More Like this

1. Impact of Semantic Representations on Analogical Mapping with Transitive Relations

   Linford, B. ; Ichien, N. ; Holyoak, K. J. ; Lu, H. ( January 2022 , Proceedings of the 44th Annual Meeting of the Cognitive Science Society)

   Analogy problems involving multiple ordered relations of the same type create mapping ambiguity, requiring some mechanism for relational integration to achieve mapping accuracy. We address the question of whether the integration of ordered relations depends on their logical form alone, or on semantic representations that differ across relation types. We developed a triplet mapping task that provides a basic paradigm to investigate analogical reasoning with simple relational structures. Experimental results showed that mapping performance differed across orderings based on category, linear order, and causal relations, providing evidence that each transitive relation has its own semantic representation. Hence, human analogical mapping of ordered relations does not depend solely on their formal property of transitivity. Instead, human ability to solve mapping problems by integrating relations relies on the semantics of relation representations. We also compared human performance to the performance of several vector-based computational models of analogy. These models performed above chance but fell short of human performance for some relations, highlighting the need for further model development. 

   more » « less
2. Relation representations in analogical reasoning and recognition memory

   Ichien, N. ; Alfred, K. L. ; Baia, S. ; Kraemer, D. J. ; Bunge, S. A. ; Lu, H. ; Holyoak, K. J. ( January 2022 , Proceedings of the 44th Annual Meeting of the Cognitive Science Society)

   Many computational models of reasoning rely on explicit relation representations to account for human cognitive capacities such as analogical reasoning. Relational luring, a phenomenon observed in recognition memory, has been interpreted as evidence that explicit relation representations also impact episodic memory; however, this assumption has not been rigorously assessed by computational modeling. We implemented an established model of recognition memory, the Generalized Context Model (GCM), as a framework for simulating human performance on an old/new recognition task that elicits relational luring. Within this basic theoretical framework, we compared representations based on explicit relations, lexical semantics (i.e., individual word meanings), and a combination of the two. We compared the same alternative representations as predictors of accuracy in solving explicit verbal analogies. In accord with previous work, we found that explicit relation representations are necessary for modeling analogical reasoning. In contrast, preliminary simulations incorporating model parameters optimized to fit human data reproduce relational luring using any of the alternative representations, including one based on non-relational lexical semantics. Further work on model comparisons is needed to examine the contributions of lexical semantics and relations on the luring effect in recognition memory. 

   more » « less
3. Generative Inferences in Relational and Analogical Reasoning: A Comparison of Computational Models

   Ichien, N. ; Kan, A. ; Holyoak, K. J. ; Lu, H. ( January 2022 , Proceedings of the 44th Annual Meeting of the Cognitive Science Society)

   A key property of human cognition is its ability to generate novel predictions about unfamiliar situations by completing a partially-specified relation or an analogy. Here, we present a computational model capable of producing generative inferences from relations and analogs. This model, BART-Gen, operates on explicit representations of relations learned by BART (Bayesian Analogy with Relational Transformations), to achieve two related forms of generative inference: reasoning from a single relation, and reasoning from an analog. In the first form, a reasoner completes a partially-specified instance of a stated relation (e.g., robin is a type of ____). In the second, a reasoner completes a target analog based on a stated source analog (e.g., sedan:car :: robin:____). We compare the performance of BART-Gen with that of BERT, a popular model for Natural Language Processing (NLP) that is trained on sentence completion tasks and that does not rely on explicit representations of relations. Across simulations and human experiments, we show that BART-Gen produces more human-like responses for generative inferences from relations and analogs than does the NLP model. These results demonstrate the essential role of explicit relation representations in human generative reasoning. 

   more » « less
4. Emergence of analogy from relation learning

   https://doi.org/10.1073/pnas.1814779116  

   Lu, Hongjing ; Wu, Ying Nian ; Holyoak, Keith J. ( February 2019 , Proceedings of the National Academy of Sciences)

   By middle childhood, humans are able to learn abstract semantic relations (e.g., antonym, synonym, category membership) and use them to reason by analogy. A deep theoretical challenge is to show how such abstract relations can arise from nonrelational inputs, thereby providing key elements of a protosymbolic representation system. We have developed a computational model that exploits the potential synergy between deep learning from “big data” (to create semantic features for individual words) and supervised learning from “small data” (to create representations of semantic relations between words). Given as inputs labeled pairs of lexical representations extracted by deep learning, the model creates augmented representations by remapping features according to the rank of differences between values for the two words in each pair. These augmented representations aid in coping with the feature alignment problem (e.g., matching those features that make “love-hate” an antonym with the different features that make “rich-poor” an antonym). The model extracts weight distributions that are used to estimate the probabilities that new word pairs instantiate each relation, capturing the pattern of human typicality judgments for a broad range of abstract semantic relations. A measure of relational similarity can be derived and used to solve simple verbal analogies with human-level accuracy. Because each acquired relation has a modular representation, basic symbolic operations are enabled (notably, the converse of any learned relation can be formed without additional training). Abstract semantic relations can be induced by bootstrapping from nonrelational inputs, thereby enabling relational generalization and analogical reasoning.

    

   more » « less
5. Eye gaze patterns reveal how reasoning skills improve with experience

   https://doi.org/10.1038/s41539-018-0035-8  

   Guerra-Carrillo, Belén_C ; Bunge, Silvia_A ( October 2018 , npj Science of Learning)

   Reasoning, our ability to solve novel problems, has been shown to improve as a result of learning experiences. However, the underlying mechanisms of change in this high-level cognitive ability are unclear. We hypothesized that possible mechanisms include improvements in the encoding, maintenance, and/or integration of relations among mental representations – i.e., relational thinking. Here, we developed several eye gaze metrics to pinpoint learning mechanisms that underpin improved reasoning performance. We collected behavioral and eyetracking data from young adults who participated in a Law School Admission Test preparation course involving word-based reasoning problems or reading comprehension. The Reasoning group improved more than the Comprehension group on a composite measure of four visuospatial reasoning assessments. Both groups improved similarly on an eyetracking paradigm involving transitive inference problems, exhibiting faster response times while maintaining high accuracy levels; nevertheless, the Reasoning group exhibited a larger change than the Comprehension group on an ocular metric of relational thinking. Across the full sample, individual differences in response time reductions were associated with increased efficiency of relational thinking. Accounting for changes in visual search and a more specific measure of relational integration improved the prediction accuracy of the model, but changes in these two processes alone did not adequately explain behavioral improvements. These findings provide evidence of transfer of learning across different kinds of reasoning problems after completing a brief but intensive course. More broadly, the high temporal precision and rich derivable parameters of eyetracking make it a powerful approach for probing learning mechanisms.

    

   more » « less