Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, LINGUISTICS (linguistics.oxfordre.com). (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 27 May 2017

Experimental Semiotics

Summary and Keywords

Experimental Semiotics (ES) is a burgeoning new discipline aimed at investigating in the laboratory the development of novel forms of human communication. Conceptually connected to experimental research on language use, ES provides a scientific complement to field studies of spontaneously emerging new languages and studies on the emergence of communication systems among artificial agents.

ES researchers have created quite a few research paradigms to investigate the development of novel forms of human communication. Despite their diversity, these paradigms all rely on the use of semiotic games, that is, games in which people can succeed reliably only after they have developed novel communication systems. Some of these games involve creating novel signs for pre-specified meanings. These games are particularly suitable for studying relatively large communication systems and their structural properties. Other semiotic games involve establishing shared meanings as well as novel signs to communicate about them. These games are typically rather challenging and are particularly suitable for investigating the processes through which novel forms of communication are created.

Considering that ES is a methodological stance rather than a well-defined research theme, researchers have used it to address a greatly heterogeneous set of research questions. Despite this, and despite the recent origins of ES, two of these questions have begun to coalesce into relatively coherent research themes.

The first theme originates from the observation that novel communication systems developed in the laboratory tend to acquire features that are similar to key features of natural language. Most notably, they tend (a) to rely on the use of symbols—that is purely conventional signs—and (b) to adopt a combinatorial design, using a few basic units to express a large number of meanings. ES researchers have begun investigating some of the factors that lead to the acquisition of such features. These investigations suggest two conclusions. The first is that the emergence of symbols depends on the fact that, when repeatedly using non-symbolic signs, people tend to progressively abstract them. The second conclusion is that novel communication systems tend to adopt a combinatorial design more readily when their signs have low degrees of motivation and fade rapidly.

The second research theme originates from the observation that novel communication systems developed in the laboratory tend to begin systematically with motivated—that is non-symbolic—signs. ES investigations of this tendency suggest that it occurs because motivation helps people bootstrap novel forms of communication. Put it another way, these investigations show that it is very difficult for people to bootstrap communication through arbitrary signs.

Keywords: experimental semiotics, language emergence, language evolution, experimental linguistics, human communication

1. Experimental Semiotics: A Working Definition

The term experimental semiotics (henceforth ES) was first introduced in the literature in the late 2000s’ (Galantucci, 2009; Galantucci & Garrod, 2010, 2011; Galantucci, Garrod, & Roberts, 2012) to name a set of laboratory studies in which humans developed novel forms of communication. Considering the novelty of ES as a discipline, its precise definition is still under development. For instance, according to the most recent review of ES (Galantucci, Garrod, & Roberts, 2012, p. 477), ES involves conducting “controlled studies in which human adults develop novel communication systems or impose novel structure on systems provided to them.” The disjunctive definition was chosen because, at that time, the literature contained only a few studies that fitted it (the review included a total of 16 studies). Today, there is a much large number of studies that would fit such a definition, and covering them all properly in a single encyclopedic entry is no longer feasible. Hence, I adopt here a more restrictive definition, excluding the second part of the disjunction.1 In other words, this article is about “controlled studies in which humans develop novel forms of communication.”2 Notice that this wording is slightly different from the one used in 2012, in that “humans” has replaced “human adults.” This is because, although no ES study has involved children as participants yet (but see Newman-Norlund et al., 2009 for a partial exception), there is no principled reason why this cannot be done. Indeed, as we shall see soon, some of the precursors of ES involve precisely children.

2. Ancient and Modern Precursors of ES

The idea of studying how humans develop novel forms of communication has roots that go all the way into antiquity. As far as we know, it was first conceived about 26 centuries ago, by the Egyptian pharaoh Psamtik I. As reported by Herodotus (who heard about this a couple of centuries later), Psamtik I had two children raised in linguistic isolation to see which language they would speak. Variants of this cruel experiment—which is now known as the “forbidden experiment”—have been performed a few more times in history.3 However not much can be learned from them. The only sound lesson they provide boils down to the notion that language is not completely innate: If a child is truly deprived of any linguistic input, s/he will not speak any language.

ES has also much more modern precursors. In particular, it is related to three contemporary lines of research (Figure 1). First, it is connected to experimental research on how humans use pre-established forms of communication such as spoken language (e.g., Clark & Wilkes-Gibbs, 1986; Garrod & Anderson, 1987; Garrod & Doherty, 1994; Horton & Keysar, 1996; Krauss & Weinheimer, 1964). ES and this line of research—which for convenience I will label experimental pragmatics (Noveck & Sperber, 2006)—share important features. For instance, both aim to uncover the causal relations behind the phenomena they observe by using methods that afford experimental manipulation and control. They also both share the assumption that, to understand human communication, researchers must investigate actual communicative interactions as well as individual cognitive processes. The main difference between ES and experimental pragmatics is that the former focuses on how humans develop novel forms of communication, whereas the latter focuses on how humans adapt existing forms of communication to various contextual situations. Although this is a clear difference, the extent to which it separates the two lines of research is tricky to assess because, as we shall see later in the article, the processes that lead to the development of novel forms of communication can be rather similar in kind to the processes that govern the pragmatic use of existing forms of communication.

Second, ES is connected to field studies of naturally occurring situations in which humans spontaneously develop new forms of communication. Most prominently, these studies focus on home sign systems developed by deaf children raised by non-signing parents (see Goldin-Meadow, 2003 for an overview), or on novel sign languages that emerge in relatively isolated populations (see Meir, Sandler, Padden, & Aronoff, 2010 for an overview). ES is an ideal complement to these studies because it can obviate two of their typical limitations: (a) the sparseness of the records, and (b) the impossibility, or impracticality, of performing controlled experiments.

Third, ES is related to research on the emergence of novel forms of communication among artificial agents (e.g., Cangelosi & Parisi, 2002; Kirby, 2002; Steels, 1997). Again, ES provides an ideal complement to this line of research, particularly in terms of ecological validity. Since there still remains quite a wide gap in behavioral complexity between artificial agents and humans, inferring human behaviors from the behaviors of artificial agents often requires a number of ad-hoc assumptions. ES can provide empirical tests for these assumptions, enhancing the ecological validity of the research done with artificial agents (Galantucci & Roberts, 2012). Furthermore, ES can provide novel hypotheses about human communication, the implications of which can then be tested with artificial agents (Galantucci & Roberts, 2012).

Experimental SemioticsClick to view larger

Figure 1. Schematic of the connections experimental semiotics has with its modern precursors.

It is important to notice that, in turn, the last two lines of research described above provide important complements to ES. While studies with artificial agents can allow experimental manipulations that are out of reach for ES researchers (e.g., radical modifications of the perceptual system as in Baronchelli, Gong, Puglisi, & Loreto, 2010), field studies provide knowledge about the development of novel forms of communication that is more ecologically valid than that provided by ES studies. Also, both field studies (e.g., Sandler, Aronoff, Meir, & Padden, 2011; Senghas, Kita, & Ozyurek, 2004) and studies with artificial agents (e.g., Puglisi, Baronchelli, & Loreto, 2008; Steels & Belpaeme, 2005) can surpass considerably the scope of ES investigations in terms of timescale and population size.

3. Main Research Paradigms in ES

Perhaps because it is still in its very early stages, ES exhibits a great methodological variety. Indeed, pretty much every research group that performs ES studies utilizes its own methods. However, there is a common core to these methods as they all involve the use of semiotic games, that is, games that require the development of novel forms of communication. The semiotic games utilized thus far in ES can be classified through two general features: the type of communicative tasks they involve and the type of actions through which people can communicate. In what follows, I first discuss these general features and then describe in more detail four specific examples of ES research paradigms.

3.1. Communicative Tasks

ES researchers have opted for two communicative tasks: Referential Communication and Coordination.

3.1.1. Referential Communication

This task is a modified version of a task commonly used in experimental pragmatics: the Referential Communication Task (e.g., Clark & Wilkes-Gibbs, 1986; Horton & Keysar, 1996; Krauss & Weinheimer, 1964). In the standard version of this task, people use natural language to communicate instructions on how to perform specific actions on a set of referents (typically objects or images of objects). The task has two desirable features for studying human communication. First, since the set of referents and the actions to be performed are determined by the experimenter, the task allows researchers to study communication in a tightly controlled context. Second, the task provides a simple measure of communicative success: The more closely people perform the operations in agreement with the instructions they are given, the more they are considered successful in communicating. ES researchers modified this task by preventing people from using natural language or other pre-established means of communication such as written text (Fay, Arbib, & Garrod, 2013; Garrod, Fay, Lee, Oberlander, & MacLeod, 2007; Healey, Swoboda, Umata, & King, 2007; Iizuka, Marocco, Ando, & Maeda, 2013; Newman-Norlund et al., 2009; Roberts & Galantucci, 2012; Roberts, Lewandowski, & Galantucci, 2015; Selten & Warglien, 2007). Also, ES researchers often used referents other than objects or images, such as pieces of music (Healey et al., 2007; Healey, Swoboda, Umata, & Katagiri, 2002), written words (Fay et al., 2013; Fay & Ellison, 2013; Fay, Garrod, & Roberts, 2008; Fay, Garrod, Roberts, & Swoboda, 2010; Fay, Lister, Ellison, & Goldin-Meadow, 2014; Garrod et al., 2007; Garrod, Fay, Rogers, Walker, & Swoboda, 2010; Theisen, Oberlander, & Kirby, 2010), or specific spatial arrangements (de Ruiter et al., 2010; de Weerd, Verbrugge, & Verheij, 2015; Noordzij, Newman-Norlund, de Ruiter, Hagoort, Levinson, & Toni, 2009; Stolk et al., 2013; Willems, Benn, Hagoort, Toni, & Varley, 2011; Blokpoel, van Kesteren, Stolk, Haselager, Toni, & van Rooij, 2012; Newman-Norlund et al., 2009).

3.1.2. Coordination

Although coordination tasks have also been developed before in experimental pragmatics (e.g., Garrod & Anderson, 1987) and other fields (Camerer, 2003), ES researchers opted for developing their own coordination tasks (Galantucci, 2005; Konno, Morita, & Hashimoto, 2013; Scott-Phillips, Kirby, & Ritchie, 2009). Thus far, these tasks have all involved coordinating movements in an artificial environment, which is a convenient way to ensure that people are not provided with pre-specified meanings to communicate about. This leads to a critical difference between referential communication tasks and coordination tasks. While the former pre-specify the meanings people are to communicate about, the latter force people to create their own meanings and jointly adopt them. In consequence, the two tasks afford rather different opportunities for research. Since people are usually good at referential communication tasks, these tasks are well-suited to study the development of relatively complex communication systems (e.g., Fay et al., 2008; Garrod et al., 2007; Theisen et al., 2010). In contrast, people typically find coordination tasks more challenging, and failures at developing functional forms of communication are not infrequent. These failures can provide useful information concerning what is necessary for successfully bootstrapping communication (e.g., Galantucci, 2009; Galantucci & Roberts, 2012; Scott-Phillips et al., 2009).

3.2. Types of Communicative Events

Two main types of communicative actions have been used in ES research: action in a dedicated channel and actions in the task space.

3.2.1. Actions in a Dedicated Channel

In a number of ES studies, people communicate in a dedicated channel through actions that are distinct from those performed in the task space (e.g., selecting a referent or moving to a game location). In some studies, people exchanged signals through digitizing pads that produced graphical outputs. To prevent people from using pre-established graphical forms such as letters or numbers, ES researchers either explicitly proscribed their use (Fay & Ellison, 2013; Fay et al., 2008, 2010; Garrod et al., 2007, 2010; Healey et al., 2007; Theisen et al., 2010) or transformed the tracings produced on the pads before broadcasting them as graphical signals (Galantucci, 2005, 2009; Galantucci, Kroos, & Rhodes, 2010; Galantucci, Theisen, Gutierrez, Kroos, & Rhodes, 2012; Roberts & Galantucci, 2012; Roberts et al., 2015; see Figure 3 for a description of the transformation). ES researchers also performed studies in which people used their own bodies to produce novel gestures or vocalizations (Fay et al., 2013; Fay, Lister, Ellison, & Goldin-Meadow, 2014) or composed novel messages combining pre-assigned units that have no pre-established meaning (Konno et al., 2013; Selten & Warglien, 2007).

3.2.2. Actions in the Task Space

In other ES studies, people communicate through the very actions they perform in the task space (e.g., Blokpoel et al., 2012; de Ruiter et al., 2010; de Weerd et al., 2015; Iizuka et al., 2013; Newman-Norlund et al., 2009; Noordzij et al., 2009; Scott-Phillips et al., 2009; Stolk et al., 2013; Willems et al., 2011). Because of this, people are faced with the challenge of identifying which actions have communicative value and which do not. In other words, people must find ways to differentiate their actions in the task space so that some of them can be understood as having the purpose of conveying a message.

3.3. Four Examples

To better illustrate typical research paradigms in ES, I will briefly describe four of them, one for each of the possible combinations between the type of communicative task involved and the type of actions through which people can communicate.

3.3.1. Referential Communication Task: Actions in a Dedicated Channel

Two participants are given a set of referents expressed by words such as “parliament” or “computer.” At any turn of the game, one participant has the task to communicate one of these referents to the other participant, who has the task to correctly identify it. This must be done exclusively through a digitizing pad, with the proscription to use pre-established symbols such as letters or numbers. In other words, to win at the game, participants must develop novel forms of graphical communication to identify the referents. This paradigm, labeled Pictionary task, has been used in a number of studies (Fay & Ellison, 2013; Fay et al., 2008, 2010; Garrod et al., 2007, 2010; Theisen et al., 2010).

3.3.2. Referential Communication Task: Actions in the Task Space

Two participants operate in a virtual space in which they have to move one object each to reproduce an arrangement pre-specified by the experimenter (Figure 2). At any round of the game, however, only one participant—the communicator—sees the arrangement. The communicator must move her object to match the arrangement while also using her movements in the virtual space to communicate to the other participant—the addressee—how to move his object to match the arrangement. Once the communicator has done this, the addressee must move his object to correctly complete the arrangement (Figure 2). In other words, to win at the game, participants must learn how to convert some of the simple movements they produce in the virtual space into signals to guide actions. This paradigm, labeled Tacit Communication Game, has been used in a number of studies (Blokpoel et al., 2012; de Ruiter et al., 2010; Newman-Norlund et al., 2009; Noordzij et al., 2009; Stolk et al., 2014, 2013; Volman, Noordzij, & Toni, 2012; Willems et al., 2011).

Experimental SemioticsClick to view larger

Figure 2. Schematic of the Tacit Communication Game (adapted from Stolk et al., 2013).

3.3.3. Coordination Task: Actions in a Dedicated Channel

Two participants operate in a virtual space in which they each move an agent on a grid. At any round of the game, the agents start at different grid locations and the participants must move them to make them meet through the minimal number of moves on the grid. This requires participants to tightly coordinate the moves and, in turn, this can be achieved only by communicating through a digitizing pad. However, the digitizing pad transforms the tracings produced on the pads before broadcasting them as graphical signals, preventing the use of pre-established symbols such as letters or numbers (Figure 3). In other words, to win at the game, participants must converge onto a set of shared meanings to coordinate their moves (e.g., a set of landmarks), as well as on a set of signs to communicate about the meanings. This paradigm has been used in a number of studies (Galantucci, 2005, 2009; Galantucci, Theisen, et al., 2012; Galantucci et al., 2010).

3.3.4. Coordination Task: Actions in the Task Space

Experimental SemioticsClick to view larger

Figure 3. (A) Schematic of the transformation from tracings on the digitizing pad to signals on screen; (B) How common graphic symbols drawn on the digitizing pad appeared on the screen.

Two participants operate in a virtual space in which they each move an agent on a grid. The locations of the grid are marked by color. At any round of the game, participants can move their agents an unlimited number of times to make them reach grid locations of the same color. However, participants do not see each other’s colors until the end of the turn. Once they both decide that they have reached the correct grid location, the turn ends and the colors are revealed. If the agents are on grid locations marked by the same color, participants win the round; otherwise they lose it. As in the tacit communication game, participants can communicate with each other only by moving their agents on the grid. In other words, to win at the game, they have to turn some of the simple movements they produce in the virtual space into signals that guide their moves. Thus far, this paradigm, labeled Embodied Communication Game, has been used in one study (Scott-Phillips et al., 2009).

4. Two Research Themes in ES

Since ES is a methodological stance rather than a well-defined research theme, researchers have used it to address a greatly heterogeneous set of research questions. Despite this and despite the recent origins of ES, some of these questions have begun to coalesce into relatively coherent research themes. To provide a sense of the opportunities for research that ES offers, I will survey two such themes. The first theme concerns the emergence of key linguistic properties in novel communication systems. The second theme concerns the role motivation plays for the bootstrapping of novel communication systems.

4.1. The Emergence of Key Linguistic Properties in Novel Communication Systems

One of the common findings in ES research is that, sometimes, novel communication systems acquire features that resemble key features of natural language. Quite naturally, ES researchers have been interested in investigating which factors govern this process. In particular, two key features of natural language have raised quite a bit of interest among ES researchers: the use of symbols and combinatoriality.

4.1.1. The Use of Symbols

One of the distinguishing features of human language is the use of symbols, that is signs that are related to their meaning only through arbitrary conventions (de Saussure, 1998; Hockett, 1960a). Garrod and colleagues (Garrod et al., 2007) investigated how novel communication systems might acquire such features through an ES study. In particular, they used the “Pictionary” task described above to collect a sample of novel communication systems. These systems typically started with non-symbolic signs, that is, with signs that were directly motivated by some feature of the referents to be communicated (see Figure 4). Over time, however, many of these signs turned into symbols. This happened because, when people started re-using the motivated signs to play new blocks of the game, they began simplifying them, getting rid of some of their parts. After a number of iterations of this process—which resembles the phenomenon of grounding in natural language use (Clark & Brennan, 1991; Schober & Clark, 1989)—the signs lost a good part of their initial parts. Since these parts played a pivotal role for the signs’ motivation, the process bleached the motivation, leading to the emergence of symbols. Crucially, however, this process of abstraction happened only when the experimenters allowed real time interactions between the people who played the game (a similar finding is also reported by Healey et al., 2007). These interactions allowed players to shift the grounding of their signs from the game’s referents (e.g., the word “computer”) to the signs’ history of use. In other words, instead of using signs referring directly to referent words such as “computer,” players began using signs referring to the signs that indicated those referent words in recent interactions, effectively abstracting them.

Experimental SemioticsClick to view larger

Figure 4. Examples of drawings for “computer monitor” over six blocks (adapted from Garrod, Fay, Lee, Oberlander, & MacLeod, 2007).

Two further studies with the “Pictionary” task expanded on this research theme, focusing on the factors that may prevent novel communication systems from becoming completely arbitrary. The first study (Theisen et al., 2010) showed that, when a communication system refers to a structured set of referents, some of that structure ends up being reflected in its symbols. This introduces systematic relations across the symbols, curtailing their overall level of arbitrariness. The second study (Fay et al., 2008) showed that symbols developed by small communities are more fit for efficient communication than the symbols developed by isolated dyads. This is so because small communities develop symbols that, although they undergo the same process of abstraction described above, end up retaining their original motivation more than symbols developed by isolated dyads. In other words, the diversity of small communities tweaks the process of abstraction, leading to symbols that are less arbitrary.

4.1.2. Combinatoriality

A second distinguishing feature of human language is combinatoriality, that is the use of a relatively small set of meaningless forms (e.g., the phonemes of a spoken language) to express an indefinite number of meanings (Hockett, 1960a; Martinet, 1960). Three ES studies focused on investigating how such a feature might emerge in novel communication system. The first one (Galantucci et al., 2010) used a coordination task with a dedicated communication channel (Figure 3) to show that, when people develop novel communication systems with signals that fade more rapidly, the systems are more likely to become combinatorial than when signals fade less rapidly. In other words, the study showed that rapidity of fading—another design feature of human language (Hockett, 1960b)—leads to combinatoriality.

The other two studies (Roberts & Galantucci, 2012; Roberts et al., 2015) used a referential communication task with the same communication channel as in the previous study to show that, when people develop novel communication systems with signs that are easier to motivate, the systems are less likely to become combinatorial than when motivation is more difficult. In other words, these studies connect the two lines of research surveyed in this section because they show that arbitrariness and combinatoriality are somehow related: The less arbitrary the signs of a communication system are, the less likely it is that the system will become combinatorial. Intriguingly, a similar conclusion has been reached by studying the Al-Sayyid Bedouin Sign Language, a novel sign language spontaneously developed by a relatively isolated community in Israel (Sandler et al., 2011).

Finally, it is worth noticing that two of the studies presented above (Galantucci et al., 2010; Roberts et al., 2015) cast doubt on a long-standing hypothesis concerning the origins of combinatoriality, namely that combinatoriality is a feature that communication systems acquire when they become very large (Hockett, 1960b; Lindblom, MacNeilage, & Studdert-Kennedy, 1984; Nowak, Krakauer, & Dress, 1999). In contrast with the prediction of this hypothesis, in fact, these studies showed a negative correlation between the size of a communication system and its level of combinatoriality.

4.2. Motivation Facilitates the Bootstrapping of a Novel Communication System

Another common finding in ES studies is that, when people are faced with the task of creating novel communication systems, they tend to rely on motivated signs (Galantucci, 2005; Garrod et al., 2007; Healey et al., 2007; Fay et al., 2008; Theisen et al., 2010; Roberts & Galantucci, 2012; Roberts et al., 2015). This finding is in line with observations from field studies of spontaneously emerging sign languages in which, especially in the early stages, the degree of sign motivation is very high (e.g., Sandler et al., 2011; Fusellier-Souza, 2006).

Four ES studies provide experimental evidence that people tend to rely on motivated signs because these signs facilitate the bootstrapping of a communication system (Fay et al., 2013, 2014; Roberts et al., 2015; Scott-Phillips et al., 2009).

In the first two studies, Fay and colleagues (Fay et al., 2013, 2014) manipulated the modality used by people to develop signs for a set of referents denoted by words. In particular, they compared body gesturing—which is well suited for the creation of motivated signs—with non-linguistic vocalizing—which is less suited for the creation of motivated signs. The result was clear: People were more effective at developing novel signs through body gesturing than through non-linguistic vocalizing. Furthermore, the communication systems created through body gesturing were more efficient and coherent than those created through non-linguistic vocalizing.

In the third study, Roberts and colleagues (2015) manipulated the set of referents people communicated about through the use of a dedicated channel (Figure 3). In particular, they compared linear profiles—which were very well suited for the creation of motivated signs—with discs colored in different shades of green—which were ill suited for the creation of motivated signs (Figure 5). The result was again clear: People were more effective at developing novel signs to communicate about the linear profiles than to communicate about the colored discs.

Taken together, these three studies provide ground for a broader conclusion. In any of their conditions, nothing prevented people from develop arbitrary signs. Yet, people opted for motivated signs whenever that was possible, showing a clear resistance against the creation of arbitrary signs.

Experimental SemioticsClick to view larger

Figure 5. The two sets of referents used by Roberts and colleagues (2015).

In the fourth study, Scott-Phillips and colleagues (2009) provided evidence corroborating this conclusion. In particular, they used the Embodied Communication Game described above and compared two conditions. In one of them, the grid was colored at random at each round of the game. In this condition, people often adopted a “default-color” strategy to bootstrap communication: After winning a few rounds by choosing a given color, they begun to systematically opt for that color. In other words, people bootstrapped the communication system by grounding it in the history of their previous interactions, a form of indirect motivation. Then, when the default color was not present on the grid, players began signaling its absence, and this led to the creation of new conventions. In the second condition, whenever players won a round by choosing a color, that color was not present in the following round, systematically frustrating people’s attempts to adopt a default-color strategy. This led to a dramatic decrease in success at the game, demonstrating that the more people are pushed toward the adoption of genuinely arbitrary signs, the harder it is for them to bootstrap communication.

Further Reading

de Ruiter, J. P., Noordzij, M. L., Newman-Norlund, S., Newman-Norlund, R., Hagoort, P., Levinson, S. C., et al. (2010). Exploring the cognitive infrastructure of communication. Interaction Studies, 11(1), 51–77.Find this resource:

Fay, N., Arbib, M., & Garrod, S. (2013). How to bootstrap a human communication system. Cognitive Science, 37(7), 1356–1367.Find this resource:

Fay, N., Garrod, S., & Roberts, L. (2008). The fitness and functionality of culturally evolved communication systems. Philosophical Transactions of the Royal Society B—Biological Sciences, 363(1509), 3553–3561.Find this resource:

Galantucci, B. (2005). An experimental study of the emergence of human communication systems. Cognitive Science, 29(5), 737–767.Find this resource:

Galantucci, B., & Garrod, S. (2011). Experimental semiotics: A review. Frontiers in Human Neuroscience, 5.Find this resource:

Galantucci, B., Garrod, S., & Roberts, G. (2012). Experimental semiotics. Language and Linguistics Compass, 6(8), 477–493.Find this resource:

Galantucci, B., & Roberts, G. (2012). Experimental semiotics: An engine of discovery for understanding human communication. Advances in Complex Systems, 15(3–4).Find this resource:

Garrod, S., Fay, N., Lee, J., Oberlander, J., & MacLeod, T. (2007). Foundations of representation: Where might graphical symbol systems come from? Cognitive Science, 31(6), 961–987.Find this resource:

Nielbo, F. (2014). Semiotics put to the test. Cognitive Semiotics, 7(1), 133–137.Find this resource:

Noordzij, M. L., Newman-Norlund, S. E., de Ruiter, J. P., Hagoort, P., Levinson, S. C., & Toni, I. (2009). Brain mechanisms underlying human communication. Frontiers in Human Neuroscience, 3.Find this resource:

Roberts, G., Lewandowski, J., & Galantucci, B. (2015). How communication changes when we cannot mime the world: Experimental evidence for the effect of iconicity on combinatoriality. Cognition, 141, 52–66.Find this resource:

Scott-Phillips, T. C., Kirby, S., & Ritchie, G. R. S. (2009). Signalling signalhood and the emergence of communication. Cognition, 113(2), 226–233.Find this resource:

Stolk, A., Blokpoel, M., van Rooij, I., & Toni, I. (2015). On the generation of shared symbols. In R. M. Willems (Ed.), Cognitive neuroscience of natural language use (pp. 201–227). Cambridge, U.K.: Cambridge University Press.Find this resource:

Stolk, A., Verhagen, L., Schoffelen, J.-M., Oostenveld, R., Blokpoel, M., Hagoort, P., et al. (2013). Neural mechanisms of communicative innovation. Proceedings of the National Academy of Sciences USA, 110(36), 14574–14579.Find this resource:

Theisen, C. A., Oberlander, J., & Kirby, S. (2010). Systematicity and arbitrariness in novel communication systems. Interaction Studies, 11(1), 14–32.Find this resource:

References

Baronchelli, A., Gong, T., Puglisi, A., & Loreto, V. (2010). Modeling the emergence of universality in color naming patterns. Proceedings of the National Academy of Sciences USA, 107(6), 2403–2407.Find this resource:

Blokpoel, M., van Kesteren, M., Stolk, A., Haselager, P., Toni, I., & van Rooij, I. (2012). Recipient design in human communication: Simple heuristics or perspective taking? Frontiers in Human Neuroscience, 6.Find this resource:

Camerer, C. (2003). Behavioral game theory: Experiments in strategic interaction. Princeton, NJ: Princeton University Press.Find this resource:

Cangelosi, A., & Parisi, D. (Eds.). (2002). Simulating the evolution of language. London: Springer-Verlag.Find this resource:

Clark, H. H., & Brennan, S. E. (1991). Grounding in communication. In L. B. Resnick, J. M. Levine, & S. D. Teasley (Eds.), Perspectives on socially shared cognition (pp. 127–149). Washington, DC: American Psychological Association.Find this resource:

Clark, H. H., & Wilkes-Gibbs, D. (1986). Referring as a collaborative process. Cognition, 22(1), 1–39.Find this resource:

de Ruiter, J. P., Noordzij, M. L., Newman-Norlund, S., Newman-Norlund, R., Hagoort, P., Levinson, S. C. et al. (2010). Exploring the cognitive infrastructure of communication. Interaction Studies, 11(1), 51–77.Find this resource:

De Jaegher, K., Rosenkranz, S., & Weitzel, U. (2014). Economic principles in communication: An experimental study. Journal of Theoretical Biology, 363, 62–73. doi:10.1016/j.jtbi.2014.07.035Find this resource:

de Saussure, F. (1998). Course in general linguistics (R. Harris, Trans.). Chicago: Open Court. Originally published in 1916.Find this resource:

de Weerd, H., Verbrugge, R., & Verheij, B. (2015). Higher-order theory of mind in the Tacit Communication Game. Biologically Inspired Cognitive Architectures, 11, 10–21.Find this resource:

dos Santos, M., Matias Rodrigues, J. F., Wedekind, C., & Rankin, D. J. (2012). The establishment of communication systems depends on the scale of competition. Evolution and Human Behavior, 33(3), 232–240. doi:10.1016/j.evolhumbehav.2011.09.007Find this resource:

Fay, N., Arbib, M., & Garrod, S. (2013). How to bootstrap a human communication system. Cognitive Science, 37(7), 1356–1367.Find this resource:

Fay, N., & Ellison, T. M. (2013). The cultural evolution of human communication systems in different sized populations: Usability trumps learnability[V]. Plos One, 8(8).Find this resource:

Fay, N., Garrod, S., & Roberts, L. (2008). The fitness and functionality of culturally evolved communication systems. Philosophical Transactions of the Royal Society B—Biological Sciences, 363(1509), 3553–3561.Find this resource:

Fay, N., Garrod, S., Roberts, L., & Swoboda, N. (2010). The interactive evolution of human communication systems. Cognitive Science, 34(3), 351–386.Find this resource:

Fay, N., Lister, C. J., Ellison, T. M., & Goldin-Meadow, S. (2014). Creating a communication system from scratch: Gesture beats vocalization hands down. Frontiers in Psychology, 5.Find this resource:

Fusellier-Souza, I. (2006). Emergence and development of signed languages: From a semiogenetic point of view. Sign language studies, 7(1), 30–56.Find this resource:

Galantucci, B. (2005). An experimental study of the emergence of human communication systems. Cognitive Science, 29(5), 737–767.Find this resource:

Galantucci, B. (2009). Experimental semiotics: A new approach for studying communication as a form of joint action. Topics in Cognitive Science, 1(2), 393–410.Find this resource:

Galantucci, B., & Garrod, S. (2010). Experimental semiotics: A new approach for studying the emergence and the evolution of human communication. Interaction Studies, 11(1), 1–13.Find this resource:

Galantucci, B., & Garrod, S. (2011). Experimental semiotics: A review. Frontiers in Human Neuroscience, 5.Find this resource:

Galantucci, B., Garrod, S., & Roberts, G. (2012). Experimental semiotics. Language and Linguistics Compass, 6(8), 477–493.Find this resource:

Galantucci, B., Kroos, C., & Rhodes, T. (2010). The effects of rapidity of fading on communication systems. Interaction Studies, 11(1), 100–111.Find this resource:

Galantucci, B., & Roberts, G. (2012). Experimental semiotics: An engine of discovery for understanding human communication. Advances in Complex Systems, 15(3–4).Find this resource:

Galantucci, B., Theisen, C. A., Gutierrez, E. D., Kroos, C., & Rhodes, T. (2012). The diffusion of novel signs beyond the dyad. Language Sciences, 34(5), 583–590.Find this resource:

Garrod, S., & Anderson, A. (1987). Saying what you mean in dialog: A study in conceptual and semantic coordination. Cognition, 27(2), 181–218.Find this resource:

Garrod, S., & Doherty, G. (1994). Conversation, coordination and convention: An empirical investigation of how groups establish linguistic conventions. Cognition, 53(3), 181–215.Find this resource:

Garrod, S., Fay, N., Lee, J., Oberlander, J., & MacLeod, T. (2007). Foundations of representation: Where might graphical symbol systems come from? Cognitive Science, 31(6), 961–987.Find this resource:

Garrod, S., Fay, N., Rogers, S., Walker, B., & Swoboda, N. (2010). Can iterated learning explain the emergence of graphical symbols? Interaction Studies, 11(1), 33–50.Find this resource:

Goldin-Meadow, S. (2003). The resilience of language:What gesture creation in deaf children can tell us about how all children learn language. New York: Psychology Press.Find this resource:

Healey, P. G. T., Swoboda, N., Umata, I., & Katagiri, Y. (2002). Graphical representation in graphical dialogue. International Journal of Human-Computer Studies, 57(4), 375–395.Find this resource:

Healey, P. G. T., Swoboda, N., Umata, I., & King, J. (2007). Graphical language games: Interactional constraints on representational form. Cognitive Science, 31(2), 285–309.Find this resource:

Hockett, C. F. (1960a). Logical considerations in the study of animal communication. In W. Lanyon & W. Tavolga (Eds.), Animal sounds and communications (pp. 392–430). Washington, DC: American Institute of Biological Sciences.Find this resource:

Hockett, C. F. (1960b). The origin of speech. Scientific American, 203(3), 88–96.Find this resource:

Horton, W. S., & Keysar, B. (1996). When do speakers take into account common ground? Cognition, 59(1), 91–117.Find this resource:

Iizuka, H., Marocco, D., Ando, H., & Maeda, T. (2013). Experimental study on co-evolution of categorical perception and communication systems in humans. Psychological Research, 77(1), 53–63.Find this resource:

Kirby, S. (2002). Natural language from artificial life. Artificial Life, 8(2), 185–215.Find this resource:

Kirby, S., Cornish, H., & Smith, K. (2008). Cumulative cultural evolution in the laboratory: An experimental approach to the origins of structure in human language. Proceedings of the National Academy of Sciences USA, 105(31), 10681–10686.Find this resource:

Kirby, S., Griffiths, T., & Smith, K. (2014). Iterated learning and the evolution of language. Current Opinion in Neurobiology, 28, 108–114.Find this resource:

Kirby, S., Tamariz, M., Cornish, H., & Smith, K. (2015). Compression and communication in the cultural evolution of linguistic structure. Cognition, 141, 87–102.Find this resource:

Konno, T., Morita, J., & Hashimoto, T. (2013). Symbol communication systems integrate implicit information in coordination tasks. In T. Omori, Y. Yamaguchi, Y. Sakaguchi, N. Sato, & I. Tsuda (Eds.), Advances in cognitive neurodynamics (III): Proceedings of the Third International Conference on Cognitive Neurodynamics—2011 (pp. 453–459). Dordrecht, The Netherlands: Springer.Find this resource:

Krauss, R. M., & Weinheimer, S. (1964). Changes in reference phrases as a function of frequency of usage in social interaction: A preliminary study. Psychonomic Science, 1(5), 113–114.Find this resource:

Leibfried, F., Grau-Moya, J., & Braun, D. A. (2015). Signaling equilibria in sensorimotor interactions. Cognition, 141, 73–86. doi:10.1016/j.cognition.2015.03.008Find this resource:

Lindblom, B., MacNeilage, P., & Studdert-Kennedy, M. (1984). Self-organizing processes and the explanation of phonological universals. In B. Butterworth, B. Comrie, & O. Dahl (Eds.), Explanations for language universals (pp. 181–203). Berlin: Mouton.Find this resource:

Martinet, A. (1960). Elements of general linguistics (E. Palmer, Trans.). Chicago: University of Chicago Press.Find this resource:

Meir, I., Sandler, W., Padden, C., & Aronoff, M. (2010). Emerging sign languages. In M. Marschark, & P. E. Spencer (Eds.), Oxford handbook of deaf studies, language, and education (pp. 267–280). New York: Oxford University Press.Find this resource:

Newman-Norlund, S. E., Noordzij, M. L., Newman-Norlund, R. D., Volman, I. A. C., de Ruiter, J. P., Hagoort, P., et al. (2009). Recipient design in tacit communication. Cognition, 111(1), 46–54.Find this resource:

Noordzij, M. L., Newman-Norlund, S. E., de Ruiter, J. P., Hagoort, P., Levinson, S. C., & Toni, I. (2009). Brain mechanisms underlying human communication. Frontiers in Human Neuroscience, 3.Find this resource:

Noveck, I. A., & Sperber, D. (Eds.). (2006). Experimental pragmatics. New York: Palgrave Macmillan.Find this resource:

Nowak, M. A., Krakauer, D. C., & Dress, A. (1999). An error limit for the evolution of language. Proceedings of the Royal Society of London Series B—Biological Sciences, 266(1433), 2131–2136.Find this resource:

Puglisi, A., Baronchelli, A., & Loreto, V. (2008). Cultural route to the emergence of linguistic categories. PNAS, 105(23), 7936–7940.Find this resource:

Roberts, G. (2010). An experimental study of social selection and frequency of interaction in linguistic diversity. Interaction Studies, 11(1), 138–159.Find this resource:

Roberts, G., & Galantucci, B. (2012). The emergence of duality of patterning: Insights from the laboratory. Language & Cognition, 4(4), 297–318.Find this resource:

Roberts, G., Lewandowski, J., & Galantucci, B. (2015). How communication changes when we cannot mime the world: Experimental evidence for the effect of iconicity on combinatoriality. Cognition, 141, 52–66.Find this resource:

Sandler, W., Aronoff, M., Meir, I., & Padden, C. (2011). The gradual emergence of phonological form in a new language. Natural Language & Linguistic Theory, 29(2), 503–543.Find this resource:

Schober, M. F., & Clark, H. H. (1989). Understanding by addressees and overhearers. Cognitive Psychology, 21, 211–232.Find this resource:

Scott-Phillips, T. C., & Kirby, S. (2010). Language evolution in the laboratory. Trends in Cognitive Sciences, 14(9), 411–417.Find this resource:

Scott-Phillips, T. C., Kirby, S., & Ritchie, G. R. S. (2009). Signalling signalhood and the emergence of communication. Cognition, 113(2), 226–233.Find this resource:

Selten, R., & Warglien, M. (2007). The emergence of simple languages in an experimental coordination game. Proceedings of the National Academy of Sciences USA, 104(18), 7361–7366.Find this resource:

Senghas, A., Kita, S., & Ozyurek, A. (2004). Children creating core properties of language: Evidence from an emerging sign language in Nicaragua. Science, 305(5691), 1779–1782.Find this resource:

Steels, L. (1997). The synthetic modeling of language origins. Evolution of Communication: An International Multidisciplinary Journal, 1(1), 1–34.Find this resource:

Steels, L., & Belpaeme, T. (2005). Coordinating perceptually grounded categories through language: A case study for colour. Behavioral and Brain Sciences, 28(4), 469–489.Find this resource:

Stolk, A., Blokpoel, M., van Rooij, I., & Toni, I. (2015). On the generation of shared symbols. In R. M. Willems (Ed.), Cognitive neuroscience of natural language use (pp. 201–227). Cambridge, U.K.: Cambridge University Press.Find this resource:

Stolk, A., Noordzij, M. L., Volman, I., Verhagen, L., Overeem, S., van Elswijk, G., et al. (2014). Understanding communicative actions: A repetitive TMS study. Cortex, 51, 25–34.Find this resource:

Stolk, A., Verhagen, L., Schoffelen, J.-M., Oostenveld, R., Blokpoel, M., Hagoort, P., et al. (2013). Neural mechanisms of communicative innovation. Proceedings of the National Academy of Sciences USA, 110(36), 14574–14579.Find this resource:

Theisen, C. A., Oberlander, J., & Kirby, S. (2010). Systematicity and arbitrariness in novel communication systems. Interaction Studies, 11(1), 14–32.Find this resource:

Uno, R., Suzuki, K., & Ikegami, T. (2012). An experimental approach to speakers’ perspectives in a paired wall game. Advances in Complex Systems, 15(3–4). doi:10.1142/S0219525911500202Find this resource:

Verhoef, T., Kirby, S., & de Boer, B. (2015). Iconicity and the emergence of combinatorial structure in language. Cognitive Science, 40, 1969–1994.Find this resource:

Volman, I., Noordzij, M. L., & Toni, I. (2012). Sources of variability in human communicative skills. Frontiers in Human Neuroscience, 6.Find this resource:

Willems, R. M., Benn, Y., Hagoort, P., Toni, I., & Varley, R. (2011). Communicating without a functioning language system: Implications for the role of language in mentalizing. Neuropsychologia, 49(11), 3130–3135.Find this resource:

Notes:

(1.) Also, this article provides limited coverage of studies in which (a) communication entails explicit costs (e.g., De Jaegher et al., 2014; dos Santos et al., 2012; Leibfried et al., 2015; Selten & Warglien, 2007), (b) communication has no precise purpose (e.g., Uno, Suzuki, & Ikegami, 2012) or (c) the primary measures are neural correlates (e.g., Noordzij et al., 2009; Stolk et al., 2013; see Stolk, Blokpoel, van Rooij, & Toni, 2015 for a review). These further restrictions have the goal to keep this article focused on studies directly connected to research in linguistics.

(2.) Readers interested in an overview of studies in which people impose novel structure on systems provided to them (e.g., Kirby, Cornish, & Smith, 2008; Roberts, 2010; Selten & Warglien, 2007; Verhoef, Kirby, & de Boer, 2015) are referred to recent reviews (e.g., Kirby, Griffiths, & Smith, 2014; Kirby, Tamariz, Cornish, & Smith, 2015; Scott-Phillips & Kirby, 2010).

(3.) For example, by the Holy Roman emperor Frederik II (13th century), by James IV of Scotland (16th century), and by the Mughal emperor Akbar the great (17th century).