A low-coercion, open-loop design provides preliminary evidence of visually guided discrimination and individual differences in two Neotropical otters (Lontra longicaudis)

A low-coercion, open-loop design can extend discrimination testing to captive animals that resist standard methods, such as the Neotropical otter (Lontra longicaudis), for which no data exist.We tested visually guided instrumental behaviour in two individuals using a two-choice task in which an incorrect first choice cost only a delay.One otter reached abovechance accuracy with full correction; the other did not exceed chance but reduced an initial side preference and corrected more often, differing in latency and item handling.The results give preliminary evidence of this capacity and a design suited to other hard-to-test species.