doi:10.1016/j.neuroimage.2006.06.057
Copyright © 2006 Elsevier Inc. All rights reserved.
How metaphors influence semantic relatedness judgments: The role of the right frontal cortex
Argyris K. Stringarisa,
, 1,
, Nicholas Medforda, 1, Rachel Giorac, Vincent C. Giampietrob, Michael J. Brammerb and Anthony S. Davida
aSection of Cognitive Neuropsychiatry, Institute of Psychiatry at the Maudsley, King’s College London, PO 68 Denmark Hill, London SE5 8AF, UK
bBrain Image Analysis Unit, Institute of Psychiatry at the Maudsley, King’s College London, UK
cDepartment of Linguistics, Tel Aviv University, Israel
Received 4 February 2006;
revised 22 June 2006;
accepted 29 June 2006.
Available online 11 September 2006.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
We used event-related fMRI (ER-fMRI) to test the hypothesis that metaphors bias cognitive processing of semantic relatedness towards a search for a wider range of associations. Twelve right-handed male volunteers read a mixture of metaphoric and literal sentences, each sentence being followed by a single word, which could be semantically related or not to the preceding sentence context. We found that judging unrelated words as contextually irrelevant was associated with increased blood oxygenation level-dependent (BOLD) signal in the right ventrolateral prefrontal cortex in the metaphoric but not in the literal condition. The same region was also activated when subjects endorsed a semantic relation between words and metaphoric sentence primes but not between words and literal sentence primes. We argue that these results are consistent with the notion of semantic open-endedness, whereby figurative statements bias cognitive processing towards a search for a wider range of semantic relationships compared to literal statements, and thus lend further support to the view that coarse semantic coding occurs preferentially in the right hemisphere.
Fig. 1. Activations in the Right Frontal Cortex: (a) rejection of words preceded by a metaphors, compared to the literal condition (IRM > IRL); (b) rejection of a word following literal sentences (IRL > IRM) compared to the metaphoric condition; (c) endorsement of words preceded by metaphors, compared to the literal condition (RM > RL); (d) endorsement of words following literal sentences compared to the metaphoric condition (RL > RM). Note that the right side (R) of this picture corresponds to the viewer's left hand side. N = 12, activations observed at voxel p value < 0.05, cluster p value < 0.0025.
Table 1.

(a) Descriptive statistics for the reaction times (RT) of literal and metaphoric sentences followed by a semantically relevant word (RL/RM) and the respective t test results. Reaction times are presented in milliseconds (ms) and also following logarithmic transformation (log); the latter were used for the t statistics.
(b) Descriptive statistics for the reaction times (RT) of literal and metaphoric sentences followed by a semantically irrelevant word (IRL/IRM) and the respective t test results. Reaction times are presented in milliseconds (ms) and also following logarithmic transformation (log); the latter were used for the t statistics.
Table 2.
Activations obtained for contrasts; IRL/IRM denote non-relevant words following a literal or metaphoric sentence, respectively, whereas RL/RM stand for related word following a literal or metaphoric sentence, respectively

The p value, voxel size, Talairach coordinates (Tal), Brodmann areas, laterality (Side), and cerebral location are given for each activation. Activations listed here were obtained at a voxel level of p < 0.05 and cluster level of p < 0.0025.