Information Need Awareness : an EEG study

Michalkova, Dominika and Parra-Rodriguez, Mario and Moshfeghi, Yashar; (2022) Information Need Awareness : an EEG study. In: SIGIR 2022 - Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. SIGIR 2022 - Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval . ACM, New York, 610–621. ISBN 9781450387323 (https://doi.org/10.1145/3477495.3531999)

[thumbnail of Michalkova-etal-SIGIR-2022-Information-need-awareness-an-EEG-study]
Preview
 Text. Filename: Michalkova_etal_SIGIR_2022_Information_need_awareness_an_EEG_study.pdf
Accepted Author Manuscript
License: Strathprints license 1.0
Download (3MB)| Preview

Abstract

A fundamental goal of Information Retrieval (IR) is to satisfy search­ers' information need (IN). Advances in neuroimaging technologies have allowed for interdisciplinary research to investigate the brain activity associated with the realisation of IN. While these studies have been informative, they were not able to capture the cognitive processes underlying the realisation of IN and the interplay between them with a high temporal resolution. This paper aims to investigate this research question by inferring the variability of brain activity based on the contrast of a state of IN with the two other (no-IN) scenarios. To do so, we employed Electroencephalography (EEG) and constructed an Event-Related Potential (ERP) analysis of the brain signals captured while the participants were experiencing the realisation of IN. In particular, the brain signals of 24 healthy participants were captured while performing a Question-Answering (Q/A) Task. Our results show a link between the early stages of processing, corresponding to awareness and the late activity, meaning memory control mechanisms. Our findings also show that participants exhibited early N1-P2 complex indexing awareness processes and indicate, thus, that the realisation of IN is manifested in the brain before it reaches the user's consciousness. This research contributes novel insights into a better understanding of IN and informs the design of IR systems to better satisfy it.

CORE (COnnecting REpositories)