Modelling serial position curves with temporal distinctiveness

We offer a critique of the temporal distinctiveness model of serial positio n effects (Nairne, Neath, Serra, & Byun, 1997). The temporal distinctivenes s model combines a precise definition of stimulus distinctiveness with a me mory perturbation process. The critique is empirically motivated-we show th at with a more complete analysis, the temporal distinctiveness model does n ot adequately account for Nairne et al.'s experimental data. To better acco unt for the data, we independently modified two components of Nairne et al. 's model: the mathematical form of the definition of temporal distinctivene ss and the mathematical form of the mapping from distinctiveness to free-re call probabilities. Both of these modifications provided for better fits. Y et both Nairne et al.'s definition and our modified definition are fairly a rbitrary. We show that a significant challenge to this approach is to find theoretically motivated constraints of the temporal distinctiveness model w hile providing for adequate fits to data.