22 May: Marieke Woensdregt


A model of cultural co-evolution of language and perspective-taking

Marieke Woensdregt (University of Edinburgh)

Tuesday 22 May 2018, 11:00–12:30
G32, 7 George Square

Language relies on mindreading (a.k.a. theory of mind), as language users have to entertain and recognise communicative intentions. Mindreading abilities in turn profit from language, as language provides a means for expressing mental states explicitly, and for transmitting one’s understanding of minds to others (e.g. younger members of the population). Given this interdependence, it has been hypothesised that language and mindreading have co-evolved. I will present an agent-based model to formalise this hypothesis, which combines referential signalling with perspective-taking. In this model, agents’ communicative behaviour is probabilistically determined by an interplay between their language and their perspective on the world. In order to learn the language, learners thus have to simultaneously infer both the language and the perspective of the speaker they’re receiving input from (using Bayesian inference). Simulation results show that learners can solve this task by bootstrapping one from the other, but only if the speaker uses a language that is at least somewhat informative.
The question then becomes under what circumstances a population of these agents can evolve such an informative language from scratch. We explore two different selection pressures: a pressure for successful communication and a pressure for accurate perspective-inference. We also compare two different types of agents: literal communicators and pragmatic communicators. Pragmatic speakers optimise their communication behaviour by maximising the probability that their interlocutor will interpret their signals correctly, given their model of the interlocutor’s language and perspective. Iterated learning results show that literal agents can evolve meaningful linguistic conventions both under a pressure for communication and under a pressure on perspective-inference. The same results were found for pragmatic agents, except that pragmatic agents can achieve equally high levels of success at communicating and inferring perspectives with much more ambiguous languages, because these agents can compensate for suboptimal languages using their pragmatic ability. Time permitting, I will also present some preliminary results on the evolvability of this pragmatic ability.

 

15 May: Mark Steedman


Combinatory Universal Grammar

Mark Steedman (Informatics Edinburgh)

Tuesday 15 May 2018, 11:00–12:30
3.10 Dugald Stewart Building

Greenberg proposed as his 20th Universal a generalization about the possible language-specific orders over the elements of the noun-phrase (NP). Greenberg’s original statement has been modified a number of times, and a number of attempts have been made to explain its various reformulations in terms of “constraints on movement” of those elements within a single primary ordering corresponding to a universal order of merger or dominance, defined ultimately by their semantic types

The present paper begins by proposing a new generalization concerning the orderings allowed over the elements of the NP in rigid and more freely ordered languages. This generalization can be parsimonously captured in a theory of grammar without movement or other syntactic “action-at-a-distance” between non-contiguous elements. This theory predicts that only two of the twenty four permutations over these four elements are universally excluded. This prediction constitutes a formal universal, in that it follows from the theory of grammar itself, and appears to be both qualitatively and statistically confirmed by the data

The paper goes on to show that the same generalization appears to hold over a number of cases of order alternations in clausal serial-verb constructions in a number of languages.

24 April: Rob Truswell


What’s ‘that’?

Rob Truswell (University of Edinburgh)

Tuesday 24 April 2018, 11:00–12:30
G32, 7 George Square

This is a talk in two parts. The first part is a case study in syntactic change, trying to make sense of a brief window in Middle English when virtually all free relatives were introduced by “that”. This period can only be understood by a generalization of standard models of grammar change, aiming to capture the following insight: Grammars are largely composed of associations between pieces of form and pieces of meaning, so grammar change is change in those sets of associations.

The second part of the talk describes some pilot work with Richard Blythe and Simon Kirby, modelling the dynamics of change in those terms. I give some preliminary results from those models, and discuss next steps.

10 April: EvoLang dry-runs by Jon Carr and Stella Frank


Conceptual Structure Is Shaped By Competing Pressures For Simplicity And Informativeness

Jon Carr (University of Edinburgh)

Tuesday 10 April 2018, 11:00–12:30
G32 7 George Square

Languages are shaped by competing pressures from learning and communication. Learning favours simple languages, while communication favours informative ones, giving rise to the simplicity–informativeness tradeoff. Languages that evolve under this tradeoff are both maximally simple (learnable) and maximally informative (communicatively useful). This has been shown in natural language and in experimental settings. For example, Kemp and Regier (2012) showed that kinship systems exist at the optimal frontier of simplicity and informativeness. In a separate line of experimental work, Kirby, Tamariz, Cornish, and Smith (2015) showed that when artificial languages evolve under a learning pressure alone, they become simple and uninformative, and when languages evolve during communication, they become complex and informative; it is only when both pressures are at play that we find languages at the optimal frontier.

However, a recent iterated learning experiment by Carstensen, Xu, Smith, and Regier (2015) showed that artificial languages expressing spatial relationships tended to become more informative when subjected to a pressure from learning. This is a surprising result given the previous work briefly reviewed above, which says that informativeness is driven by the pressure from communication, not from learning. One potential explanation for this result lies in their measure of informativeness, communicative cost, which is sensitive to (a) the number of words that the language is comprised of (expressivity) and (b) the extent to which similar meanings are expressed by the same word (which we will term convexity). In their experiment expressivity was fixed at four words. As a result, the reduction in communicative cost they found must be due to categories evolving to become more convex, i.e. picking out increasingly tightly-clustered sets of meanings.

To demonstrate that learning favours convex categories, we conducted two experiments in which participants learned and produced a category system for stimuli varying on two dimensions, size and angle. In Experiment 1 participants were trained on one of three systems: One marking a distinction in angle, one marking a distinction in size, and one marking a distinction on both dimensions (see Fig. 1). The results indicated that the Angle-only system was easiest to learn, followed by the Size-only system; the Angle & Size system was hardest to learn, despite having the lowest communicative cost.

In Experiment 2, the output of one participant became the input to the following participant in a standard iterated learning design. An example chain is shown in Fig. 2. Over 12 generations the category system became increasingly easy to learn, as indicated by decreasing intergenerational transmission error. Furthermore, in the majority of chains, the language converged on a system marking only a distinction on the angle dimension, which participants found easiest in Experiment 1. This increase in simplicity is driven by increasing convexity.

We also found that most chains converged on fewer than four categories. This suggests that iterated learning acts as a pressure for simplicity by simultaneously decreasing expressivity and increasing convexity. However, if, as in Carstensen et al. (2015), expressivity is held constant, the learning pressure can only act through convexity: Although languages may become more informative under iterated learning, they do so not because of a pressure to be more communicatively useful, which in Carstensen et al.’s study necessarily decreases communicative cost as a side-effect of increasing convexity. This therefore suggests that, contra Carstensen et al. (2015), languages which are both simple yet informative will only emerge when pressures from learning and communication are at play. We support these conclusions with a Bayesian iterated learning model that displays strikingly similar results.

 

A rational model of linguistic accommodation and its potential role in language simplification

Stella Frank (University of Edinburgh)

Tuesday 10 April 2018, 11:00–12:30
G32 7 George Square

Languages with large numbers of adult learners tend to be less morphosyntactically complex than languages where adult learners are rare (Wray & Grace, attributed to deficiencies in adult language learning. Here we investigate an additional or alternative mechanism: rational accommodation by native speakers to non-native interlocutors.

Humans have a general aptitude for reasoning about the knowledge, beliefs and motivations of other individuals, including their linguistic knowledge (e.g. Clark, 1996; Ferguson, 1981). While our interlocutors’ linguistic knowledge will often be close to our own, this may not be the case in a population with many non- native speakers. We introduce a rational model of interactions between individuals capable of reasoning about the linguistic knowledge of others, and investigate the case of a non-native speaker interacting with an native speaker who reasons about their linguistic knowledge and accommodates accordingly. Our model shows that this accommodation mechanism can lead to the non-native speaker acquiring a language variant that is less complex than the original language.

We assume a simple model in which a language consists of a distribution over linguistic variants (e.g. past tense forms). Language simplification is modelled as regularisation, whereby the most frequent variant becomes more frequent; this corresponds to, and can be measured as, entropy reduction. We model the interaction between a non-native speaker and a native speaker as interaction between two rational (Bayesian) agents. Both agents have the same initial priors and update their beliefs about the language from data in the same way, but the non-native speaker has simply seen much less data. Within an interaction, the native speaker has a parametrisable tendency to accommodate to the non-native speaker: instead of simply using their own language, they use the version of the language that they believe the non-native speaker may have acquired at this stage of their learning, given limited exposure. Importantly, the native speaker does not know exactly what data the non-native has seen. Instead, the native speaker models the non- native speaker’s linguistic knowledge by integrating over possible datasets the non-native speaker might have seen.

Representative model results for a sample language are shown in Figure 1. While learners interacting with non-accommodating speakers eventually learn the original language, non-native speakers interacting with accommodating native speakers end up learning a more regular language. This is due to the combination of the limited exposure of the non-native individual, which results in highly skewed initial distributions and some probability of not having seen low-frequency variants (Hertwig, Barron, Weber, & Erev, 2004; Hahn, 2014), in conjunction with a native speaker who is aware of and accommodates this initial bias in the non- native speaker’s input, therefore providing the non-native speaker with further data which ‘locks in’ their biased starting point.

This model shows that accommodation by native speakers to non-native speakers during interaction can lead to language simplification, and therefore suggests how accommodation can explain the link between population makeup and lin- guistic complexity. The model assumes that individuals are capable of reasoning rationally about their interlocutors’ linguistic knowledge, an assumption we are currently testing empirically with human learners.

3 April: EvoLang dry-runs by Marieke Schouwstra and Jonas Nölle


Building meanings: compositionality of human language, and its evolution

Marieke Schouwstra (University of Edinburgh)

Tuesday 3 April 2018, 11:00–12:30
G32, 7 George Square

Compositionality is seen as a key feature of human language, It describes the mechanism by which complex structures and complex meanings are related to each other: the meaning of a complex expression is determined by the meanings of its constituents and the way in which they are put together. This characterisation has been a firm assumption among many linguists, but it is not the only possible view on language. I will describe the linguistic and philosophical background of the principle of compositionality, by sketching why human language is often described as compositional, and what the alternatives might be. Subsequently, I will look briefly at how cultural evolution experiments can help us answer the question how compositionality came about.

This is a talk I will give in Wendy Sandler’s workshop on Compositionality. See https://evolang.cles.umk.pl/pdf/WorkshopCompositionality.pdf

Environmental and social factors motivate the emergence of systematic categories and signs

Jonas Nölle , Marlene Staib, Riccardo Fusaroli and Kristian Tylén

Tuesday 3 April 2018, 11:00–12:30
G32, 7 George Square

While arbitrariness has long been considered a hallmark of human language, there have been increasing discussions about non-arbitrary relationships between form and meaning such as iconicity and systematicity (Dingemanse et al., 2015). We argue that these phenomena are not just two facets of non-arbitrariness, but serve orthogonal functions in the scaffolding of an efficient communication system. Iconicity is usually associated with learning and bootstrapping (Imai & Kita, is inhibited (Roberts et al., 2015; Verhoef et al., 2016) or when participants were provided with pre-established combinatorial categories (Theisen et al., 2010).

Building on this work, we tested in a series of dyadic silent gesture experiments whether systematicity is functionally adaptive and could also emerge in competition to iconicity in response to particular environmental and social factors. We hypothesized that structure and openness of the environment as well as working memory constrains (Christiansen & Chater, 2016) could affect the degree of systematicity as well as which referential features become systematized. In two experiments, participants had to communicate stimuli depicting stylized characters that afforded both iconic and systematic gestures (see Fig. 1A). Experiment 1 tested a) whether functional adaptivity of a given trait, based on its distribution in the referential environment (consisting of 14 stimuli drawn from a set of 24) and thus its discriminative value, would affect the likelihood of this trait being systematized and b) the impact of an expanding referential environment. In a 2×2 design we varied the distribution of PET vs. GENDER (7:7 vs. 10:4) and the openness of the environment (an expanding meaning space in the open vs. constant meanings across trials in the closed condition). Logistic mixed regression models (see supplementary materials) indicated that functionally more adaptive traits were indeed more likely to be systematized (p<.01) while openness of the environment had no significant effect (see Fig. 1B). Debriefing revealed that participants in the open condition frequently realized that the changing competitor images never had to be signed, which narrowed down the search space rather than enlarging it.

Experiment 2 was designed to further explore the roles of informational bottlenecks. We improved the ”openness” manipulation by increasing the number of stimuli (n=32) and referents per trial (n=16) and randomly sampling both targets and competitors for each trial in the open condition, while the set remained the same across trials in the closed condition. In addition, we added ”displacement” as a second factor. As in experiment 1, in the co-present condition, participant pairs could gesture while seeing the stimuli, whereas dyads in the displaced condition had to wait 3 seconds after the stimuli disappeared before they could gesture—effectively displacing the communicative from the referential context and simulating communication about absent entities. Our results indicate that displacement increased the tendency to systematize overall, while openness of the environment affected the temporal development of systematicity (see Fig. 1C). In the open condition systematicity kept increasing until the end of the experiment, while there was a decline in the closed condition that resembled the slope of both the closed and ”pseudo”-open condition in experiment 1, suggesting that a true open environment can override the tendency to fall back on simple iconic signs as communication becomes more efficient over time (Kirby et al., 2015).

Taken together, our results indicate that, given certain environmental affordances, systematicity can emerge in a novel communication system at the interaction level even in the presence of competing iconic solutions. In our experiment both systematic categories and signs emerged in response to their functional adaptivity to the environment and working memory constraints of the communicative situation (openness of the referential context, displacement). This could lead to variation that is amplified over cultural transmission (Kirby et al., 2015). Lastly, our study provides the first experimental semiotics study investigating the effect of displacement, a core property of language (Hockett, 1960) that has curiously been neglected in experimental studies so far.

27 March: EvoLang dry-runs by Molly Flaherty & Marieke Schouwstra and Christine Cuskley


Alien symbols for alien language: iterated learning in a unique, novel signal space

Christine Cuskley (University of Edinburgh)

Tuesday 27 March 2018, 11:00–12:30
G32, 7 George Square

Studies in iterated learning of artificial ‘alien’ languages show that structure in language can arise from the process of cultural transmission (Kirby, Cornish, & Smith, 2008). Structure can accumulate not only in iterated artificial language learning (ALL) tasks, but also in iterated sequence learning tasks which are not explicitly linguistic (e.g., they contain no meanings, and do not use linguistic stimuli). Meaningless sequences of colours become more structured and more learnable over time (Cornish, Smith, & Kirby, 2013), reproduction of rhythmic sequences results in increased structure, learnability, and emergence of ‘musical universals’ (Ravignani, Delgado, & Kirby, 2016), and transmission of slide whistles can result increased structure (Verhoef, Kirby, & Boer, 2014). Where signals are mapped onto meanings, the relationships between these two spaces can have significant effects on the resulting structure (Little, Erylmaz, & Boer, 2017). While the signals and modalities with which iterated learning studies are conducted have expanded considerably in recent years, how properties of signal spaces interact with biases of learners in the emergence of structure is relatively under-studied.

To further investigate the effect of signal space on the evolution of structure and learnability, we used a completely novel set of graphical symbols called Ferros: a set of 137 abstract graphemes created using ferrofluid ink and visually unlike e.g., Roman orthography. To articulate in Ferro, participants move in a virtual two dimensional palette which produces a different symbol de- pending on their location within a square. The space is structured along two axes corresponding to the number of contours and nodes in each Ferro. Much like the phonetic space of consonants, distance between Ferros encodes similarity, but the forms are discrete. Ferros are signals which are entirely foreign to participants, who have to learn not only to use the apparatus which produces Ferros, but also what the relevant features of Ferros are.

The Ferro palette was used in an iterated sequence copying task involving hundreds of participants: each participant had three minutes to copy as many 3- character Ferro sequences as they could. Each sequence faded slowly after presentation, providing an additional pressure for quick reproduction. Participants were confined to producing three characters (i.e., sequences could not vary in length), and were provided with feedback on accuracy at each trial.

The initial sequences were pseudo-random, consisting of graphemes with ei- ther large articulation spaces in the palette or small, specific articulation spaces. Overall, 672 sequences were produced across 76 chains, each chain being be- tween 5 and 10 generations. Reproduction error was measured as mean Euclidean distance in the palette from the target sequence to the produced sequence, and ex- amined using a linear mixed effects model with generation and articulation space size as fixed effects.

Inclusion of generation and articulation space size improved the model fit sig- nificantly over a null model (χ2 = 4.53, p = 0.03). Error was higher overall in chains which started with small articulation spaces (β = 0.1, SE= 0.04, t = 2.5), indicating that participants found these harder to reproduce. Error decreased over generations in the small articulation space condition (β = -0.018, SE = 0.006, t = −3.035), but there was no such effect of generation for sequences which started with large articulation spaces.

Error decreased despite an overall retention of high sequence entropy: in other words, most sequences retained three different characters throughout a chain. The size of the area of the articulation space of a particular Ferro grapheme was sig- nificantly correlated with copying accuracy, and sequences in small articulation chains moved towards the larger articulation spaces over time, suggesting that larger articulation spaces make for greater ease of articulation.

The current study shows transmission can lead to increased learnability even in a completely novel signal system that participants have to learn from scratch in the process of transmission. In this case, the biases which drive increased learnability seem to derive primarily from the shape of the signal space: articulation area of a particular Ferro correlates significantly with its learnability (operationalised as reproduction error), and sequences gravitated towards larger articulation spaces over ‘time’. The shape of the space was perhaps especially influential in the cur- rent task because participants had no prior experience with Ferros, and so had fewer (or less influential) prior biases relative to other iterated learning tasks, par- ticularly those that utilise non-words and leverage existing linguistic knowledge. This has potential implications for the co-evolution of language and speech: con- straints of the signal space may be more influential in early stages of emergence as ‘speakers’ are not only reproducing utterances, but also learning how to ma- nipulate a novel signal space. This study shows that the Ferro palette has broad applications in language evolution research as a system of truly ‘alien’ symbols to study ‘alien’ language learning, emergence, and evolution.

Do we see word order patterns from silent gesture studies in a new natural language?

Molly Flaherty & Marieke Schouwstra (University of Edinburgh) (joint paper with Susan Goldin-Meadow)

Tuesday 27 March 2018, 11:00–12:30
G32, 7 George Square

Typological analysis clearly shows that the world’s languages are not evenly distributed among all logically possible patterns. Of the six possible orderings of Subject (S), Object (O), and Verb (V), SOV and SVO orders are vastly overrepresented in the world’s languages. Studies on the emergence of word order regularities in silent gesture by hearing non-signers (e.g., Goldin- Meadow, et al., 2008; Gibson et al., 2013) overwhelmingly find evidence for SOV ordering. Based on this type of evidence, it has been proposed that SOV ordering is the most basic ordering from which all other orders emerged. However, semantic properties of the meanings to be conveyed also influence word order in silent gesture. For instance, for intensional events (in which the object is possibly non-existent or dependent on the action; e.g., ‘man thinks of guitar’, ‘woman builds house’) a cross-linguistic preference for SVO was found (Schouwstra & de Swart, 2014). Recent work finds that meaning-dependent word order patterns typical of silent gesture disappear under the influence of interaction (Christensen et al., 2016) and cultural transmission (Schouwstra et al., 2016), in favor of more consistent word order usage. However, in these studies, word order usage never becomes completely regular.

Here we investigate whether traces of the SOV/SVO pattern found in silent gesture can be observed in a new natural language: Nicaraguan Sign Language. This sign language, one of the youngest languages known to science, was born in the late 1970s with the founding of a new school for special education. Though instruction was in Spanish, students soon began to communicate with one another manually. As succeeding cohorts of students learn NSL, the language itself is changing rapidly. Though somewhat variable, NSL word order is strongly verb-final and predominantly SOV (Flaherty, 2014). However, these data are based exclusively on analysis of extensional events. If NSL word order is also influenced by semantic properties of the utterance’s intended meaning, we would expect to see deviation from this SOV patterning. Participants viewed a series of events depicting eight extensional events (i.e. woman pop bubble) and eight intensional events (i.e. woman blow bubble) involving the same object. Participants were asked to describe what they saw to a peer. Twenty-six NSL signers participated. All signers were exposed to NSL before age 7, upon school entry between the early 1980s and early 2000s.

When we analyzed SOV and SVO strings (which accounted for only 39% of strings with 1 verb and 2 arguments), we did not observe the pattern typical of silent gesture: SOV was dominant for both extensional and intensional events, and very few SVO strings were observed (13 total, 10 for intensional events). NSL’s preference for verb-finalness (Flaherty, 2014) may not have allowed the SVO pattern to emerge. However, NSL signers tend to provide more detail than silent gesturers. As a result, many NSL strings were longer than strings observed in silent gesture. When we took into account all strings (including those with several verbs) and asked whether the Object preceded or followed target the Verb, we found more utterances with VO sub-strings (as opposed to OV) for intensional events than for extensional events (Fig 1). A logit mixed effects regression (with event type as fixed effect and random effects for item and signer) confirmed that strings containing VO were uncommon for extensional events (β=-2.9, SE=0.40, p<0.001), but significantly more likely for intensional events (β=1.8, SE=0.45, p<0.001). Thus, objects of intensional verbs are more likely to follow those verbs than are objects of extensionals not only in silent gesture, but also in an emerging sign language.

In this study, we find evidence for lab-documented word order preferences in an emergent natural language: objects precede verbs for extensional events, but follow verbs for intensional events. However, this word order pattern is manifested differently in Nicaraguan Sign because it interacts with NSL’s language-internal constraint for verb finalness. A combination of lab and field-based methodologies made this finding possible: without laboratory results, we would not have looked at a wider semantic range of events in the field; without field data, we would not have discovered the interaction between VO ordering preference and existing natural language constraints.

9 March: Vanessa Ferdinand


On the cultural evolution of public and private replicators

Vanessa Ferdinand (Santa Fe Institute)

Friday 9 March 2018, 12:00–13:30
3.01, 21 Buccleuch Place

When we replicate cultural artifacts, we transform them. But we are also selective when choosing which artifacts to copy in the first place. In this talk, I will discuss the California and Paris schools of cultural evolution as compatible descriptions of evolutionary processes acting on public and private representations, respectively. First, using a unique corpus of digital image evolution, I will show how public and private selection processes co-determine the evolution of image complexity in this system. These results demonstrate that public and private selection processes are both required for a full understanding of the evolution of cultural artifacts. Second, in a novel model of Bayesian cultural evolution, I will explore how the mappings between public and private representations determine the co-evolutionary dynamics of these two classes of replicators (i.e. data and hypotheses). This model provides a richer set of dynamics than those found in existing models of cultural evolution where public representations adapt to fixed private representations. Both of these projects highlight culture as a special evolutionary system that is composed of two classes of replicators (following Sperber, 1996): public structures in the world, such as artifacts and behaviors, and private structures in the mind, such as brain states and grammars.

30 January: Marieke Schouwstra


Improvisation and semantic biases: silent gesture goes cross-modal

Marieke Schouwstra (University of Edinburgh)
(collaborative work with Katie Mudd, Ross Towns, and Simon Kirby)

Tuesday 30 January 2018, 11:00–12:30
G32, 7 George Square

Silent gesture, a laboratory method in which naive participants improvise, using their hands and no speech to convey information, has been taken up as a way to investigate cognitive preferences for constituent ordering when communication occurs in the absence of a conventional system. Recent years have seen a steady increase in silent gesture experiments, and the method has made it possible to observe semantic biases that might play a role in the emergence of conventional word order in human language.

But can we draw any conclusions about general linguistic mechanisms, based on results from the gestural modality? Much silent gesture work is presented under the assumption that the results are (at least partly) modality-independent. However, recent work suggests that some word order patterns can only be explained by appealing to the gestural modality specifically.

I will present two studies that address the issue of modality dependence, one in which participants interpret silent gesture strings, and one in which participants improvise in the vocal modality. Together, the experiments show that different semantic biases interact differently with the affordances of the modality, and that at least some patterns occur independently of the modality (phew!).

23 January: Molly Flaherty


How Developing Minds Build a New Language: the Emergence of Nicaraguan Sign Language

Molly Flaherty (University of Edinburgh)

Tuesday 23 January 2018, 11:00–12:30
G32, 7 George Square

This is a dry run for my job talk at UCSB for their job in “First Language Learning.” In this talk, I will give an overview of my research investigating the interaction between language input and the developing mind. I will closely examine a case of atypical first language learning: that of deaf children learning and creating a new language in Nicaragua. In this presentation, I will explore how the same forces at work in typical child language learning also build language structure when typical language input is not available. I use a range of methods and tools to explore the relative contributions of language input, language learning, and community interaction in building Nicaragua Sign Language (NSL). I will begin by discussing the growth of syntactic structure in homesign (manual communication systems created by isolated deaf individuals) and NSL by tracking the emergence of devices to mark argument structure. I find that the emergence of argument structure marking is neither instantaneous nor monolithic: some devices can be created by individual learners: homesigners and NSL signers of all generations employ these devices. Other markers show up only later with the addition of the language community and iterated transmission/learning of the language system. Next, I will turn to the emergence of structure at the morphological level and ask whether NSL signers have introduced overt marking of the morphological categories of noun and verb. I find that this distinction grows gradually: it appears in early Nicaraguan Sign Language and even in homesign, but increases in frequency as generations of children learn and pass down the language. Additionally, I will describe the gestural raw materials to Nicaraguan homesign and sign, and examine whether the gesture-rich communicative culture in Nicaragua may have been a particularly fertile environment in which to grow a new language. Finally, I investigate the cognitive consequences of not learning a mature language in childhood. I will present results indicating that in the domain of numerical cognition, individuals who do not learn number words, even if they are immersed in a world full of number, are unable to track exact quantity above five. In conclusion, I will show that the shape of language derives from human learning and use, and that one’s specific childhood language environment can have lifelong cognitive consequences.

5 December: Jennifer Culbertson


Children’s sensitivity to phonological and semantic cues during noun class learning: evidence for a phonological bias

Jennifer Culbertson (University of Edinburgh)

Tuesday 5 December 2017, 11:00–12:30
1.17 Dugald Stewart Building

Previous research on natural language acquisition of noun classification systems, such as grammatical gender, has shown that child learners appear to rely disproportionately on phonological cues (e.g., Gagliardi & Lidz, 2014; Karmiloff-Smith, 1981). Surprisingly, this occurs even when competing semantic cues are more reliable predictors of class. Culbertson, Gagliardi & Smith (2017) present evidence from artificial language learning experiments with adults suggesting that the over-reliance on phonology may be due to the fact that phonological cues are generally available earlier than semantic cues; learners acquire early representations of phonological dependencies (e.g., between a gendered determiner and a noun) before acquiring the semantic referents of nouns. In other words, Culbertson et al. (2017) suggest there is no a priori bias in favor of phonological cues to noun class. In this talk, I will present follow-up work investigating whether our results hold for child learners. In a series of experiments, we show that two cues–one semantic and one phonological–which children are equally sensitive to in isolation, are in fact treated differently when they are in conflict. In particular, unlike adults, children prioritize phonological cues regardless of when cues are available. This suggests the possibility that children are in fact biased to attend to phonological cues when acquiring noun classification systems.