Matthew Crocker - Academia.edu (original) (raw)

Papers by Matthew Crocker

Cognitive Technologies, 2010

... A. Bundy JG Carbonell M. Pinkal H. Uszkoreit M. Veloso W. Wahlster MJ Wooldridge Advisory Boa... more ... A. Bundy JG Carbonell M. Pinkal H. Uszkoreit M. Veloso W. Wahlster MJ Wooldridge Advisory Board: Luigia Carlucci Aiello Franz ... Sharon Oviatt Luis Pereira Lu Ruqian Stuart Russell Erik Sandewall Luc Steels Oliviero Stock Peter Stone Gerhard Strube Katia Sycara Milind ...

Studies in Theoretical Psycholinguistics, 1996

Studies in Theoretical Psycholinguistics, 1996

Studies in Theoretical Psycholinguistics, 1996

Previous research has shown that listeners follow speaker gaze to mentioned objects in a shared e... more Previous research has shown that listeners follow speaker gaze to mentioned objects in a shared environment to ground referring expressions, both for human and robot speakers. What is less clear is whether the benefit of speaker gaze is due to the inference of referential intentions (Staudte and Crocker, 2011) or simply the (reflexive) shifts in visual attention. That is, is gaze special in how it affects simultaneous utterance comprehension? In four eye-tracking studies we directly contrast speech-aligned speaker gaze of a virtual agent with a non-gaze visual cue (arrow). Our findings show that both cues similarly direct listeners' attention and that listeners can benefit in utterance comprehension from both cues. Only when they are similarly precise, however, does this equality extend to incongruent cueing sequences: that is, even when the cue sequence does not match the concurrent sequence of spoken referents can listeners benefit from gaze as well as arrows. The results suggest that listeners are able to learn a counter-predictive mapping of both cues to the sequence of referents. Thus, gaze and arrows can in principle be applied with equal flexibility and efficiency during language comprehension.

Studies in Theoretical Psycholinguistics, 1996

Empirical evidence demonstrating that sentence meaning is rapidly reconciled with the visual envi... more Empirical evidence demonstrating that sentence meaning is rapidly reconciled with the visual environment has been broadly construed as supporting the seamless interaction of visual and linguistic representations during situated comprehension. Based on recent behavioral and neuroscientific findings, however, we argue for the more deeply rooted coordination of the mechanisms underlying visual and linguistic processing, and for jointly considering the behavioral and neural correlates of scene-sentence reconciliation during situated comprehension. The Coordinated Interplay Account (CIA; Knoeferle, P., & Crocker, M. W. (2007). The influence of recent scene events on spoken comprehension: Evidence from eye movements. Journal of Memory and Language, 57(4), 519-543) asserts that incremental linguistic interpretation actively directs attention in the visual environment, thereby increasing the salience of attended scene information for comprehension. We review behavioral and neuroscientific findings in support of the CIA's three processing stages: (i) incremental sentence interpretation, (ii) language-mediated visual attention, and (iii) the on-line influence of non-linguistic visual context. We then describe a recently developed connectionist model which both embodies the central CIA proposals and has been successfully applied in modeling a range of behavioral findings from the visual world paradigm (Mayberry, M. R., Crocker, M. W., & Knoeferle, P. (2009). Learning to attend: A connectionist model of situated language comprehension. Cognitive Science). Results from a new simulation suggest the model also correlates with eventrelated brain potentials elicited by the immediate use of visual context for linguistic disambiguation . Visual scenes trigger immediate syntactic reanalysis: Evidence from ERPs during situated spoken comprehension. Cerebral Cortex, 18(4), 789-795). Finally, we argue that the mechanisms underlying interpretation, visual attention, and scene apprehension are not only in close temporal synchronization, but have co-adapted to optimize real-time visual grounding of situated spoken language, thus facilitating the association of linguistic, visual and motor representations that emerge during the course of our embodied linguistic experience in the world.

As potential candidates for explaining human cognition, connectionist models of sentence processi... more As potential candidates for explaining human cognition, connectionist models of sentence processing must demonstrate their ability to behave systematically, generalizing from a small training set. It has recently been shown that simple recurrent networks and, to a greater extent, echo-state networks possess some ability to generalize in artificial language learning tasks. We investigate this capacity for a recently introduced model that consists of separately trained modules: a recursive self-organizing module for learning temporal context representations and a feedforward two-layer perceptron module for next-word prediction. We show that the performance of this architecture is comparable with echo-state networks. Taken together, these results weaken the criticism of connectionist approaches, showing that various general recursive connectionist architectures share the potential of behaving systematically. r Bode´n and van Gelder [3] proposed a more fine-grained taxonomy of the levels of systematicity, but since here we only focus on weak systematicity, there is no need to introduce this taxonomy here. . His research areas include neural networks and cognitive science with focus on connectionist natural language modeling.

European Conference on Artificial Intelligence, 1992

Proceedings of the 18th conference on Computational linguistics -, 2000

Given the recent evidence for prot)abilistic mechanisms in models of hmnan aml)iguity resolution,... more Given the recent evidence for prot)abilistic mechanisms in models of hmnan aml)iguity resolution, this paper investigates the plausibility of exl)loiting current wide-coverage, 1)robal)ilistic 1)arsing techniques to model hmnan linguistic t)ert'orman(:e. In l)arl.i(:ulm ', we investigate the, t)crforlnance of stan(tar(l stoclmstic parsers when they arc revis(;(l to el)crate incrementally, and with reduced nlenlory resources. We t)resent techniques for ranking and filtering mlMyses, together with exl)erimental results. Our results confirm that stochastic parsers which a(lhere to these 1)sy('hologically lnotivated constraints achieve goo(l l)erf()rman(:e. Memory cast t)e reduce(t (lown to 1% ((:Oml)are(l to exhausitve search) without reducing recall an(l 1)rox:ision. A(lditionally, thes(; models exhil)it substamtially faster l)ertbrmance. FinMly, we ~rgue that this generM result is likely to hold for more sophisticated, nnd i)sycholinguistically plausil)le, probal)ilistic parsing models.

Architectures and Mechanisms for Language Processing, 1999

The Modular Statistical Hypothesis: Exploring Lexical Category Ambiguity STEFFAN CORLEY AND MATTH... more The Modular Statistical Hypothesis: Exploring Lexical Category Ambiguity STEFFAN CORLEY AND MATTHEW W. CROCKER 6.1 Introduction A central topic of debate in the sentence processing community has been whether or not the Human Sentence-Processing Mechanism ...

Skull base : official journal of North American Skull Base Society ... [et al.], 2008

To report a new case of cavernous hemangioma of the optic chiasm and to review all previously pub... more To report a new case of cavernous hemangioma of the optic chiasm and to review all previously published cases with regard to presentation, surgical treatment, and outcomes. Case report and literature review. Cases identified though PubMed and published literature. Presentation states of patients in terms of visual loss and pituitary function. Surgical approaches, operations performed, and outcomes of the surgery. thirty-nine previously reported cases were studied with the present case. All 40 patients presented with visual failure. Where documented, there was an 20% pituitary dysfunction rate. A total of 32 craniotomies were reported. Seventy-eight percent of patients underwent decompression including hematoma evacuation and partial or complete removal of the cavernoma, with improvement in visual function in 87% of these patients. The patients undergoing only biopsy showed stable visual function in 50% with further deterioration in 50%. We conclude that this rare condition can be ma...

Architectures and Mechanisms for Language Processing, 1999

Neuro-oncology, 2012

Tumor vasculature is a promising therapeutic target in glioblastoma. Imaging tumor blood flow may... more Tumor vasculature is a promising therapeutic target in glioblastoma. Imaging tumor blood flow may help assess the efficacy of anti-angiogenic treatments. We determined the clinical usefulness of stable xenon CT performed preoperatively in patients with glioblastoma. This is a prospective cohort study. We determined absolute tumor blood flow before surgery in 38 patients with glioblastoma using stable xenon CT. We also histologically examined tumor specimens obtained from surgery and quantified their vascularity (by CD31 and CD105 immunostain), necrosis (by hematoxylin and eosin stain), and the presence of neuronal processes (by neurofilament immunostain). According to the xenon CT blood flow map, there are 3 types of glioblastoma. Type I glioblastomas have unimodal high blood flow histograms; histologically there is little necrosis and vascular proliferation. Type II glioblastomas have unimodal low blood flow histograms; histologically there is prominent necrosis and vascular prolif...

This article deals with gradience in human sentence processing. We review the experimental eviden... more This article deals with gradience in human sentence processing. We review the experimental evidence for the role of experience in guiding the decisions of the sentence processor. Based on this evidence, we argue that the gradient behavior observed in the processing of certain syntactic constructions can be traced back to the amount of past experience that the processor has had with these constructions. In modeling terms, linguistic experience can be approximated using large, balanced corpora. We give an overview of corpus-based and probabilistic models in the literature that have exploited this fact, and hence are well placed to make gradient predictions about processing behavior. Finally, we discuss a number of questions regarding the relationship between gradience in sentence processing and gradient grammaticality, and come to the conclusion that these two phenomena should be treated separately in conceptual and modeling terms.

A model of sentence-picture integration developed by Carpenter and Just (1975) predicts that pict... more A model of sentence-picture integration developed by Carpenter and Just (1975) predicts that picture- sentence integration ease/diculty depends on picture- sentence match/mismatch respectively. Recent find- ings by Underwood, Jebbet, and Roberts (2004), how- ever, fail to find a match/mismatch dierence for se- rial picture-sentence presentation in a sentence veri- fication study. In a sentence comprehension study with serial picture-sentence presentation we find no match/mismatch eect in total sentence inspection times. However, inspection times for individual sentence regions reveal a mismatch eect at the very sentence constituent for which the corresponding picture con- stituent mismatches, and this in a study with a sentence comprehension rather than verification task. Drawing on insights about spoken sentence comprehension dur- ing the inspection of concurrent scenes, we suggest that the absence of a mismatch eect in the Underwood et al. studies might be due to grain size of gaz...

Journal of Psycholinguistic Research, 2012

We present two eye-tracking experiments that investigate lexical frequency and semantic context c... more We present two eye-tracking experiments that investigate lexical frequency and semantic context constraints in spoken-word recognition in German. In both experiments, the pivotal words were pairs of nouns overlapping at onset but varying in lexical frequency. In Experiment 1, German listeners showed an expected frequency bias towards high-frequency competitors (e.g., Blume, 'flower') when instructed to click on low-frequency targets (e.g., Bluse, 'blouse'). In Experiment 2, semantically constraining context increased the availability of appropriate low-frequency target words prior to word onset, but did not influence the availability of semantically inappropriate high-frequency competitors at the same time. Immediately after target word onset, however, the activation of high-frequency competitors was reduced in semantically constraining sentences, but still exceeded that of unrelated distractor words significantly. The results suggest that (1) semantic context acts to downgrade activation of inappropriate competitors rather than to exclude them from competition, and (2) semantic context influences spoken-word recognition, over and above anticipation of upcoming referents.

Narrow focus refers to accent placement that forces one interpretation of a sentence, which is th... more Narrow focus refers to accent placement that forces one interpretation of a sentence, which is then often perceived contrastively. Narrow focus is formalised in terms of alternative sets, i.e. contextually or situationally salient alternatives. In this paper, we investigate whether this model is valid also in human utterance processing. We present an eye-tracking experiment to study listeners' expectations (i.e. eye-movements) with respect to upcoming referents. Some of the objects contrast in colour with objects that were previously referred to, others do not; the objects are referred to with either a narrow focus on the colour adjective or with broad focus on the noun. Results show that narrow focus on the adjective increases early fixations to contrastive referents. Narrow focus hence activates alternative referents in human utterance processing.

This paper describes a computational model of human sentence processing based on the principles a... more This paper describes a computational model of human sentence processing based on the principles and parameters paradigm of current linguistic theory. The syntactic processing model posits four modules, recovering phrase structure, long-distance dependencies, coreference, and thematic structure. These four modules are implemented as recta-interpreters over their relevant components of the grammar, permitting variation in the deductive strategies employed by each module. These four interpreters are also 'coroutined' via the freeze directive of constraint logic programruing to achieve incremental interpretation across the modules.

In this study, a technique called semantic self-organization is used to scale up the subsymbolic ... more In this study, a technique called semantic self-organization is used to scale up the subsymbolic approach by allowing a network to optimally allocate frame representations from a semantic dependency graph. The resulting architecture, INSOMNet, was trained on semantic representations of the newly-released LinGO Redwoods HPSG Treebank of annotated sentences from the VerbMobil project. The results show that INSOMNet is able to accurately represent the semantic dependencies while demonstrating expectations and defaults, coactivation of multiple interpretations, and robust processing of noisy input. The cognitive plausibility of the model is underscored by the collective modelling of four experiments from the visual worlds paradigm to show the model's ability to adapt to context.

Cognitive Technologies, 2010

... A. Bundy JG Carbonell M. Pinkal H. Uszkoreit M. Veloso W. Wahlster MJ Wooldridge Advisory Boa... more ... A. Bundy JG Carbonell M. Pinkal H. Uszkoreit M. Veloso W. Wahlster MJ Wooldridge Advisory Board: Luigia Carlucci Aiello Franz ... Sharon Oviatt Luis Pereira Lu Ruqian Stuart Russell Erik Sandewall Luc Steels Oliviero Stock Peter Stone Gerhard Strube Katia Sycara Milind ...

Studies in Theoretical Psycholinguistics, 1996

Studies in Theoretical Psycholinguistics, 1996

Studies in Theoretical Psycholinguistics, 1996

Previous research has shown that listeners follow speaker gaze to mentioned objects in a shared e... more Previous research has shown that listeners follow speaker gaze to mentioned objects in a shared environment to ground referring expressions, both for human and robot speakers. What is less clear is whether the benefit of speaker gaze is due to the inference of referential intentions (Staudte and Crocker, 2011) or simply the (reflexive) shifts in visual attention. That is, is gaze special in how it affects simultaneous utterance comprehension? In four eye-tracking studies we directly contrast speech-aligned speaker gaze of a virtual agent with a non-gaze visual cue (arrow). Our findings show that both cues similarly direct listeners' attention and that listeners can benefit in utterance comprehension from both cues. Only when they are similarly precise, however, does this equality extend to incongruent cueing sequences: that is, even when the cue sequence does not match the concurrent sequence of spoken referents can listeners benefit from gaze as well as arrows. The results suggest that listeners are able to learn a counter-predictive mapping of both cues to the sequence of referents. Thus, gaze and arrows can in principle be applied with equal flexibility and efficiency during language comprehension.

Studies in Theoretical Psycholinguistics, 1996

Empirical evidence demonstrating that sentence meaning is rapidly reconciled with the visual envi... more Empirical evidence demonstrating that sentence meaning is rapidly reconciled with the visual environment has been broadly construed as supporting the seamless interaction of visual and linguistic representations during situated comprehension. Based on recent behavioral and neuroscientific findings, however, we argue for the more deeply rooted coordination of the mechanisms underlying visual and linguistic processing, and for jointly considering the behavioral and neural correlates of scene-sentence reconciliation during situated comprehension. The Coordinated Interplay Account (CIA; Knoeferle, P., & Crocker, M. W. (2007). The influence of recent scene events on spoken comprehension: Evidence from eye movements. Journal of Memory and Language, 57(4), 519-543) asserts that incremental linguistic interpretation actively directs attention in the visual environment, thereby increasing the salience of attended scene information for comprehension. We review behavioral and neuroscientific findings in support of the CIA's three processing stages: (i) incremental sentence interpretation, (ii) language-mediated visual attention, and (iii) the on-line influence of non-linguistic visual context. We then describe a recently developed connectionist model which both embodies the central CIA proposals and has been successfully applied in modeling a range of behavioral findings from the visual world paradigm (Mayberry, M. R., Crocker, M. W., & Knoeferle, P. (2009). Learning to attend: A connectionist model of situated language comprehension. Cognitive Science). Results from a new simulation suggest the model also correlates with eventrelated brain potentials elicited by the immediate use of visual context for linguistic disambiguation . Visual scenes trigger immediate syntactic reanalysis: Evidence from ERPs during situated spoken comprehension. Cerebral Cortex, 18(4), 789-795). Finally, we argue that the mechanisms underlying interpretation, visual attention, and scene apprehension are not only in close temporal synchronization, but have co-adapted to optimize real-time visual grounding of situated spoken language, thus facilitating the association of linguistic, visual and motor representations that emerge during the course of our embodied linguistic experience in the world.

As potential candidates for explaining human cognition, connectionist models of sentence processi... more As potential candidates for explaining human cognition, connectionist models of sentence processing must demonstrate their ability to behave systematically, generalizing from a small training set. It has recently been shown that simple recurrent networks and, to a greater extent, echo-state networks possess some ability to generalize in artificial language learning tasks. We investigate this capacity for a recently introduced model that consists of separately trained modules: a recursive self-organizing module for learning temporal context representations and a feedforward two-layer perceptron module for next-word prediction. We show that the performance of this architecture is comparable with echo-state networks. Taken together, these results weaken the criticism of connectionist approaches, showing that various general recursive connectionist architectures share the potential of behaving systematically. r Bode´n and van Gelder [3] proposed a more fine-grained taxonomy of the levels of systematicity, but since here we only focus on weak systematicity, there is no need to introduce this taxonomy here. . His research areas include neural networks and cognitive science with focus on connectionist natural language modeling.

European Conference on Artificial Intelligence, 1992

Proceedings of the 18th conference on Computational linguistics -, 2000

Given the recent evidence for prot)abilistic mechanisms in models of hmnan aml)iguity resolution,... more Given the recent evidence for prot)abilistic mechanisms in models of hmnan aml)iguity resolution, this paper investigates the plausibility of exl)loiting current wide-coverage, 1)robal)ilistic 1)arsing techniques to model hmnan linguistic t)ert'orman(:e. In l)arl.i(:ulm ', we investigate the, t)crforlnance of stan(tar(l stoclmstic parsers when they arc revis(;(l to el)crate incrementally, and with reduced nlenlory resources. We t)resent techniques for ranking and filtering mlMyses, together with exl)erimental results. Our results confirm that stochastic parsers which a(lhere to these 1)sy('hologically lnotivated constraints achieve goo(l l)erf()rman(:e. Memory cast t)e reduce(t (lown to 1% ((:Oml)are(l to exhausitve search) without reducing recall an(l 1)rox:ision. A(lditionally, thes(; models exhil)it substamtially faster l)ertbrmance. FinMly, we ~rgue that this generM result is likely to hold for more sophisticated, nnd i)sycholinguistically plausil)le, probal)ilistic parsing models.

Architectures and Mechanisms for Language Processing, 1999

The Modular Statistical Hypothesis: Exploring Lexical Category Ambiguity STEFFAN CORLEY AND MATTH... more The Modular Statistical Hypothesis: Exploring Lexical Category Ambiguity STEFFAN CORLEY AND MATTHEW W. CROCKER 6.1 Introduction A central topic of debate in the sentence processing community has been whether or not the Human Sentence-Processing Mechanism ...

Skull base : official journal of North American Skull Base Society ... [et al.], 2008

To report a new case of cavernous hemangioma of the optic chiasm and to review all previously pub... more To report a new case of cavernous hemangioma of the optic chiasm and to review all previously published cases with regard to presentation, surgical treatment, and outcomes. Case report and literature review. Cases identified though PubMed and published literature. Presentation states of patients in terms of visual loss and pituitary function. Surgical approaches, operations performed, and outcomes of the surgery. thirty-nine previously reported cases were studied with the present case. All 40 patients presented with visual failure. Where documented, there was an 20% pituitary dysfunction rate. A total of 32 craniotomies were reported. Seventy-eight percent of patients underwent decompression including hematoma evacuation and partial or complete removal of the cavernoma, with improvement in visual function in 87% of these patients. The patients undergoing only biopsy showed stable visual function in 50% with further deterioration in 50%. We conclude that this rare condition can be ma...

Architectures and Mechanisms for Language Processing, 1999

Neuro-oncology, 2012

Tumor vasculature is a promising therapeutic target in glioblastoma. Imaging tumor blood flow may... more Tumor vasculature is a promising therapeutic target in glioblastoma. Imaging tumor blood flow may help assess the efficacy of anti-angiogenic treatments. We determined the clinical usefulness of stable xenon CT performed preoperatively in patients with glioblastoma. This is a prospective cohort study. We determined absolute tumor blood flow before surgery in 38 patients with glioblastoma using stable xenon CT. We also histologically examined tumor specimens obtained from surgery and quantified their vascularity (by CD31 and CD105 immunostain), necrosis (by hematoxylin and eosin stain), and the presence of neuronal processes (by neurofilament immunostain). According to the xenon CT blood flow map, there are 3 types of glioblastoma. Type I glioblastomas have unimodal high blood flow histograms; histologically there is little necrosis and vascular proliferation. Type II glioblastomas have unimodal low blood flow histograms; histologically there is prominent necrosis and vascular prolif...

This article deals with gradience in human sentence processing. We review the experimental eviden... more This article deals with gradience in human sentence processing. We review the experimental evidence for the role of experience in guiding the decisions of the sentence processor. Based on this evidence, we argue that the gradient behavior observed in the processing of certain syntactic constructions can be traced back to the amount of past experience that the processor has had with these constructions. In modeling terms, linguistic experience can be approximated using large, balanced corpora. We give an overview of corpus-based and probabilistic models in the literature that have exploited this fact, and hence are well placed to make gradient predictions about processing behavior. Finally, we discuss a number of questions regarding the relationship between gradience in sentence processing and gradient grammaticality, and come to the conclusion that these two phenomena should be treated separately in conceptual and modeling terms.

A model of sentence-picture integration developed by Carpenter and Just (1975) predicts that pict... more A model of sentence-picture integration developed by Carpenter and Just (1975) predicts that picture- sentence integration ease/diculty depends on picture- sentence match/mismatch respectively. Recent find- ings by Underwood, Jebbet, and Roberts (2004), how- ever, fail to find a match/mismatch dierence for se- rial picture-sentence presentation in a sentence veri- fication study. In a sentence comprehension study with serial picture-sentence presentation we find no match/mismatch eect in total sentence inspection times. However, inspection times for individual sentence regions reveal a mismatch eect at the very sentence constituent for which the corresponding picture con- stituent mismatches, and this in a study with a sentence comprehension rather than verification task. Drawing on insights about spoken sentence comprehension dur- ing the inspection of concurrent scenes, we suggest that the absence of a mismatch eect in the Underwood et al. studies might be due to grain size of gaz...

Journal of Psycholinguistic Research, 2012

We present two eye-tracking experiments that investigate lexical frequency and semantic context c... more We present two eye-tracking experiments that investigate lexical frequency and semantic context constraints in spoken-word recognition in German. In both experiments, the pivotal words were pairs of nouns overlapping at onset but varying in lexical frequency. In Experiment 1, German listeners showed an expected frequency bias towards high-frequency competitors (e.g., Blume, 'flower') when instructed to click on low-frequency targets (e.g., Bluse, 'blouse'). In Experiment 2, semantically constraining context increased the availability of appropriate low-frequency target words prior to word onset, but did not influence the availability of semantically inappropriate high-frequency competitors at the same time. Immediately after target word onset, however, the activation of high-frequency competitors was reduced in semantically constraining sentences, but still exceeded that of unrelated distractor words significantly. The results suggest that (1) semantic context acts to downgrade activation of inappropriate competitors rather than to exclude them from competition, and (2) semantic context influences spoken-word recognition, over and above anticipation of upcoming referents.

Narrow focus refers to accent placement that forces one interpretation of a sentence, which is th... more Narrow focus refers to accent placement that forces one interpretation of a sentence, which is then often perceived contrastively. Narrow focus is formalised in terms of alternative sets, i.e. contextually or situationally salient alternatives. In this paper, we investigate whether this model is valid also in human utterance processing. We present an eye-tracking experiment to study listeners' expectations (i.e. eye-movements) with respect to upcoming referents. Some of the objects contrast in colour with objects that were previously referred to, others do not; the objects are referred to with either a narrow focus on the colour adjective or with broad focus on the noun. Results show that narrow focus on the adjective increases early fixations to contrastive referents. Narrow focus hence activates alternative referents in human utterance processing.

This paper describes a computational model of human sentence processing based on the principles a... more This paper describes a computational model of human sentence processing based on the principles and parameters paradigm of current linguistic theory. The syntactic processing model posits four modules, recovering phrase structure, long-distance dependencies, coreference, and thematic structure. These four modules are implemented as recta-interpreters over their relevant components of the grammar, permitting variation in the deductive strategies employed by each module. These four interpreters are also 'coroutined' via the freeze directive of constraint logic programruing to achieve incremental interpretation across the modules.

In this study, a technique called semantic self-organization is used to scale up the subsymbolic ... more In this study, a technique called semantic self-organization is used to scale up the subsymbolic approach by allowing a network to optimally allocate frame representations from a semantic dependency graph. The resulting architecture, INSOMNet, was trained on semantic representations of the newly-released LinGO Redwoods HPSG Treebank of annotated sentences from the VerbMobil project. The results show that INSOMNet is able to accurately represent the semantic dependencies while demonstrating expectations and defaults, coactivation of multiple interpretations, and robust processing of noisy input. The cognitive plausibility of the model is underscored by the collective modelling of four experiments from the visual worlds paradigm to show the model's ability to adapt to context.