Anjan Chatterjee & Eileen Cardillo – Why the right hemisphere’s role in metaphor processing might not be right

Why the right hemisphere’s role in metaphor processing might not be right 

Anjan Chatterjee & Eileen Cardillopenn_logo
The Center for Cognitive Neuroscience
The University of Pennsylvania

Are metaphors processed in the left or right hemisphere of the brain? Cardillo, Chatterjee and their colleagues have investigated the neural underpinnings of metaphors in order to find out.

Since Broca, Wernicke, and Lichtheim (Roth, 2014), the idea that the left hemisphere specializes in processing language is well accepted. However, in the late 1970s and 80s the right hemisphere began to be viewed as playing a unique role in language processing, specifically for figurative language. This view was bolstered by the hypothesis that the right hemisphere specializes in coarse coding of semantics (Beeman & Chiarello, 1998). This proposal offered a sensible account of the right hemisphere’s role in processing metaphors in so far as concepts represented by base and target terms that are only coarsely related become linked in metaphors. Here, we argue that the right hemisphere hypothesis for metaphor processing, despite being motivated by this attractive model, does not bear up to empirical scrutiny.

Initial investigations with brain-injured patients that supported the right hemisphere conjecture suffered from several methodological weaknesses (Schmidt, Kranjec, Cardillo & Chatterjee, 2009). These studies typically had a small numbers of test items, made gross anatomical claims, and most importantly included confounding variables in their stimuli or tasks. Subsequent neuroimaging studies did not clarify matters. Past imaging investigations often failed to consider non-figurative factors that can vary between metaphoric and literal conditions. These include lexical properties that might make metaphors more difficult to process, such as frequency, concreteness, and length. Similarly, other non-figurative features of sentences such as imageability, naturalness, valence, syntactic complexity, and ease of interpretation also likely alter neural demands in areas both within and beyond the classic language areas of the brain.

Beyond these methodological shortcomings, the early and influential studies did not incorporate theoretical advancements in our understanding of metaphor processing. Metaphors do not represent a unitary construct (Cardillo, Schmidt, Kranjec & Chatterjee, 2010; Chen, Widick & Chatterjee, 2008). Differences in the class of words used figuratively distinguish different types of metaphor. While most metaphor research has focused on figurative extensions of nouns, or nominal metaphors (“The stock is a rollercoaster”), other parts of speech are frequently used figuratively. For example, we extend verbs metaphorically in predicate metaphors (“The stock soared”), prepositions in locative metaphors (“The stock is down”), and adjectives in attributive metaphors (“the hot stock”). Studies and models typically do not address this variety of forms, nor the altered neural demands entailed by their different syntaxes and different semantic domains. Another critical factor is that metaphor processing changes over time. Novel metaphoric uses of words that were creative initially become familiar and unremarkable with continued use. As pointed out by Bowdle and Gentner (2005) in the “career of metaphor” hypothesis, a shift from novel to conventional use is accompanied by a shift in how metaphors are understood and presumably by a change in how they are instantiated neurally.

We suspect these methodological and conceptual issues have contributed to inconsistent support for the right hemisphere hypothesis. Ambiguity concerning metaphor’s neural substrates is further confused by an over-reliance on reverse inference to interpret the functional contributions of the many implicated brain areas. The richly specified cognitive models of metaphor were not developed with brains in mind, and do not neatly extend to the neural level. Our strategy to sort through this empirical muddle has been to establish well-normed sentential stimuli that are optimized for testing neuroscientific hypotheses. Armed with such stimuli, we conduct both imaging and patient studies – since each method has different inferential limitations (Chatterjee, 2005) – to gather converging evidence for the neural bases of metaphors processing.

We developed a large, flexible, extensively normed, and theoretically motivated set of metaphoric and literal sentences. Across two reports (Cardillo et al., 2010; Cardillo, Watson & Chatterjee, 2016), we published 400 metaphoric sentences and 400 matched literal sentences that are characterized along 10 relevant dimensions: length, frequency, concreteness, familiarity, naturalness, imageability, figurativeness, interpretability, valence, and an online measure of comprehension difficulty (valence judgment reaction time). The stimuli distinguish metaphors along three theoretically-motivated dimensions: the grammatical class of their base terms, the sensorimotor features of their base terms (visual, motion, and auditory imagery), and the syntactic form in which the base terms appear. In addition to controlling potentially confounding lexical and sentential factors, these stimuli can be used flexibly to address questions about the role of novelty, metaphor type, and sensory grounding in determining the neural basis of metaphor comprehension.

We investigated the neural career of metaphors (Cardillo, Watson, Schmidt, Ktanjec & Chatterjee, 2012) using these stimuli, simulating the gradual experience of metaphor conventionalization by manipulating participants’ exposure to novel metaphors. We found that metaphor processing primarily occurred within classic left hemisphere language areas and its conventionalization specifically tuned activity within the left inferior frontal gyrus and posterior middle temporal gyrus, as well as the right inferior gyrus and postero-lateral occipital cortex. These results run counter to the idea that the right hemisphere is dominant for metaphor processing, or even the more restricted proposal that novel metaphors are preferentially processed in the right hemisphere.

We also used these stimuli in a preliminary study of metaphor comprehension with three patients with focal brain damage (Ianni, Cardillo, McQuire & Chatterjee, 2014). After reading each sentence, patients selected its meaning from four adjective-noun answer choices (target + three foil types). Each patient demonstrated a distinct metaphor comprehension deficit, none of which were captured by traditional neuropsychological language assessments. Such deficits suggest an unknown prevalence of hidden communication difficulties not generally appreciated by clinicians. More relevant to this discussion, two of the three patients reported had left hemisphere damage.

We are in the midst of analyzing data from a larger group of patients with focal brain damage and in another group with the logopenic variant of primary progressive aphasia (focal degeneration typically within left peri-Sylvian cortex). Stay tuned for full details. However, in brief: results from both data sets confirm our suspicion that the right hemisphere’s putative privileged role in metaphor processing is not right.


Beeman, M., & Chiarello, C. (Eds.) (1998). Right hemipshere language comprehension: Perspectives from cognitive neuroscience. Mahwah, NJ: Lawrence Erlbaum Associates.

Bowdle, B. F., & Gentner, D. (2005). The career of metaphor. Psychological Review, 112, 193-216.

Cardillo, E. R., Schmidt, G. L., Kranjec, A., & Chatterjee, A. (2010). Stimulus design is an obstacle course: 560 matched literal and metaphorical sentences for testing neural hypotheses about metaphor. Behavior Research Methods, 42(3), 651-664.

Cardillo, E. R., Watson, C., & Chatterjee, A. (2016). Stimulus needs are a moving target: 240 additional matched literal and metaphorical sentences for testing neural hypotheses about metaphor. Behavior Research Methods, 1-13. doi:10.3758/s13428-016-0717-1

Cardillo, E. R., Watson, C. E., Schmidt, G. L., Ktanjec, A., & Chatterjee, A. (2012). From novel to familiar: Tuning the brain for metaphors. NeuroImage, 59, 3212-3221.

Chatterjee, A. (2005). A madness to the methods in cognitive neuroscience? Journal of Cognitive Neuroscience, 17(6), 847-849. doi:10.1162/0898929054021085

Chen, E., Widick, P., & Chatterjee, A. (2008). Functional-anatomical organization of predicate metaphor processing. Brain and Language, 107(3), 194-202.

Ianni, G. R., Cardillo, E. R., McQuire, M., & Chatterjee, A. (2014). Flying under the radar: Figurative language impairments in focal lesion patients. Frontiers in Human Neuroscience, 8, 871.

Roth, H. (2014). We stand on the shoulders of giants: The golden era of behavioral neurology 1860-1950 and its relevance to cognitive neuroscience today. In A. Chatterjee & H. Coslett (Eds.), The roots of cognitive neuroscience: Behavioral neurology and neuropsychiatry (pp. 11-52). New York: Oxford University Press.

Schmidt, G. L., Kranjec, A., Cardillo, E. R., & Chatterjee, A. (2009). Beyond laterality: A critical assessment of research on the neural basis of metaphor. Journal of the International Neuropsychological Society, 16, 1-5.