Why some surprises are more surprising than others: Surprise as a metacognitive sense of explanatory difficulty
|Title:||Why some surprises are more surprising than others: Surprise as a metacognitive sense of explanatory difficulty||Authors:||Keane, Mark T.
Foster, Meadhbh I.
|Permanent link:||http://hdl.handle.net/10197/8532||Date:||Sep-2015||Abstract:||Early theories of surprise, including Darwin's, argued that it was predominantly a basic emotion. Recently, theories have taken a more cognitive view of surprise, casting it as a process of 'making sense of surprising events'. The current paper advances the view that the essence of this sense-making process is explanation; specifically, that people's perception of surprise is a metacognitive estimate of the cognitive work involved in explaining an abnormal event. So, some surprises are more surprising because they are harder to explain. This proposal is tested in eight experiments that explore how (i) the contents of memory can influence surprise, (ii) different classes of scenarios can retrieve more/less relevant knowledge from memory to explain surprising outcomes, (iii) how partial explanations constrain the explanation process, reducing surprise, and (iv) how, overall, any factor that acts to increase the cognitive work in explaining a surprising event, results in higher levels of surprise (e.g., task demands to find three rather than one explanations). Across the present studies, using different materials, paradigms and measures, it is consistently and repeatedly found that the difficulty of explaining a surprising outcome is the best predictor for people’s perceptions of the surprisingness of events. Alternative accounts of these results are considered, as are future directions for this research.||Type of material:||Journal Article||Publisher:||Elsevier||Copyright (published version):||2015 Elsevier||Keywords:||Machine learning;Statistics;Surprise judgments;Comprehension;Explanation;Difficulty||DOI:||10.1016/j.cogpsych.2015.08.004||Language:||en||Status of Item:||Peer reviewed|
|Appears in Collections:||Computer Science Research Collection|
Insight Research Collection
Show full item record
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. For other possible restrictions on use please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.