Thesis
An interdisciplinary exploration of information overload
- Creator
- Rights statement
- Awarding institution
- University of Strathclyde
- Date of award
- 2022
- Thesis identifier
- T16444
- Person Identifier (Local)
- 201871083
- Qualification Level
- Qualification Name
- Department, School or Faculty
- Abstract
- While the propagation of information can lead to innovation and reduce inequality, the fast propagation of information, and the lack of gatekeepers which can support individuals online has led to the proliferation of the problem of information overload. The problem of information overload has always been with us, however, with the development of the world wide web and of new applications serving information, new dimensions to information overload appeared. Attempts have been made to tackle it through various approaches, the most notable being recommender systems. Yet, the lack of a clear definition makes it difficult to scope the problem and clearly identify which issue is being tackled and whether it is tackled efficiently. In the academic context, recommender systems are usually evaluated on past datasets. Hence, using a method from nursing science called concept analysis, we provide a clear definition of information overload. The concept analysis offers a clear outline composed of causes, manifestations, and consequences of information overload. This breakdown enables better scoping of the problem, and the operationalisation of the concept with the aim to conduct more empirical studies. Our work further provided outcome measures which can lead to a scalable way for future systems to detect information overload. These outcome measures were identified during a user study based on the social network Twitter. They were derived from the interaction of our participants with our Twitter clone. By leveraging machine learning methods, we developed classifiers to detect whether individuals were under IO, these classifiers can reach up to 80% accuracy. We found that outcome measures related to mouse tracking solely could lead to detect when individuals were under IO. Our work also showed that while individuals might feel information overload through its emotional and cognitive manifestations, it is not a phenomenon which affects their behaviour instantly. Rather there is a time gap before its onset and its influence on the behaviour of the individuals. We established that individuals can identify that they are in fact subject to information overload; however, there is a delay before it affects them. Complementing our attempt at tackling IO through developing smarter systems, this work investigated how information overload can be tackled from the side of the users of a system. To do so, we designed a search system (a clone of Google) and leveraged advances in cognitive neuroscience and in the research area of neuromodulation to stimulate a brain region which was previously shown to be involved in the online search process. The brain region stimulated was the left dorsolateral prefrontal cortex, part of the prefrontal cortex which is associated with executive functions. Our work highlighted the important considerations needed to be given to executive functions when studying the concept of information overload. Finally, for both of our Twitter-based study and the search-based study, we open-source both systems to offer the research community the opportunity to develop further studies using similar experimental paradigms to those that are used in this thesis.
- Advisor / supervisor
- Learmouth, Gemma
- Pennington, Diane Rasmussen
- Resource Type
- Note
- This thesis was previously held under moratorium from 2nd December 2022 to 2nd December 2023.
- DOI
Relations
Items
| Thumbnail | Title | Date Uploaded | Visibility | Actions |
|---|---|---|---|---|
|
|
PDF of thesis T16444 | 2023-01-09 | Public | Download |