VDVera Demberg

Vera Demberg

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Vera Demberg has a total of 148 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Polarity and the speed of scalar implicature processing

    by Michael Vrazitulis, J E Loy, Vera Demberg

    Scalar implicatures (SIs) such as the inference from 'some' to 'some but not all' are known to vary in processing cost. According to the Polarity Hypothesis (van Tiel et al., 2019; van Tiel & Pankratz, 2021), this cost depends on the polarity of the scalar term: SIs from positively polar expressions (e.g., 'some' implicating 'some but not all') are expected to be slower to compute than literal interpretations, while the reverse should hold for SIs from negatively polar expressions, sometimes referred to as indirect SIs (e.g., 'not all' implicating 'not all but some'). We tested this prediction in a sentence–picture verification experiment with 538 English-speaking participants. Each participant judged whether sentences containing weak or strong scalar terms matched a corresponding picture, allowing us to measure response times as an index of processing effort. Our results showed no clear evidence for the Polarity Hypothesis effect. Specifically, we found that the effect strongly varies across the different scales included in our experiment. We additionally conducted a meta-analysis pooling our estimate with all existing evidence so far: the pooled effect is small and in the predicted direction, but the study-specific estimates behind it vary widely in magnitude. Surprisingly, our results do also not clearly validate the traditional assumption of a B&N effect (Bott & Noveck, 2004), i.e., a general polarity-independent processing slowdown under SIs: a subsequent analysis including trial order revealed that the B&N effect was present in early trials of our study, whereas later trials do not show the effect. Overall, we conclude that polarity may influence the speed of scalar implicature processing, but that this effect is not robust across all types of scales.

  • Open Access

    Comprehenders’ Expectations During Discourse Processing: Disentangling Relation, Content, and Form Predictions From Connective Facilitation

    by Marian Marchal, Merel Scholman, Ted Sanders, Vera Demberg

    Comprehenders have been shown to make predictions about upcoming information at multiple levels, but the role of prediction in the processing of discourse relations remains unclear. This study investigated the role of prediction in discourse relation processing by disentangling three levels of predictability which are typically confounded: relation type, semantic content, and lexical-syntactic form. Furthermore, connectives are known to facilitate reading of subsequent material; here we examine whether this effect can be explained by connectives enhancing predictability at multiple levels, or whether they exert processing benefits beyond their effects on prediction. We derive measures of relation type, content predictability, and form predictability from a continuation task (Exp. 1) and show that the presence of a connective increases the predictability of the relation type and, in turn, of the semantic content. In a self-paced reading (Exp. 2) and an eye-tracking experiment (Exp. 3), we find that relation type predictability is indirectly associated with reading times through content predictability. However, relation type predictability did not directly facilitate processing and was associated with longer initial reading times in Exp. 3, suggesting that its effects on processing may be more complex than straightforward facilitation. The connective showed direct effects on reading times even after controlling for all three levels of predictability, suggesting that connective facilitation operates through mechanisms beyond enhanced prediction. These findings support accounts in which predictions at different linguistic levels constrain one another and show that both predictive and integrative mechanisms contribute to discourse comprehension.

  • Open Access

    Working memory and vocabulary knowledge sustain activation of the literal meaning of metaphors during sentence comprehension

    by Laura Pissani, Merel C. J. Scholman, Vera Demberg

    Metaphors convey meanings beyond their literal interpretation, requiring individuals to rely on linguistic and cognitive resources beyond those used for literal language. Models of metaphor comprehension propose different routes to deriving metaphorical content. The indirect-access model suggests that literal meaning is accessed first and then rejected if context demands, whereas the direct-access model argues that metaphorical content is retrieved automatically without an initial literal interpretation. However, previous research has yielded mixed findings on this issue. We examine whether individual differences in metaphor processing explain these inconsistencies. We hypothesize that individuals may differ in their ability to retrieve metaphorical content or inhibit literal meaning directly. In Experiment 1, participants read sentences such as John is an early bird so he can... and were slower and less likely to select a target continuation (e.g., attend ) when it was paired with a distractor related to the literal meaning of the metaphor (e.g., fly ) compared to an unrelated distractor (e.g., cry ). This literal awakening effect was more pronounced in individuals with higher vocabulary knowledge and working memory capacity. In Experiment 2, we examined whether the effect persisted when distractors were no longer grammatical in context (e.g., attend versus flying or crying ). The literal awakening effect remained but was no longer modulated by individual differences. These findings suggest that, under semantic and syntactic competition, individuals with higher vocabulary knowledge and working memory capacity may attempt to reinterpret the related distractor as an elaboration of the metaphor, indicating that activation of the literal meaning of metaphors is sustained by vocabulary knowledge and working memory capacity.

  • Open Access

    Two-Part Discourse Markers in Chinese: The Processing of Causal and Concessive Relations

    by Merel C. J. Scholman, Jingqi Yan, Frances Yung, Vera Demberg

    Prior work has shown that comprehenders can predict upcoming discourse relations based on two-part markers such as on the one hand…on the other hand or true/sure…but in English, Dutch and German. Chinese has a larger inventory of such two-part markers, and these pairs thus might have larger effects on processing. In the present study, we investigated whether Chinese readers show a similar processing advantage for two-part connectives, and whether this facilitation differs across relation types. We compare processing of causal relations to concessive relations, which are cognitively more complex. Specifically, we conducted two self-paced reading experiments testing whether concessive two-part connectives yield a greater processing advantage than causal connectives. Across both experiments, we found no evidence that the presence of the first marker facilitated processing of the second marker, nor that concessive relations elicited greater facilitation than causal relations. Bayes factor analyses provided evidence against the hypothesized interaction. These findings suggest that, unlike previous findings in Indo-European languages, two-part connectives in Chinese may not consistently enhance processing, which might be due to the greater reliance on contextual and pragmatic cues in discourse comprehension. Our results contribute to the growing body of cross-linguistic research indicating that the facilitative role of connectives is language- and context-dependent.

  • Open Access

    Referring Effectively to a Group of Objects

    by Emilia Ellsiepen, Alexandra Mayn, Natalia Bila, Vera Demberg

    In communication, we often need to refer to multiple objects. We have several possibilities: enumerating the objects one by one or referring to a shared feature. This is the first study to systematically compare different ways of referring to groups of objects in terms of response times, accuracy, and memory encoding. In a within-subjects design, participants listened to instructions either using a shared feature or enumerating coordinates of individual objects We find that there is no one way of referring that is universally more effective, but that the optimal way of referring depends on the type of common feature and the number of objects. When the shared feature is prominent, such as color, and multiple objects need to be identified, referring to it results in shorter identification times and fewer errors. In contrast, when the shared feature is less salient, such as pattern, referring by location tends to be better.

  • Open Access

    Speech Cues Influence the Interpretation of Metaphoric Gestures

    by Laura Pissani, Vera Demberg

    Language and gesture are argued to form an integrated system where each modality contributes to the meaning of utterances. While gestures are known to influence language comprehension, less is known about how language affects the interpretation of gestures. We investigated whether the interpretation of metaphoric gestures is modulated by multimodal cues. Participants viewed hand gestures depicting the figurative interpretation of metaphors such as Men are oaks, where the actor flexed both arms to convey robustness while speaking. Gestures were presented with no additional cues, only lip movements, or audible speech. Participants then rated how well each gesture matched the gestured interpretation (e.g., robust), an alternative interpretation related to the metaphor but not to the gesture (e.g., tall), and an unrelated distractor (e.g., hikers). We found that ratings for alternative interpretations increased when gestures were paired with speech cues, compared to conditions without speech cues, suggesting that gesture comprehension is influenced by verbal content.


  • Open Access

    Modeling Working Memory Capacity in Sentence Processing with Lossy-Context Surprisal

    by Jiaxin Li, Michael Hahn, Vera Demberg

    Understanding how comprehenders interpret linguistic input in real time has long been a focus of sentence processing research. The recently-proposed Resource-Rational Lossy-context Surprisal (RR-LCS) model can be applied for making predictions at the level of individual readers: it predicts that individuals with low working memory capacity (WMC) have a lossy representation of the previous context and may therefore experience high surprisal at a long-distance dependency, while individuals with high WMC will experience a lower surprisal effect as this dependent to them may still be highly predictable from previous context. To date, these predictions have not been empirically tested. Here, we evaluate them using a controlled A-Maze task (Experiment 1) and an analysis of reading more naturalistic text based on the InDiCo corpus (Experiment 2). In Experiment 1, we investigated how WMC influences sentence processing across short, long, and very long dependency conditions. The results revealed that low-WMC participants initially showed a strong locality effect between the short and long conditions, which decreased over time due to adaptation. High-WMC participants demonstrated locality effects between both the short-long and long-very long conditions from the start, with these effects gradually diminishing as their memory allocation became more efficient. These findings align with the RR-LCS model's predictions, showing that high-WMC individuals maintained larger context sizes, such that they could more easily process long-distance dependencies, whereas low-WMC individuals, constrained by smaller context sizes, experienced pronounced locality effects. Experiment 2 confirmed that WMC modulates sensitivity to lossy-context surprisal (LCS), with LCS improving model fit particularly strongly for low-WMC participants. These findings underscore the significance of individual differences in WMC for processing complex syntactic structures. By combining experimental data with computational modeling, we provide a deeper understanding of how WMC shapes language comprehension.

  • Open Access

    Bidirectional effects of speech and gesture: Evidence from a novel levels-of-congruence paradigm

    by Laura Pissani, Vera Demberg

    Previous research has shown that speech and gesture are integrated during comprehension, with differences in response times and electrophysiological activity observed between incongruent and congruent gesture–speech pairings. Beyond these incongruency effects, it remains unclear whether comprehenders use gestures to adapt or extend conventional interpretations for novel communicative purposes, and whether gestures are equally influenced by speech cues. The current study examined the bidirectional influence of speech and gesture by employing a novel levels-of-congruence paradigm, in which the same metaphor (e.g., Men are oaks) was paired with a gesture matching its preferred interpretation (men are robust) or a less preferred but possible interpretation (men are tall). In Experiment 1, participants rated the meaningfulness of the preferred and possible interpretations under different gesture conditions (gesture-of-preferred, gesture-of-possible, non-gestured). In Experiment 2, participants chose a single interpretation under the same gesture conditions. In Experiment 3, video clips of an actor producing the same gesture and utterance (e.g., flexing both arms to convey robustness, while saying Men are oaks) was presented with no additional cues (muted, blurred lips), with only lip movements (muted, visible lips), and with audible speech. Participants rated how well each gesture matched the gestured interpretation (e.g., robust), an alternative interpretation related to the speech but not the gesture (e.g., tall), and an unrelated interpretation (e.g., hikers). Overall, our findings are consistent with the integrated-systems hypothesis, showing that speech and gesture mutually and obligatorily interact in ways that enhance comprehension.

  • Open Access

    The Role of Reinforcement Learning in Pragmatic Reasoning Tasks: Modeling and Validating the Sources of Individual Differences

    by John Duff, Alexandra Mayn, Vera Demberg

    In Gricean pragmatics, inference during communication is regarded as a form of rational, domain-general reasoning about the intentions of other agents. Studies using the pictorial communication “reference game” task are sometimes used in support of this hypothesis. Yet, measures of pragmatic reasoning in this task sometimes reveal poor performance, with participants requiring many rounds of play before they exhibit patterns which match Gricean inferences, and demonstrating substantial individual differences in behavior. Do these results challenge the idea of widespread inferencing via fundamental social competence? We advance an alternative proposal here, which posits that these patterns emerge as a factor of the way participants perform pragmatic reasoning in a task: namely, they prefer to use simpler interpretation strategies until experience motivates the use of additional resources. Building off of work modeling task adaptation as reinforcement learning, we use the cognitive architecture ACT-R to simulate the expected behavior of individuals with this kind of resource-rational performance algorithm, subject to individualized parameters for reinforcement learning. These simulations provide a proof-of-concept for our adaptation proposal, recreating known patterns and generating new concrete predictions for the particular domain-general sources of individual variance in reference game tasks. We then go on to validate some of these new predictions in a pre-registered experiment, and find that pragmatic response behavior is indeed related to a participant’s general persistence in self-directed exploration of strategies for task completion. Our results offer a path to reconcile variable empirical data with models of core pragmatic competence. From a broader perspective, we see this as an important step towards more robust theories of performance factors in pragmatic reasoning, and ultimately, a case study in the value of process-level computational modeling.

  • Open Access

    Listeners Adapt Dynamically to Speakers' Pragmatic Competence

    by Kata Naszadi, Alexandra Mayn, John Duff, Christof Monz, Vera Demberg

    Speakers are typically assumed to adhere to Gricean communicative principles, which allows comprehenders to make inferences about the intended meaning. However, such inferences depend on the speaker being a pragmatic communicator, and overestimating the speaker's pragmatic competence may result in incorrect interpretations. We investigate whether comprehenders track speakers' pragmatic competence during interaction and adjust their interpretations accordingly. Across two paradigms, we find strong evidence of such an adaptation effect, whereby comprehenders lower their confidence in the pragmatic interpretation when interacting with a speaker who behaves less optimally. We propose a Bayesian belief updating model of this adaptation process. Our modeling results suggest that in some situations listeners may learn not through direct observation of speaker behavior, but by reasoning about their own interpretations: adaptation occurs on trials that are ambiguous from the listener’s perspective, even when they are unambiguous from the speaker’s perspective. Additionally, we observe evidence for different adaptation strategies, which may be explained by different prior beliefs about the speaker and tendency to generalize across interlocutors.