Skip to content
printicon

Measuring the costs and benefits of curiosity: Exploratory behavior and Learning outcomes

Research project The aim of this project is to measure curiosity and assess its role in learning.

Curiosity causes humans to invest much time on exploratory activities. We may watch TV, play computer games, gossip with friends, leisurely browse the web and get involved in a host of other activities mainly for satisfying curiosity and by extension induce learning. This pervasiveness of curiosity in our daily lives as well as its apparent role in motivating learning suggests curiosity is largely beneficial. But curiosity can be a double-edged sword. It can both stimulate and drive depth learning, but also distract from the completion of important tasks. In this project, we investigate the role of expected information gain, topical interest, as well as cognitive and personality traits to assess the basic principles of curiosity and how they determine learning outcomes.

Head of project

Linus Holm
Senior lecturer (associate professor)
E-mail
Email

Project overview

Project period

2015-01-01 2018-12-31

Funding

Marianne och Marcus Wallenbergs stiftelse

Research subject

Psychology

Project description

The aim of this project is to both measure curiosity, and assess its role in driving learning outcomes. Curiosity can be a double-edged sword. On the one hand, it can both stimulate and drive depth learning. On the other hand, curiosity can distract from the completion of important tasks. It forms a fundamental drive that causes humans to invest much time on activities that do not directly satisfy life sustaining needs. We may watch TV, play computer games, gossip with friends, leisurely browse the web and get involved in a host of other exploratory activities mainly for satisfying curiosity and by extension induce learning. This pervasiveness of curiosity in our daily lives as well as its apparent role in motivating learning certainly merits scientific investigation of curiosity. But curiosity has many diverse properties making it hard to characterize and current understanding is correspondingly confused. For instance, curiosity can operate as a drive in competing with our basic needs. We sometimes forego both food and sleep for seeing the latest show or for continuing to play the game through the night. Curiosity might also appear as an emotion, such as anticipation. Moreover, topical curiosity differs greatly in the population, and seemingly arbitrary topics might instill curiosity. Furthermore, some people appear intrinsically more curious than others, suggesting that curiosity is also a personality trait. Additionally, curiosity can take on the appearance of rashness and impulsivity as well as focused attention on a set knowledge goal. This variety of curious or explorative behaviors might then be expected to have very different learning consequences. The overall aim of the project is to test whether this wide range of proposed properties of curiosity can be unified in a single, parsimonious information theoretic account that produces strong quantitative predictions and should have far reaching impact on our understanding of how curiosity can be recruited to motivate learning. Specifically, the project investigates the basic principles of curiosity, how it can be instilled, how it differs between people and its impact on learning.

SPECIFIC AIMS
We break down the overall aim into a series of specific aims that proceed sequentially from developing a rigorous measure of curiosity to exploiting this measure to inform and improve learning outcomes.
AIM 1: Develop a rigorous methodology to quantify the incentive value of curiosity via economic trade-offs between actions that increase knowledge versus actions that provide e.g., a monetary reward.
AIM 2: Use the curiosity measure to profile participant’s information preferences over a range of tasks and topics.
AIM 3: Quantify individual differences in the incentive value of task relevant vs. task-irrelevant information.
AIM 4: Quantify the learning effects of information congruence: does aligning information reward (action that increases knowledge) with the task reward (action that increases e.g., monetary reward) produce learning gains that facilitate task learning (i.e., faster learning of what action produces the highest monetary reward)?

Keywords: Curiosity, Learning, Information theory, Exploration, Exploitation, Personality, Cognitive control