The leading models of learning in repeated choice tasks rest on the
assumption that decision makers tend to select the alternatives that led to the
best recent outcomes. The current research highlights three boundaries of this
"recency assumption." Analysis of the stock market and simple
laboratory experiments suggests that positively surprising obtained payoffs, and
negatively surprising forgone payoffs reduced the rate of repeating the
previous choice. In addition all previous outcomes, but the most recent, have
similar effect of future choices. We show that these results, and other robust
properties of decisions from experience, can be captured with a simple addition
to the leading models: the assumption that surprise triggers change.
