Theories of Meaning for the Internet of Things

In this chapter, we consider the theoretical foundations for representing knowledge in the Internet of Things context. Specifically, we consider (1) the model-theoretic semantics (i.e., extensional semantics), (2) the possible-world semantics (i.e., intensional semantics), (3) the situation semantics, and (4) the cognitive/distributional semantics. Given the peculiarities of the Internet of Things, we pay particular attention to (a) perception (i.e., how to establish a connection to the world), (b) intersubjectivity (i.e., how to align world representations), and (c) the dynamics of world knowledge (i.e., how to model events). We come to the conclusion that each of the semantic theories helps in modeling specific aspects, but does not sufficiently address all three aspects simultaneously.

Grounding Abstract Concepts in Action

Sensory-motor information is linguistically encoded by action verbs. Such verbs are not only used to express action concepts and events, but they are also pervasively exploited in the linguistic representation of abstract concepts and figurative meanings. In the light of several theoretical approaches (i.e., Embodied Theories, Conceptual Metaphor Theory, Image Schema Theory), this paper analyzes the mechanisms that enable action verbs to acquire abstract meanings and that motivate the symmetries (or asymmetries) in the semantic variations of locally equivalent verbs (e.g., premere and spingere; Eng., to press and to push). The research is carried out within the IMAGACT framework and focuses on a set of four Italian action verbs encoding force (i.e., premere, spingere, tirare, and trascinare; Eng., to press, to push, to pull, and to drag). The results confirm that metaphorical extensions of action verbs are constrained by the image schemas involved in the core meaning of the verbs. Additionally, the paper shows that these image schemas are responsible for the asymmetries in the metaphorical variation of action verbs pertaining to the same semantic class (i.e., force).

Concepts in Action: Introduction

It is impossible to talk about human cognition without talking about concepts—there simply is no human cognition without concepts. Concepts form an abstraction of reality that is central to the functioning of the human mind. Conceptual knowledge (of e.g., APPLE, LOVE and BEFORE) is crucial for us to categorize, understand, and reason about the world. Only equipped with concepts and words for them can we successfully communicate and carry out actions. But what exactly are concepts? How are concepts acquired? How does the human mind use concepts? In this introduction chapter, we elaborate on these questions (i.e., representation, learning, application), and provide summaries of the seven individual chapters in the volume, which from different scientific disciplines relate to one or the other of the questions.

Evaluating Semantic Co-creation by Using a Marker as a Linguistic Constraint Tool in Shared Cognitive Representation Models

Semantic co-creation occurs in the process of communication between two or more people, where human cognitive representation models of the topic of discussion converge. The use of linguistic constraint tools (for example a shared marker) enable participants to focus on communication, improving communicative success. Recent results state that the best communicative success can be achieved if two users can interact in a restricted way, so called team focused interaction hypothesis. Even though the advantage of team focused interaction sounds plausible, it needs to be noted that previous studies enforce the constraint usage. Our study aims at investigating the advantage of using shared markers as a linguistic constraint tool in semantic co-creation, while moving them becomes optional. In our experimental task, based on a shared geographic map as a cognitive representation model, the two participants have to identify a target location, which is only known to a third participant. We assess two main factors, the teams’ use of a shared marker and the two complexity levels of the cognitive representation model. We had hypothesized that sharing a marker should improve communicative success, as communication is more focused. However, our results indicated no general benefit by using a marker as well as team interaction, itself. Our results suggest that the use of a shared marker is an efficient linguistic constraint at higher levels of complexity of the cognitive representation than those tested in our study. Based on this consideration, the team focused interaction hypothesis should be further developed to include a control parameter for the perceived decision complexity of the cognitive representation model.

Generalizing Psychological Similarity Spaces to Unseen Stimuli

The cognitive framework of conceptual spaces proposes to represent concepts as regions in psychological similarity spaces. These similarity spaces are typically obtained through multidimensional scaling (MDS), which converts human dissimilarity ratings for a fixed set of stimuli into a spatial representation. One can distinguish metric MDS (which assumes that the dissimilarity ratings are interval or ratio scaled) from nonmetric MDS (which only assumes an ordinal scale). In our first study, we show that despite its additional assumptions, metric MDS does not necessarily yield better solutions than nonmetric MDS. In this chapter, we furthermore propose to learn a mapping from raw stimuli into the similarity space using artificial neural networks (ANNs) in order to generalize the similarity space to unseen inputs. In our second study, we show that a linear regression from the activation vectors of a convolutional ANN to similarity spaces obtained by MDS can be successful and that the results are sensitive to the number of dimensions of the similarity space.

A Qualitative Similarity Framework for the Interpretation of Natural Language Similarity Expressions

In this paper, a representational framework is presented featuring a qualitative notion of similarity. It is aimed at issues of natural language semantics, in particular the semantics of expressions of similarity and sameness and their role in comparison and ad-hoc kind formation. The framework makes use of attribute spaces, which are well-established in AI and also in some branches of natural language semantics, e.g., frame-based approaches (Barsalou 1992). What distinguishes attribute spaces and representations as proposed in this paper is the idea of systems of predicates on attribute spaces corresponding to predicates on the domain. On the worldy side, a domain includes a set of relevant predicates talking about individuals. These predicates have counterparts on the representational side talking about points of an attribute space. Counterpart predicates are required to be consistent with their originals; more precisely, they have to agree in truth-value on the set of positive and negative exemplars thereby approximating the original predicates. Moreover, counterpart predicates will be assumed to have convex and open extensions. This system facilitates a qualitative notion of similarity which is suited to account for the meaning of natural language similarity expressions and, furthermore, their role in comparison and ad-hoc kind formation.

Numerical Concepts in Context

Numerical concepts are an integral part of everyday conversation and communication. Expressions relating to numbers in natural language can have precise or imprecise interpretations. While the precise interpretation most prominently appears in mathematical contexts, the imprecise interpretation seems to arise when numbers (as quantities) are applied to real world contexts (e.g., the rope is 50 m long). Earlier literature shows that the (im)precise interpretation can depend on different factors, e.g., the kind of approximator a numeral appears with (precise vs. imprecise, e.g., exactly vs. roughly) or the kind of numeral itself (round vs. non-round, e.g., 50 vs. 47). We report on a corpus-linguistic study and a rating experiment of English numerical expressions. The results confirm the effects of both factors and additionally an effect of the kind of unit (discrete vs. continuous, e.g., people vs. meters). This shows the contextual variability in the interpretation of numerical concepts in natural language.

Does the Activation of Motor Information Affect Semantic Processing?

Several behavioral studies show that semantic content influences reach-to-grasp movement responses. However, not much is known about the influence of motor activation on semantic processing. The present study aimed at filling this gap by examining the influence of pre-activated motor information on a subsequent lexical decision task. Participants were instructed to observe a prime object (e.g., the image of a frying pan) and then judge whether the following target was a known word in the lexicon or not. They were required to make a keypress response to target words describing properties either relevant (e.g., handle) or irrelevant (e.g., ceramic) for action or unrelated to the prime object (e.g., eyelash). Response key could be located on the same side as the depicted action-relevant property of the prime object (i.e., spatially compatible key) or on the opposite side (i.e., spatially incompatible key). Results showed a facilitation in terms of lower percentage errors when the target word was action-relevant (e.g., handle) and there was spatial compatibility between the orientation of the action-relevant component of the prime object and the response. This preliminary finding suggests that the activation of motor information may affect semantic processing. We discuss implications of these results for current theories of action knowledge representation.