This article shows that the Korean Extrinsic Plural Marker (EPM) may be acquired by children on the basis of very little evidence. The EPM marks distributivity, unlike the Instrinsic Plural Marker, which marks plurality. Thirty monolingual learners of Korean aged 5;03 to 6;09 (mean age 6;01) were tested using a series of Truth Value Judgment Tasks (Crain & Thornton 1998). The children were split into three groups: Experimental-1, Experimental-2, and Control. All children received a pretest, establishing that they are unaware of the properties of EPM. Children in the two experimental groups then received two interventions, each employing a single instance of a felicitous interaction between a child and a mother involving the EPM. One intervention presented positive evidence of the meaning of the EPM, and the other presented negative evidence. Each intervention was followed by an assessment of the effect of the intervention (using TVJT tests similar to the pretest). The experimental groups differed in the order of the interventions, and the control group received no interventions but otherwise received exactly the same regimen of tests as the experimental groups. The results show that fourteen of the twenty experimental children responded to the intervention that involved positive evidence; none responded to the intervention involving negative evidence; and none of the children in the control group acquired the meaning of the EPM. We conclude that inflectional patterns may be acquired on the basis of very little evidence, perhaps employing mechanisms of fast mapping akin to those found in the domain of the lexicon.