Taboo language across the globe: A multi-lab study.

The use of taboo words represents one of the most common and arguably universal linguistic behaviors, fulfilling a wide range of psychological and social functions. However, in the scientific literature, taboo language is poorly characterized, and how it is realized in different languages and populations remains largely unexplored. Here we provide a database of taboo words, collected from different linguistic communities (Study 1, N = 1046), along with their speaker-centered semantic characterization (Study 2, N = 455 for each of six rating dimensions), covering 13 languages and 17 countries from all five permanently inhabited continents. Our results show that, in all languages, taboo words are mainly characterized by extremely low valence and high arousal, and very low written frequency. However, a significant amount of cross-country variability in words' tabooness and offensiveness proves the importance of community-specific sociocultural knowledge in the study of taboo language.

The Persian Lexicon Project: minimized orthographic neighbourhood effects in a dense language.

In recent years large datasets of lexical processing times have been released for several languages, including English, French, Spanish, and Dutch. Such datasets have enabled us to study, compare, and model the global effects of many psycholinguistic measures such as word frequency, orthographic neighborhood (ON) size, and word length. We have compiled and publicly released a frequency and ON dictionary of 64,546 words and 1800 plausible NWs from a language that has been relatively little studied by psycholinguists: Persian. We have also collected visual lexical decision reaction times for 1800 Persian words and nonwords. Persian offers an interesting psycholinguistic environment for several reasons, including that it has few long words and has resultantly dense orthographic neighborhoods. These characteristics provide us with an opportunity to contrast how these factors affect lexical access by comparing them to several other languages. The results suggest that sensitivity to word length and orthographic neighbourhood may reflect the statistical structure of a particular language, rather than being a universal element of lexical processing. The dictionary and LDRT data are available from https://osf.io/tb4m6/ .

Sound symbolism shapes the English language: The maluma/takete effect in English nouns.

Sound symbolism refers to associations between language sounds (i.e., phonemes) and perceptual and/or semantic features. One example is the maluma/takete effect: an association between certain phonemes (e.g., /m/, /u/) and roundness, and others (e.g., /k/, /ɪ/) and spikiness. While this association has been demonstrated in laboratory tasks with nonword stimuli, its presence in existing spoken language is unknown. Here we examined whether the maluma/takete effect is attested in English, across a broad sample of words. Best-worst judgments from 171 university students were used to quantify the shape of 1,757 objects, from spiky to round. We then examined whether the presence of certain phonemes in words predicted the shape of the objects to which they refer. We found evidence that phonemes associated with roundness are more common in words referring to round objects, and phonemes associated with spikiness are more common in words referring to spiky objects. This represents an instance of iconicity, and thus nonarbitrariness, in human language.

A pompous snack: On the unreasonable complexity of the world's third-worst jokes.

Although studies of humour are as old as the Western academic tradition, most theories are too vague to allow for modelling and prediction of humour judgments. Previous work in modelling humour judgments has succeeded by focusing on the world's worst jokes: the slight humour of single nonwords (Westbury, Shaoul, Moroschan, & Ramscar, 2016) and single words (Westbury & Hollis, 2019). Here that work is extended to the world's third-worst jokes, adjective-noun pairs such as dancing dildo, flabby goldfish, and pompous snack. Participants used best-worst scaling to rate the humour of random word pairs. Those judgments were modelled using both linear regression and genetic programming, which is not constrained by assumptions of linearity. The linear regression models were as successful as the nonlinear models at predicting humour judgments, accounting for 27% of the variance in a 540-item validation set. Predictors associated only with the noun and with the relationship between the adjective and noun accounted for much more variance (over 14% each) than predictors associated only with the adjective (6.3%). Greater cosine distance of the adjective word2vec vector from the vectors of the shared neighbors of the noun and adjective is associated with higher humour ratings, whereas the opposite relationship is true for the noun. This captures a form of incongruity not seen in single items, by which neighbours of the adjective become unexpectedly relevant only when the noun brings them into focus. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

The role of number of items per trial in best-worst scaling experiments.

Best-worst scaling is a judgment format in which participants are presented with K items and must choose the best and worst items from that set, along some underlying latent dimension. Best-worst scaling has seen recent use in natural-language processing and psychology to collect lexical semantic norms. In such applications, four items have always been presented on each trial. The present study provides reasoning that values other than 4 might provide better estimates of latent values. The results from simulation experiments and behavioral research confirmed this: Both suggest that, in the general case, six items per trial better reduces errors in the latent value estimates.

Learning about things that never happened: A critique and refinement of the Rescorla-Wagner update rule when many outcomes are possible.

A vector-based model of discriminative learning is presented. It is demonstrated to learn association strengths identical to the Rescorla-Wagner model under certain parameter settings (Rescorla & Wagner, 1972, Classical Conditioning II: Current Research and Theory, 2, 64-99). For other parameter settings, it approximates the association strengths learned by the Rescorla-Wagner model. I argue that the Rescorla-Wagner model has conceptual details that exclude it as an algorithmically plausible model of learning. The vector learning model, however, does not suffer from the same conceptual issues. Finally, we demonstrate that the vector learning model provides insight into how animals might learn the semantics of stimuli rather than just their associations. Results for simulations of language processing experiments are reported.

Wriggly, squiffy, lummox, and boobs: What makes some words funny?

Theories of humor tend to be post hoc descriptions, suffering from insufficient operationalization and a subsequent inability to make predictions about what will be found humorous and to what extent. Here we build on the Engelthaler & Hills' (2017) humor rating norms for 4,997 words, by analyzing the semantic, phonological, orthographic, and frequency factors that play a role in the judgments. We were able to predict the original humor rating norms and ratings for previously unrated words with greater reliability than the split half reliability in the original norms, as estimated from splitting those norms along gender or age lines. Our findings are consistent with several theories of humor, while suggesting that those theories are too narrow. In particular, they are consistent with incongruity theory, which suggests that experienced humor is proportional to the degree to which expectations are violated. We demonstrate that words are judged funnier if they are less common and have an improbable orthographic or phonological structure. We also describe and quantify the semantic attributes of words that are judged funny and show that they are partly compatible with the superiority theory of humor, which focuses on humor as scorn. Several other specific semantic attributes are also associated with humor. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Conceptualizing syntactic categories as semantic categories: Unifying part-of-speech identification and semantics using co-occurrence vector averaging.

Co-occurrence models have been of considerable interest to psychologists because they are built on very simple functionality. This is particularly clear in the case of prediction models, such as the continuous skip-gram model introduced in Mikolov, Chen, Corrado, and Dean (2013), because these models depend on functionality closely related to the simple Rescorla-Wagner model of discriminant learning in nonhuman animals (Rescorla & Wagner, 1972), which has a rich history within psychology as a model of many animal learning processes. We replicate and extend earlier work showing that it is possible to extract accurate information about syntactic category and morphological family membership directly from patterns of word co-occurrence, and provide evidence from four experiments showing that this information predicts human reaction times and accuracy for class membership decisions.

When is best-worst best? A comparison of best-worst scaling, numeric estimation, and rating scales for collection of semantic norms.

Large-scale semantic norms have become both prevalent and influential in recent psycholinguistic research. However, little attention has been directed towards understanding the methodological best practices of such norm collection efforts. We compared the quality of semantic norms obtained through rating scales, numeric estimation, and a less commonly used judgment format called best-worst scaling. We found that best-worst scaling usually produces norms with higher predictive validities than other response formats, and does so requiring less data to be collected overall. We also found evidence that the various response formats may be producing qualitatively, rather than just quantitatively, different data. This raises the issue of potential response format bias, which has not been addressed by previous efforts to collect semantic norms, likely because of previous reliance on a single type of response format for a single type of semantic judgment. We have made available software for creating best-worst stimuli and scoring best-worst data. We also made available new norms for age of acquisition, valence, arousal, and concreteness collected using best-worst scaling. These norms include entries for 1,040 words, of which 1,034 are also contained in the ANEW norms (Bradley & Lang, Affective norms for English words (ANEW): Instruction manual and affective ratings (pp. 1-45). Technical report C-1, the center for research in psychophysiology, University of Florida, 1999).

Scoring best-worst data in unbalanced many-item designs, with applications to crowdsourcing semantic judgments.

Best-worst scaling is a judgment format in which participants are presented with a set of items and have to choose the superior and inferior items in the set. Best-worst scaling generates a large quantity of information per judgment because each judgment allows for inferences about the rank value of all unjudged items. This property of best-worst scaling makes it a promising judgment format for research in psychology and natural language processing concerned with estimating the semantic properties of tens of thousands of words. A variety of different scoring algorithms have been devised in the previous literature on best-worst scaling. However, due to problems of computational efficiency, these scoring algorithms cannot be applied efficiently to cases in which thousands of items need to be scored. New algorithms are presented here for converting responses from best-worst scaling into item scores for thousands of items (many-item scoring problems). These scoring algorithms are validated through simulation and empirical experiments, and considerations related to noise, the underlying distribution of true values, and trial design are identified that can affect the relative quality of the derived item scores. The newly introduced scoring algorithms consistently outperformed scoring algorithms used in the previous literature on scoring many-item best-worst data.

Estimating the average need of semantic knowledge from distributional semantic models.

Continuous bag of words (CBOW) and skip-gram are two recently developed models of lexical semantics (Mikolov, Chen, Corrado, & Dean, Advances in Neural Information Processing Systems, 26, 3111-3119, 2013). Each has been demonstrated to perform markedly better at capturing human judgments about semantic relatedness than competing models (e.g., latent semantic analysis; Landauer & Dumais, Psychological Review, 104(2), 1997 211; hyperspace analogue to language; Lund & Burgess, Behavior Research Methods, Instruments, & Computers, 28(2), 203-208, 1996). The new models were largely developed to address practical problems of meaning representation in natural language processing. Consequently, very little attention has been paid to the psychological implications of the performance of these models. We describe the relationship between the learning algorithms employed by these models and Anderson's rational theory of memory (J. R. Anderson & Milson, Psychological Review, 96(4), 703, 1989) and argue that CBOW is learning word meanings according to Anderson's concept of needs probability. We also demonstrate that CBOW can account for nearly all of the variation in lexical access measures typically attributable to word frequency and contextual diversity-two measures that are conceptually related to needs probability. These results suggest two conclusions: One, CBOW is a psychologically plausible model of lexical semantics. Two, word frequency and contextual diversity do not capture learning effects but rather memory retrieval effects.

Extrapolating human judgments from skip-gram vector representations of word meaning.

There is a growing body of research in psychology that attempts to extrapolate human lexical judgments from computational models of semantics. This research can be used to help develop comprehensive norm sets for experimental research, it has applications to large-scale statistical modelling of lexical access and has broad value within natural language processing and sentiment analysis. However, the value of extrapolated human judgments has recently been questioned within psychological research. Of primary concern is the fact that extrapolated judgments may not share the same pattern of statistical relationship with lexical and semantic variables as do actual human judgments; often the error component in extrapolated judgments is not psychologically inert, making such judgments problematic to use for psychological research. We present a new methodology for extrapolating human judgments that partially addresses prior concerns of validity. We use this methodology to extrapolate human judgments of valence, arousal, dominance, and concreteness for 78,286 words. We also provide resources for users to extrapolate these human judgments for three million English words and short phrases. Applications for large sets of extrapolated human judgments are demonstrated and discussed.

The principals of meaning: Extracting semantic dimensions from co-occurrence models of semantics.

Notable progress has been made recently on computational models of semantics using vector representations for word meaning (Mikolov, Chen, Corrado, & Dean, 2013; Mikolov, Sutskever, Chen, Corrado, & Dean, 2013). As representations of meaning, recent models presumably hone in on plausible organizational principles for meaning. We performed an analysis on the organization of the skip-gram model's semantic space. Consistent with human performance (Osgood, Suci, & Tannenbaum, 1957), the skip-gram model primarily relies on affective distinctions to organize meaning. We showed that the skip-gram model accounts for unique variance in behavioral measures of lexical access above and beyond that accounted for by affective and lexical measures. We also raised the possibility that word frequency predicts behavioral measures of lexical access due to the fact that word use is organized by semantics. Deconstruction of the semantic representations in semantic models has the potential to reveal organizing principles of human semantics.

Connected text reading and differences in text reading fluency in adult readers.

The process of connected text reading has received very little attention in contemporary cognitive psychology. This lack of attention is in parts due to a research tradition that emphasizes the role of basic lexical constituents, which can be studied in isolated words or sentences. However, this lack of attention is in parts also due to the lack of statistical analysis techniques, which accommodate interdependent time series. In this study, we investigate text reading performance with traditional and nonlinear analysis techniques and show how outcomes from multiple analyses can used to create a more detailed picture of the process of text reading. Specifically, we investigate reading performance of groups of literate adult readers that differ in reading fluency during a self-paced text reading task. Our results indicate that classical metrics of reading (such as word frequency) do not capture text reading very well, and that classical measures of reading fluency (such as average reading time) distinguish relatively poorly between participant groups. Nonlinear analyses of distribution tails and reading time fluctuations provide more fine-grained information about the reading process and reading fluency.

Now you see it, now you don't: on emotion, context, and the algorithmic prediction of human imageability judgments.

Many studies have shown that behavioral measures are affected by manipulating the imageability of words. Though imageability is usually measured by human judgment, little is known about what factors underlie those judgments. We demonstrate that imageability judgments can be largely or entirely accounted for by two computable measures that have previously been associated with imageability, the size and density of a word's context and the emotional associations of the word. We outline an algorithmic method for predicting imageability judgments using co-occurrence distances in a large corpus. Our computed judgments account for 58% of the variance in a set of nearly two thousand imageability judgments, for words that span the entire range of imageability. The two factors account for 43% of the variance in lexical decision reaction times (LDRTs) that is attributable to imageability in a large database of 3697 LDRTs spanning the range of imageability. We document variances in the distribution of our measures across the range of imageability that suggest that they will account for more variance at the extremes, from which most imageability-manipulating stimulus sets are drawn. The two predictors account for 100% of the variance that is attributable to imageability in newly-collected LDRTs using a previously-published stimulus set of 100 items. We argue that our model of imageability is neurobiologically plausible by showing it is consistent with brain imaging data. The evidence we present suggests that behavioral effects in the lexical decision task that are usually attributed to the abstract/concrete distinction between words can be wholly explained by objective characteristics of the word that are not directly related to the semantic distinction. We provide computed imageability estimates for over 29,000 words.

The blue-collar brain.

Much effort has gone into elucidating control of the body by the brain, less so the role of the body in controlling the brain. This essay develops the idea that the brain does a great deal of work in the service of behavior that is controlled by the body, a blue-collar role compared to the white-collar control exercised by the body. The argument that supports a blue-collar role for the brain is also consistent with recent discoveries clarifying the white-collar role of synergies across the body's tensegrity structure, and the evidence of critical phenomena in brain and behavior.

NUANCE 3.0: using genetic programming to model variable relationships.

Previously, we introduced a new computational tool for nonlinear curve fitting and data set exploration: the Naturalistic University of Alberta Nonlinear Correlation Explorer (NUANCE) (Hollis & Westbury, 2006). We demonstrated that NUANCE was capable of providing useful descriptions of data for two toy problems. Since then, we have extended the functionality of NUANCE in a new release (NUANCE 3.0) and fruitfully applied the tool to real psychological problems. Here, we discuss the results of two studies carried out with the aid of NUANCE 3.0. We demonstrate that NUANCE can be a useful tool to aid research in psychology in at least two ways: It can be harnessed to simplify complex models of human behavior, and it is capable of highlighting useful knowledge that might be overlooked by more traditional analytical and factorial approaches. NUANCE 3.0 can be downloaded from the Psychonomic Society Archive of Norms, Stimuli, and Data at www.psychonomic.org/archive.

NUANCE: Naturalistic University of Alberta Nonlinear Correlation Explorer.

In this article, we describe the Naturalistic University of Alberta Nonlinear Correlation Explorer (NUANCE), a computer program for data exploration and analysis. NUANCE is specialized for finding nonlinear relations between any number of predictors and a dependent value to be predicted. It searches the space of possible relations between the predictors and the dependent value by using natural selection to evolve equations that maximize the correlation between their output and the dependent value. In this article, we introduce the program, describe how to use it, and provide illustrative examples. NUANCE is written in Java, which runs on most computer platforms. We have contributed NUANCE to the archival Web site of the Psychonomic Society (www.psychonomic.org/archive), from which it may be freely downloaded.