Markus Hofmann - Academia.edu (original) (raw)
Papers by Markus Hofmann
Neuroscience Letters, 2006
NeuroReport, 2006
A recent eye-tracking study reported a reverse effect of a noun&a... more A recent eye-tracking study reported a reverse effect of a noun's lexical frequency in the context of the resolution of coreferring pronouns. Investigating the neurophysiological basis of this effect, the present electroencephalographic study found differential patterns in theta activation when participants read pronouns referring to nouns of different frequency classes. Evoked theta power after pronoun onset increased with the frequency of the critical noun. This finding suggests differential load on memory resources depending on the nouns' frequency. Elevated attention promoting memory encoding for low-frequency words is assumed to facilitate the resolution of pronouns. Location of sources of differential theta activity in the parahippocampal region is accounted for by its role in an association network that mediates memory processes.
Neuroreport, 2006
A recent eye tracking study reported a reverse effect of a nouns lexical frequency in the context... more A recent eye tracking study reported a reverse effect of a nouns lexical frequency in the context of the resolution of coreferring pronouns. Investigating the neurophysiological basis of this effect, the present electroencephalogram study found differential pat- terns in theta activation when participants read pronouns refer- ring to nouns of different frequency classes. Evoked theta power after pronoun onset increased
Previous neurocognitive approaches to word predictability from sentence context in electroencepha... more Previous neurocognitive approaches to word predictability from sentence context in electroencephalographic (EEG) and eye movement (EM) data relied on cloze completion probability (CCP) data effortly collected from up to 100 human participants. Here we test whether two well-established techniques in computational linguistics can predict these data. Together with baseline pre- dictors of word position and frequency, we found that n-gram language models but not topic models provide an approach to EEG and EM data that is not significantly inferior to the CCP-based predictability data. This is the case for the three corpora we used. Most strikingly, our models accounted for about half of the variance of the CCP-based predictability estimates, thus suggesting that it provides a computational framework to explain the predictability of a word from sentence context. This can help to generalize neurocognitive models to all possible novel word combinations.
Frontiers in Neuroscience, 2014
ABSTRACT Roediger and McDermott (1995) showed that words are falsely recognized, when they provid... more ABSTRACT Roediger and McDermott (1995) showed that words are falsely recognized, when they provide many associated items in the stimulus set. The Associative Read-Out Model (AROM) is the first interactive model with an implemented semantic layer (Hofmann et al., 2011). It suggests that two words are associated, when they co-occur significantly more often together in sentences than predictable by chance (see Hofmann et al., 2011). We conducted a recognition memory task, in which 80 words were presented in a study phase. These studied words had to be discriminated from 80 non-studied (N=33). We crossed this Oldness factor with the number of associated words in a 2x2 rmANOVA design. High co-occurrence words had at least 8 associated words in the stimulus set, and low co-occurrence words less than 8. We not only replicated that high co-occurrence words increased the “yes” response in studied and non-studied words (Hofmann & Jacobs, in press), but we also found a significant interaction of Oldness and Co-occurrence in the reaction times, showing slowest reaction times for high co-occurrence non-studied words and fastest for high co-occurrence studied words. The Event-related potentials recorded from 32-channels during the test phase showed significant Oldness and Co-occurrence main effects for the N400 amplitude. The N400 was smaller for high co-occurrence and studied words and larger for low co-occurrence and non-studied words, which was predicted by the neurobiologically plausible AROM (Hofmann & Jacobs, in press).
Frontiers in Psychology, 2015
Physics of Life Reviews, 2015
Neuroscience and biobehavioral reviews, 2014
Interactive activation and competition models (IAMs) cannot only account for behavioral data from... more Interactive activation and competition models (IAMs) cannot only account for behavioral data from implicit memory tasks, but also for brain data. We start by a discussion of standards for developing and evaluating cognitive models, followed by example demonstrations. In doing so, we relate IAM representations to word length, sequence, frequency, repetition, and orthographic neighborhood effects in behavioral, electrophysiological, and neuroimaging studies along the ventral visual stream. We then examine to what extent lexical competition can account for anterior cingulate cortex (ACC) activation and the N2/N400 complex. The subsequent section presents the Associative Read-Out Model (AROM), which extends the scope of IAMs by introducing explicit memory and semantic representations. Thereby, it can account for false memories, and familiarity and recollection - explaining why memory signal variances are greater for studied than non-studied items. Since the AROM captures associative spr...
Abstract The current study investigated the role played by conflict monitoring in a lexical-decis... more Abstract The current study investigated the role played by conflict monitoring in a lexical-decision task involving competing word representations, using event-related potentials. We extended the multiple read-out model (Grainger and Jacobs, 1996), a connectionist model of word recognition, to quantify conflict by means of Hopfield Energy, which is defined as the sum of the products of all orthographic word node pair activations within the artificial mental lexicon of this model.
Psychological Research Psychologische Forschung, 2009
The conflict monitoring theory (CMT) assumes that word-stems associated with several completions ... more The conflict monitoring theory (CMT) assumes that word-stems associated with several completions should lead to crosstalk and conflict due to underdetermined responding situation (Botvinick et al. in Psychol Rev 108(3):624-652, 2001). In contrast, the Multiple-Read-Out-Model (MROM) of Jacobs and Grainger (J Exp Psychol 20(6): 1311-1334, 1994) predicts a high level of general lexical activity (GLA) for word-stems with many completions, indicating a higher stimulus familiarity because these stems are more probable to be read. We compared word-stems with several completions against word-stems with one possible completion while measuring response times and electrophysiological recordings. Slowest response times and a distinct FN400 component, which has previously been related to the concept of familiarity (Curran in Memory Cogn 28(6):923-938, 2000), were apparent for word-stems that could only be associated with a single response. These findings support the claims of the MROM. Furthermore, the lack of the N2-component for word-stems with several completions continues to challenge the EEG-extension of the CMT (Yeung et al. in Psychol Rev 111(4):2004).
Neuroscience Letters, 2009
Recent neurocognitive studies of visual word recognition provide information about neuronal netwo... more Recent neurocognitive studies of visual word recognition provide information about neuronal networks correlated with processes involved in lexical access and their time course (eg,[Holcomb, Ph. J., Grainger J. and O'Rourke, T.(2002). An Electrophysiological Study of the Effects of Orthographic Neighborhood Size on Printed Word Perception, J. of Cogn. Neurosci. 14 938–950; Binder, JR, McKiernan, KA, Parsons, ME, Westbury, CF, Possing, ET, Kaufman, JN and Buchanan, L.(2003).
Behavior Research Methods, 2007
Neuroscience Letters, 2006
NeuroReport, 2006
A recent eye-tracking study reported a reverse effect of a noun&a... more A recent eye-tracking study reported a reverse effect of a noun's lexical frequency in the context of the resolution of coreferring pronouns. Investigating the neurophysiological basis of this effect, the present electroencephalographic study found differential patterns in theta activation when participants read pronouns referring to nouns of different frequency classes. Evoked theta power after pronoun onset increased with the frequency of the critical noun. This finding suggests differential load on memory resources depending on the nouns' frequency. Elevated attention promoting memory encoding for low-frequency words is assumed to facilitate the resolution of pronouns. Location of sources of differential theta activity in the parahippocampal region is accounted for by its role in an association network that mediates memory processes.
Neuroreport, 2006
A recent eye tracking study reported a reverse effect of a nouns lexical frequency in the context... more A recent eye tracking study reported a reverse effect of a nouns lexical frequency in the context of the resolution of coreferring pronouns. Investigating the neurophysiological basis of this effect, the present electroencephalogram study found differential pat- terns in theta activation when participants read pronouns refer- ring to nouns of different frequency classes. Evoked theta power after pronoun onset increased
Previous neurocognitive approaches to word predictability from sentence context in electroencepha... more Previous neurocognitive approaches to word predictability from sentence context in electroencephalographic (EEG) and eye movement (EM) data relied on cloze completion probability (CCP) data effortly collected from up to 100 human participants. Here we test whether two well-established techniques in computational linguistics can predict these data. Together with baseline pre- dictors of word position and frequency, we found that n-gram language models but not topic models provide an approach to EEG and EM data that is not significantly inferior to the CCP-based predictability data. This is the case for the three corpora we used. Most strikingly, our models accounted for about half of the variance of the CCP-based predictability estimates, thus suggesting that it provides a computational framework to explain the predictability of a word from sentence context. This can help to generalize neurocognitive models to all possible novel word combinations.
Frontiers in Neuroscience, 2014
ABSTRACT Roediger and McDermott (1995) showed that words are falsely recognized, when they provid... more ABSTRACT Roediger and McDermott (1995) showed that words are falsely recognized, when they provide many associated items in the stimulus set. The Associative Read-Out Model (AROM) is the first interactive model with an implemented semantic layer (Hofmann et al., 2011). It suggests that two words are associated, when they co-occur significantly more often together in sentences than predictable by chance (see Hofmann et al., 2011). We conducted a recognition memory task, in which 80 words were presented in a study phase. These studied words had to be discriminated from 80 non-studied (N=33). We crossed this Oldness factor with the number of associated words in a 2x2 rmANOVA design. High co-occurrence words had at least 8 associated words in the stimulus set, and low co-occurrence words less than 8. We not only replicated that high co-occurrence words increased the “yes” response in studied and non-studied words (Hofmann & Jacobs, in press), but we also found a significant interaction of Oldness and Co-occurrence in the reaction times, showing slowest reaction times for high co-occurrence non-studied words and fastest for high co-occurrence studied words. The Event-related potentials recorded from 32-channels during the test phase showed significant Oldness and Co-occurrence main effects for the N400 amplitude. The N400 was smaller for high co-occurrence and studied words and larger for low co-occurrence and non-studied words, which was predicted by the neurobiologically plausible AROM (Hofmann & Jacobs, in press).
Frontiers in Psychology, 2015
Physics of Life Reviews, 2015
Neuroscience and biobehavioral reviews, 2014
Interactive activation and competition models (IAMs) cannot only account for behavioral data from... more Interactive activation and competition models (IAMs) cannot only account for behavioral data from implicit memory tasks, but also for brain data. We start by a discussion of standards for developing and evaluating cognitive models, followed by example demonstrations. In doing so, we relate IAM representations to word length, sequence, frequency, repetition, and orthographic neighborhood effects in behavioral, electrophysiological, and neuroimaging studies along the ventral visual stream. We then examine to what extent lexical competition can account for anterior cingulate cortex (ACC) activation and the N2/N400 complex. The subsequent section presents the Associative Read-Out Model (AROM), which extends the scope of IAMs by introducing explicit memory and semantic representations. Thereby, it can account for false memories, and familiarity and recollection - explaining why memory signal variances are greater for studied than non-studied items. Since the AROM captures associative spr...
Abstract The current study investigated the role played by conflict monitoring in a lexical-decis... more Abstract The current study investigated the role played by conflict monitoring in a lexical-decision task involving competing word representations, using event-related potentials. We extended the multiple read-out model (Grainger and Jacobs, 1996), a connectionist model of word recognition, to quantify conflict by means of Hopfield Energy, which is defined as the sum of the products of all orthographic word node pair activations within the artificial mental lexicon of this model.
Psychological Research Psychologische Forschung, 2009
The conflict monitoring theory (CMT) assumes that word-stems associated with several completions ... more The conflict monitoring theory (CMT) assumes that word-stems associated with several completions should lead to crosstalk and conflict due to underdetermined responding situation (Botvinick et al. in Psychol Rev 108(3):624-652, 2001). In contrast, the Multiple-Read-Out-Model (MROM) of Jacobs and Grainger (J Exp Psychol 20(6): 1311-1334, 1994) predicts a high level of general lexical activity (GLA) for word-stems with many completions, indicating a higher stimulus familiarity because these stems are more probable to be read. We compared word-stems with several completions against word-stems with one possible completion while measuring response times and electrophysiological recordings. Slowest response times and a distinct FN400 component, which has previously been related to the concept of familiarity (Curran in Memory Cogn 28(6):923-938, 2000), were apparent for word-stems that could only be associated with a single response. These findings support the claims of the MROM. Furthermore, the lack of the N2-component for word-stems with several completions continues to challenge the EEG-extension of the CMT (Yeung et al. in Psychol Rev 111(4):2004).
Neuroscience Letters, 2009
Recent neurocognitive studies of visual word recognition provide information about neuronal netwo... more Recent neurocognitive studies of visual word recognition provide information about neuronal networks correlated with processes involved in lexical access and their time course (eg,[Holcomb, Ph. J., Grainger J. and O'Rourke, T.(2002). An Electrophysiological Study of the Effects of Orthographic Neighborhood Size on Printed Word Perception, J. of Cogn. Neurosci. 14 938–950; Binder, JR, McKiernan, KA, Parsons, ME, Westbury, CF, Possing, ET, Kaufman, JN and Buchanan, L.(2003).
Behavior Research Methods, 2007