A Master’s thesis submitted in partial fulfillment of the requirements for the degree of

EFFECT OF SPATIAL PRIMING ON AFFECTIVE CONCEPTS ACCESS: A LEXICAL DECISION AND REACTION TIMES STUDY

by

Ernesto E. Guerra

A Master’s thesis submitted in partial fulfillment of the requirements for the degree of

Master of Science in Clinical Linguistics

at the Joint European Erasmus Mundus Master’s Programme in Clinical Linguistics (EMCL)

UNIVERSITY OF POTSDAM

January, 2010

EFFECT OF SPATIAL PRIMING ON AFFECTIVE CONCEPTS ACCESS:

LEXICAL DECISION AND REACTION TIMES STUDY

Ernesto E. Guerra

Under the supervision of Robin Hörnig at the University of Potsdam and

Laura D’Odorico at Universita degli Studio di Milano-Bicocca

ABSTRACT

Background: The organization of the human conceptual system is a central issue for cognitive sciences. Conceptual Metaphors Theory, proposed that this system is fundamentally metaphorical, based on the description of linguistic data that systematically linked concrete concepts with abstract ones. However, advances in neurosciences and cognitive psychology confronted the theory with new challenges. Intending to inspect two aspects that still controversial; the psychological reality and the cross-linguistic validity of conceptual metaphors, the present study examines the relation between verticality and affective concepts.

Methods: A spatial concept (verticality) was primed before words with positive and negative

affective information (valence) were presented. A sample of 41 participants performed a

lexical-decision task in Spanish with, while reaction times were recorded. Results: Statistical

analysis showed a significant interaction effect between position and valence, in a way that

was coherent with the description provided by conceptual metaphors. Discussion: The

findings of the present study give support to the main assumptions of this theory. However,

they demonstrated as well that this field of investigation is far to a conclusion and further

developments with appropriate research approaches can still deliver helpful information for

central questions in cognitive sciences.

ACKNOWLEDGMENTS

I would like to give special thanks to all participants at Universität Potsdam and at Max

Planck Institute of Molecular Plant Physiology in Golm, that participate voluntarily in this

study, in particular to those that selflessly build up human links in order to help me achieve

my purposes. To them I owe the success of this project. I also would like to thank to Robin

Hörning for his valuable help during the experimental design and data analysis, to Claudia

Gil for her indispensable comments on early versions of this thesis and finally to all the

Faculty of the European Master in Clinical Linguistics in Milan, Groningen and Potsdam for

their warm support, encouragement and advisement.

TABLE OF CONTENTS

ABSTRACT……….. iii

ACKNOWLEDGMENTS………. iv

LIST OF TABLES……….vii

LIST OF FIGURES……….……….. viii

1 Chapter 1: Introduction 1.1 Background……… 1

1.2 Research Focus……….. 3

1.3 Overall research aim and individual research objectives……….. 4

1.4 Value of this Research………... 5

1.5 Outline Structure……… 5

2 Chapter 2: Literature Review 2.1 Introduction……… 8

2.2 Conceptual Metaphors Theory and the Cognitive Linguistics: Theoretical and Epistemological Context……….. 9

2.1.1 The Two Commitments………. 9

2.3 Toward a Neural Theory of Language: Psychological Reality………..12

2.4 Conceptual Theory of Metaphors………... 14

2.4.1 Target and source: directionality, hiding-and-highlighting, and entailment………...16

2.4.2 Metaphor systems and image schemas: inheritance relations and invariance………. 17

2.4.3 Primary Metaphors and a Neural Theory of Metaphors……… 20

2.5 Summary……… 27

3 Chapter 3: Experimental Approach to Conceptual Metaphors 3.1 Introduction………29

3.2 Experimental Background on the Study of Conceptual Metaphors……….. 29

3.3 Verticality and Emotional Conceptual Metaphors……….35

3.4 Preview of the Study………..39

3.5 Summary……… 41

4 Chapter 4: Research Methods and Empirical Findings

4.1 Research Methods...………42

4.1.1 Participants……….42

4.1.2 Materials……… 43

4.1.3 Procedure………... 44

4.1.4 Research Hypothesis………. 45

4.2 Empirical Finding……….. 48

4.2.1 Framework for Data Analysis………48

4.2.2 Statistical Analysis of the Data……….. 50

5 Chapter 5: Conclusions and Discussion. 5.1 Introduction………52

5.2 Empirical Research Finding Interpretation………53

5.3 Summary of Findings and Conclusions………. 56

5.3.1 Objective 1: The emerging context Conceptual Metaphors……….. 57

5.3.2 Objective 2: The psychological reality of HAPPY IS UP and SAD IS DOWN ………... 58

5.3.3 Objective 3: Experimental Approach to HAPPY IS UP , SAD IS DOWN in Spanish……… 59

5.3.4 Objective 4: The validity of metaphorical properties ………... 60

5.3.5 Synthesis of Literature and Experimental Findings………...61

5.4 Embodied Cognition and Affective Conceptualization………. 62

5.5 Final Remarks……… 65

REFERENCES……….66

APPENDIX Appendix A: Word and Pseudo Words………..75

Appendix B: Instructions………... 77

LIST OF TABLES

4.1 Summary of factors’ facilitation effects...…... 47

5.1 Summary of average responses latencies………... 55

LIST OF FIGURES

4.1 Possible scenarios for different factors’ contribution ………... 48

4.2 Lexical Decision Latencies for Words………...50

Chapter 1: Introduction

1.1 Background

As simple as it could sound, just by observing people’s use of everyday language one can realize that it is frequently metaphorical. We use a wide range of expressions such as “I feel up today” to communicate that “today I feel happy”, or “he fell into a depression”, when we want to express that someone is affected by a mood disorder. Generally speaking, we make use of metaphorical expressions whenever we are able to understand one concept (object, subject, experience, etc) in terms of another concept; in other words X is Y, which entails a fundamental pragmatic communicative value (Reddy, 1979). For example, when we want to transmit an abstract complex concept, for instance affectivity, we use more concrete concepts, such as temperature (e.g. she is very warm, but her husband seems to be a cold person). One dimension of human experience in which people use a lot of metaphorical

expressions is the emotional world, and this is because it is usually highly complex and, therefore, hard to express in a few simple words. Figurative language helps to effectively communicate and share our internal states with others.

One well known and influential approach to the study of metaphors is the Conceptual

Metaphors Theory (CMT) by Lakoff and Johnson (1980b, 1999). From this perspective, the

extensive use of metaphorical expressions reflects that the human conceptual system itself is fundamentally metaphorical. It contains metaphorical and nonmetaphorical concepts.

Metaphorical concepts are abstract and complex and they are understood in terms of other concepts, as in the former example affectivity is temperature. On the simple equation X is Y, where X recalls a source domain and Y a target domain; metaphorical concepts are the target domain. Nonmetaphorical concepts instead, cannot be understood in terms of other concepts.

They emerge early in development directly from highly recurrent sensoperceptual experiences; they are source domains (Lakoff and Johnson, 1980a).

CMT is a theory among Cognitive Linguistics, a framework concerned with the relationship between human language, the mind and socio-physical experience (Evans, 2009).

From this framework, it is argued that thinking (and language as its manifestation) is highly dependent on perceptions, actions and experience; the embodied hypothesis. Human body, consequently, should play a crucial role in influencing and constraining human cognition (Hurtienne, 2009; Zlatev, 2007). CMT, and Cognitive Linguistics, appears to be strongly associated with Embodied Cognition framework. It describes metaphors not only as a speech figure, but indeed a figure of thought. Abstract concepts, for instance emotions like happiness or sadness, are mentally represented and structured in terms of metaphors (Gibbs

and Steen, 1999). Those recurrent meaningful body-based concepts are brought and extended to abstract thought by means of metaphors (Johnson, 2008).

Although CMT has demonstrated to be an influential and robust theory, there are still

some detractors (Keysar, Shen, Glucksberg & Horton, 2000) and critics (Murphy, 1993,

1997). In addition, it also had confronted changes on its foundation thanks to empirical

(Feldman & Narayanan, 2004) and theoretical advances (Grady, 1997). This means that on

one hand, CMT is still a controversial theory. On the other hand, it means that CMT has

showed itself as a theory which has been capable of enriching itself from new findings

(Gallese and Lakoff, 2005). In this sense, Conceptual Metaphors Theory has had to deal with two main sources of challenge, the neural networks approach and the cognitive-behavioral approach to the study of language. The major question arising from these two trends is about the psychological reality of the metaphorical organization of human conceptual system.

1.2 Research Focus

As exemplified above, emotional states are often communicated metaphorically. They can be expressed in terms of temperature, distance, brightness, even flavour and verticality among other source domains. The present study addresses the relation between affective concepts and spatial cognition. More concretely it concentrates on the primary metaphors

HAPPY IS UP, SAD IS DOWN . Albeit early described by Lakoff and Johnson (1980b), this topic has only lately being studied from a behavioral approach (Meier & Robinson, 2004, 2006). Following Joseph Grady, the enterprise of testing the cognitive reality of the mapping described by CMT is of great importance for Cognitive Linguistics, since it considers the study of language as a way to access to other cognitive functions (Generalization Commitment, see section 2.1.1). The systematic and predictable association between concepts in linguistics expression shows connection on some levels of understanding (see Lakoff, 1987), but “evidence from experimental psychology also helps confirm the cognitive reality of conceptual metaphors” (Grady, 2007: p. 196).

For example, Meier and Robinson (2004) designed a set of experiments that

emphasizes the theory of conceptual metaphors in a reaction times paradigm. Their results

give support to various concepts of the theory. However, in a subsequent article the same

authors suggest that “Although the studies reviewed [,] provide support for the metaphorical

representation of affect, there is plenty of room for future work [and] future research may

also seek cross-cultural verification” (Meier and Robinson, 2005: p. 253). Consequently,

weather as corroboration or as the exploration of new aspects of it, the study of the

relationship between affectivity and verticality within the framework of Cognitive Linguistics is still necessary.

Two main current challenges for research on CMT are to confirm the psychological reality of metaphorical conceptualization and then to explore the cross-linguistic validity of them. The implementation of a behavioral paradigm in Spanish that tests this hypothesis, should deliver valuable information to contribute to answer both fundamental questions. The present study is intended to evaluate the primary metaphors concept and some key principle within CMT. In addition, it aims to be a suitable contribution to the notion of embodied cognition.

1.3 Overall research aim and individual research objectives

The main objective of the present study is to examine the psychological reality of conceptual metaphors and its cross-linguistic validity. In order to do that is crucial to use a suitable methodology approaching. First of all, the review of relevant theoretical and empirical literature is necessary to provide an adequate background to understand the main objective. Subsequently, a sound experimental approach is needed to collect reliable data that shed some lights on these questions. Therefore, a lexical- decision task in with reaction times measures was implemented. A description of the specific objectives that appear to be highly relevant to achieve the aim of the research is given as follows:

OBJECTIVES OF THE STUDY

1 Describe the context in which Conceptual Metaphors Theory was developed.

2 Examine the psychological reality of primary metaphors of affect and vertical spatial perception.

3 Explore the interaction between affect conceptualization and verticality in Spanish.

4 Evaluate the validity of Conceptual Metaphors’ description of metaphorical properties and primary metaphors.

5 Illustrate the role that embodied cognition plays in the affective

conceptualization system.

A review of relevant literature situates the reader in an adequate position to understand the main assumptions and purposes of the Conceptual Metaphors Theory.

Objective 1 is aimed to provide of this ground. Objectives 2 and 3 worked as the operationalization of the main research objective providing access to the two central problems in question. Objectives 4 and 5 are more specifically orientated. Objective 4 orientates the analysis of the empirical data while, and finally Objective 5 should be achieved by the critical valuation of the results.

1.4 Value of the Research

On one hand, as it has been outlined and is described in more details in the following chapter, CMT is not free of controversy, and research groups within the framework and Cognitive Linguistics, insist on the necessity of more research on the topic. On the other hand, the latest study on affective concepts and its relation with verticality appears to be unclear still, and is hard to generate strong conclusion over the available data. The value of the present research is that it addresses both mentioned points. First, it is an empirical approach to CMT, in a widely spoken language that has been not yet analyzed in this domain thus it moves towards a cross-cultural comparison. Second, it implements a sound experiment design intended to generate reliable data on the way of exploring the relation between affect conceptualization and verticality.

1.5 Outline Structure

Chapter 1: Introduction

This chapter provides the reader with background information on some of the current

study of metaphors, including illustration of the importance of this topic for linguistic and

cognitive research, a relevant approach to it and its open questions. The focus of this research

is discussed and justified and the overall research aim and individual research objectives are identified. At the end, the possible contributions of this study are briefly commented.

Chapter 2: Literature Review

Chapter 2 presents an examination of the relevant literature on the cognitive study of metaphors, concentrating in three main issues: the epistemological context in which the theory of conceptual metaphors emerged; the psychological reality of this linguistic recurrence; and the main assumptions that give support to the theory. It provides a historic overview of the theory, highlighting improvements and current challenges.

Chapter 3: Experimental Approach to Conceptual Metaphor

As a necessary complement of the previous one, this chapter presents an experimental review. It provides a general view of experimental approaches to conceptual metaphors and

an in-depth review of the experimental study of the relation between affective conceptualization and verticality. At the end, Chapter 3 provides a study’s preview, aimed to

draw the fundamental connection between relevant methodologies and current questions.

Chapter 4: Research Methods and Empirical Findings

A detailed description of the participants, material and procedures is supplied, followed by consistent hypothesis concerning potential outcomes. Before the review of the main empirical findings, a framework for data analysis explaining the necessary steps to provide a statistical conclusion is given. The second part of Chapter 4 reports on the findings from the experimental research, which is intended to provide a dense account of the statistical result.

Chapter 5: Conclusions and Discussion

The last chapter is aimed to tackle Objectives 4 and 5 of the present thesis. It begins

with the interpretation of the empirical findings that provides a theoretically coherent account

for the results. Consequently, a summary of findings and conclusions supplies an overview of

the results for each individual objective, in addition to an integrative conclusion at the end.

The account for individual results offers substantial arguments to discuss CMT assumptions and to depict the connections between embodied cognition and affective conceptualization.

At the end, some final remarks on the main conclusion of the study work as a reflection about

the impact of the results and further possible steps.

Chapter 2: Literature Review

2.1 Introduction

The present literature examination is intended to revise three main levels: a) the paradigmatic context in which CMT emerged, b) the central theoretical assumptions and concepts, and c) some of the remaining questions within the theory. Chapters 2 and 3 focus on the first two objectives of the present research. Therefore, they offer the theoretical bases for the correct empirical approach to address the Objectives 1 and 2.

This review contributes to the study by presenting an up-to-date assessment of the

state of art in metaphors research from a logical and focused perspective, particularly from an

embodied and cognitive perspective. The most important aspects of theory and paradigm are

described. Moreover, a critical in-depth evaluation of the relevant research on the area is

presented. First, it examines in which levels the psychological reality of conceptual

metaphors has found support, and in which levels is still in need of being revised. Second, the

review explores the research made in these levels, from a cross-linguistics perspective. At the

end, this revision provides a synthesis of what has been done and what is needed to be done in the field, by exhibiting the crucial issues and plausible answers, articulating the justification of the present empirical research study. In order to achieve all these purposes, the literature review starts with the description of the main framework in which Conceptual Metaphors Theory is situated: Cognitive Linguistics.

2.2 Conceptual Metaphors Theory and the Cognitive Linguistics: An Epistemological and Theoretical Context.

Cognitive Linguistics is an interdisciplinary approach to the analysis of natural language that treats it as an instrument for organizing, processing and conveying information.

It is distinguished from other approaches to natural language for studying the formal structures of language not as autonomous and independent processes, separated from cognition, but instead as expression of more general conceptual organization, categorization principles, processing mechanisms, and experiential, environmental and cultural influences (Geeraerts & Cuyckens, 2007). As an interdisciplinary framework on language research, it has been always influenced by other cognitive sciences, especially by cognitive psychology (Evans, 2009). It has demonstrated to be highly tolerant towards both internal variety and external interaction. It is concerned with the relations between cognition and social, psychological, pragmatic and discourse orientated dimensions of language (Ruiz de Mendoza and Peña, 2005). All these trends convey the importance that Cognitive Linguistics attributes to conceptual and experiential basis of linguistic categories which appears to be relevant for the present study.

2.2.1 The Two Commitments

Cognitive Linguistics does not intend to present itself as a single theory, but rather as

flexible framework for the study of language. Nevertheless, theoretically speaking, any strand

of Cognitive Linguistics should be guided by two fundamental principles or “commitments”

after Lakoff (1990, 1991). The first is the Generalization Commitment, which essentially goes against the understanding of “modules” of language, independently organized in significant different ways such as phonology, syntax and semantics. It recognizes that this division might be useful for practical purposes and that in fact, the study of syntactic organization involves slightly different cognitive and linguistic phenomena than the study of phonological organization. However, these formal approaches explicitly hypothesize that they are mechanical devices that operate these subsystems in order to produce the complete set of linguistic possibilities in a given language (Lakoff, 1990, 1991; Evans and Green, 2006). Instead, Cognitive Linguistics posits that different areas of human language share organizing principles. For example, as mentioned above, metaphor entails the equation X is Y, therefore this phenomenon is a meaning extension, but, should it be exclusive material of

“lexical-semantic domain”? As Goldberg (1992) illustrates, metaphors appear in other

“modules” as well, namely in syntax: The case of ditransitive construction generally requires a volitional AGENT in subject position, because it expresses an intentional transfer (Goldberg, 1992). Even so, the following instances show a non-volitional AGENT in subject position.

(1) a. [

The sun] gave me a beautiful day.

b. [

This thesis] gives you a new perspective.

As it can be seen, this ditransitive construction uses objects in the subject position and still is both, grammatical and semantically, correct.

The second principle is the Cognitive Commitment. Language and linguistic

organization must reflect general cognitive principles, and not only specific cognitive

principles for language. In this sense, it is crucial that linguistic structure principles follow

findings on human cognition in the other cognitive sciences (Lakoff, 1990, 1991), namely

cognitive psychology, neurosciences, philosophy and artificial intelligence and subdisciplines

(Varela, Thompson and Rosch, 1991). As in the first commitment, which rejected a modular

theory of language, the second rejects a modular theory of mind. This formal theory states

that the organization of human mind is based on different independent subsystems of knowledge, language among others. If modules should serve as sensorial input filters for a central cognitive system, which involves reasoning, deduction, and so forth, so it can process that information (Evans and Green, 2006), an approach that goes along with the embodied thesis does not agree with a modular view.

Cognitive Linguistics affirms that “metaphor is a matter of thinking, not a matter of language […] and allows one to reason about, not just talk about, one thing in terms of another” (DesCamp, 2007: p. 21) and supports this understanding on extensive evidence.

This view of many aspects of our cognition as essentially metaphorical has been strongly influential (see Gibbs, 2008; Kövecses, 2006). For example, we conceptualize emotions in terms of verticality (Meier and Robinson, 2004, 2006). A schema of emotion as such, describes someone who is happy in an upper position than someone who is sad, who is metaphorically located down (or falling, or in a hole). Therefore emotions, or at least some aspects of it, are conceptualized metaphorically and represented non-linguistically in terms of the conceptual metaphors HAPPY IS UP and SAD IS DOWN. Being emotions such a complex phenomena, there are multiple ways to refer to them ( GOOD IS BRIGHTNESS , Meier, Robinson and Clore, 2004) depending on what type of emotion we experience ( ANGER IS THE HEAT OF THE FLUID IN A CONTAINER , Kövecses; 1986) and the cultural constrains (Kövecses, 2000, 2003), nevertheless all of these ways to communicate our affective internal state are metaphorical. The present study concentrates in two conceptual metaphors mentioned above:

HAPPY IS UP and SAD IS DOWN.

Why should one ascribe to this approach? The first advantage of approaching the

study of language from Cognitive Linguistics is that they are profoundly compromised with

the empirical study of language. This means that no a priori assumptions about the nature of

its object of study are made. Instead, it treats its judgments as empirical matters. The second

benefit is its interdisciplinary nature. In this sense, Cognitive Linguistics is constantly enriched by findings in cognitive and developmental psychology, cognitive anthropology, neurobiology, and so on (Lakoff, 1991). On one hand, this exercise helps to have an up-to- date state of art in the field. On the other hand, it promotes the integrative research and gives linguistics a deserved place within cognitive sciences. Following these two strong reasons, it is assumed that this framework is appropriate to address the current research problem.

Cognitive Linguistics approaches, and more specifically Conceptual Metaphors Theory, based their claims on empirical findings. It is open to engage in integrative and interdisciplinary explanations and it goes along, theoretically and empirically, with the embodied hypothesis. All this together make it possible to achieve models of human language that are psychologically, socially and biologically coherent with each other. The following section describes the paradigmatic viewpoint and the “evolutionary path” of the CMT.

2.3 Towards a Neural Theory of Language: The Psychological Reality.

In 1980, George Lakoff and Mark Johnson released Metaphors that We Live By, an influential book that changed profoundly the existing understanding of metaphors. In terms of the paradigmatic change, they criticized the assumption that most of our thinking is literal and directly corresponding to the external world. Instead, they proposed that we construct meaning in and through the language, and that this is a reflection of our experiences as a whole (Geeraerts, 2006). This distinction is important as evidence of the transition from first generation to second generation cognitive science in linguistics (Lakoff and Johnson, 1999).

In cognitive science, there is a separation between disembodied and embodied paradigms.

Early cognitive science assumed a dualistic perspective in which the mind is disembodied

and defined by its formal structures. Contrary, second generation cognitive sciences assume a

embodied approach to the mind, therefore recognizing the significant function of body and

experience in all aspects of meaning and in the structure and content of human thought

(Varela et al., 1991). According to Rohrer (2007), this strand of Cognitive Linguistics, the cognitive semantics approach to the study of metaphors or CMT is, in this sense, specially associated with the Embodiment Hypothesis.

Historically the concept of Embodiment has had different uses. Referring to the embodied hypothesis in its broad sense it could be said that it has two main characteristics.

First, it considers that the conceptual and linguistic human system is grounded in our human

physical, cognitive and social embodiment; second, although it is recognized that such a

theoretical claim is difficult to prove to be true or false with one research or experiment, the

embodied hypothesis intends always to be an empirical hypothesis. These two aspects lead to

the use of a great variety of the term “embodiment”, for example, following Rohrer (2007)

there must be at least twelve ways to refer to embodied cognition. Naturally, it is not the

purpose of the present study to determinate which of these uses is the most appropriate (in

fact that is highly dependent on the purpose of any particular research project). With a

practical purpose, these definitions can be divided in two main domains or sense of use. The

first group considers Embodiment as the bodily substrate: more orientated to physiological,

neurophysiological and neurocomputational substrate. The second contains all definitions

where Embodiment is considered broadly experiential: more orientated to context, culture

and history of the languages’ speakers. Good research in Cognitive Linguistic frame would

include at least these two levels, providing a non-reductionist way to join them (Rohrer,

2007). However, not all definitions can be clustered that radically. For instance, two

important notions can be merely put in both levels and even work complementary. One is

Embodiment accompanying developmental changes, which refer to the constant ontogenetic

process that every being goes through (MacWhinney, 1999; Tomasello, 2003). The second is

the Embodiment accompanying evolutionary changes, linked to the human phylogenetic path

(Tomasello, 2008); they both involve neurophysiological substrate and environmental constrains as well.

The present project included one of each level in order to fulfill Tim Rohrer’s requirement for good research. From the first sense of use, Embodiment is taken as referring to neurocomputational model of language (Feldman, 2006; Gallese and Lakoff, 2005). This level appears to be particularly relevant for the interaction of conceptual metaphors and spatial language. Following Feldman and Narayanan (2004), a neural network of abstract nature is embodied since it uses as its input-structures, outputs from other embodied neural networks that are better understood, and commonly those are perceptual modalities. The examination of on-line behavioral data implies the assumption that the underlying process is a neural process and therefore can provide information of associated neural network (Elman, 1991, 1993). From the second dimension, Embodied is understood in relation to the Social and Cultural Context in which cognition, body and language are immersed. This level is represented by the influence of historical events on individual learners, institutions and cultural artefacts; the examination of a widely observed linguistic phenomenon in a less studied language can contribute to this point.

From the specific epistemological perspective brought by the embodied hypothesis, the inclusion and consideration of these different levels is a way to address the ongoing organism-environment interaction, therefore seriously addressing their psychological reality them (Johnson and Rohrer, 2007). After having portrayed the epistemological scenario, the following subsection describes the main assumptions and principles of the Conceptual Metaphors Theory.

2.4 Conceptual Metaphors Theory.

Undoubtedly, the origins of human conceptualization are central in cognitive sciences.

Of the same interest is the question about how are we capable to communicate and actually

share our internal conceptual world with other humans. Lakoff and Johnson (1980b, 1999) postulate that conceptual systems are fundamentally metaphorical, since abstract concepts (target Y) are defined and understood in terms of other concepts (source X; X is Y). But what are these “other” concepts? Nonmetaphorical concepts have their origin in purely senso- motor knowledge, which is to say pure experience. They are images schemas, a dense redescription of perceptual experience intended to map spatial structures onto abstract structure (Oakley, 2007). They cannot be use as target because they are defined in their own terms. Lakoff and Johnson (1980a) gave a description of at least, three dimensions in which nonmetaphorical concepts can be clustered: a) spatial orientations, which include notions as

UP-DOWN, IN-OUT, NEAR-FAR , and FRONT-BACK; b) ontological concepts, like ENTITY, SUBSTANCE, PERSON , and CONTAINER; c) structured experiences and activities, including

EATING, MOVING and TRANSFERRING OBJECT FROM PLACE TO PLACE.

Contrary to nonmetaphorical concepts, metaphorical concepts are not fixed. They are creative, rich and flexible, although topologically limited. With a schematic purpose, it can be said that these nonmetaphorical levels, described above act like based for three types of

METAPHORICAL concepts: a) orientational metaphors b) ontological metaphors and c) structural metaphors. These three groups offer a wide number of metaphorical uses. For instance, in a) we found HAPPY IS UP, SAD IS DOWN (of which examples appeared above) ; MORE IS UP, LESS IS DOWN and FUTURE IS IN FRONT, PAST IS IN THE BACK.

(2) a. The amount of money in your account is going up (or down).

b. My birthday is only two weeks ahead (or away).

c. This movie takes me back in time.

While in b) we observe that THE MIND IS A MACHINE (ENTITY), VITALITY IS A SUBSTANCE, INFLATION IS AN ADVERSARY (PERSON) , and VISUAL FIELDS ARE CONTAINERS.

(3) a. My head is not working very well.

b. She is full of vitality.

c. The inflation is strong but it will not defeat us.

d. The land came into the captain’s view.

Finally, c) presents us structural metaphors like UNDERSTANDING IS SEEING or LIFE IS A JOURNEY (Lakoff and Johnson, 1980a, 1999; Lakoff and Turner, 1989; Grady, 2007).

(4) a. Now I can see what you mean.

b. His life turned into a new direction

What have been described are just some instances, among many, of common use of metaphorical expression in terms of corresponding nonmetaphorical concepts. These all come from conversational analysis and corpus data analysis. The following two sections describe some of the main concepts and principles among the theory.

2.4.1 Target and source: directionality, hiding-and-highlighting, and entailment.

So far the concepts of TARGET DOMAIN and SOURCES DOMAIN have been mentioned quite a bit. Nevertheless, an important aspect has not been remarked yet: the directionality of metaphors. This principle states that metaphors can map a structure from a source domain to

a target domain but not reversely (Evans and Green, 2006). By mapping, what is meant is the projection of a frame, which is a detailed knowledge structure emerging from everyday experiences (Gentner and Markman, 1997). This reasoning can easily be illustrated with a few linguistics examples. Taking another prototypical conceptual metaphor one can say that

THEORIES ARE BUILDINGS , yet BUILDINGS ARE THEORIES does not apply. Using this metaphor the sentences in (5a) is easily understood, but it would appear odd and unusual to use it in the inverse sense like in (5b).

(5) a. Its foundations made it a strong theory.

b. The postulates of my house made it strong.

This example appeared to be clear-cut from a linguistic point of view; THEORIES ARE BUILDINGS . It is just that houses cannot be made of postulates. However, there are some other instances where this becomes apparently problematic. Let’s consider the conceptual metaphor

PEOPLE ARE ANIMALS , and ANIMALS ARE PEOPLE.

(6) a. The king of England is a lion.

b. The lion is the king of the jungle.

In (6a) and (6b) we observe that these metaphors seem to be correct in both directions; and they are. But in fact they are two different metaphors since they are based on different mappings. In (6a) the lion’s attributes, such as physical strength and aggressiveness are mapped onto the king of England to exalt his characteristics. In (6b) what is actually mapped onto the lion are the features that the royal title of king provides, such as being highly respected (or feared) and to have power over others (other animals in this case). It also involves another metaphors as JUNGLES ARE KINGDOMS, whereas the mapping corresponding to order and hierarchies, among other features. Same with KINGDOMS ARE JUNGLES , the features mapped are distinct; perhaps this kingdom is ruled by the law of the strongest.

As can be observed from the latter description, metaphors are much more complex phenomena than just X is Y. They involved generally a complex system of mapped structures and they convey a lot of information. This leads to two other important characteristics of metaphors. In addition to the explicit information, metaphors infer implicit knowledge carried out by the metaphorical mapping. These two levels of conveyed information give rise to the principle of hide-and-highlight. Metaphors always standout information that is more relevant or meaningful in a specific context but that is not all the information that it supplies.

Moreover, both hided and highlighted contents are metaphorical entailments. The metaphor

LION IS A KING entails that for example, JUNGLES ARE KINGDOMS implicitly , but explicitly that PHYSICAL STRENGTH IS SOCIAL STATUS and so forth. From this overt complexity another set of concepts emerge.

2.4.2 Metaphor systems and image schemas: inheritance relations and invariance

Since metaphors operate in complex ways and interact with each other, the result of

this interaction is a metaphor system. Metaphors systems are a set of schematic metaphorical

mapping that works in such manner that gives structure to more specific metaphors in a great

diversity. According to Lakoff (1993), this generates particular complex metaphorical systems, for instance, LIFE IS A JOURNEY. By means of this example, Evans and Green (2006) offer an illustrative decomposition of the individual metaphors contained in this metaphorical system as follows:

(7) a. STATES ARE LOCATIONS:

He is stuck in his adolescence.

b. CHANGE IS MOTION:

We better move on from this topic.

c. CAUSES ARE FORCES:

The lack of scholarship pushes him to find a job.

d. PURPOSES ARE DESTINATIONS:

She is arriving to the peak of her career.

e. MEANS ARE PATHS:

They go the exact way to succeed.

f. DIFFICULTIES ARE IMPEDIMENTS TO MOTION:

That problem on the experiment paralyzed my thesis for a while.

g. PURPOSEFUL ACTIVITIES ARE JOURNEYS:

At the end of the road, writing the thesis was a lot of fun.

The entailment of all of these metaphors by this event structured metaphor helps to comprehend it. At the same time it provides an extended meaning to utterances that engaged with it. This is because of the very nature of hierarchies of specificity on metaphorical structures. They are organized in a way that the more specific patterns are inherited in a more general one (Grady, 2007). All metaphorical expressions in (7) inherit structure from the metaphor LIFE IS A JOURNEY. As it could be inferred by (7g), even more specific metaphors can inherit its structure, and they entail a set of other metaphors as well.

(8) a. AN ARGUMENT IS A JOURNEY

She went on her reasoning from the beginning to the end.

b. LOVE IS A JOURNEY

We have got very far together, but I just ran out of gas.

c. A CAREER IS A JOURNEY

She is arriving to the peak of her career.

An important feature for metaphorical systems are images schemas. They were already defined as redescriptions of perceptual experience that map spatial structure onto abstract structure. According to CMT images schemas serve as sources domain for metaphorical mapping precisely because, since they emerge mainly from our objects’

manipulation, our perceptual interaction and our bodily movements through space, they are essentially meaningful structures in a pre-conceptual level (Johnson, 1987). For example, when people refer to the metaphor ANGER IS A BOILING FLUID LIQUID like in (9) , they do so motivated by the embodied experience of being angry, involving higher blood pressure and body temperature.

(9) a. English

He was filled with anger.

b. Chinese

Man qiang fen nu

“To have one’s body cavities full of anger”

c. Spanish

Él estaba lleno de ira.

“He was full of rage.”

Naturally, the presentation of this metaphor received influences from the cultural context and they are differently realized (Kövecses, 2000). Finding these expressions in such different languages and cultures (English, Chinese and Spanish) is strong evidence that there must be some source of relationship between the linguistic expression and bodily- experienced emotions. This is explained by another important principle of CMT, the invariance principle according to which the metaphorical mapping between source and target

should always go along with the topological structure (the image schema) of the target domain (Lakoff, 1990, 1993). As stated before, the mapping from one conceptual domain to other is a complex process that can be represented as X is Y, nevertheless not explained that way. Metaphorical mapping involves much more than just objects and shared characteristics;

it does involves contexts (physical, social and psychological environments), and events and

relationships (metaphorical entailments and inheritance relations) as well (Grady, 2007).

Therefore, all of these elements act to produce and constrain metaphors at the same time. As noticed by Feldman and Narayanan (2004), we all share the same semantic potential due to the same shared neural circuit and, at the same time, different languages explicitly label aspects of life in different way. But there is something saying what can be said and not said about those aspects, which works as a complexity barrier. Because metaphors are based on images schemas and, images schemas are based on the interaction of different perceptive channels that human from different cultures share, then metaphors should be flexibly shaped by culture but inescapably embodied.

Based on the described principles, CMT became an influential theory among Cognitive Linguistic and cognitive sciences, causing that many researchers and research groups kept studying issues on metaphors from a cognitive semantic perspective. CMT offered strong arguments for its account; however, some methodological and theoretical challenges stressed the theory. Next section describes the update of concepts and the computational approach to the study of metaphors that followed the second half of the ‘90s.

2.4.3 Primary Metaphors and a Neural Theory of Metaphors

After mid-‘90s, the classical version of CMT, suffered some modifications. This

update included some new concepts (and the rejection of some conflicting ones) that were

significant to the further development of the theory. This actualization appears to be proving

of the strength of the cognitive commitment. On one hand, CMT needed to act in accordance

with the requirement from psycholinguistics, cognitive psychology and developmental

psychology for independent empirical evidence on the psychological reality of its main

aspects (Gibbs and Colston, 1995). As consequence, some of the core concepts of the initial

version of CMT were reviewed and updated (Grady, 1997, 2007; Johnson, 1999) to provide

better explanations to the nature of metaphorical conceptualization. On the other hand,

Conceptual Metaphors Theory had to confront the rapid development of neurosciences, both from computational neural networks (Feldman and Narayanan, 2004; Feldman, 2006) and brain imaging studies (Rizzolatti and Arbib, 1998).

Cognitive Studies. One of the important critics was centred in the invariance principle (Murphy, 1996; Grady, Taub, and Morgan, 1996). It was found problematic because it claimed that inherent structures of target domains constrain the possible mapping and entailments for it and at the same time, target domains were abstract and not clearly delimited entities. This appears to be at some level contradictory, since a target domain that already comprises a structure needs no other structure to be map onto it, as CMT stated. Confronted with this, Joseph Grady (1997, 2007) proposed a plausible explanation to solve this issue.

According to him, there are actually two types or level of metaphors, primary metaphors and compound metaphors. They have remarkable differences but are intimately related. Primary

metaphors work as ground for compound metaphors; they are foundational subjective interpretation of direct experience. For example, one can experience “anger” and make a fast and subjective interpretation about an embodied state, but one cannot experience “argument”

in this way. Instead argument is cognitively and consciously understood. Both concepts can be expressed in a metaphorical manner, but the first would be a primary metaphor and the second a compound one. Compound metaphors are constructed from the unification of various primary metaphors.

This distinction leads to an important theoretical change for CMT. The role of the

level of abstraction of a concept involved appears to be not as relevant for distinguishing

between target and sources as it is the level of subjectivity. This means that the most

important distinctive aspect between primary source and target domains is that the sources

are direct senso-motor experiences, while the targets are the individual judgments,

assessments, evaluation and so on, of the particular sensory perception. Let’s consider some primary metaphors examples.

(10) a. SIMILARITY IS CLOSENESS

Your master thesis is close to mine b. SIGNIFICANCE IS SIZE

Today is a big day for us c. HAPPINESS IS UP

That grade on my thesis cheered me up

On one hand, primary target domains SIMILARITY, SIGNIFICANCE and HAPPINESS

have no direct relation with sensory perceptive processing

; instead they can be identified in various domains and that is the reason why they can be targets; but in addition there are personal judgments (about shared characteristics between A and B (10a); about the importance or relevance of an object or an event (10b); about an emotional state (10c)), and that is what makes them not only target but primary target. On the other hand, primary sources can only be identified in direct connection with sensoperceptive phenomena, being the case in these three examples of spatial perception.

Grady’s distinction between types of metaphors supplies an interesting explanation for the understanding of universal and cultural-depending metaphors. Primary metaphors are evaluations of some image schemas that all humans share and they provide the primary matter for compound metaphors. That explains why examples in (9), coming from different cultures tend to have the same underlying logic of construction. All human beings have the potential structure (body awareness) to evaluate their own ANGER , and the potential to physically perceive A FLUID IN A CONTAINER and also to perceive A FLUID BOILING . In contrast, since compound metaphors are flexibly constructed in base of these primary metaphors the particular way that this occurs is influenced by a particular cultural context.

1

Note: it appears very clear that emotions like happiness do have a direct relation with sensoperceptive levels as

physiological activation (e.g. heart rates), but as a concept or category, there is no overt reason why it should be

more connected with one specific sensorial level than another.

Nonetheless, even the use of specific primary metaphors changes between cultures and within cultures at different moments of time (Gevaert, 2001).

Following Kövecses (2005), the specific manner of awareness varies between people and it is strongly mediated by culture. With the concept of differential experiential focus, Kövecses intends to explain this variation in the use of conceptual metaphors. Since the target domains are complex conceptual structures, they are generally constituted by several aspects or components. The emerging conceptual metaphors can be based on one or more of these different aspects depending on cultural factors, or in different aspects depending on the moment on time, due to the dynamism of cultures. Culture plays such a mediating role, that even a universal or biological embodied experience can be completely excluded from the conceptualization of a specific domain (see Lutz, 1988). The main implications of the concept of differential experiential focus are that on one hand, embodiment does not play a mechanic role in conceptualization, but instead works as universal motivation, subject of cultural influences. On the other hand, embodiment should not be taken as a fixed phenomenon linked to a specific domain of experience but rather as a wide set of components that are candidates as source domains (Kövecses, 2006).

The merit of the distinction between primary metaphors and compound metaphors and the notion of differential experiential focus is that they allowed the theory of conceptual metaphors to account for linguistic examples that seem to be contradictories with the embodied hypothesis. Furthermore, it provides a framework that is sensitive to cultural sense of embodiment, understanding that biological/structural determinism and cultural/physical environment come naturally together shaping the way humans think.

Neural Studies. Current cognitive sciences states that every action that we perform, and every stimulus that we perceive is computed by our brain and body (Varela et al, 1991;

Gibbs, 2006). Based on the state of art of cognitive sciences, this claim is just undeniable. It

is simple not the case that thought and language can be something else but physical, and caused by the physiological structure of the nervous system. Neurons communicate by means of ions throughout synapses and every time the brain works this is happening. For example, when we process a word it takes as average half a second (500 milliseconds). If we know that the signal from one neuron to another take approximately 5 ms. that means that to process a word it takes around 100 steps, which are more likely to work in parallel and by means of direct connections than sequentially, due to the amount of information that a word is able to carry and the speed in which our conceptual system is able to process (Lakoff, 2008).

Another important property of the nervous system is that neurons form anatomical groups that are generally modelled as nodes, and the connection between different nodes gives origin to neural networks. Due to the large amount of neural connections that neurons build up these nodes can overlap, meaning that same neurons or groups of cells can function for different neural networks (Arbib, 2003). From a general neural networks frame, a concept is treated as neural circuit that is formed when neurons co-activate significantly, the Hebbian principle. The circuit of a conceptual metaphor is an especially complex one, since it is a network that links two (or more) networks. The primary metaphor mapping links source with a target. These types of circuit are formed when a perceptual and a non-perceptual concept are repeatedly co-activated; “if they fire together, they wire together” (Hebb, 1949).

According to Grady (2007; see also Feldman, 2006), this process is certainly crucial for the

origins of language and thought. From early stages of development, target and sources of

primary metaphors appear so closely associated that a combined representation is expected to

appear as one in the conceptual (neural) structure. In this sense, the framework offered by

primary metaphors theory, not only explains the surface of linguistic data; it presents

explanations that are coherent with what is known in artificial intelligence (AI),

neurosciences and other cognitive sciences.

Based on these findings Gallese and Lakoff (2005) argued that concepts (words) and conceptual knowledge are embodied, which is to say, they are mapped in the sensory-motor system. Strong support to this idea comes from studies in neurobiology that showed how mental simulation of actions or perceptions uses many of the same neurons as the actual action or perception (Gallese, 2003). Mirror neurons are multimodal neurons that are activated when a specific action is performed or when a specific action is perceived (e.g.

grasping something or seeing someone grasping something), but at the same time these neurons get activated when imaging the action or perception, that is a mental simulation of acting or perceiving the same action (Rizzolatti and Arbib, 1998). Under the same logic, a neural theory of language assumes that the same neural substrates uses for imaging are use for understanding. According to Feldman (2006), one simply cannot understand what does mean the sentence “she took the apple”, if one is not capable to mentally simulate the perception of this action. Therefore words have a profoundly embodied meaning; in order to access to their semantics a network that simulates action or perception must be activated (Lakoff, 2008).

Attempts to integrate these findings (and many others) about the neural correlation of language have been made by the Neural Theory of Language (NTL) group at least since the late ‘80s. Feldman insists in his book From Molecule to Metaphor – a Neural Theory of Language (2006) that in order to understand language and thought it is absolutely necessary

to work in an interdisciplinary manner integrating psychology, AI, biology and linguistics.

More than two decades of research in neurosciences have shown how the many motor areas

respond to multimodal sensory stimulation (e.g. mirror neurons), and in the same way

somatosensorial cortex is involved in motor control. From a NTL perspective, this neural

complexity gives support to the meaning of actions words (e.g. verbs like grasp). NTL

(Bailey, Chang, Feldman, & Narayanan, 1998; Feldman, 2002) has contributed to the

understanding of embodied meaning of the words. It formulates an explicit neural theory of metaphors’ acquisition in the language and thought of children, from simple learning of verbs (action words) to abstract concepts and to narratives (see Feldman and Narayanan, 2004).

In this approach, the learning process of meaning and language is assumed to be grounded in direct interaction with adult’s context (Tomasello, 2000; Langacker, 1990;

Johnson, 1999). Therefore the computer program builds a scenario in which the children perform an action and hear a word (parent’s label). The program (child) learns to associate perception while carrying out the action. When confronted with the problem of the wide spectrum of different labels that different languages give to the same action, Bailey (1997) focus on what is shared by humans from different cultures; the neural circuitry. Any action involves coordinated neural firing; however we are not aware of that. Instead, what we are aware of are different parameters of action, like speed, position, direction, orientation, etc.

and we can talk about them. Therefore parameterization appears as a key concept, meaning that languages only labels action properties, “a fixed set of embodied features that determine the semantic space for any set of concepts, such as motor actions” (Feldman and Narayanan, 2004: p. 387). Another key concept for this computational model of embodied learning of words is the schematic representation of actions or X-schemas. This concept appears in many other computational models and can be parameterized as well.

After training the model, it was tested and gave a result of 80% of correct responses

both for requested action or for labelling of new actions on the base of the association of

sensorial stimuli (hearing words) and performed action sequence (schemas). Hence, this

model that follows biological neural properties appears as plausible for embodied verb

meaning learning, reason why this development was extended to abstract words and

narratives. For NTL, the learning of abstract words and the use of metaphors are both

profoundly connected to concrete words. According to Johnson (1999) children may not

construct different representations for literal and metaphorical sense of many commonly used words, as for example see (recalling KNOWING IS SEEING ). In early stages of language acquisition, children represent words like see both for literal and metaphorical way, for instance when adults say “let’s see what happened”; the child knows that she or he is supposed to both learn and understand an event, and actually visually perceive it as well.

Based on the same principles (simulation, parameterization, etc) Narayanan (1997, 1999), developed KARMA (Knowledge-based active representations for metaphor and aspect), a computational model capable of understanding the metaphorical use of words. In this model, the source domain (spatial motion) is characterized by the connection between X- schemas, which interact activating, inhibiting or interrupting one to another depending on experience. Simulation framework assumed that x-schemas have the potentiality of modifying associated networks (other x-schemas) causing parallel triggering or inhibition.

This model was tested for narratives using target domains like economics projected by embodied actions, giving more support to the idea of simulation semantics (see also Chang, Narayanan, and Petruck, 2002).

2.4 Summary

In this chapter, an extensive review of the principal concepts among conceptual

metaphor theory and cognitive linguistics is offered. This is aimed to provide the reader with

the sufficient knowledge on the framework necessary to understand the theoretical and

epistemological context of their origins. In addition, this review is intended to supply critical

evidence to evaluate the psychological reality of the conceptual metaphors. In order to

achieve these objectives the chapter is organized temporally, from the initial hypothesis to the

main actualizations. At the end, evidence from other disciplines among cognitive sciences

gives support to the main assumptions of conceptual metaphors theory.

Consequently, these ideas provide a framework in which first, sensoperceptual concepts can play a fundamental role in understanding more abstract concepts, thanks to the importance of simulation in language understanding and second, they showed how different concepts of different natures are naturally linked (by neural structures) through experience. In addition, linguistic evidence showed how much of our ways to understand (simulate) language is metaphorical, that is, one concept is better understood in terms of another.

Altogether, this is evidence for the idea of metaphor as a way of thinking and reasoning, and

not only as a matter of language.

Chapter 3: Experimental Research of Conceptual Metaphors

3.1 Introduction

With a few exceptions (Richards, 1936; Black, 1962; Ortony, 1979) the study of metaphors was very poor before CMT became visible in the field of linguistics in 1980, perhaps because the most predominant theory on the field, the Generative Approach considered this phenomenon as anomalous and meaningless. In his well-cited book Syntactic Structures, Chomsky (1957: p. 15) stated that “colourless green ideas sleep furiously” and

“furiously sleep ideas green colourless” were equally nonsensical, standing out the primacy of syntax and leaving metaphorical figures in a secondary position among the study of language. However, with the works of Reddy (1979) and Lakoff and Johnson (1980a), a new valuation of the role of metaphors and semantics occurred. Section 3.2 describes experimental studies on primary metaphors while 3.3 concentrates on the experimental research on verticality and affect. At the end, a preview of the present study is portrayed.

3.2 Experimental Background on the Study of Conceptual Metaphors

As mentioned before, Metaphors that We Live By based its postulates in a systematic

and conscious analysis of linguistic expression in English; this non-experimental empirical

approach led research to more controlled studies. One of the first inquiries that rose in experimental analysis was the comprehension process of figurative language (Searle, 1979;

Glucksberg, Gildea and Bookin, 1982). Do people understand non-literal language in the same way that they cope with literal language? Experimental data suggested that nominative metaphors are easily automatically understood, and more complex ones need of context to provide information (Gildea and Glucksberg, 1983). This idea, made more suitable to state that metaphors were an everyday phenomenon and not isolated or anomalous. Moreover, this experimental approach allows constructing more refined hypothesis not only about processing but about the relations between metaphor and thought. As illustrated in previews chapters, the range of Conceptual Metaphors is fairly wide; almost any domain of human conceptualization can be expressed through a metaphor. For primary metaphors, what is more often expressed in these terms, are highly subjective states mapped into less abstract, sensory related concepts. The following section concentrates on the description of experimental studies on primary metaphors.

In the early ‘80s, Sam Glucksberg and colleagues (Gildea and Glucksberg, 1983;

Glucksberg et al., 1982; Camac & Glucksberg, 1984) started a more systematic study of metaphorical language from a new approach; an empirical paradigm based on reaction times responses. This initiative was an important change in the study of metaphors because it went one step further on accounting for the underlying cognitive processes in language usage.

Several other studies followed this initiative (Blasko & Connine, 1993; Allbritton, McKoon

& Gerrig 1995; Johnson, 1996) using same on-line strategies. Off-line methods were still

used, however with a much more experimental spirit (Nayak and Gibbs, 1990; McGlone,

1996; Shen, 1997, Gibbs, Lenz Costa Lima and Francozo, 2004). Nevertheless, it was not

until the last decade, experimental research turned from conceptual metaphors’

comprehension studies to primary metaphors studies that inquire about the cognitive processes underlying conceptual mapping of this kind.

An insightful set of experiments by Borodistky and Ramscar (2002), accounts for a common conceptual metaphor; TIME IS SPACE . The first study, participants were requested to response an ambiguous question about time. The question was preceded by one of two different diagrams with a spatial prime, in which participants actively solve a given problem.

In both cases, participants were asked to imagine how they would move a chair through space, but one was designed to make people think on moving together with the chair, and the other to move the chair towards them. After this, they were given the following statement;

“Next Wednesday’s meeting has been moved forward two days. What day is the meeting now that is has been rescheduled?” The research question was if the spatial primes would have an effect on the answers that participants give to the ambiguous inquiry. As a result, participants primed to think on themselves moving through space were more likely to see themselves moving through time, answering Friday. Instead, when they were primed to see objects coming through space towards them, participants were more likely to see time approaching to them, answering Monday. With these results Boroditsky suggest that the way in which people think about time is linked to the way that they think about space.

Three consecutive experiments give support to Boroditsky’s ideas (see also

Borodistky, 2000). In a second experiment, people in a lunch line were asked the same

question, in order to observe the influence of actual forward movement in people’s thinking

about time. Coherently with the first study, the more actual motion participants had

experienced in the line, the more likely they thought of themselves moving through time. The

third experiment was aimed to investigate if spatial thinking was enough prime for the way of

thinking about time, or if actual movement was required. In San Francisco International

airport, people were given the ambiguous question about the meeting. People who had just