The best game in town: The reemergence of the language-of-thought hypothesis across the cognitive sciences

Abstract

Mental representations remain the central posits of psychology after many decades of scrutiny. However, there is no consensus about the representational format(s) of biological cognition. This paper provides a survey of evidence from computational cognitive psychology, perceptual psychology, developmental psychology, comparative psychology, and social psychology, and concludes that one type of format that routinely crops up is the language of thought (LoT). We outline six core properties of LoTs: (i) discrete constituents; (ii) role-filler independence; (iii) predicate-argument structure; (iv) logical operators; (v) inferential promiscuity; and (vi) abstract content. These properties cluster together throughout cognitive science. Bayesian computational modeling, compositional features of object perception, complex infant and animal reasoning, and automatic, intuitive cognition in adults all implicate LoT-like structures. Instead of regarding LoT as a relic of the previous century, researchers in cognitive science and philosophy of mind must take seriously the explanatory breadth of LoT-based architectures. We grant that the mind may harbor many formats and architectures, including iconic and associative structures as well as deep-neural-network-like architectures. However, as computational/representational approaches to the mind continue to advance, classical compositional symbolic structures—i.e., LoTs—only prove more flexible and well-supported over time.

Undercompression errors as evidence for conceptual primitives

“…In other words, the MFA assumes that structured meaningful representations can be available to a child (as well as non-human animals) before learning a language. A recent review article by Quilty-Dunn et al (2022) discusses several pieces of evidence from both infants and non-human animals' studies, among other data, in favor of a language of thought as one of the representational formats of cognition and thus in line with our view that structured meaningful representation can be available independently of language. In fact, research has accumulated showing that infants, as young as 4-month-old can use abstract content and reason about it under certain conditions (when their attention is attracted to the relevant dimension through priming, e.g., Lin et al, 2021 or when some categories are relevant for them or salient, e.g., Bonatti et al, 2002;Surian and Caldi, 2010), suggesting that failures noticed in the literatures are due to testing conditions, rather than to lack of competence (Stavans et al, 2019).…”

Towards a unifying theory of linguistic meaning

“…This way symbolic conceptual/cognitive representations can facilitate the operation of cognitive tasks in perception, reasoning, planning, etc. in relation to things in the outside world (see [ 22 ].…”

Break It Down: Evidence for Structural Compositionality in Neural Networks

“…Compositionality has long been touted as a key property of human cognition, enabling humans to exhibit flexible and abstract language processing and visual processing, among other cognitive processes (Marcus, 2003;Piantadosi et al, 2016;Lake et al, 2017;Smolensky et al, 2022). According to common definitions (Quilty-Dunn et al, 2022;Fodor & Lepore, 2002), a rep-resentation system is compositional if it implements a set of discrete constituent functions which exhibit some degree of modularity. That is, blue circle is represented compositionally if a system is able to entertain the concept blue independently of circle, and vice-versa.…”

“…It is an open question whether neural networks require explicit symbolic mechanisms to implement compositional solutions, or whether they implicitly learn compositional solutions during training. Historically, neural networks have been considered non-compositional systems, instead solving tasks by matching new inputs to memorized or "iconic" representations (Marcus, 2003;Quilty-Dunn et al, 2022). Neural networks' apparent lack of compositionality has served as a key point in favor of integrating explicit symbolic mechanisms into artificial intelligence systems (Andreas et al, 2016;Koh et al, 2020;Ellis et al, 2020).…”