This series will include monographs and collections of studies devoted to the investigation and exploration of knowledge, information and data-processing systems of all kinds, no matter whether human, (other) animal or machine. Its scope is intended to span the full range of interests from classical problems in the philosophy of mind and phi losophical psychology through issues in cognitive psychology and socio biology (concerning the mental capabilities of other species) to ideas related to artificial intelligence and computer science. While (...) primary emphasis will be placed upon theoretical, conceptual and epistemologi cal aspects of these problems and domains, empirical, experimental and methodological studies will also appear from time to time. Among the most challenging and difficult projects within the scope of artificial intelligence is the development and implementation of com puter programs suitable for processing natural language. Our purpose in compiling the present volume has been to contribute to the foundations of this enterprise by bringing together classic papers devoted to crucial problems involved in understanding natural language, which range from issues of formal syntax and logical form to those of possible-worlds and situation semantics. The book begins with a comprehensive introduc tion composed by Jack Kulas, the senior editor of this work, which pro vides a systematic orientation to this complex field, and ends with a selected bibliography intended to promote further research. If our efforts assist others in dealing with these problems, they will have been worthwhile. J. H. F. (shrink)

Abstract Artificial language philosophy (also called ‘ideal language philosophy’) is the position that philosophical problems are best solved or dissolved through a reform of language. Its underlying methodology—the development of languages for specific purposes—leads to a conventionalist view of language in general and of concepts in particular. I argue that many philosophical practices can be reinterpreted as applications of artificial language philosophy. In addition, many factually occurring interrelations between the sciences and philosophy of science are justified and clarified (...) by the assumption of an artificial language methodology. (shrink)

The growing advancement of Artificial Intelligence models based on deep learning and the consequent popularization of large language models (LLMs), such as ChatGPT, place the academic community facing unprecedented dilemmas, in addition to corroborating questions involving research activities and human beings. In this work, Content Analysis was chosen as the object of study, an important technique for analyzing qualitative data and frequently used among Brazilian researchers. The objective of this work was to compare the process of categorization by themes (...) carried out by human researchers on material from the Educational area, at the end of 2022, with the help of the ATLAS.TI software and the same process of categorization by themes, now carried out by the version 2023 of the ATLAS.TI software, carried out in an automated way by Artificial Intelligence, for the same material. Through lexical and semantic analysis, the themes used in the two processes were compared, observing similarities and differences between the two proposals, and verifying the consistency between the two analyses. The main result of this study is the difficulty of Artificial Intelligence in carrying out a broader categorization, eliminating specificities that were considered irrelevant by the authors for understanding the material. It is concluded, therefore, that the automated categorization process proposed by ATLAS.TI through Natural Language Processing (NLP) is consistent from a lexical point of view but is still insufficient from a semantic aspect. (shrink)

Wittgenstein’s Private Language Argument holds that language requires rule-following, rule following requires the possibility of error, error is precluded in pure introspection, and inner mental life is known only by pure introspection, thus language cannot exist entirely within inner mental life. Fodor defends his Language of Thought program against the Private Language Argument with a dilemma: either privacy is so narrow that internal mental life can be known outside of introspection, or so broad that computer language serves as a counter-example. (...) I suggest that the developing field of artificial intelligence (deep learning neural networks) tends to vitiate Fodor’s defense and hence vindicate the Private Language Argument. The first horn of Fodor’s dilemma requires language to encompass genuinely internal mental life, i.e. non-projected intentional states, which are not exhibited in classical machine learning but only by deep learning neural networks (artificial intelligence). Such networks act as black boxes, however, whose state cannot be understood by tracking the changes in their supervenience bases without shared context, and that shared context introduces the possibility of error. The language of artificial intelligence is not private. (shrink)

The article examines the dialogue between British cognitive science expert Murray Shanahan and the large language model Claude 3 Opus about “self-awareness” of artificial intelligence (AI). Adopting a text-centric approach, the author analyzes AI’s discourse through a hermeneutic lens from a reader’s perspective, irrespective of whether AI possesses consciousness or personhood. The article draws parallels between AI’s reasoning about the nature of consciousness and Buddhist concepts, especially the doctrine of dharmas, which underpins the Buddhist concept of anātman (“non-Self”). Basic (...) classifications of dharmas and their justification are examined in light of the Buddhist system of ideas about the foundations of an individual’s cognitive experience in the world. The author emphasizes that the problem of the Self as a linguistic and conceptual construct, rather than a real ontological category, was first formulated in the teachings of Buddha Shakyamuni who also proposed an “experimental” application of this concept in practices of systematic introspection (smṛti). The article contends that Claude’s discourse on self-awareness, even if it is just a tapestry of linguistic constructs woven by preset algorithms, could prove to be a source inspiring new approaches to the enigma of consciousness. This potential stems from its vast database, which is a melting pot of textual heritage from diverse human cultures. The author posits that examining AI-generated texts through the prism of Indian and Buddhist thought traditions can be eye-opening. Such an approach might help shed light on and overcome the unconscious cognitive biases and cultural blind spots within Western consciousness studies that have hindered their engagement with the full spectrum of human intellectual traditions. The author concludes that discovering different cultural sources in AI discourse and examining it from the perspective of various cultural traditions can: firstly, enrich the conceptual apparatus of cognitive studies; secondly, reveal universal cross-cultural patterns in understanding consciousness; thirdly, generate new research hypotheses and directions in studying not only artificial but also natural intelligence; fourthly, contribute to rethinking our understanding of the Other, by expanding the boundaries of what we today consider conscious or sentient. (shrink)

Artificial language learning experiments have become an important tool in exploring principles of language and language learning. A persistent question in all of this work, however, is whether ALL engages the linguistic system and whether ALL studies are ecologically valid assessments of natural language ability. In the present study, we considered these questions by examining the relationship between performance in an ALL task and second language learning ability. Participants enrolled in a Spanish language class were evaluated using a number (...) of different measures of Spanish ability and classroom performance, which was compared to IQ and a number of different measures of ALL performance. The results show that success in ALL experiments, particularly more complex artificial languages, correlates positively with indices of L2 learning even after controlling for IQ. These findings provide a key link between studies involving ALL and our understanding of second language learning in the classroom. (shrink)

This paper studies the relationship of Artificial Intelligence to the study of language and the representation of the underlying knowledge which supports the comprehension process. It develops the view that intelligence is based on the ability to use large amounts of diverse kinds of knowledge in procedural ways, rather than on the possession of a few general and uniform principles. The paper also provides a unifying thread to a variety of recent approaches to natural language comprehension. We conclude with (...) a brief discussion of how Artificial Intelligence may have a radical impact on education if the principles which it utilizes to explore the representation and use of knowledge are made available to the student to use in his own learning experiences. (shrink)

Wittgenstein and AI (Volume I): Mind and Language. This is the first of two edited collections, exploring Wittgensteinian themes in AI. The issues covered by the various chapters of this volume range over a number of topics, with a specific focus on mind and language.

The foundations of AI design discourse are worth analyzing. Here, attention is paid to the nature of theory languages used in designing new AI technologies because the limits of these languages can clarify some fundamental questions in the development of AI. We discuss three types of theory language used in designing AI products: formal, computational, and natural. Formal languages, such as mathematics, logic, and programming languages, have fixed meanings and no actual-world semantics. They are context- and practically content-free. Computational languages (...) use terms referring to the actual world, i.e., to entities, events, and thoughts. Thus, computational languages have actual-world references and semantics. They are thus no longer context- or content-free. However, computational languages always have fixed meanings and, for this reason, limited domains of reference. Finally, unlike formal and computational languages, natural languages are creative, dynamic, and productive. Consequently, they can refer to an unlimited number of objects and their attributes in an unlimited number of domains. The differences between the three theory languages enable us to reflect on the traditional problems of strong and weak AI. (shrink)

How did human vocalizations come to acquire meaning in the evolution of our species? Charles Darwin proposed that language and music originated from a common emotional signal system based on the imitation and modification of sounds in nature. This protolanguage is thought to have diverged into two separate systems, with speech prioritizing referential functionality and music prioritizing emotional functionality. However, there has never been an attempt to empirically evaluate the hypothesis that a single communication system can split into two functionally (...) distinct systems that are characterized by music- and languagelike properties. Here, we demonstrate that when referential and emotional functions are introduced into an artificial communication system, that system will diverge into vocalization forms with speech- and music-like properties, respectively. Participants heard novel vocalizations as part of a learning task. Half referred to physical entities and half functioned to communicate emotional states. Participants then reproduced each sound with the defined communicative intention in mind. Each recorded vocalization was used as the input for another participant in a serial reproduction paradigm, and this procedure was iterated to create 15 chains of five participants each. Referential vocalizations were rated as more speech-like, whereas emotional vocalizations were rated as more music-like, and this association was observed cross-culturally. In addition, a stable separation of the acoustic profiles of referential and emotional vocalizations emerged, with some attributes diverging immediately and others diverging gradually across iterations. The findings align with Darwin’s hypothesis and provide insight into the roles of biological and cultural evolution in the divergence of language and music. (shrink)

Human beings express affinity (Shinwa‐kan in Japanese language) in communicating transactive engagements among healthcare providers, patients and healthcare robots. The appearance of healthcare robots and their language capabilities often feature characteristic and appropriate compassionate dialogical functions in human–robot interactions. Elements of healthcare robot configurations comprising its physiognomy and communication properties are founded on the positivist philosophical perspective of being the summation of composite parts, thereby mimicking human persons. This article reviews Mori's theory of the Uncanny Valley and its consequent debates, (...) and examines "Uncanny" relations with generating healthcare robot conversational content with artificial affective communication (AAC) using natural language processing. With healthcare robots provoking influential physical composition and sensory expressions, the relations in human–healthcare robot transactive engagements are argued as supportive of the design and development in natural language processing. This implies that maintaining human–healthcare robot interaction and assessing the eeriness situations explained in the Uncanny Valley theory are crucial positions for healthcare robot functioning as a valuable commodity in health care. As such, physical features, language capabilities and mobility of healthcare robots establish the primacy of the AAC with natural language processing as integral to healthcare robot–human healthcare practice. (shrink)

Artificial intelligence, conceived either as an attempt to provide models of human cognition or as the development of programs able to perform intelligent tasks, is primarily interested in theuses of language. It should be concerned, therefore, withpragmatics. But its concern with pragmatics should not be restricted to the narrow, traditional conception of pragmatics as the theory of communication (or of the social uses of language). In addition to that, AI should take into account also the mental uses of language (...) (in reasoning, for example) and the existential dimensions of language as a determiner of the world we (and our computers) live in. In this paper, the relevance of these three branches of pragmatics-sociopragmatics, psychopragmatics, and ontopragmatics-for AI are explored. (shrink)

Linguistics. Artificial intelligence. Related fields. Computation.

Large-scale language technologies are increasingly used in various forms of communication with humans across different contexts. One particular use case for these technologies is conversational agents, which output natural language text in response to prompts and queries. This mode of engagement raises a number of social and ethical questions. For example, what does it mean to align conversational agents with human norms or values? Which norms or values should they be aligned with? And how can this be accomplished? In this (...) paper, we propose a number of steps that help answer these questions. We start by developing a philosophical analysis of the building blocks of linguistic communication between conversational agents and human interlocutors. We then use this analysis to identify and formulate ideal norms of conversation that can govern successful linguistic communication between humans and conversational agents. Furthermore, we explore how these norms can be used to align conversational agents with human values across a range of different discursive domains. We conclude by discussing the practical implications of our proposal for the design of conversational agents that are aligned with these norms and values. (shrink)

This paper explores the challenges posed by the rapid advancement of artificial intelligence specifically Large Language Models (LLMs). I show that traditional linguistic theories and corpus studies are being outpaced by LLMs’ computational sophistication and low perplexity levels. In order to address these challenges, I suggest a focus on language as a cognitive tool shaped by embodied-environmental imperatives in the context of Agentive Cognitive Construction Grammar. To that end, I introduce an Embodied Human Language Model (EHLM), inspired by Active (...) Inference research, as a promising alternative that integrates sensory input, embodied representations, and adaptive strategies for contextualized analysis and conceptual utility maximization. By incorporating Active Inference, which sees perception as inferring the world's state from sensory data, the findings reveal that the characterization of the English modal be able to, as a triadic construction encoding biological intelligent agency, introduces a more plausible theoretical basis for the positing of linguistic constructions. This emphasizes the crucial role of embodied human language models in the comprehension of how humans construct preferred futures through language. (shrink)

Wagman examines the emulation of human cognition by artificial intelligence systems. The book provides detailed examples of artificial intelligence programs (such as the FERMI System and KEKADA program) accomplishing highly intellectual tasks.

Presupposing no familiarity with the technical concepts of either philosophy or computing, this clear introduction reviews the progress made in AI since the inception of the field in 1956. Copeland goes on to analyze what those working in AI must achieve before they can claim to have built a thinking machine and appraises their prospects of succeeding.There are clear introductions to connectionism and to the language of thought hypothesis which weave together material from philosophy, artificial intelligence and neuroscience. John (...) Searle's attacks on AI and cognitive science are countered and close attention is given to foundational issues, including the nature of computation, Turing Machines, the Church-Turing Thesis and the difference between classical symbol processing and parallel distributed processing. The book also explores the possibility of machines having free will and consciousness and concludes with a discussion of in what sense the human brain may be a computer. (shrink)

We investigated the emergence of sociolinguistic indexicality using an artificial-language-learning paradigm. Sociolinguistic indexicality involves the association of linguistic variants with nonlinguistic social or contextual features. Any linguistic variant can acquire “constellations” of such indexical meanings, though they also exhibit an ordering, with first-order indices associated with particular speaker groups and higher-order indices targeting stereotypical attributes of those speakers. Much natural-language research has been conducted on this phenomenon, but little experimental work has focused on how indexicality emerges. Here, we present (...) three miniature artificial-language experiments designed to break ground on this question. Results show ready formation of first-order indexicality based on co-occurrence alone, with higher-order indexicality emerging as a result of extension to new speaker groups, modulated by the perceived practical importance of the indexed social feature. (shrink)

Natural languages contain many layers of sequential structure, from the distribution of phonemes within words to the distribution of phrases within utterances. However, most research modeling language acquisition using artificial languages has focused on only one type of distributional structure at a time. In two experiments, we investigated adult learning of an artificial language that contains dependencies between both adjacent and non-adjacent words. We found that learners rapidly acquired both types of regularities and that the strength of the (...) adjacent statistics influenced learning of both adjacent and non-adjacent dependencies. Additionally, though accuracy was similar for both types of structure, participants’ knowledge of the deterministic non-adjacent dependencies was more explicit than their knowledge of the probabilistic adjacent dependencies. The results are discussed in the context of current theories of statistical learning and language acquisition. (shrink)

We expose the implications of lexical innovation as supported by the ONOMATURGE knowledge-based paradigm, for policies intended to foster native or national languages. In certain cases, the survival of native cultures, as supported by their language, depends on their ability to fill lexical gaps due to the technological gap. In certain other cases, the native culture itself is not in crisis (e.g., in the case of the national language of a sovereign country), but the local technologists or translators participate in (...) lexical innovation in a specific technical domain, and can benefit from the ONOMATURGE paradigm. (shrink)

When one looks into the role of artificial languages in philosophy of language it seems appropriate to start with making a distinction between philosophy of language proper and formal semantics of natural language. Although the distinction between the two disciplines may not always be easy to make since there arguably exist substantial historical and systematic relationships between the two, it nevertheless pays to keep the two apart, at least initially, since the motivation commonly given for the use of (...) class='Hi'>artificial languages in philosophy of language is often rather different from the one that drives the use of such languages in semantics. Of course, this difference in motivation should not blind us for the commonalities that exists between the two disciplines. Philosophy of language and formal semantics have a common history, and arguably also share some of their substance. Philosophy of language is by and large an outgrowth of work in the analytical tradition in philosophy in the first half of the twentieth century. Both ordinary language philosophy, with its emphasis on the description of actual language use, as well as the more logic-oriented and formally inclined school of logical positivism contributed to the definition of philosophy of language as a separate philosophical discipline, with its own set of problems and methods to solve them. Another major contributor to the establishment of philosophy of language as a distinct discipline has been modern linguistics, in particular generative grammar in the tradition of Chomsky that became the dominant paradigm in linguisticsin the fifties and sixties of the previous century. And as it happens, both the generative tradition of Chomsky and analytic philosophy in its formal and less formal guises have been important factors in the development of formal semantics as well. Thus, it should come as no surprise that the two have something in common. That the communalities go beyond a common ancestry, but are reflected in substance and methods as well, will be argued later on. However, be that as it may, it is still a good idea to keep philosophy of language and formal semantics separate, at least initially, since the role that is assigned to artificial languages and the ways in which these languages are employed in both does differ in a number of respects that are worth keeping in mind.. (shrink)

J. M. E. McTaggart famously divided time into two time series, which he entitled A-series and B-series. Although he was proponent of neither of them, his division initiated a discussion as to which of the series is prior or real. This paper follows Clifford Williams’s claim that these series are not as distant as their proponents argue they are. It demonstrates their translatability in the case of examples from temporal logic and natural language. It argues that, if there are any (...) deep-rooted differences between these two series, they are only metaphysical. These metaphysical differences could be overcome by focusing on logic or natural language. These two series of time are also used in computer science and in the development of Artificial Intelligence. However, no attempt at unifying them has been suggested in these fields of study. J. M. E. McTaggart rozdělil čas na dvě časové série, A-série a B-série. Třebaže nebyl zastáncem žádné z nich, jeho rozdělení podnítilo diskuzi, která z těchto časových sérií je přednější nebo jediná skutečně reálná. Tento článek navazuje na názor Clifforda Williamse, že tyto časové série nejsou tak rozdílné, jak tvrdí jejich zastánci. V článku je demonstrována jejich přeložitelnost na příkladech z temporální logiky a přirozeného jazyka. Článek dále tvrdí, že jestliže je zde nějaký hluboce zakořeněný rozdíl mezi těmito dvěma sériemi, pak je pouze metafyzický. Tyto metafyzické rozdíly mohou být překonány, zaměříme-li se na logiku a přirozený jazyk. Tyto dvě časové série jsou také použity v informatice a rozvoji umělé inteligence, třebaže v těchto oborech dosud nebyly navrženy žádné pokusy o jejich propojení. (shrink)

In this article, we develop a hierarchical Bayesian model of learning in a general type of artificial language‐learning experiment in which learners are exposed to a mixture of grammars representing the variation present in real learners’ input, particularly at times of language change. The modeling goal is to formalize and quantify hypothesized learning biases. The test case is an experiment (Culbertson, Smolensky, & Legendre, 2012) targeting the learning of word‐order patterns in the nominal domain. The model identifies internal biases (...) of the experimental participants, providing evidence that learners impose (possibly arbitrary) properties on the grammars they learn, potentially resulting in the cross‐linguistic regularities known as typological universals. Learners exposed to mixtures of artificial grammars tended to shift those mixtures in certain ways rather than others; the model reveals how learners’ inferences are systematically affected by specific prior biases. These biases are in line with a typological generalization—Greenberg's Universal 18—which bans a particular word‐order pattern relating nouns, adjectives, and numerals. (shrink)

To a greater extent than in other technical domains, research and progress in Artificial Intelligence has always been entwined with the fictional. Its language echoes strongly with other forms of cultural narratives, such as fairytales, myth and religion. In this essay we present varied examples that illustrate how these analogies have guided not only readings of the AI enterprise by commentators outside the community but also inspired AI researchers themselves. Owing to their influence, we pay particular attention to the (...) similarities between religious language and the way in which the potential advent of greater than human intelligence is presented contemporarily. We then move on to the role that fiction, science fiction most of all, has historically played and is still playing in the discussion of AI by influencing researchers and the public, shifting the weights of different scenarios in our collectively perceived probability space. We sum up by arguing that the lore surrounding AI research, ancient and modern, points to the ancestral and shared human motivations that drive researchers in their pursuit and fascinate humanity at large. These points of narrative entanglement where AI meets the wider culture should serve to amplify the call to engage ourselves with the discussion of the potential destination of this technology. (shrink)

Are rules processes or similarity processes the default for acquisition of grammatical knowledge during natural second language acquisition? Whereas Pothos argues similarity processes are the default in the many areas he reviews, including artificial grammar learning and first language development, I suggest, citing evidence, that in second language acquisition of grammatical morphology “rules processes” may be the default.

Artificial Knowing challenges the masculine slant in the Artificial Intelligence (AI) view of the world. Alison Adam admirably fills the large gap in science and technology studies by showing us that gender bias is inscribed in AI-based computer systems. Her treatment of feminist epistemology, focusing on the ideas of the knowing subject, the nature of knowledge, rationality and language, are bound to make a significant and powerful contribution to AI studies. Drawing from theories by Donna Haraway and Sherry (...) Turkle, and using tools of feminist epistemology, Adam provides a sustained critique of AI which interestingly re-enforces many of the traditional criticisms of the AI project. Artificial Knowing is an esential read for those interested in gender studies, science and technology studies, and philosophical debates in AI. (shrink)

This study is intended to create an innovative contextual English learning environment making use of the widely used communication software, LINE ChatBot, based on the Artificial Intelligence Markup Language, in order to improve speaking and listening ability among learners. A total of 73 students were invited to participate in learning activities involving a 4-week English conversation exercise including both speaking and listening. Additionally, in order to explore the influence of competition on language acquisition, we added competition characteristics to the (...) learning activities in the experimental group to enhance learning motivation and learning outcomes. The results showed that with the help of the LINE ChatBot contextual learning environment, the performance of both groups of students was slightly enhanced, but no significant differences were found. Meanwhile, extrinsic motivation in both the experimental and control group was improved if they spoke anonymously. That is, the contextual learning environment based on the LINE ChatBot significantly improved the learners’ English speaking and listening ability. In addition, the results showed that the addition of a competition element effectively enhanced the learners’ intrinsic motivation to learn English on the LINE ChatBot. (shrink)

The purpose of this paper is to explore three alternative frameworks for understanding the nature of language and mentality, which accent syntactical, semantical, and pragmatical aspects of the phenomena with which they are concerned, respectively. Although the computational conception currently exerts considerable appeal, its defensibility appears to hinge upon an extremely implausible theory of the relation of form to content. Similarly, while the representational approach has much to recommend it, its range is essentially restricted to those units of language that (...) can be understood in terms of undefined units. Thus, the only alternative among these three that can account for the meaning of primitive units of language is one emphasizing the basic role of skills, habits, and tendencies in relating signs and dispositions. (shrink)

Medieval Arabic algebra is a good example of an artificial language.Yet despite its abstract, formal structure, its utility was restricted to problem solving. Geometry was the branch of mathematics used for expressing theories. While algebra was an art concerned with finding specific unknown numbers, geometry dealtwith generalmagnitudes.Algebra did possess the generosity needed to raise it to a more theoretical level—in the ninth century Abū Kāmil reinterpreted the algebraic unknown “thing” to prove a general result. But mathematicians had no motive (...) to rework their theories in algebraic form. Because it offered no advantage over geometry, algebra remained a practical art in both the Islamic world and in Europe until the scientific uphevals of the 17th–18th centuries. (shrink)

From a leading authority in artificial intelligence, this book delivers a synthesis of the major modern techniques and the most current research in natural language processing. The approach is unique in its coverage of semantic interpretation and discourse alongside the foundational material in syntactic processing.

Some empirical evidence in the artificial language acquisition literature has been taken to suggest that statistical learning mechanisms are insufficient for extracting structural information from an artificial language. According to the more than one mechanism (MOM) hypothesis, at least two mechanisms are required in order to acquire language from speech: (a) a statistical mechanism for speech segmentation; and (b) an additional rule-following mechanism in order to induce grammatical regularities. In this article, we present a set of neural network (...) studies demonstrating that a single statistical mechanism can mimic the apparent discovery of structural regularities, beyond the segmentation of speech. We argue that our results undermine one argument for the MOM hypothesis. (shrink)

Statistical learning (SL) involving sensitivity to distributional regularities in the environment has been suggested to be an important factor in many aspects of cognition, including language. However, the degree to which statistically-learned information is retained over time is not well understood. To establish whether or not learners are able to preserve such regularities over time, we examined performance on an artificial second language learning task both immediately after training and also at a follow-up session 2 weeks later. Participants were (...) exposed to an artificial language (Brocanto2), half of them receiving simplified training items in which only 20% of sequences contained complex structures, whereas the other half were exposed to a training set in which 80% of the items were composed of complex sequences. Overall, participants showed signs of learning at the first session and retention at the second, but the degree of learning was affected by the nature of the training they received. Participants exposed to the simplified input outperformed those in the more complex training condition. A GLMM was used to model the relationship between stimulus properties and participants’ endorsement strategies across both sessions. The results indicate that participants in the complex training condition relied more on an item’s chunk strength than those in the simple training condition. Taken together, this set of findings shows that statistically learned regularities are retained over the course of 2 weeks. The results also demonstrate that training on input featuring simple items leads to improved learning and retention of grammatical regularities. (shrink)