Exemplar and prototype models revisited: response strategies, selective attention, and stimulus generalization.

摘要: J. D. Smith and colleagues (J. P. Minda & Smith, 2001; Minda, 1998, 2000; M. Murray, 1997) presented evidence that they claimed challenged the predictions of exemplar models supported prototype models. In authors’ view, this confounded issue nature category representation with type response rule (probabilistic vs. deterministic) was used. Also, their designs did not test whether correctly predicted generalization performance. The present work demonstrates an model includes a response-scaling mechanism provides natural account all et al.’s experimental results. Furthermore, predicts classification performance better than when novel transfer stimuli are included in designs. A classic cognitive psychology concerns manner which people represent categories memory. According to (Homa, 1984; Posner Keele, 1968; Reed, 1972), by forming summary is central tendency experienced members category. Classification decisions based on similarity item alternative prototypes. By contrast, according (Hintzman, 1986; Medin Schaffer, 1978; Nosofsky, 1986), storing individual or exemplars as separate traces classify items these stored exemplars.