On language and connectionism: analysis of a parallel distributed processing model of language acquisition

By Steven Pinker. 1988

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Relational inductive biases, deep learning, and graph networks (2018, background): “…connectionist (Rumelhart et al., 1987) forebears were faced with analogous critiques from structured, symbolic positions (Fodor and Pylyshyn, 1988; Pinker and Prince, 1988), there was a constructive eﬀort (Bobrow and Hinton, 1990; Marcus, 2001) to address the challenges directly and carefully.”
Deep Learning: A Critical Appraisal (2018, background): “…10 Deep learning’s predecessors were vulnerable to similar problems, as Pinker and Prince (1988)pointed out, in a 11 discussion of neural networks that produced bizarre past tense forms for a subset of its inputs.”
Rethinking eliminative connectionism. (1998, methodology): “For example, an extension of the RAAM networks of Pollack (1990) that was proposed by Chalmers (1990) and Niklasson and Gelder (1994) uses two separate networks, one to encode (and decode) all possible instances of variable in a fixed input bank,”
Recursive Distributed Representations (1990, background): “¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ 2 To point out this "Banana Problem" with Rumelhart & McClelland's actual representation, which was phonological rather than orthographic, Pinker and Prince [24] discovered words with enough internal duplication in the Oykangand language.”
Darwin's mistake: Explaining the discontinuity between human and nonhuman minds (2008, unknown): “…to generalize rule-like relations over abstract categorical variables, to reason in an inferentially coherent fashion, and to use the artificial symbols of a natural language in a systematic, recursive, and generative manner (Fodor & Pylyshyn 1988; Marcus 2001; Newell 1980; Pinker & Prince 1988).”
The neural architecture of language: Integrative modeling converges on predictive processing (2020, background): “Cognitive scientists have long treated neural network models of language processing with skepticism (Marcus, 2018; Pinker & Prince, 1988) given that these systems lack (and often deliberately attempt to do without) explicit symbolic representation – traditionally seen as a core feature of…”
The Acquisition of Anaphora by Simple Recurrent Networks (2013, background): “There has been some work in this connection in the domains of phonology and morphology, most famously the work spawned by Rumelhart and McClelland’s (1986) model of the English past tense (see, for example, Pinker & Prince 1988, Pinker & Ullman 2002, and McClelland & Patterson 2002, for discussion).”
Connectionism and cognitive architecture: a critical analysis. (1988, background): “This is, for example, one way of understanding what’s going on in the argument between Rumelhart and McClelland (1986b) and Prince and Pinker (1988), though neither paper puts it in quite these terms.”
Generalization in Interactive Networks: The Benefits of Inhibitory Competition and Hebbian Learning (2001, background): “Such results should help to counter the persistent claims that neural networks are incapable of producing systematic behavior based on statistical learning of the environment (e.g., Marcus, 1998; Pinker & Prince, 1988).”