Where does reversal collide?

In behavioral neuroscience, reversal learning is what an organism must do when a well-learned rule is flipped: having learned that option A is rewarded and B is not, the contingencies are swapped, and the speed with which it stops choosing A and switches to B indexes cognitive and behavioural flexibility. Thompson's 1954 Discrimination learning and habit reversal studied this relearning, and Izquierdo and colleagues' 2017 The neural basis of reversal learning reviews the circuitry that supports it. The referent is the capacity to update behaviour when contingencies change, measured by errors after a switch, and it is a different construct from a price reversal in markets or a reversing entry in the ledger.[1][2]

Unit

The cost, in errors or trials, of relearning after a rule is reversed

Recovery means

Not applicable

Measured by

Quantified by perseverative errors and trials-to-criterion after the reward contingencies are switched

In behavioral finance, a reversal is the tendency of securities to undo their prior relative performance: portfolios of prior-period losers tend to outperform prior-period winners over the following years, a pattern read as a correction of investors' overreaction to news. De Bondt and Thaler's 1985 Does the stock market overreact? documented this long-horizon reversal and launched the overreaction literature behind contrarian strategies. The referent is a predictable turn in relative returns, measured over subsequent holding periods, and it is unrelated to an animal switching a learned rule or an accountant undoing an adjusting entry.[3]

Unit

The subsequent return difference between prior losers and prior winners

Recovery means

Not applicable

Measured by

Estimated from long-horizon returns of loser-minus-winner portfolios formed on past performance