Free Association Norms Research Articles

Contrasting the semantic typology biases of Deaf and hearing nonsigners in their conceptualization of time and space

AbstractThe mental lexicon offers a window into the configuration of conceptual domains such as space and time, which has been labeled as concrete the former and abstract the latter in the current embodiment approach to cognition. Space has a phonological and semantic value in sign languages, but not in spoken languages. Additionally, the representation of time by spatial means is robust in oral and sign languages. This research asks if Deaf signers and hearing nonsigners have the same conceptual organization of those domains. In their respective languages, sixty-two participants made a repeated free word association task. These results showed that the studied populations have a little overlap in the associates evocated for each clue. The analysis of the preferences of the semantic relations of the pairs clue-associate showed a greater tendency of the Deaf signers to establish thematic relations. In contrast, the hearing participants indicated a bias toward taxonomic relations. The results suggest that the abstractness or concreteness of concepts may be modulated by factors associated with linguistic modalities. However, in this compared free association norms factors related to the language deprivation of Deaf, the asymmetries in the cross-modal language contact and cross-modal borrowing were not exhaustively controlled.

The contribution of affective content to cue-response correspondence in a word association task: Focus on emotion words and emotion-laden words

AbstractThis study aimed at examining the contribution of affective content to the organization of words in the lexicon. Based on existing free association norms and on a series of questionnaires we developed, we examined the characteristics of the words produced as associates to 840 Spanish cue words. Half of them were affective words and the other half were neutral (non-affective) words. Among the affective cue words, some words directly labeled an emotion (emotion words, EM) and others did not label an emotion but could elicit it (emotion-laden words, EL). The words produced as associates were also classified according to this distinction. Furthermore, we examined the relationship between the lexico-semantic and affective properties of the cue words and the associated words. The results revealed that EM, EL, and neutral associated words were elicited to a greater extent by cue words of the same type than by other types of cue words. Furthermore, the degree of correspondence between the affective properties of the cues and their associates was higher than that of lexico-semantic variables. These results have methodological implications for research on semantic memory and are of interest for applied studies focused on affective word organization in specific populations.

Free association in a neural network.

Free association among words is a fundamental and ubiquitous memory task. Although distributed semantics (DS) models can predict the association between pairs of words, and semantic network (SN) models can describe transition probabilities in free association data, there have been few attempts to apply established cognitive process models of memory search to free association data. Thus, researchers are currently unable to explain the dynamics of free association using memory mechanisms known to be at play in other retrieval tasks, such as free recall from lists. We address this issue using a popular neural network model of free recall, the context maintenance and retrieval (CMR) model, which we fit using stochastic gradient descent on a large data set of free association norms. Special cases of CMR mimic existing DS and SN models of free association, and we find that CMR outperforms these models on out-of-sample free association data. We also show that training CMR on free association data generates improved predictions for free recall from lists, demonstrating the value of free association for the study of many different types of memory phenomena. Overall, our analysis provides a new account of the dynamics of free association, predicts free association with increased accuracy, integrates theories of free association with established models of memory, and shows how large data sets and neural network training methods can be used to model complex cognitive processes that operate over thousands of representations. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

False memory in a second language: The importance of controlling the knowledge of word meaning.

In the globalized world we live in, it is increasingly common for people to speak more than one language. Although research in psychology has been widely interested in the study of false memories with the Deese/Roediger-McDermott (DRM) paradigm, to date, there is a scarcity of studies comparing false memories in the first and the second language (L1 and L2, respectively). It is noteworthy that one of the most studied variables in the DRM paradigm, the backward associative strength (BAS), has hardly been studied in the L2. Moreover, the only study that recently examined this matter found differences in the knowledge of L2-word meaning between the high-BAS and low-BAS lists, which would hinder the interpretation of the BAS effect in L2 false memories. Taking all this into account, the current work examined false memories in the L1 (Spanish) and the L2 (English) as a function of BAS overcoming the limitations of the previous study. We selected DRM lists using both Spanish and English free association norms and lists were constructed to vary in BAS values while controlling the knowledge of word meaning. Results showed that false recognition was greater in the L1 or dominant language than in the L2 or non-dominant language. Furthermore, BAS modulated the false recognition in both the L1 and the L2. That is, false recognition was higher in high-BAS than low-BAS lists in both languages. Sensitivity index from the signal-detection theory helped us gain further insight into these results. The main findings are discussed in the light of theoretical models from both the false memory and the second language processing literature. Finally, practical implications and future research are provided.

The "Small World of Words" free association norms for Rioplatense Spanish.

Large-scale word association datasets are both important tools used in psycholinguistics and used as models that capture meaning when considered as semantic networks. Here, we present word association norms for Rioplatense Spanish, a variant spoken in Argentina and Uruguay. The norms were derived through a large-scale crowd-sourced continued word association task in which participants give three associations to a list of cue words. Covering over 13,000 words and +3.6 M responses, it is currently the most extensive dataset available for Spanish. We compare the obtained dataset with previous studies in Dutch and English to investigate the role of grammatical gender and studies that used Iberian Spanish to test generalizability to other Spanish variants. Finally, we evaluated the validity of our data in word processing (lexical decision reaction times) and semantic (similarity judgment) tasks. Our results demonstrate that network measures such as in-degree provide a good prediction of lexical decision response times. Analyzing semantic similarity judgments showed that results replicate and extend previous findings demonstrating that semantic similarity derived using spreading activation or spectral methods outperform word embeddings trained on text corpora.

Lexical connectivity effects in immediate serial recall of words.

In six experiments, we tested whether immediate serial recall is influenced by a word's degree centrality, an index of lexical connectivity. Words of high degree centrality are associated with more words in free association norms than those of low degree centrality. Experiment 1 analyzed secondary data to explore the effect of degree centrality in wordlists containing a mixture of high- and low-degree words. High-degree words were advantaged across all serial positions, independently of other variables including word frequency. Experiment 2 replicated this finding using an expanded stimulus set. Experiment 3 used pure lists with each list containing high- or low-degree words only (e.g., HHHHHH vs. LLLLLL). Once again, high-degree words were better recalled across all serial positions. In Experiment 4, each wordlist alternated between high and low-degree words (e.g., HLHLHL and LHLHLH). Recall of low-degree words was facilitated by the neighboring high-degree words, abolishing the overall high-degree advantage. Experiment 5 used a within-participant design and replicated the findings from Experiments 3 and 4 such that the high-degree advantage in pure lists disappeared in alternating lists. Experiment 6 compared high and low frequency words in pure lists while controlling for degree centrality between the item sets. A high-frequency advantage emerged, suggesting that the effects of frequency and degree centrality are separable. We conclude that degree centrality is a distinct psycholinguistic variable that affects serial recall as both (a) an item-level characteristic such that high (vs. low) degree words have greater accessibility in the lexicon and (b) an interitem property such that high-degree words facilitate the recall of neighboring words by enhancing the formation of associative links. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Semantic Memory Search and Retrieval in a Novel Cooperative Word Game: A Comparison of Associative and Distributional Semantic Models.

Considerable work during the past two decades has focused on modeling the structure of semantic memory, although the performance of these models in complex and unconstrained semantic tasks remains relatively understudied. We introduce a two-player cooperative word game, Connector (based on the boardgame Codenames), and investigate whether similarity metrics derived from two large databases of human free association norms, the University of South Florida norms and the Small World of Words norms, and two distributional semantic models based on large language corpora (word2vec and GloVe) predict performance in this game. Participant dyads were presented with 20-item word boards with word pairs of varying relatedness. The speaker received a word pair from the board (e.g., exam-algebra) and generated a one-word semantic clue (e.g., math), which was used by the guesser to identify the word pair on the board across three attempts. Response times to generate the clue, as well as accuracy and latencies for the guessed word pair, were strongly predicted by the cosine similarity between word pairs and clues in random walk-based associative models, and to a lesser degree by the distributional models, suggesting that conceptual representations activated during free association were better able to capture search and retrieval processes in the game. Further, the speaker adjusted subsequent clues based on the first attempt by the guesser, who in turn benefited from the adjustment in clues, suggesting a cooperative influence in the game that was effectively captured by both associative and distributional models. These results indicate that both associative and distributional models can capture relatively unconstrained search processes in a cooperative game setting, and Connector is particularly suited to examine communication and semantic search processes.

Similarity Judgment Within and Across Categories: A Comprehensive Model Comparison.

Similarity is one of the most important relations humans perceive, arguably subserving category learning and categorization, generalization and discrimination, judgment and decision making, and other cognitive functions. Researchers have proposed a wide range of representations and metrics that could be at play in similarity judgment, yet have not comprehensively compared the power of these representations and metrics for predicting similarity within and across different semantic categories. We performed such a comparison by pairing nine prominent vector semantic representations with seven established similarity metrics that could operate on these representations, as well as supervised methods for dimensional weighting in the similarity function. This approach yields a factorial model structure with 126 distinct representation-metric pairs, which we tested on a novel dataset of similarity judgments between pairs of cohyponymic words in eight categories. We found that cosine similarity and Pearson correlation were the overall best performing unweighted similarity functions, and that word vectors derived from free association norms often outperformed word vectors derived from text (including those specialized for similarity). Importantly, models that used human similarity judgments to learn category-specific weights on dimensions yielded substantially better predictions than all unweighted approaches across all types of similarity functions and representations, although dimension weights did not generalize well across semantic categories, suggesting strong category context effects in similarity judgment. We discuss implications of these results for cognitive modeling and natural language processing, as well as for theories of the representations and metrics involved in similarity.

Norming studies of European Portuguese words: A literature review

Words are widely used as stimuli in cognitive and linguistic research. As words may vary on various domains (e.g., lexicosemantic and affective), which can influence performance in many ways, it is essential to select them carefully. However, databases of European Portuguese words are still relatively scarce, and their presentation is spread in various sources. The aims of this work were: (1) to provide a review of the databases containing information on lexicosemantic, affective, and free association norming data for European Portuguese words published since the year 2000; (2) to summarize the definitions of the different variables; and (3) to combine in a single database the norming data available at this point. We found 25 databases with 18 containing lexicosemantic and affective norming data and seven containing word free association norms. These databases include words from various categories (e.g., nouns and adjectives), characterized on different variables (e.g., familiarity, imagery, valence, age of acquisition and animacy). This review provides a useful tool for researchers looking for Portuguese word-related resources. A thorough review of the databases, including some details about each one, is presented. The combined database is shared with the scientific community via the Open Science Framework: https://osf.io/9ta3y/.

The effect of language proficiency and associative strength on false memory.

In two experiments we examined the role of language proficiency and associative strength in the production of false memory. We constructed Deese-Roediger-McDermott lists using both Spanish and English free association norms. Lists were constructed to vary in backward associative strength (BAS). Experiment 1 participants were native Spanish speakers with some proficiency in English while Experiment 2 participants were native Spanish participants that had either high, intermediate, or low English proficiency. Results showed that, in both Experiment 1 and Experiment 2, false recognition was greater in participants' dominant language (L1 or Spanish) than in their non-dominant language (L2 or English), and false recognition in L2 increased with L2 proficiency when low-BAS lists were studied (Experiment 2). Further, false recognition was higher in high-BAS lists than in low-BAS lists in both L1 and L2. Finally, we collected a measure of participants' knowledge of our stimulus words in L2. These data showed that participants had far from perfect knowledge of all L2 stimuli. Analyses that factored out the effects of L2 word knowledge failed to alter the effects of L1 vs. L2, L2 proficiency and BAS on false recognition.

First steps towards the Lithuanian word association database

This paper introduces the first version of the Lithuanian database of free association norms. This is an attempt to provide an open-access resource, which would be helpful for psycholinguists, linguists, computational linguists, and students. This version of the database includes 277 cue word forms. The responses were collected from 304 participants. In total 15,612 association pairs were recorded. The paper presents the procedure of collecting free associations and additional data available for researchers. It also provides a list of all cue words with their five most frequent associates and some summary statistics.

The role of language proficiency in producing false memories

Abstract We report three experiments examining the role that language proficiency plays in the production of false memory. We constructed Deese-Roediger-McDermott paradigm lists using both English and Spanish free association norms, which enabled us to control the associations between studied items and critical words. Experiment 1 showed that native English speakers who were learning Spanish produced more false memory when DRM critical words were studied and tested in English compared to Spanish. Experiment 2 showed that native Spanish speakers who were learning English produced more false memory when DRM critical words were studied and tested in Spanish compared to English. Experiment 3 showed that native Spanish speakers who were highly proficient in English produced more false memory for DRM critical words studied and tested in English compared to native Spanish speakers who were lower in English proficiency. Collectively, these results support the role that the automaticity of concept access plays in producing false memory.

Brain morphometry predicts individual creative potential and the ability to combine remote ideas

For complex mental functions such as creative thinking, inter-individual variability is useful to better understand the underlying cognitive components and brain anatomy. Associative theories propose that creative individuals have flexible semantic associations, which allows remote elements to be formed into new combinations. However, the structural brain variability associated with the ability to combine remote associates has not been explored. To address this question, we performed a voxel-based morphometry (VBM) study and explored the anatomical connectivity of significant regions. We developed a Remote Combination Association Task adapted from Mednick's test, in which subjects had to find a solution word related to three cue words presented to them. In our adaptation of the task, we used free association norms to quantify the associative distance between the cue words and solution words, and we varied this distance. The tendency to solve the task with insight and the ability to evaluate the appropriateness of a proposed solution were also analysed. Fifty-four healthy volunteers performed this task and underwent a structural MRI. Structure–function relationships were analysed using regression models between grey matter (GM) volume and task performance. Significant clusters were mapped onto an atlas of white matter (WM) tracts. The ability to solve the task, which depended on the associative distance of the solution word, was associated with structural variation in the left rostrolateral prefrontal and posterior parietal regions; the left rostral prefrontal region was connected to distant regions through long-range pathways. By using a creative combination task in which the semantic distance between words varied, we revealed a brain network centred on the left frontal pole that appears to support the ability to combine information in new ways by bridging the semantic distance between pieces of information.

An Association Network for Computing Semantic Relatedness

To judge how much a pair of words (or texts) are semantically related is acognitive process. However, previous algorithms for computing semanticrelatedness are largely based on co-occurrences within textualwindows, and do not actively leverage cognitive human perceptions ofrelatedness. To bridge this perceptional gap, we propose to utilizefree association as signals to capture such human perceptions.However, free association, being manually evaluated,has limited lexical coverage and is inherently sparse. We propose to expand lexical coverage and overcome sparseness by constructing an association network of terms and concepts that combines signals from free association norms and five types of co-occurrences extracted from therich structures of Wikipedia. Our evaluation results validate thatsimple algorithms on this network give competitive results incomputing semantic relatedness between words and between shorttexts.

Priming and competition of associated memory representations: A comparison between response times and event-related potentials following lesions to left temporal cortex

Priming and competition of associated memory representations: A comparison between response times and event-related potentials following lesions to left temporal cortex

Hidden processes in structural representations: A reply to Abbott, Austerweil, and Griffiths (2015).

In recent work exploring the semantic fluency task, we found evidence indicative of optimal foraging policies in memory search that mirror search in physical environments. We determined that a 2-stage cue-switching model applied to a memory representation from a semantic space model best explained the human data. Abbott, Austerweil, and Griffiths demonstrate how these patterns could also emerge from a random walk applied to a network representation of memory based on human free-association norms. However, a major representational issue limits any conclusions that can be drawn about the process model comparison: Our process model operated on a memory space constructed from a learning model, whereas their model used human behavioral data from a task that is quite similar to the behavior they attempt to explain. Predicting semantic fluency (e.g., how likely it is to say cat after dog in a sequence of animals) from free association (how likely it is to say cat when given dog as a cue) should be possible with a relatively simple retrieval mechanism. The 2 tasks both tap memory, but they also share a common process of retrieval. Assuming that semantic memory is a network from free-association behavior embeds variance due to the shared retrieval process directly into the representation. A simple process mechanism is then sufficient to simulate semantic fluency because much of the requisite process complexity may already be hidden in the representation. (PsycINFO Database Record

How activation, entanglement, and searching a semantic network contribute to event memory

Free-association norms indicate that words are organized into semantic/associative neighborhoods within a larger network of words and links that bind the net together. We present evidence indicating that memory for a recent word event can depend on implicitly and simultaneously activating related words in its neighborhood. Processing a word during encoding primes its network representation as a function of the density of the links in its neighborhood. Such priming increases recall and recognition and can have long-lasting effects when the word is processed in working memory. Evidence for this phenomenon is reviewed in extralist-cuing, primed free-association, intralist-cuing, and single-item recognition tasks. The findings also show that when a related word is presented in order to cue the recall of a studied word, the cue activates the target in an array of related words that distract and reduce the probability of the target's selection. The activation of the semantic network produces priming benefits during encoding, and search costs during retrieval. In extralist cuing, recall is a negative function of cue-to-distractor strength, and a positive function of neighborhood density, cue-to-target strength, and target-to-cue strength. We show how these four measures derived from the network can be combined and used to predict memory performance. These measures play different roles in different tasks, indicating that the contribution of the semantic network varies with the context provided by the task. Finally, we evaluate spreading-activation and quantum-like entanglement explanations for the priming effects produced by neighborhood density.

Evocation: analyzing and propagating a semantic link based on free word association

Studies of lexical---semantic relations aim to understand the mechanism of semantic memory and the organization of the mental lexicon. However, standard paradigmatic relations such as hypernym and hyponym cannot capture connections among concepts from different parts of speech. WordNet, which organizes synsets (i.e., synonym sets) using these lexical---semantic relations, is rather sparse in its connectivity. According to WordNet statistics, the average number of outgoing/incoming arcs for the hypernym/hyponym relation per synset is 1.33. Evocation, defined as how much a concept (expressed by one or more words) brings to mind another, is proposed as a new directed and weighted measure for the semantic relatedness among concepts. Commonly applied semantic relations and relatedness measures do not seem to be fully compatible with data that reflect evocations among concepts. They are compatible but evocation captures MORE. This work aims to provide a reliable and extendable dataset of concepts evoked by, and evoking, other concepts to enrich WordNet, the existing semantic network. We propose the use of disambiguated free word association data (first responses to verbal stimuli) to infer and collect evocation ratings. WordNet aims to represent the organization of mental lexicon, and free word association which has been used by psycholinguists to explore semantic organization can contribute to the understanding. This work was carried out in two phases. In the first phase, it was confirmed that existing free word association norms can be converted into evocation data computationally. In the second phase, a two-stage association-annotation procedure of collecting evocation data from human judgment was compared to the state-of-the-art method, showing that introducing free association can greatly improve the quality of the evocation data generated. Evocation can be incorporated into WordNet as directed links with scales, and benefits various natural language processing applications.

The contributions of language and experience to the representation of abstract and concrete words: Different weights but similar organizations

In the present studies, we evaluated the contributions of language and sensorimotor information to the representation of abstract and concrete words and the possibility that the organizations of the two types of concepts follow different organizational principles: association, for abstract concepts, and semantic similarity, for concrete concepts. In Study 1, we examined the two strongest associates of concrete and abstract words from published free association norms. Study 2 then extended this analysis to individual data collected with a free association task. Language associations were more important for abstract than for concrete words, but for sensorimotor information no differences were observed between the two types of concepts. Also, no clear evidence was found for different qualitative organizational principles for abstract and concrete concepts. Multiple representational systems thus seem to be engaged in the conceptual processing of abstract and concrete concepts, while it remains to be investigated whether their representations follow different organizational principles.

COMPLEX STRUCTURES AND SEMANTICS IN FREE WORD ASSOCIATION

We investigate the directed and weighted complex network of free word associations in which players write a word in response to another word given as input. We analyze in details two large datasets resulting from two very different experiments: On the one hand the massive multiplayer web-based Word Association Game known as Human Brain Cloud, and on the other hand the South Florida Free Association Norms experiment. In both cases, the networks of associations exhibit quite robust properties like the small world property, a slight assortativity and a strong asymmetry between in-degree and out-degree distributions. A particularly interesting result concerns the existence of a characteristic scale for the word association process, arguably related to specific conceptual contexts for each word. After mapping, the Human Brain Cloud network onto the WordNet semantics network, we point out the basic cognitive mechanisms underlying word associations when they are represented as paths in an underlying semantic network. We derive in particular an expression describing the growth of the HBC graph and we highlight the existence of a characteristic scale for the word association process.