From partners to populations: A hierarchical Bayesian account of coordination and convention.

Languages are powerful solutions to coordination problems: They provide stable, shared expectations about how the words we say correspond to the beliefs and intentions in our heads. Yet, language use in a variable and nonstationary social environment requires linguistic representations to be flexible: Old words acquire new ad hoc or partner-specific meanings on the fly. In this article, we introduce continual hierarchical adaptation through inference (CHAI), a hierarchical Bayesian theory of coordination and convention formation that aims to reconcile the long-standing tension between these two basic observations. We argue that the central computational problem of communication is not simply transmission, as in classical formulations, but continual learning and adaptation over multiple timescales. Partner-specific common ground quickly emerges from social inferences within dyadic interactions, while community-wide social conventions are stable priors that have been abstracted away from interactions with multiple partners. We present new empirical data alongside simulations showing how our model provides a computational foundation for several phenomena that have posed a challenge for previous accounts: (a) the convergence to more efficient referring expressions across repeated interaction with the same partner, (b) the gradual transfer of partner-specific common ground to strangers, and (c) the influence of communicative context on which conventions eventually form. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

This Better Be Interesting: A Speaker's Decision to Speak Cues Listeners to Expect Informative Content.

In anticipating upcoming content, comprehenders are known to rely on real-world knowledge. This knowledge can be deployed directly in favor of upcoming content about typical situations (implying a transparent mapping between the world and what speakers say about the world). Such knowledge can also be used to estimate the likelihood of speech, whereby atypical situations are the ones newsworthy enough to merit reporting (i.e., a nontransparent mapping in which improbable situations yield likely utterances). We report four forced-choice studies (three preregistered) testing this distinction between situation knowledge and speech production likelihood. Comprehenders are shown to anticipate situation-atypical meanings more when guessing content (a) that a speaker announces (rather than thinks), (b) that is said out of the blue (rather than produced when prompted), and (c) that is addressed to a large audience (rather than a single listener). The findings contrast with prior work that emphasizes a comprehension bias in favor of typicality, and they highlight the need for comprehension models that incorporate expectations for informativity (as one of a set of inferred speaker goals) alongside expectations for content plausibility.

Mutual influence between language and perception in multi-agent communication games.

Language interfaces with many other cognitive domains. This paper explores how interactions at these interfaces can be studied with deep learning methods, focusing on the relation between language emergence and visual perception. To model the emergence of language, a sender and a receiver agent are trained on a reference game. The agents are implemented as deep neural networks, with dedicated vision and language modules. Motivated by the mutual influence between language and perception in cognition, we apply systematic manipulations to the agents' (i) visual representations, to analyze the effects on emergent communication, and (ii) communication protocols, to analyze the effects on visual representations. Our analyses show that perceptual biases shape semantic categorization and communicative content. Conversely, if the communication protocol partitions object space along certain attributes, agents learn to represent visual information about these attributes more accurately, and the representations of communication partners align. Finally, an evolutionary analysis suggests that visual representations may be shaped in part to facilitate the communication of environmentally relevant distinctions. Aside from accounting for co-adaptation effects between language and perception, our results point out ways to modulate and improve visual representation learning and emergent communication in artificial agents.

Meaning and Use in the Expression of Estimative Probability.

Words of estimative probability (WEPs), such as 'possible' and 'a good chance', provide an efficient means for expressing probability under uncertainty. Current semantic theories assume that WEPs denote crisp thresholds on the probability scale, but experimental data indicate that their use is characterised by gradience and focality. Here, we implement and compare computational models of the use of WEPs to explain novel production data. We find that, among models incorporating cognitive limitations and assumptions about goal-directed speech, a model that implements a threshold-based semantics explains the data equally well as a model that semantically encodes patterns of gradience and focality. We further validate the model by distinguishing between participants with more or fewer autistic traits, as measured with the Autism Spectrum Quotient test. These traits include communicative difficulties. We show that these difficulties are reflected in the rationality parameter of the model, which modulates the probability that the speaker selects the pragmatically optimal message.

Corrigendum: Pragmatic Prediction in the Processing of Referring Expressions Containing Scalar Quantifiers.

[This corrects the article DOI: 10.3389/fpsyg.2021.662050.].

Pragmatic Prediction in the Processing of Referring Expressions Containing Scalar Quantifiers.

Previous research in cognitive science and psycholinguistics has shown that language users are able to predict upcoming linguistic input probabilistically, pre-activating material on the basis of cues emerging from different levels of linguistic abstraction, from phonology to semantics. Current evidence suggests that linguistic prediction also operates at the level of pragmatics, where processing is strongly constrained by context. To test a specific theory of contextually-constrained processing, termed pragmatic surprisal theory here, we used a self-paced reading task where participants were asked to view visual scenes and then read descriptions of those same scenes. Crucially, we manipulated whether the visual context biased readers into specific pragmatic expectations about how the description might unfold word by word. Contrary to the predictions of pragmatic surprisal theory, we found that participants took longer reading the main critical term in scenarios where they were biased by context and pragmatic constraints to expect a given word, as opposed to scenarios where there was no pragmatic expectation for any particular referent.

Probabilistic pragmatics explains gradience and focality in natural language quantification.

An influential view in philosophy and linguistics equates the meaning of a sentence to the conditions under which it is true. But it has been argued that this truth-conditional view is too rigid and that meaning is inherently gradient and revolves around prototypes. Neither of these abstract semantic theories makes direct predictions about quantitative aspects of language use. Hence, we compare these semantic theories empirically by applying probabilistic pragmatic models as a link function connecting linguistic meaning and language use. We consider the use of quantity words (e.g., "some," "all"), which are fundamental to human language and thought. Data from a large-scale production study suggest that quantity words are understood via prototypes. We formulate and compare computational models based on the two views on linguistic meaning. These models also take into account cognitive factors, such as salience and numerosity representation. Statistical and empirical model comparison show that the truth-conditional model explains the production data just as well as the prototype-based model, when the semantics are complemented by a pragmatic module that encodes probabilistic reasoning about the listener's uptake.

Mutual Exclusivity in Pragmatic Agents.

One of the great challenges in word learning is that words are typically uttered in a context with many potential referents. Children's tendency to associate novel words with novel referents, which is taken to reflect a mutual exclusivity (ME) bias, forms a useful disambiguation mechanism. We study semantic learning in pragmatic agents-combining the Rational Speech Act model with gradient-based learning-and explore the conditions under which such agents show an ME bias. This approach provides a framework for investigating a pragmatic account of the ME bias in humans but also for building artificial agents that display an ME bias. A series of analyses demonstrates striking parallels between our model and human word learning regarding several aspects relevant to the ME bias phenomenon: online inference, long-term learning, and developmental effects. By testing different implementations, we find that two components, pragmatic online inference and incremental collection of evidence for one-to-one correspondences between words and referents, play an important role in modeling the developmental trajectory of the ME bias. Finally, we outline an extension of our model to a deep neural network architecture that can process more naturalistic visual and linguistic input. Until now, in contrast to children, deep neural networks have needed indirect access to (supposed to be novel) test inputs during training to display an ME bias. Our model is the first one to do so without using this manipulation.

Presuppositions of determiners are immediately used to disambiguate utterance meaning: A mouse-tracking study on the German language.

The present study investigated how listeners understand and process the definite and the indefinite determiner. While the definite determiner clearly conveys a uniqueness presupposition, the status of the anti-uniqueness inference associated with the indefinite determiner is less clear. In a forced choice production task, we observed that participants make use of the information about number usually associated with the two determiners to convey a message. In a subsequent mouse-tracking task, participants had to select one of two potential referents presented on screen according to an auditorily presented stimulus sentence. The data revealed that participants use the information about uniqueness or anti-uniqueness encoded in determiners to disambiguate sentence meaning as early as possible, but only when they are exclusively faced with felicitous uses of determiners.

The PANGAEA mineralogical database.

Future human missions to the surface of the Moon and Mars will involve scientific exploration requiring new support tools to enable rapid and high quality science decision-making. Here, we describe the PANGAEA (Planetary ANalogue Geological and Astrobiological Exercise for Astronauts) Mineralogical Database developed by ESA (European Space Agency): a catalog of petrographic and spectroscopic information on all currently known minerals identified on the Moon, Mars, and associated with meteorites. The catalog also includes minerals found in the analog field sites used for ESA's geology and astrobiology training course PANGAEA, to broaden the database coverage. The Mineralogical Database is composed of the Summary Catalog of Planetary Analog Minerals and of the Spectral Archive and is freely available in the public repository of ESA PANGAEA. The Summary Catalog provides essential descriptive information for each mineral, including name (based on the International Mineralogical Association recommendation), chemical formula, mineral group, surface abundance on planetary bodies, geological significance in the context of planetary exploration, number of collected VNIR and Raman spectra, likelihood of detection using different spectral methods, and bibliographic references evidencing their detection in extraterrestrial or terrestrial analog environments. The Spectral Archive provides a standard library for planetary in-situ human and robotic exploration covering Visual-Near-Infrared reflective (VNIR) and Raman spectroscopy (Raman). To populate this library, we collected VNIR and Raman spectra for mineral entries in the Summary Catalog from open-access archives and analyzed them to select the ones with the best spectral features. We also supplemented this collection with our own bespoke measurements. Additionally, we compiled the chemical compositions for all the minerals based on their empirical formula, to allow identification using the measured abundances provided by LIBS and XRF analytical instruments. When integrated into an operational support system like ESA's Electronic Fieldbook (EFB) system, the Mineralogical Database can be used as a real-time and autonomous decision support tool for sampling operations on the Moon, Mars and during astronaut geological field training. It provides both robust spectral libraries to support mineral identification from instrument outputs, and relevant contextualized information on detected minerals.

Strategies of Deception: Under-Informativity, Uninformativity, and Lies-Misleading With Different Kinds of Implicature.

Conversation is often cast as a cooperative effort, and some aspects of it, such as implicatures, have been claimed to depend on an assumption of cooperation (Grice, 1989). But any systematic class of inference derived from assumptions of cooperation, such as implicatures, could also be, on occasion, used to deceive listeners strategically. Here, we explore the extent to which speakers might choose different kinds of implicature triggers in an uncooperative game of communication. Concretely, we present a study in the form of a cooperative or competitive signaling game where communicators can exploit three kinds of implicatures: exact reading of numeral expressions, scalar implicatures linked to the quantifier some and ad hoc scalar implicatures. We compare how these implicatures are used depending on whether the participants' co-player is cooperative, a strategic opponent, or a non-strategic opponent. We find that when the strategy of the co-player is clear to the participants, the three types of implicatures are used to exploit the co-player's interpretation strategy. Indeed, participants use numeral implicatures as reliably as truth conditional content in all three conditions, while scalar quantifiers and ad hoc implicature elicit different strategies. We interpret these findings as evidence that speakers expect their interlocutors to infer implicatures from their utterances even in contexts where they know that they will be perceived as uncooperative.

Gricean Expectations in Online Sentence Comprehension: An ERP Study on the Processing of Scalar Inferences.

There is substantial support for the general idea that a formalization of comprehenders' expectations about the likely next word in a sentence helps explaining data related to online sentence processing. While much research has focused on syntactic, semantic, and discourse expectations, the present event-related potentials (ERPs) study investigates neurolinguistic correlates of pragmatic expectations, which arise when comprehenders expect a sentence to conform to Gricean Maxims of Conversation. For predicting brain responses associated with pragmatic processing, we introduce a formal model of such Gricean pragmatic expectations, using an idealized incremental interpreter. We examine whether pragmatic expectancies derived from this model modulate the amplitude of the N400, a component that has been associated with predictive processing. As part of its parameterization, the model distinguishes genuine pragmatic interpreters, who expect maximally informative true utterances, from literal interpreters, who only expect truthfulness. We explore the model's non-trivial predictions for an experimental setup which uses picture-sentence verification with ERPs recorded at several critical positions in sentences containing the scalar implicature trigger some. We find that Gricean expectations indeed affect the N400, largely in line with the predictions of our model, but also discuss discrepancies between model predictions and observations critically.

Pragmatic processing: An investigation of the (anti-)presuppositions of determiners using mouse-tracking.

A presupposition is a condition that has to be met in order for a linguistic expression to be appropriate. The definite determiner (as in the banana) triggers the uniqueness-presupposition that there is a uniquely identifiable banana in the relevant discourse context. The indefinite determiner (as in a banana) is similarly associated with anti-uniqueness (that there are several bananas). Application of the Maximize Presupposition principle to the indefinite determiner suggests that this latter effect results indirectly as an anti-presupposition from considering the uniqueness-presupposition of the definite determiner, which is then negated. This results in increased processing difficulty. We utilized mouse-tracking to compare processing of definite and indefinite determiners when used felicitously and infelicitously in a particular context. First, processing of the indefinite determiner was associated with more processing difficulties compared with the definite determiner. Second, we also observed evidence for an initial temporary activation and evaluation of the uniqueness-presupposition, just as derived from anti-presupposition theory and the Maximize Presupposition principle.

Evidential Strength of Intonational Cues and Rational Adaptation to (Un-)Reliable Intonation.

Intonation plays an integral role in comprehending spoken language. Listeners can rapidly integrate intonational information to predictively map a given pitch accent onto the speaker's likely referential intentions. We use mouse tracking to investigate two questions: (a) how listeners draw predictive inferences based on information from intonation? and (b) how listeners adapt their online interpretation of intonational cues when these are reliable or unreliable? We formulate a novel Bayesian model of rational predictive cue integration and explore predictions derived under a concrete linking hypothesis relating a quantitative notion of evidential strength of a cue to the moment in time, relative to the unfolding speech signal, at which mouse trajectories turn towards the eventually selected option. In order to capture rational belief updates after concrete observations of a speaker's behavior, we formulate and explore an extension of this model that includes the listener's hierarchical beliefs about the speaker's likely production behavior. Our results are compatible with the assumption that listeners rapidly and rationally integrate all available intonational information, that they expect reliable intonational information initially, and that they adapt these initial expectations gradually during exposition to unreliable input. All materials, data, and scripts can be retrieved here: https://osf.io/dnbuk/.

Complex probability expressions & higher-order uncertainty: Compositional semantics, probabilistic pragmatics & experimental data.

We present novel experimental data pertaining to the use and interpretation of simple probability expressions (such as possible or likely) and complex ones (such as possibly likely or certainly possible) in situations of higher-order uncertainty, i.e., where speakers may be uncertain about the probability of a chance event. The data is used to critically assess a probabilistic pragmatics model in the vein of Rational Speech Act approaches (e.g., Frank and Goodman, 2012; Franke and Jäger, 2016; Goodman and Frank, 2016). The model embeds a simple compositional threshold-semantics for probability expressions, following recent work in formal linguistics (Swanson, 2006; Yalcin, 2007, 2010; Lassiter, 2010, 2017; Moss, 2015).

Coevolution of Lexical Meaning and Pragmatic Use.

According to standard linguistic theory, the meaning of an utterance is the product of conventional semantic meaning and general pragmatic rules on language use. We investigate how such a division of labor between semantics and pragmatics could evolve under general processes of selection and learning. We present a game-theoretic model of the competition between types of language users, each endowed with certain lexical representations and a particular pragmatic disposition to act on them. Our model traces two evolutionary forces and their interaction: (i) pressure toward communicative efficiency and (ii) transmission perturbations during the acquisition of linguistic knowledge. We illustrate the model based on a case study on scalar implicatures, which suggests that the relationship between underspecified semantics and pragmatic inference is one of coevolution.

Transient surface photovoltage measurement over 12 orders of magnitude in time.

The measurement of transient surface photovoltage (SPV) signals in a fixed capacitor arrangement over 12 orders of magnitude in time has been demonstrated for a SnO2:F/TiO2/In2S3 layer system under high vacuum. For this purpose, a high impedance buffer with a bandwidth above 200 MHz and an effective input resistance of 200-700 TΩ has been developed. Fast separation of photo generated charge carriers within ns and very slow relaxation of SPV signals excited with short laser pulses and the measurement of SPV spectra under continuous illumination with a halogen lamp were demonstrated.

Reasoning in Reference Games: Individual- vs. Population-Level Probabilistic Modeling.

Recent advances in probabilistic pragmatics have achieved considerable success in modeling speakers' and listeners' pragmatic reasoning as probabilistic inference. However, these models are usually applied to population-level data, and so implicitly suggest a homogeneous population without individual differences. Here we investigate potential individual differences in Theory-of-Mind related depth of pragmatic reasoning in so-called reference games that require drawing ad hoc Quantity implicatures of varying complexity. We show by Bayesian model comparison that a model that assumes a heterogenous population is a better predictor of our data, especially for comprehension. We discuss the implications for the treatment of individual differences in probabilistic models of language use.

Predictive parameters for return to pre-injury level of sport 6 months following anterior cruciate ligament reconstruction surgery.

PURPOSE: The aim of the study was to find predictive parameters for a successful resumption of pre-injury level of sport 6 months post anterior cruciate ligament (ACL) reconstruction. METHODS: In a prospective study, 40 patients with a ruptured ACL were surgically treated with semitendinosus tendon autograft. Six months after surgery, strength of knee extensors and flexors, four single-leg hop tests, Anterior Cruciate Ligament-Return to Sport after Injury Scale (ACL-RSI), subjective International Knee Documentation Committee (IKDC) 2000 and the Tampa Scale of Kinesiophobia-11 (TSK-11) were assessed. Seven months post-operatively, a standardized interview was conducted to identify "return to sport" (RS) and "non-return to sport" (nRS) patients. Logistic regression and "Receiver Operating Characteristic" (ROC) analyses were used to determine predictive parameters. RESULTS: No significant differences could be detected between RS and nRS patients concerning socio-demographic data, muscle tests, square hop and TSK-11. In nRS patients, the Limb Symmetry Index (LSI) of single hop for distance (p = 0.005), crossover hop (p = 0.008) and triple hop (p = 0.001) were significantly lower, in addition to the ACL-RSI (p = 0.013) and IKDC 2000 (p = 0.037). The cut-off points for LSI single hop for distance were 75.4 % (sensitivity 0.74; specificity 0.88), and for ACL-RSI 51.3 points (sensitivity 0.97; specificity 0.63). Logistic regression distinguished between RS and nRS subjects (sensitivity 0.97; specificity 0.63). CONCLUSIONS: The single hop for distance and ACL-RSI were found to be the strongest predictive parameters, assessing both the objective functional and the subjective psychological aspects of returning to sport. Both tests may help to identify patients at risk of not returning to pre-injury sport. LEVEL OF EVIDENCE: II.

Quantitative uncertainty analysis of Life Cycle Assessment for algal biofuel production.

As a result of algae's promise as a renewable energy feedstock, numerous studies have used Life Cycle Assessment (LCA) to quantify the environmental performance of algal biofuels, yet there is no consensus of results among them. Our work, motivated by the lack of comprehensive uncertainty analysis in previous studies, uses a Monte Carlo approach to estimate ranges of expected values of LCA metrics by incorporating parameter variability with empirically specified distribution functions. Results show that large uncertainties exist at virtually all steps of the biofuel production process. Although our findings agree with a number of earlier studies on matters such as the need for wet lipid extraction, nutrients recovered from waste streams, and high energy coproducts, the ranges of reported LCA metrics show that uncertainty analysis is crucial for developing technologies, such as algal biofuels. In addition, the ranges of energy return on (energy) invested (EROI) values resulting from our analysis help explain the high variability in EROI values from earlier studies. Reporting results from LCA models as ranges, and not single values, will more reliably inform industry and policy makers on expected energetic and environmental performance of biofuels produced from microalgae.