| Abstract | In a character decision task, phonetic compound targets (composed of a semantic radical and a
phonetic component) followed primes that shared (a) the target’s radical and were semantically
related (R1S1), (b) the target’s radical and were not semantically related (R1S2), (c) no radical but
were semantically related (R2S1), and (d) no radical and were not semantically related (R2S2).
Target radicals also varied as to the number of compounds in which they appeared (i.e., combin-ability). When targets followed primes immediately (Experiment 1; SOA 243 ms), target latencies following R1S2 primes were slowed relative to R2S2 controls but those following R1S1 and R2S1 primes were facilitated equivalently. Increases in combinability significantly reduced decision latencies. When 10 items separated primes and targets (Experiment 2), facilitation was evident only after R1S1 primes. Results indicate that one type of component, the semantic radical, is processed in the course of Chinese character recognition and that orthographic similarity due to repetition of a radical is not an adequate account. |