Fig. 2.

Simulation of the Kamin blocking effect applied to acoustic cue learning, as in Experiment 1. The connection weight (y-axis) over trials (x-axis) is shown for the nasality cue to the diminutive outcome in the control (VOT) pre-training condition (left) and blocking pre-training condition (right). In the blocking condition, a single cue is first learned as a predictor to a particular outcome, in this case tone as a cue to diminutive size (pre-training phase). Later (Phase 2; onset indicated by the vertical dashed line), a second cue is presented simultaneously with the original cue: tone and nasality as cues to diminutive size. The control condition is the same except that a control cue (VOT) is presented in the pre-training phase, instead of a critical cue. In Phase 2, the connection weight of nasality increases in the control condition, but remains low in the blocking condition. In the blocking condition, learning of the second cue, nasality, is ‘blocked’ by the first cue, tone. The simulation assumes equal salience of tone and nasality cues.