Temporal dynamics of categorization: forgetting as the basis of abstraction and generalization

Abstract

Historically, models of categorization have focused on how learners track frequencies and co-occurrence information to abstract relevant category features for generalization. The current study takes a different approach by examining how the temporal dynamics of categorization affect abstraction and generalization. In the learning phase of the experiment, all relevant category features were presented an equal number of times across category exemplars. However, the relevant features were presented on one of two learning schedules: massed or interleaved. At a series of immediate and delayed tests, learners were asked to generalize to novel exemplars that contained massed features, interleaved features, or all novel features. The results of this experiment revealed that, at an immediate test, learners more readily generalized based upon features presented on a massed schedule. Conversely, at a delayed test, learners more readily generalized based upon features presented on an interleaved schedule, until information was no longer readily retrievable from memory. These findings suggest that forgetting and retrieval processes engendered by the temporal dynamics of learning are used as a basis of abstraction, implicating forgetting as a central mechanism of generalization.

The effects of learning and retrieval contexts on cross-situational word learning

“…On average, participants in the Non-themed conditions learned over 16 word-object pairs (out of 24). This learning rate was comparable to a design used in Romberg & Yu [19], in which participants learned around 16 word-object pairs (out of 18) after being exposed to each pair 12 times. Moreover, participants in the Themed conditions, on average, learned over 20 word-object pairs, a rate higher than seen in many other studies.…”

“…This finding pointed to the possibility that semantically-themed learning contexts facilitated word learning. However, since words referring to semantically-related objects were also presented in consecutive trials in that study, it was not clear whether the blocked presentation also contributed to the observed facilitative effect, as massed presentation has been shown to enhance performance in immediate tests [19]. Another noteworthy finding of that study was that artificially induced categories (i.e., semantically-unrelated objects that co-occurred frequently) had similar, albeit weaker, facilitative effect.…”

Moving Word Learning to a Novel Space: A Dynamic Systems View of Referent Selection and Retention

“…In a similar vein, the work reviewed above on the use of space to bind names to objects, as well as related work (see, e.g., Richardson & Kirkham, 2004;Vlach & Kalish, 2014), highlights that understanding the role of visual memory can lead to novel insights into how children solve referential ambiguity. Critically, theories of early word learning must understand multiple senses of visual memory: how visual memory modulates attention via habituation to novelty and how visual memory is used to build representations that track which objects are where.…”

A Test of Three Models of Character Reversal in Typically Developing Children’s Writing

“…Third, mirror confusion could then occur because the perceived orientation information is lost in the memory code. Because this forgetting leads to non-discrimination of an image from its mirror, it appears as a generalization process, forgetting being a central mechanism of generalization (Vlach and Kalish, 2014). The credibility of this temporary forgetting is strengthened by sensitivity to mirror reversals in an earlier visual processing object-selective region, the lateral occipital sulcus, followed by tolerance to mirror reversals in one object-selective region, the posterior fusiform sulcus (Dilks et al, 2011).…”