Neural mechanisms of curiosity and curiosity cycles

First, we discuss the neural mechanisms of curiosity. In the PACE framework, prediction errors trigger the curiosity state. Then appraisal could trigger curiosity and exploration. The exploration could trigger further prediction errors. Then we propose questions that we could ask ourselves in the PACE cycle to continue the curiosity state.

Neural mechanisms of curiosity #

Gruber et al. (2014) asked

1. What are the brain areas that support curiosity?
2. How does curiosity benefit learning of interesting material?
3. How does a curious state modulate learning of incidental items?
4. Do curiosity-driven memory benefits for incidental events persist after a long delay?

While the participants are in the fMRI scanner, the researcher presented trivia questions and asked participants to make an incidental judgment to a face. Afterwards, the participants completed memory tests on both the trivia answers and faces that were presented during the scan.

Finding 1: What are the brain areas that support curiosity?

Curiosity influences activities in the nucleus accumbens and substantia nigra/ventral tegmental area complex.

Finding 2: How does curiosity benefit learning of interesting material?

Curiosity-driven memory benefits were driven by anticipatory (of trivia answers) rather than processing (memory tests) of such interesting answers.

Finding 3: How does a curious state modulate learning of incidental items?

Analysis shows that “the relationship between hippocampal memory-predicting activity and curiosity-driven memory benefits was mediated by midbrain memory-predicting activity”. This is in the SN/ VTA and hippocampus.

Finding 4: Do curiosity-driven memory benefits for incidental events persist after a long delay?

After 1 day of delay, participants recalled more answers to high-curiosity trivia questions and higher confident face recognition during states of high curiosity.

The findings show that there is a neural mechanism between curiosity, learning and memory. One important limitation is that the trivia question task might only approximate the effects of a person’s true interests and motivation to learn. We can not conclude that for all different kinds of tasks the same benefit exists. It could be smaller or greater.

Gruber and Ranganath (2019) integrate the current evidence (including above) on curiosity and propose the Prediction, Appraisal, Curiosity, and Exploration (PACE) framework. It includes research on curiosity from psychology, cognitive neuroscience, and systems neuroscience. “PACE proposes that curiosity is triggered by significant prediction errors that are appraised. This cycle enhances memory encoding through increased attention, exploration, and information seeking and enhances the consolidation of information acquired while in a curious state through dopaminergic neuromodulation of the hippocampus.”

There are two types of predictions: information-based prediction error and context-based prediction error. The former is caused by cognitive conflicts such as conflicts in requirements, which stimulate curiosity to find information to resolve the conflicts. This processing involves the anterior cingulate cortex. The latter is caused by new information or events that is unexpected, such as arriving at a novel place or noticing furniture has been moved, which, for example, people will spend more time visually exploring the scene. This processing involves the hippocampus.

Appraisal of the prediction error happens next and the lateral prefrontal cortex is involved in the processing. If the appraisal determines that one does not have the capacity to resolve the prediction error, it could trigger anxiety (via amygdala) and behavioral inhibition (such as procrastination). If deemed capable, curiosity and exploration (information seeking) could be triggered (dopaminergic midbrain and dopamine release). Dopamine enhances memory encoding and consolidation.

Trait curiosity of individuals does modulate the effect of curiosity. The authors also note that “the underlying neural substrates of curiosity-related memory enhancements are much more complex than what [they] have proposed.”

Curiosity cycles #

Exploration can trigger prediction errors which continues the PACE cycle. Here are some possible questions to ask during curiosity-driving state while learning.

1. Prediction. Are there any cognitive conflicts? For example, conflicting requirements or competing actors.
2. Appraisal. What resources (knowledge, people, financial) do I have access to?
3. Curiosity and exploration. Let curiosity guide exploration. Repeat questions 1 and 2.

References #

• Gruber, M. J., Gelman, B. D., & Ranganath, C. (2014). States of Curiosity Modulate Hippocampus-Dependent Learning via the Dopaminergic Circuit. Neuron, 84(2), 486–496. https://doi.org/10.1016/j.neuron.2014.08.060
• Gruber, M. J., & Ranganath, C. (2019). How Curiosity Enhances Hippocampus-Dependent Memory: The Prediction, Appraisal, Curiosity, and Exploration (PACE) Framework. Trends in Cognitive Sciences, 23(12), 1014–1025. https://doi.org/10.1016/j.tics.2019.10.003