English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Many researchers and policymakers have expressed excitement about algorithmicexplanations enabling more fair and responsible decision-making. However,recent experimental studies have found that explanations do not always improvehuman use of algorithmic advice. In this study, we shed light on how peopleinterpret and respond to counterfactual explanations (CFEs) -- explanationsthat show how a model's output would change with marginal changes to itsinput(s) -- in the context of pretrial risk assessment instruments (PRAIs). Weran think-aloud trials with eight sitting U.S. state court judges, providingthem with recommendations from a PRAI that includes CFEs. We found that theCFEs did not alter the judges' decisions. At first, judges misinterpreted thecounterfactuals as real -- rather than hypothetical -- changes to defendants.Once judges understood what the counterfactuals meant, they ignored them,stating their role is only to make decisions regarding the actual defendant inquestion. The judges also expressed a mix of reasons for ignoring or followingthe advice of the PRAI without CFEs. These results add to the literaturedetailing the unexpected ways in which people respond to algorithms andexplanations. They also highlight new challenges associated with improvinghuman-algorithm collaborations through explanations.

"If it didn't happen, why would I change my decision?": How Judges Respond to Counterfactual Explanations for the Public Safety Assessment