Related Experiment Video
Updated: Jun 26, 2025

Errors as a Means of Reducing Impulsive Food Choice
Published on: June 5, 2016
Surprise!-Clarifying the link between insight and prediction error
Maxi Becker1, Xinhao Wang2, Roberto Cabeza2,3
1Department of Psychology, Humboldt University Berlin, Berlin, Germany. maxi_becker@gmx.net.
Abstract:
The AHA experience, a moment of deep understanding during insightful problem-solving involving feelings of certainty, pleasure, and surprise, has captivated psychologists for more than a century. Recently, a new theoretical framework has proposed a link between the AHA experience and prediction error (PE), a popular concept in decision-making and reinforcement learning. This framework suggests that participants maintain a meta-cognitive prediction about the time it takes to solve a problem and the AHA experience arises when the problem is solved earlier than expected, resulting in a meta-cognitive PE. In our preregistered online study, we delved deeper into this idea, investigating whether prediction errors also pertain to participants' predictions regarding the solvability of the problem itself, and which dimension of the AHA experience aligns with the meta-cognitive PE. Utilizing verbal insight problems, we found a positive association between the AHA experience and the meta-cognitive PE, specifically in regards to problem solvability. Specifically, the element of surprise, a critical AHA dimension, emerged as a key indicator of the meta-cognitive PE, while other dimensions-such as pleasure, certainty, and suddenness-showed no signs for similar relationships, with suddenness exhibiting a negative correlation with meta-cognitive PE. This new finding provides further evidence that aspects of the AHA experience, surprise in particular, correspond to a meta-cognitive PE. The finding also underscores the multifaceted nature of this phenomenon, linking insights with learning theories and enhancing our understanding of this intriguing phenomenon.
More Related Videos
Related Concept Videos
Hindsight Biases
Prediction Intervals
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
Accuracy and Errors in Hypothesis Testing
In hypothesis testing, the probability of making a Type I error, denoted as α, is commonly set at 0.05. This significance level indicates a 5%...
Errors In Hypothesis Tests
Cause and Effect
Variation
When independent and dependent variables are plotted on a scatter plot, the slope of a line is a value that describes the rate of change between the two...

