The Cognitive Neuroscience of Stable and Flexible Semantic Typicality

Abstract

Typicality effects are among the most well-studied phenomena in the study of concepts. The classical notion of typicality is that typical concepts share many features with category co-members and few features with members of contrast categories. However, this notion was challenged by evidence that typicality is highly context dependent and not always dependent on central tendency. Dieciuc and Folstein (2019) argued that there is strong evidence for both views and that the two types of typicality effects might depend on different mechanisms. A recent theoretical framework, the controlled semantic cognition framework (Lamdon Ralph et al., 2017) strongly emphasizes the classical view, but includes mechanisms that could potentially account for both kinds of typicality. In contrast, the situated cognition framework (Barsalou, 2009b) articulates the context-dependent view. Here, we review evidence from cognitive neuroscience supporting the two frameworks. We also briefly evaluate the ability of computational models associated with the CSC to account for phenomena supporting SitCog (Rogers and McClelland, 2004). Many predictions of both frameworks are borne out by recent cognitive neuroscience evidence. While the CSC framework can at least potentially account for many of the typicality phenomena reviewed, challenges remain, especially with regard to ad hoc categories.

Keywords: categorization; cognitive neuroscience; semantic cognition; situated cognition; typicality.

Figure 1

Cartoon illustration of the hub and spoke model. The hub (central circle) represents stable similarity relations between concepts based on the semantic features that each concept activates. In the left-hand panel, a sparrow is used as input. A sparrow is a typical bird, because it has similar features to other birds. The representation for the name “bird” is amplified because it is reinforced by other similar birds. In the right-hand panel, the less typical penguin is used as input. The “bird” label is more weakly activated because other birds are dissimilar to penguins.

Figure 2

Cartoon illustration of the Rumelhart model. The representation layer has the same properties as the hub, shown in Figure 1. Similarity within the hidden layer changes depending on the desired semantic information. Penguins are similar to other birds if naming is required (top, ISa context) but more similar to fish when functional properties are required.

Figure 3

Cartoon illustration of the situated cognition framework. Input, including task requirements and, in this case, words, results in situation simulations, not only of the concept, but of a context in which the concept is embedded. Typicality judgments are driven by the match between concept and context. Note that the simulation poodle in the third panel is possible (cf. Barsalou, 1999) and was indeed required for the creation of the figure, but it is unlikely when the word “dog” is a cue, rendering dogs far less typical in this context than the other two contexts. Note also that the posture of the dogs in the first two panels (standing vs. sitting) is determined by the context rather than the frequency or “averageness” of the respective postures, reflecting sensitivity to causal interactions within situations.