Understanding spoken words involves a rapid mapping from speech to conceptual representations. One distributed feature-based conceptual account assumes that the statistical characteristics of concepts’ features—the number of concepts they occur in and likelihood of co-occurrence —determine conceptual activation. To test these claims, we investigated the role of distinctiveness/sharedness and correlational strength in speech-to-meaning mapping, using a lexical decision task and computational simulations. Responses were faster for concepts with higher sharedness, suggesting that shared features are facilitatory in tasks like lexical decision (...) that require access to them. Correlational strength facilitated responses for slower participants, suggesting a time-sensitive co-occurrence-driven settling mechanism. The computational simulation showed similar effects, with early effects of shared features and later effects of correlational strength. These results support a general-to-specific account of conceptual processing, whereby early activation of shared features is followed by the gradual emergence of a specific target representation. (shrink)

Traditional methods for deriving property-based representations of concepts from text have focused on either extracting only a subset of possible relation types, such as hyponymy/hypernymy (e.g., car is-a vehicle) or meronymy/metonymy (e.g., car has wheels), or unspecified relations (e.g., car—petrol). We propose a system for the challenging task of automatic, large-scale acquisition of unconstrained, human-like property norms from large text corpora, and discuss the theoretical implications of such a system. We employ syntactic, semantic, and encyclopedic information to guide our extraction, (...) yielding concept-relation-feature triples (e.g., car be fast, car require petrol, car cause pollution), which approximate property-based conceptual representations. Our novel method extracts candidate triples from parsed corpora (Wikipedia and the British National Corpus) using syntactically and grammatically motivated rules, then reweights triples with a linear combination of their frequency and four statistical metrics. We assess our system output in three ways: lexical comparison with norms derived from human-generated property norm data, direct evaluation by four human judges, and a semantic distance comparison with both WordNet similarity data and human-judged concept similarity ratings. Our system offers a viable and performant method of plausible triple extraction: Our lexical comparison shows comparable performance to the current state-of-the-art, while subsequent evaluations exhibit the human-like character of our generated properties. (shrink)

Abstract Since their inception, distributional models of semantics have been criticized as inadequate cognitive theories of human semantic learning and representation. A principal challenge is that the representations derived by distributional models are purely symbolic and are not grounded in perception and action; this challenge has led many to favor feature-based models of semantic representation. We argue that the amount of perceptual and other semantic information that can be learned from purely distributional statistics has been underappreciated. We compare the representations (...) of three feature-based and nine distributional models using a semantic clustering task. Several distributional models demonstrated semantic clustering comparable with clustering-based on feature-based representations. Furthermore, when trained on child-directed speech, the same distributional models perform as well as sensorimotor-based feature representations of children’s lexical semantic knowledge. These results suggest that, to a large extent, information relevant for extracting semantic categories is redundantly coded in perceptual and linguistic experience. Detailed analyses of the semantic clusters of the feature-based and distributional models also reveal that the models make use of complementary cues to semantic organization from the two data streams. Rather than conceptualizing feature-based and distributional models as competing theories, we argue that future focus should be on understanding the cognitive mechanisms humans use to integrate the two sources. (shrink)

Vector-based models of word meaning have become increasingly popular in cognitive science. The appeal of these models lies in their ability to represent meaning simply by using distributional information under the assumption that words occurring within similar contexts are semantically similar. Despite their widespread use, vector-based models are typically directed at representing words in isolation, and methods for constructing representations for phrases or sentences have received little attention in the literature. This is in marked contrast to experimental evidence (e.g., in (...) sentential priming) suggesting that semantic similarity is more complex than simply a relation between isolated words. This article proposes a framework for representing the meaning of word combinations in vector space. Central to our approach is vector composition, which we operationalize in terms of additive and multiplicative functions. Under this framework, we introduce a wide range of composition models that we evaluate empirically on a phrase similarity task. (shrink)

In this prcis we focus on phenomena central to the reaction against similarity-based theories that arose in the 1980s and that subsequently motivated the approach to semantic knowledge. Specifically, we consider (1) how concepts differentiate in early development, (2) why some groupings of items seem to form or coherent categories while others do not, (3) why different properties seem central or important to different concepts, (4) why children and adults sometimes attest to beliefs that seem to contradict their direct experience, (...) (5) how concepts reorganize between the ages of 4 and 10, and (6) the relationship between causal knowledge and semantic knowledge. The explanations our theory offers for these phenomena are illustrated with reference to a simple feed-forward connectionist model. The relationships between this simple model, the broader theory, and more general issues in cognitive science are discussed. (shrink)

O'Brien & Opie's (O&O's) theory demands a view of unconscious processing that is incompatible with virtually all current PDP models of neural activity. Relative to the alternatives, the theory is closer to an AI than a parallel distributed processing (PDP) perspective, and its treatment of phenomenology is ad hoc. It raises at least one important question: Could features of network relaxation be the “switch” that turns an unconscious into a conscious network?

The semantic fluency task is a widely used clinical tool in the diagnostic process of Alzheimer's disease. The task requires efficient mapping of the semantic space to produce as many items as possible within a semantic category. We examined whether healthy volunteers and patients with early Alzheimer's disease take advantage of and travel in the semantic space differently. With focus on the animal fluency task, we sought to emulate the detailed structure of the multidimensional semantic space by utilizing word2vec-method from (...) the natural language processing domain. To render the resulting multidimensional semantic space visually comprehensible, we applied a dimensionality reduction algorithm, which enabled a straightforward division of the semantic space into sub-categories. Moving in semantic space was quantified with the number of items created, sub-categories visited, and switches and returns to these sub-categories. Multinomial logistic regression models were used to predict the diagnostic group with these independent variables. We found that returning to a sub-category provided additional information, besides the number of words produced in the task, to differentiate patients with Alzheimer's dementia from both amnestic Mild Cognitive Impairment patients and healthy controls. The results suggest that the frequency of returning to a sub-category may serve as an additional aid for clinicians in diagnosing early Alzheimer's disease. Moreover, our results imply that the combination of word2vec and subsequent t-SNE-visualization may offer a valuable tool for examining the semantic space and its sub-categories. (shrink)

Topics in Cognitive Science, Volume 14, Issue 1, Page 54-77, January 2022.

Generic sentences (e.g., bare plural sentences such as “dogs have four legs” and “mosquitoes carry malaria”) are used to talk about kinds of things. Three experiments investigated the conceptual foundations of generics as well as claims within the formal semantic approaches to generics concerning the roles of prevalence, cue validity and normalcy in licensing generics. Two classes of generic sentences that pose challenges to both the conceptually based and formal semantic approaches to generics were investigated. Striking property generics (e.g. “sharks (...) bite swimmers”) are true even though only a tiny minority of instances have the property and thus pose obvious problems for quantificational approaches, and they also do not seem to characterize kinds in terms of the principled or statistical connections investigated in previous research ( Prasada and Dillingham, 2006 ; Prasada and Dillingham, 2009). The second class — minority characteristic generics (e.g. “ducks lay eggs”) — also poses serious problems for quantificational accounts, and appears to involve principled connections even though fewer than half of its instances have the relevant property. The experiments revealed three principal discoveries: first, striking generics involve neither principled nor statistical connections. Instead, they involve a causal connection between a kind and a property. Second, minority characteristic generics exhibit the characteristics of principled connections, which suggests that principled connections license the expectation that most instances will have the property, but do not require it. Finally, the experiments also provided evidence that prevalence and the acceptability of generics may be dissociated and provided data that are problematic for normalcy approaches to generics, and for the idea that cue validity licenses low prevalence generics. As such, the studies provided evidence in favor of a conceptually based approach to the semantics of generics ( Leslie, 2007 ; Leslie, 2008; see also Carlson, 2009). (shrink)

The semantic fluency task is commonly used as a measure of one’s ability to retrieve semantic concepts. While performance is typically scored by counting the total number of responses, the ordering of responses can be used to estimate how individuals or groups organize semantic concepts within a category. I provide an overview of this methodology, using Alzheimer’s disease as a case study for how the approach can help advance theoretical questions about the nature of semantic representation. However, many open questions (...) surrounding the validity and reliability of this approach remain unresolved. (shrink)

The kappa effect is a well-reported phenomenon in which spatial distance between discrete stimuli affects the perception of temporal distance demarcated by the corresponding stimuli. Here, we report a new phenomenon that we propose to designate as thelexical kappa effectin which word distance, a non-magnitude relationship of discrete stimuli that exists in the lexical space of the mental lexicon, affects the perception of temporal distance. A temporal bisection task was used to assess the subjective perception of the time interval demarcated (...) by two successively presented words. Word distance was manipulated by varying the semantic (Experiment 1) or phonological (Experiment 2) similarity between the two words. Results showed that the temporal distance between the two words was perceived to be shorter when the corresponding two words were lexically closer. We explain this effect within the internal clock framework by assuming faster detection of the word that terminated timing when it is preceded by a semantically or phonologically similar word. (shrink)

Various aspects of semantic features drive early vocabulary development, but less is known about how the global and local structure of the overall semantic feature space influences language acquisition. A feature network of English words was constructed from a large database of adult feature production norms such that edges in the network represented feature distances between words (i.e., Manhattan distances of probability distributions of features elicited for each pair of words). A word's global feature distinctiveness is measured with respect to (...) all other words in the network and a word's local feature distinctiveness is measured relative to words in sub‐networks derived from clustering analyses. This paper investigates how feature distinctiveness of individual words at local and global scales of the network influences language acquisition. Regression analyses indicate that global feature distinctiveness was associated with earlier age of acquisition ratings, and was a stronger predictor of age of acquisition than local feature distinctiveness. These results suggest that the global structure of the semantic feature network could play an important role in language acquisition, whereby globally distinctive concepts help to structure vocabulary development over the lifespan. (shrink)

Patterns of object naming often differ between languages, but bilingual speakers develop convergent naming patterns in their two languages that are distinct from those of monolingual speakers of each language. This convergence appears to reflect interactions between lexical representations for the two languages. In this study, we developed a self-organizing connectionist model to simulate semantic convergence in the bilingual lexicon and investigate the mechanisms underlying this semantic convergence. We examined the similarity of patterns in the simulated data to empirical data (...) from past research, and we identified how semantic convergence was manifested in the simulated bilingual lexical knowledge. Furthermore, we created impaired models in which components of the network were removed so as to examine the importance of the relevant components on bilingual object naming. Our results demonstrate that connections between two languages’ lexicons can be established through the simultaneous activations of related words in the two languages. These connections between languages allow the outputs of their lexicons to become more similar, that is, to converge. Our model provides a basis for future computational studies of how various input variables may affect bilingual naming patterns. (shrink)

We identify a number of principles with respect to prediction that, we argue, underpin adult language comprehension: (a) comprehension consists in realizing a mapping between the unfolding sentence and the event representation corresponding to the real‐world event being described; (b) the realization of this mapping manifests as the ability to predict both how the language will unfold, and how the real‐world event would unfold if it were being experienced directly; (c) concurrent linguistic and nonlinguistic inputs, and the prior internal states (...) of the system, each drive the predictive process; (d) the representation of prior internal states across a representational substrate common to the linguistic and nonlinguistic domains enables the predictive process to operate over variable time frames and variable levels of representational abstraction. We review empirical data exemplifying the operation of these principles and discuss the relationship between prediction, event structure, thematic role assignment, and incrementality. (shrink)

This study presents original evidence that abstract and concrete concepts are organized and represented differently in the mind, based on analyses of thousands of concepts in publicly available data sets and computational resources. First, we show that abstract and concrete concepts have differing patterns of association with other concepts. Second, we test recent hypotheses that abstract concepts are organized according to association, whereas concrete concepts are organized according to (semantic) similarity. Third, we present evidence suggesting that concrete representations are more (...) strongly feature-based than abstract concepts. We argue that degree of feature-based structure may fundamentally determine concreteness, and we discuss implications for cognitive and computational models of meaning. (shrink)

Recent metaphor research has revealed that metaphor comprehension involves both categorization and comparison processes. This finding has triggered the following central question: Which property determines the choice between these two processes for metaphor comprehension? Three competing views have been proposed to answer this question: the conventionality view (Bowdle & Gentner, 2005), aptness view (Glucksberg & Haught, 2006b), and interpretive diversity view (Utsumi, 2007); these views, respectively, argue that vehicle conventionality, metaphor aptness, and interpretive diversity determine the choice between the categorization (...) and comparison processes. This article attempts to answer the question regarding which views are plausible by using cognitive modeling and computer simulation based on a semantic space model. In the simulation experiment, categorization and comparison processes are modeled in a semantic space constructed by latent semantic analysis. These two models receive word vectors for the constituent words of a metaphor and compute a vector for the metaphorical meaning. The resulting vectors can be evaluated according to the degree to which they mimic the human interpretation of the same metaphor; the maximum likelihood estimation determines which of the two models better explains the human interpretation. The result of the model selection is then predicted by three metaphor properties (i.e., vehicle conventionality, aptness, and interpretive diversity) to test the three views. The simulation experiment for Japanese metaphors demonstrates that both interpretive diversity and vehicle conventionality affect the choice between the two processes. On the other hand, it is found that metaphor aptness does not affect this choice. This result can be treated as computational evidence supporting the interpretive diversity and conventionality views. (shrink)

O'Brien & Opie do not address the question of the psychotherapeutic role of unconscious representational states such as beliefs. A dilemma is proposed: if they accept the legitimacy of such states then they should modify what they say about dissociation, and if they do not, they owe us an account of why.

Information management systems improve the retention of information in large collections. As such they act as memory prostheses, implying an ideal basis in human memory models. Since humans process information by association, and situate it in the context of space and time, systems should maximize their effectiveness by mimicking these functions. Since human attentional capacity is limited, systems should scaffold cognitive efforts in a comprehensible manner. We propose the Principles of Mnemonic Associative Knowledge (P-MAK), which describes a framework for semantically (...) identifying, organizing, and retrieving information, and for encoding episodic events by time and stimuli. Inspired by prominent human memory models, we propose associative networks as a preferred representation. Networks are ideal for their parsimony, flexibility, and ease of inspection. Networks also possess topological properties—such as clusters, hubs, and the small world—that aid analysis and navigation in an information space. Our cognitive perspective addresses fundamental problems faced by information management systems, in particular the retrieval of related items and the representation of context. We present evidence from neuroscience and memory research in support of this approach, and discuss the implications of systems design within the constraints of P-MAK’s principles, using text documents as an illustrative semantic domain. (shrink)

An attractor network was trained to compute from word form to semantic representations that were based on subject‐generated features. The model was driven largely by higher‐order semantic structure. The network simulated two recent experiments that employed items included in its training set (McRae and Boisvert, 1998). In Simulation 1, short stimulus onset asynchrony priming was demonstrated for semantically similar items. Simulation 2 reproduced subtle effects obtained by varying degree of similarity. Two predictions from the model were then tested on human (...) subjects. In Simulation 3 and Experiment 1, the items from Simulation 1 were reversed, and both the network and subjects showed minimally different priming effects in the two directions. In Experiment 2, consistent with attractor networks but contrary to a key aspect of hierarchical spreading activation accounts priming was determined by featural similarity rather than shared superordinate category. It is concluded that semantic‐similarity priming is due to featural overlap that is a natural consequence of distributed representations of word meaning. (shrink)

Computational models of meaning trained on naturally occurring text successfully model human performance on tasks involving simple similarity measures, but they characterize meaning in terms of undifferentiated bags of words or topical dimensions. This has led some to question their psychological plausibility (Murphy, 2002;Schunn, 1999). We present here a fully automatic method for extracting a structured and comprehensive set of concept descriptions directly from an English part‐of‐speech‐tagged corpus. Concepts are characterized by weighted properties, enriched with concept–property types that approximate classical (...) relations such as hypernymy and function. Our model outperforms comparable algorithms in cognitive tasks pertaining not only to concept‐internal structures (discovering properties of concepts, grouping properties by property type) but also to inter‐concept relations (clustering into superordinates), suggesting the empirical validity of the property‐based approach. (shrink)

Conceptual knowledge is acquired through recurrent experiences, by extracting statistical regularities at different levels of granularity. At a fine level, patterns of feature co-occurrence are categorized into objects. At a coarser level, patterns of concept co-occurrence are categorized into contexts. We present and test CONCAT, a connectionist model that simultaneously learns to categorize objects and contexts. The model contains two hierarchically organized CALM modules (Murre, Phaf, & Wolters, 1992). The first module, the Object Module, forms object representations based on co-occurrences (...) between features. These representations are used as input for the second module, the Context Module, which categorizes contexts based on object co-occurrences. Feedback connections from the Context Module to the Object Module send activation from the active context to those objects that frequently occur within this context. We demonstrate that context feedback contributes to the successful categorization of objects, especially when bottom-up feature information is degraded or ambiguous. (shrink)

We propose that human reasoning relies on an inherence heuristic, an implicit cognitive process that leads people to explain observed patterns (e.g., girls wear pink) in terms of the inherent features of their constituents (e.g., pink is an inherently feminine color). We then demonstrate how this proposed heuristic can provide a unified account for a broad set of findings spanning areas of research that might at first appear unrelated (e.g., system justification, nominal realism, is–ought errors in moral reasoning). By revealing (...) the deep commonalities among the diverse phenomena that fall under its scope, our account is able to generate new insights into these phenomena, as well as new empirical predictions. A second main goal of this paper, aside from introducing the inherence heuristic, is to articulate the proposal that the heuristic serves as a foundation for the development of psychological essentialism. More specifically, we propose that essentialism—which is the common belief that natural and social categories are underlain by hidden, causally powerful “essences”—emerges over the first few years of life as an elaboration of the earlier, and more open-ended, intuitions supplied by the inherence heuristic. In the final part of the paper, we distinguish our proposal from competing accounts (e.g., Strevens' K-laws) and clarify the relationship between the inherence heuristic and related cognitive tendencies (e.g., the correspondence bias). In sum, this paper illuminates a basic cognitive process that emerges early in life and is likely to have profound effects on many aspects of human psychology. (shrink)

The structure of people’s conceptual knowledge of concrete nouns has traditionally been viewed as hierarchical (Collins & Quillian, 1969). For example, superordinate concepts (vegetable) are assumed to reside at a higher level than basic‐level concepts (carrot). A feature‐based attractor network with a single layer of semantic features developed representations of both basic‐level and superordinate concepts. No hierarchical structure was built into the network. In Experiment and Simulation 1, the graded structure of categories (typicality ratings) is accounted for by the flat (...) attractor network. Experiment and Simulation 2 show that, as with basic‐level concepts, such a network predicts feature verification latencies for superordinate concepts (vegetable ). In Experiment and Simulation 3, counterintuitive results regarding the temporal dynamics of similarity in semantic priming are explained by the model. By treating both types of concepts the same in terms of representation, learning, and computations, the model provides new insights into semantic memory. (shrink)

The commentaries reflect three core themes that pertain not just to our theory, but to the enterprise of connectionist modeling more generally. The first concerns the relationship between a cognitive theory and an implemented computer model. Specifically, how does one determine, when a model departs from the theory it exemplifies, whether the departure is a useful simplification or a critical flaw? We argue that the answer to this question depends partially upon the model's intended function, and we suggest that connectionist (...) models have important functions beyond the commonly accepted goals of fitting data and making predictions. The second theme concerns perceived in-principle limitations of the connectionist approach to cognition, and the specific concerns these perceived limitations raise for our theory. We argue that the approach is not in fact limited in the ways our critics suggest. One common misconception, that connectionist models cannot address abstract or relational structure, is corrected through new simulations showing directly that such structure can be captured. The third theme concerns the relationship between parallel distributed processing (PDP) models and structured probabilistic approaches. In this case we argue that there the difference between the approaches is not merely one of levels. Our PDP approach differs from structured statistical approaches at all of Marr's levels, including the characterization of the goals of cognitive computations, and of the representations and algorithms used. (shrink)

The human conceptual system comprises simulated information of sensorimotor experience and linguistic distributional information of how words are used in language. Moreover, the linguistic shortcut hypothesis predicts that people will use computationally cheaper linguistic distributional information where it is sufficient to inform a task response. In a pre‐registered category production study, we asked participants to verbally name members of concrete and abstract categories and tested whether performance could be predicted by a novel measure of sensorimotor similarity (based on an 11‐dimensional (...) representation of sensorimotor strength) and linguistic proximity (based on word co‐occurrence derived from a large corpus). As predicted, both measures predicted the order and frequency of category production but, critically, linguistic proximity had an effect above and beyond sensorimotor similarity. A follow‐up study using typicality ratings as an additional predictor found that typicality was often the strongest predictor of category production variables, but it did not subsume sensorimotor and linguistic effects. Finally, we created a novel, fully grounded computational model of conceptual activation during category production, which best approximated typical human performance when conceptual activation was allowed to spread indirectly between concepts, and when candidate category members came from both sensorimotor and linguistic distributional representations. Critically, model performance was indistinguishable from typical human performance. Results support the linguistic shortcut hypothesis in semantic processing and provide strong evidence that both linguistic and grounded representations are inherent to the functioning of the conceptual system. All materials, data, and code are available at https://osf.io/vaq56/. (shrink)

A view that has been gaining prevalence over the past decade is that the human conceptual system is malleable, dynamic, context‐dependent, and task‐dependent, that is, flexible. Within the flexible conceptual representation framework, conceptual representations are constructed ad hoc, forming a different, idiosyncratic instantiation upon each occurrence. In this review, we scrutinize the neurocognitive literature to better understand the nature of this flexibility. First, we identify some key characteristics of these representations. Next, we consider how these flexible representations are constructed by (...) addressing some of the open questions in this framework: We review the age‐old question of how to reconcile flexibility with the apparent need for shareable stable definitions to anchor meaning and come to mutual understanding, as well as some newer questions we find critical, namely, the nature of relations among flexible representations, the role of feature saliency in activation, and the viability of all‐or‐none feature activations. We suggest replacing the debate about the existence of a definitional stable core that is obligatorily activated with a question of the degree and probability of activation of the information constituting a conceptual representation. We rely on published works to suggest that (1) prior featural salience matters, (2) feature activation may be graded, and (3) Bayesian updating of prior information according to current demands offers a viable account of how flexible representations are constructed. This proposal provides a theoretical mechanism for incorporating a changing momentary context into a constructed representation, while still preserving some of the concept's constituent meaning. (shrink)

We compared the ability of three different contextual models of lexical semantic memory and of a simple associative model to predict the properties of semantic networks derived from word association norms. None of the semantic models were able to accurately predict all of the network properties. All three contextual models over-predicted clustering in the norms, whereas the associative model under-predicted clustering. Only a hybrid model that assumed that some of the responses were based on a contextual model and others on (...) an associative network successfully predicted all of the network properties and predicted a word's top five associates as well as or better than the better of the two constituent models. The results suggest that participants switch between a contextual representation and an associative network when generating free associations. We discuss the role that each of these representations may play in lexical semantic memory. Concordant with recent multicomponent theories of semantic memory, the associative network may encode coordinate relations between concepts, and contextual representations may be used to process information about more abstract concepts. (shrink)

Although language has long been regarded as a primarily arbitrary system, sound symbolism, or non-arbitrary correspondences between the sound of a word and its meaning, also exists in natural language. Previous research suggests that listeners are sensitive to sound symbolism. However, little is known about the specificity of these mappings. This study investigated whether sound symbolic properties correspond to specific meanings, or whether these properties generalize across semantic dimensions. In three experiments, native English-speaking adults heard sound symbolic foreign words for (...) dimensional adjective pairs and for each foreign word, selected a translation among English antonyms that either matched or mismatched with the correct meaning dimension. Listeners agreed more reliably on the English translation for matched relative to mismatched dimensions, though reliable cross-dimensional mappings did occur. These findings suggest that although sound symbolic properties generalize to meanings that may share overlapping semantic features, sound symbolic mappings offer semantic specificity. (shrink)

The Visual World Paradigm (VWP) presents listeners with a challenging problem: They must integrate two disparate signals, the spoken language and the visual context, in support of action (e.g., complex movements of the eyes across a scene). We present Impulse Processing, a dynamical systems approach to incremental eye movements in the visual world that suggests a framework for integrating language, vision, and action generally. Our approach assumes that impulses driven by the language and the visual context impinge minutely on a (...) dynamical landscape of attractors corresponding to the potential eye-movement behaviors of the system. We test three unique predictions of our approach in an empirical study in the VWP, and describe an implementation in an artificial neural network. We discuss the Impulse Processing framework in relation to other models of the VWP. (shrink)