A symbolic-connectionist theory of relational inference and generalization.

By John E. Hummel, Keith J. Holyoak. 2003

555 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Cognitive Neuroscience, Cultural Studies

How researchers see this paper

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

In Artificial Intelligence

In Cognitive Neuroscience

In Cultural Studies

Citations in the corpus

Relational inductive biases, deep learning, and graph networks (2018, background): “We draw analogies by aligning the relational structure between two domains and drawing inferences about one based on corresponding knowledge about the other (Gentner and Markman, 1997; Hummel and Holyoak, 2003).”
Relation learning in a neurocomputational architecture supports cross-domain transfer (2019, background): “During relational generalization propositions with unmapped elements enter the driver, and any propositions to which they map enter the recipient (as information is generalized from the driver to recipient; see Hummel & Holyoak, 2003).”
TARGET ARTICLE: Duality Models in Social Psychology: From Dual Processes to Interacting Systems (2006, unknown): “The IS lacks an important functional capability: It cannot flexibly combine concepts by abstract relations, which is the prerequisite for a number of higher order cognitive capabilities (Hummel & Holyoak, 2003).”
Teaching the Conceptual Structure of Mathematics (2012, background): “Moreover, even if a good source analog is provided, relational comparisons tax limited working memory (Halford, 1993 ; Hummel & Holyoak, 1997 , 2003 ; Waltz, Lau, Grewal, & Holyoak, 2000 ).”
Learning Through Case Comparisons: A Meta-Analytic Review (2013, background): “Prior work has shown that analogy making across multiple cases affords the opportunity of acquiring a schematic representation that includes the common features of the cases but does not include features unique to each given case (Gick & Holyoak, 1983; Hummel & Holyoak, 1997, 2003).”
Using matrices to model symbolic relationship (2008, background): “Some of the existing connectionist models for representing and learning relations and analogies [2, 4] are able to detect new relations and to represent hierarchical relations of high complexity.”
The Import of Knowledge Export: Connecting Findings and Theories of Transfer of Learning (2012, background): “…that emerged during the “cognitive revolution” of the 1960s (see Gardner, 1987), and much of the psychological research on transfer has become tightly linked with symbolic cognitive approaches, particularly with theories of analogical reasoning (e.g., Gentner, 1983; Hummel & Holyoak, 1997, 2003).”
Storage capacity explains fluid intelligence but executive control does not (2012, background): “…new relational structures, which seem to be crucial for both deductive reasoning (e.g., integration of the premises into mental models; Johnson-Laird, 1999) and inductive thinking (e.g., mapping elements of source and target during analogy-making; Hummel & Holyoak, 2003; Waltz et al., 1999).”
Models of Semantic Memory (2015, background): “As Hummel and Holyoak (2003) have noted, this type of intuitive modeling may have serious consequences: “The problem of hand-coded representations is the most serious problem facing computational modeling as a scientific enterprise.”