through behavior as an essential feedback signal for performance evaluation and improvement. Traditionally, click is usually treated as a positive implicit feedback signal of relevance or usefulness, while non-click is regarded as a signal of irrelevance or uselessness. However, there are many cases where users satisfy their information need with the contents shown on the Search Engine Result Page (SERP). This raises the problem of measuring the usefulness of non-click results and modeling user satisfaction in such circumstances.

For a long period, understanding non-click results is challenging owing to the lack of user interactions. In recent years, the rapid development of neuroimaging technologies constitutes a paradigm shift in various industries, e.g., search, entertainment, and education. Therefore, we benefit from these technologies and apply them to bridge the gap between the human mind and the external search system in non-click situations. To this end, we analyze the differences in brain signals between the examination of non-click search results in different usefulness levels. Inspired by these findings, we conduct supervised learning tasks to estimate the usefulness of non-click results with brain signals and conventional information (i.e., content and context factors). Furthermore, we devise two re-ranking methods, i.e., a Personalized Method (PM) and a Generalized Intent modeling Method (GIM), for search result re-ranking with the estimated usefulness. Results show that it is feasible to utilize brain signals to improve usefulness estimation performance and enhance human-computer interactions by search result re-ranking.