Upcoming LUSH talks

The following speakers have been confirmed for the coming months.
If you wish to receive regular updates via email, please subscribe to our mailinglist. If you are interested in giving a talk, please send an e-mail to one of our organisers.

  • Patricia Cabredo Hofherr (CNRS). Leiden, April 23, 15.30-17.00.
  • Mieke Slim (Ghent University). Utrecht, May 29, 11.00-12.30.

Have a look at our previous speakers in the sidebar.

May 28, Mieke Slim

We are happy to announce that on Thursday, May 28, Mieke Slim will give a LUSH talk (combined talk with LACG). We hope to see you all there!

Date: Thursday, 28.5.2020

Time: 11:15 – 12:15

Location: Online (contact lacg@hum.leidenuniv.nl to get access)

Speaker: Mieke Slim, Department of Experimental Psychology, Ghent University, Belgium

Title: How abstract are logical representations of quantificational scope? Measuring effects of priming within and between quantifiers in language comprehension

Abstract:

A doubly quantified sentence (e.g. All bears approached a tent) is ambiguous between a universal-wide interpretation (All bears approached a different tent) and an existential-wide interpretation (all bears approached the same tent). Previous studies have shown that such interpretations, mentally represented as logical representations, are susceptible to effects of priming in language comprehension (Raffray & Pickering, 2010). This finding suggests that logical representations are not merely a theoretical construct, but are mentally represented.

Feiman and Snedeker (2016) observed that these priming effects only emerge if prime and target share the same quantifiers. This finding suggests that logical representations contain quantifierspecific information on the combinatorial properties of the quantifiers involved, and thus, that logicalrepresentations are quantifier-specific. If this conclusion is correct, one would expect this pattern (priming within but not between quantifiers) to generalize to different languages, even if they use slightly different quantifiers than English (e.g. Gil, 1995).

I will present a study in which we tested within- and between-quantifier priming in Dutch. We elicited effects of priming in language comprehension across multiple experiments by means of a sentence-picture matching task. In these experiments, the critical trials always involved a doubly quantified sentences. Prime trials contained either an elkeeen (‘every…a’), iedereeen (‘every….a’), or an alleeen (‘all…a’) sentence. Target trials always involved an elkeeen sentence. Thus, quantifiers were either shared between prime and target or not. Prime trials either elicited a universal-wide reading or an existential-wide reading (manipulated within participants). In the target trials, participants could freely choose between either interpretation. In case of priming, the universal-wide response is selected more often after a universal-wide prime trial than after an existential-wide prime trial.

In Experiment 1, we manipulated the quantifier in the prime sentence between subjects (n ≈ 60 per prime quantifier condition. The results of this experiment revealed a significant effect of priming in the elke-elke and alleelke condition, but not in the iedereelke condition. These results were unexpected, given that alle-elke priming is a between-quantifier condition. However, our data showed large individual differences between subjects in all three conditions, which may have led to the unexpected pattern of priming effects. Experiment 2 therefore repeated Experiment 1 in a fully withinsubject design (n = 180). These results showed comparable effects of logical representation priming in all three conditions.

These results are in contradiction with Feiman and Snedeker’s (2016) hypothesis. Therefore, we hypothesise instead that priming of logical representations emerges once the bias for the preferred interpretation of the target sentence has weakened: The threshold to compute the less-preferred interpretation must be lowered sufficiently in order for that dispreferred interpretation to be primed. In case of within-quantifier priming, the participant is explicitly exposed to the dispreferred interpretation of the target sentence in half of the prime sentences. Therefore, the participants rapidly became familiar with the dispreferred interpretation of the target sentence.

In Experiment 3, we will test this hypothesis. Here, we manipulate the presence of elke-elke trials between blocks. For half of the participants, elke-elke trials will appear in Block 1 (besides alle-elke and iedere-elke trials). For the other half of the participants, the elke-elke trials will only appear in Block 2. If our hypothesis is correct, there should always be priming in the block with  elke-elke trials, regardless of block order, as the bias for the preferred interpretation is adapted quickly. But in the block without elke-elke trials, priming depends on block order: no strong priming if it is the first block, priming if it is the second block. Pilot data (n = 40) tentatively show that the effect of priming is indeed boosted when the elke-elke trials are presented in the first block, in line with our hypothesis.

April 23, Patricia Cabredo Hofherr

We are happy to announce that on Thursday, April 23, Patricia Cabredo Hofherr will give a LUSH talk (combined talk with ComSyn). We hope to see you all there!

Date: Thursday, April 23,  15.15 – 16.30
Location: Online (contact one of our organisers to get access)
Speaker: Patricia Cabredo Hofherr (CNRS & U. Paris-8, UMR 7023 – SFL)
Title: Two superiority comparatives in Haitian Creole

The abstract can be found here.

January 14, Gert-Jan Schoenmakers

We are happy to announce that on Tuesday, January 14, Gert-Jan Schoenmakers (Radboud University Nijmegen) will give a LUSH talk in Utrecht. We hope to see you all there!

Date: Tuesday, January 14, 11:00-12:15
Location: Janskerkhof 15A, 2.04
Speaker: Gert-Jan Schoenmakers
Title: To the left or to the right of the adverb? Escaping the focus domain to avoid conflict.

Abstract:

Direct objects in the Dutch middlefield may surface to the left or to the right of an adverb. This phenomenon, known as scrambling, has been a persistent topic in the literature, because to this day it is not entirely clear what drives the variation in sentences with a definite object. Most researchers claim that information structure and matters of scope motivate scrambling. I argue on the basis of behavioral evidence from a speeded acceptability judgment task that while these notions influence the processing of scrambling structures, the human parser is also able to resolve conflict at the interfaces. In other words, scrambling is not as restricted as commonly assumed. Participants show a distinct preference to follow surface structure when building a discourse representation of clauses with adverbs that are sensitive to focus placement. Mismatches between form and meaning lead to delays in reaction times, but nonetheless yield fully acceptable structures. Crucially, the data reveal no such pattern for structures with adverbs that are focus insensitive. The findings indicate that, even in case of imperfect mapping between the speaker’s grammar and production system, the human parser is able to reconstruct the scrambling scope structure at the pragmatics interface.

January 6, Rodica Ivan

We are happy to announce that on Monday, January 6, Rodica Ivan (University of Massachusetts Amherst) will give a LUSH talk in Utrecht. We hope to see you all there!

Date: Monday, January 6, 15:30 – 16:45
Location: Utrecht, Drift 23, room 0.13
Speaker: Rodica Ivan
Title: Talking about her(self): Ambiguity avoidance in Romanian anaphora production

Abstract:

This talk bridges the theoretical and psycholinguistic literature in a study of Romanian pronominal anaphora. Romanian is an understudied language in the psycholinguistic literature which allows bound variables to be expressed through a variety of pronominal expressions. In Romanian, pronouns like her can refer back to a local referential antecedent (Luna talked about her), as well as be bound by local quantifiers (Every girl talked about her). Although Romanian is an apparent counterexample to both classic and competition-based accounts of the Binding Theory, I provide evidence, by means of data from production and comprehension experiments, in favor of competition: Romanian speakers exhibit ambiguity avoidance behavior in bound variable and local coreference contexts alike. Despite the obvious parallels between theoretical competition-based accounts of Condition B and the processing assumptions regarding cross-sentential pronominal reference, this is among the first studies of its kind to experimentally investigate pronominal competition in intrasentential environments. Crucially (and surprisingly to the Binding Theory literature), these results also show that Gricean reasoning applies to the selection of the form of a bound variable, and not only to unbound pronouns expressing coreference.

December 17, Mats Rooth

We are happy to announce that on Tuesday, December 17, Mats Rooth (Cornell University) will give a LUSH talk in Utrecht. We hope to see you all there!

Date: Tuesday, December 17, 11:00 – 13:00
Location: Utrecht, Drift 23, room 0.13
Speaker: Mats Rooth
Title: Constructive possible worlds semantics based on event sequences and finite state representations

Abstract:

In realistic possible worlds models, the set of possible worlds corresponding to a sentence is usually too big to represent as such. But with some severe idealizations, it is possible to represent propositions (sets of worlds) and relations between worlds such as epistemic alternative relations finitely. The idealization is to identify worlds with character strings, and to set up the lexicon and compositional semantics so that the sets and relations that come up are always regular sets and regular relations. These are the sets and relations that are representable by finite state machines. In one version, a world (at a time) is constructed as a string of events, resulting in a branching-time modal space. Epistemic alternative relations on worlds are constructed from primitive alternative relations on events. This is realized in an epistemic extension of guarded string models for Kleene algebra with tests (KAT). Two code bases are presented. One is an embedding in the finite state calculus, which is the string algebra implemented in Xfst and Foma. The other is Epik, which is a special-purpose language implemented in Haskell, using techniques adapted from computer science applications of KAT such as Netkat, a language for reasoning about packet networks. The talk is based on work with Eric Campbell (Cornell CS).

December 16, Annemarie van Dooren

We are happy to announce that on Monday, December 16, Annemarie van Dooren (University of Maryland) will give a LUSH talk in Utrecht. We hope to see you all there!

Date: Monday, December 16, 15:30 – 16:45
Location: Utrecht, Drift 23, room 0.12
Speaker: Annemarie van Dooren
Title: Correlating modal syntax and modal flavor across languages and in development

Abstract:

Words like must in English and moeten in Dutch can be used to express various modal ‘flavors’, including obligations (as in employees must wash their hands, called deontic modality) and likelihoods (as in John must be home; his car isn’t in the parking lot, called epistemic modality). The availability of various modal flavors seems to be constrained, in part, by the syntactic environment in which the modals appear. In this talk I ask how robust and principled the correlations between modal flavor and modal syntax are, by looking both at cross-linguistic and acquisition data. I first argue that despite surface variation, correlations between modal flavor and modal complement size hold robustly across Dutch, English, and Hebrew. Because such correlations may not be apparent on the surface, this raises acquisition questions: If the correlations are arbitrary, how can children pick up on them? But if they are principled, in ways that learners are privy to, they may be able to exploit surface syntax to figure out modal meanings. In the second part of the talk I explore this possibility by investigating how children figure out that modals can express different flavors in the first place.

December 12, Fabian Schlotterbeck

We are happy to announce that on Thursday, December 12, Fabian Schlotterbeck (Universität Tübingen) will give a LUSH talk in Utrecht. We hope to see you all there!

Date: Thursday, December 12, 11:00 – 12:15
Location: Utrecht, Trans 10, room 0.19 (A.W. de Grootkamer)
Speaker: Fabian Schlotterbeck
Title: Decision processes in the verification of monotone quantifiers

Abstract:

Psycholinguistic studies have repeatedly demonstrated that downward entailing quantifiers are more difficult to process than upward entailing ones. Although the empirical phenomenon itself is well-documented, it is a matter of current debate what cognitive processes cause the monotonicity effect. Our main aim is to contribute to this debate by testing predictions about the underlying processes that are derived from competing theoretical proposals: two-step and pragmatic processing models. To this end, we model data from two verification experiments, in particular, reaction times and accuracy, using a well-established model of decision making from mathematical psychology, namely the diffusion decision model (DDM). In both experiments, verification of upward entailing ‘more than half’ was compared to downward entailing ‘fewer than half’. One experiment employed a sentence-picture verification task and the other one used a purely linguistic version of the task. Our initial analyses revealed the same pattern of results across tasks: Both non-decision times and drift rates, two of the free model parameters of the DDM, were affected by the monotonicity manipulation. Thus, our initial modeling results support both two-step (prediction: non-decision time is affected) and pragmatic processing models (prediction: drift rate is affected). I will discuss theoretical implications of these results.

December 9, Martín Fuchs

We are happy to announce that on Monday, December 9, Martín Fuchs (Yale University) will give a LUSH talk in Utrecht. We hope to see you all there!

Date: Monday, December 9, 9:30 – 10:45
Location: Utrecht, Trans 10, room 0.51
Speaker: Martín Fuchs
Title: The communicative underpinnings of synchronic dialectal variation: experimental studies in the Spanish Imperfective domain

Abstract:

Within the Spanish Imperfective domain, the Simple Present and the Present Progressive markers alternatively convey the event-in-progress and the habitual reading in a many-to-many mapping. In this talk, I argue that the availability of each marker to convey each of these readings ultimately responds to communicative and cognitive pressures that are reflected in features of the contexts that affect the interpretation of these markers. This proposal is supported both by a corpus study and by experimental data from a set of tasks that allow for the controlled manipulation of contextual information. I finally show that the observed synchronic variation is the manifestation of sub-stages within the Progressive-to-Imperfective shift, a process of semantic change well-documented cross-linguistically.

October 28, Evangelia Vlachou

We are happy to announce that on Monday, October 28 Evangelia Vlachou (University of Athens) will give a LUSH talk in Utrecht. We hope to see you all there!

Date: Monday, October 28, 15:15 – 17:00
Location: Utrecht, Drift 23, room 0.13
Speaker: Evangelia Vlachou
Title: High degree modification and adjectives

Abstract:

High degree modifiers present interesting properties in the languages of the world. In this talk, we will focus on their combinability with adjectives. It will be argued that high degree modifiers in the languages of the world are sensitive to the distinction between eventive and stative predicates. Evidence will be given mainly from French and Greek, but comparison to other languages will also be made.

17/18 October, special LUSH event

On Thursday 17 and Friday 18 October you are very welcome to join us at a special LUSH event in Leiden. We hope to see you there!!

Numerals in Grammar and Beyond
Workshop @ Leiden University
17-18 October 2019

Workshop on numeral semantics, morphology, syntax, processing, acquisition, cross-linguistic variation and cognition — organized by LUCL, Leiden University, as part of NWO VENI project ‘Number Words’ (PI: Lisa Bylinina, bylinina@gmail.com)

Morning sessions: Lipsius, room 147
Afternoon sessions: Eyckhof 1, room 003C

For more information, see http://bylinina.com/veni-workshop.