Quantitative modeling of the neural representation of adjective-noun phrases to account for fMRI activation
Chang, Kai-min K. and Cherkassky, Vladimir L. and Mitchell, Tom M. and Just, Marcel Adam

Article Structure

Abstract

Recent advances in functional Magnetic Resonance Imaging (fMRI) offer a significant new approach to studying semantic representations in humans by making it possible to directly observe brain activity while people comprehend words and sentences.

Introduction

How humans represent meanings of individual words and how lexical semantic knowledge is combined to form complex concepts are issues fundamental to the study of human knowledge.

Brain Imaging Experiments on Adj ec-tive-Noun Comprehension

2.1 Experimental Paradigm

Topics

semantic representation

Appears in 12 sentences as: Semantic Representation (1) semantic representation (7) semantic representations (4)
In Quantitative modeling of the neural representation of adjective-noun phrases to account for fMRI activation
  1. Recent advances in functional Magnetic Resonance Imaging (fMRI) offer a significant new approach to studying semantic representations in humans by making it possible to directly observe brain activity while people comprehend words and sentences.
    Page 1, “Abstract”
  2. There have been a variety of approaches from different scientific communities trying to characterize semantic representations .
    Page 1, “Introduction”
  3. Recent advances in functional Magnetic Resonance Imaging (fMRI) provide a significant new approach to studying semantic representations in humans by making it possible to directly observe brain activity while people comprehend words and sentences.
    Page 1, “Introduction”
  4. Given these early succesess in using fMRI to discriminate categorial information and to model lexical semantic representations of individual words, it is interesting to ask whether a similar approach can be used to study the representation of adjective-noun phrases.
    Page 2, “Introduction”
  5. 4 Using vector-based models of semantic representation to account for the systematic variances in neural activity
    Page 4, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  6. 4.1 Lexical Semantic Representation
    Page 4, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  7. Table 3 shows the semantic representation for strong and dog.
    Page 5, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  8. The lexical semantic representation for strong and dog.
    Page 5, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  9. Table 4 shows the semantic representation for strong dog under each of the four models.
    Page 5, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  10. The semantic representation for strong dog under the adjective, noun, additive, and multiplicative models.
    Page 5, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  11. An alternative approach is to model the semantic representation as a hidden variable using a generative probabilistic model that describes how neural activity is generated from some latent seman-
    Page 8, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”

See all papers in Proc. ACL 2009 that mention semantic representation.

See all papers in Proc. ACL that mention semantic representation.

Back to top.

regression model

Appears in 8 sentences as: regression model (9)
In Quantitative modeling of the neural representation of adjective-noun phrases to account for fMRI activation
  1. In this analysis, we train a regression model to fit the activation profile for the 12 phrase stimuli.
    Page 6, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  2. The regression model examined to what extent the semantic feature vectors (explanatory variables) can account for the variation in neural activity (response variable) across the 12 stimuli.
    Page 6, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  3. All explanatory variables were entered into the regression model simultaneously.
    Page 6, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  4. A regression model was trained for each of the 120 voxels and the reported R2 is the average across the 120 voxels.
    Page 6, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  5. (2008), the regression model can be used to decode mental states.
    Page 7, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  6. Specifically, for each regression model , the estimated regression weights can be used to generate the predicted activity for each word.
    Page 7, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  7. (2008), where the regression model was used to make predictions for items outside the training set, here we are just showing that the regression model can be used for classification purposes.
    Page 7, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  8. This affects the parameter-to-data-points ratio in our regression model .
    Page 7, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”

See all papers in Proc. ACL 2009 that mention regression model.

See all papers in Proc. ACL that mention regression model.

Back to top.

lexical semantic

Appears in 5 sentences as: Lexical Semantic (1) lexical semantic (4)
In Quantitative modeling of the neural representation of adjective-noun phrases to account for fMRI activation
  1. How humans represent meanings of individual words and how lexical semantic knowledge is combined to form complex concepts are issues fundamental to the study of human knowledge.
    Page 1, “Introduction”
  2. Given these early succesess in using fMRI to discriminate categorial information and to model lexical semantic representations of individual words, it is interesting to ask whether a similar approach can be used to study the representation of adjective-noun phrases.
    Page 2, “Introduction”
  3. In section 4, we discuss a vector-based approach to modeling the lexical semantic knowledge using word occurrence measures in a text corpus.
    Page 2, “Introduction”
  4. 4.1 Lexical Semantic Representation
    Page 4, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  5. The lexical semantic representation for strong and dog.
    Page 5, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”

See all papers in Proc. ACL 2009 that mention lexical semantic.

See all papers in Proc. ACL that mention lexical semantic.

Back to top.

statistically significant

Appears in 4 sentences as: statistically significant (3) statistically significantly (1)
In Quantitative modeling of the neural representation of adjective-noun phrases to account for fMRI activation
  1. They compared the composition models to human similarity ratings and found that all models were statistically significantly correlated with human judgements.
    Page 2, “Introduction”
  2. However, the difference between the multiplicative model and the noun model is not statistically significant in this case.
    Page 7, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  3. The difference is statistically significant at p < 0.05.
    Page 7, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  4. Although neither difference is statistically significant , this clearly shows a pattern different from the attribute-specifying adjectives.
    Page 7, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”

See all papers in Proc. ACL 2009 that mention statistically significant.

See all papers in Proc. ACL that mention statistically significant.

Back to top.

feature vector

Appears in 3 sentences as: feature vector (2) feature vectors (1)
In Quantitative modeling of the neural representation of adjective-noun phrases to account for fMRI activation
  1. The regression model examined to what extent the semantic feature vectors (explanatory variables) can account for the variation in neural activity (response variable) across the 12 stimuli.
    Page 6, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  2. Table 5 also supports our hypothesis that the multiplicative model should outperform the additive model, based on the assumption that adjectives are used to emphasize particular semantic features that will already be represented in the semantic feature vector of the noun.
    Page 6, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  3. We are currently exploring the infinite latent semantic feature model (ILFM; Griffiths & Ghahramani, 2005), which assumes a nonparametric Indian Buffet prior to the binary feature vector and models neural activation with a linear Gaussian model.
    Page 8, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”

See all papers in Proc. ACL 2009 that mention feature vector.

See all papers in Proc. ACL that mention feature vector.

Back to top.

latent semantic

Appears in 3 sentences as: latent semantic (3)
In Quantitative modeling of the neural representation of adjective-noun phrases to account for fMRI activation
  1. There are also efforts to recover the latent semantic structure from text corpora using techniques such as LSA (Landauer & Dumais, 1997) and topic models (Blei et al., 2003).
    Page 1, “Introduction”
  2. We are currently exploring the infinite latent semantic feature model (ILFM; Griffiths & Ghahramani, 2005), which assumes a nonparametric Indian Buffet prior to the binary feature vector and models neural activation with a linear Gaussian model.
    Page 8, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”
  3. We are investigating if the compositional models also operate in the learned latent semantic space.
    Page 8, “Brain Imaging Experiments on Adj ec-tive-Noun Comprehension”

See all papers in Proc. ACL 2009 that mention latent semantic.

See all papers in Proc. ACL that mention latent semantic.

Back to top.

PDF available from www.aclweb.org.
Send us an email