Language with Flavour

By: Elisa Bailey

Francesca Citron's current position is as Postdoctoral Research Fellow working at the Cluster of Excellence “Languages of Emotion” at the Free University of Berlin.

To start off, given that you are working at a high level with psycholinguistics, can you tell me a little bit about your own linguistic background?

My native language is Italian and I speak German and English fluently, because I studied these languages at school and then lived in Germany for six years overall and in the UK for 3 years during my PhD studies. I love learning foreign languages and about other cultures, so I took French at school between the ages of 11 to 14, then Spanish for a year and Portuguese for 3 years while at university. I used to be fluent in Portuguese too and immersed myself in the culture but now I feel like Germany is my second home country since I moved there for the first time when I was 19 and absorbed a lot from the culture. Most recently, I studied Modern Greek for a year because of my last relationship, but had previously taken classes in Ancient Greek at university, since I am very interested in how the origins of many Italian words are often unknown to many native Italian speakers. I find every language fascinating, and learning a language gives you a way to enter and explore a specific culture.

So now you have moved over to Linguistics, can you tell me a little bit about your current work, and do you still use your foreign language skills in this?

Yes, every day! In order to conduct my research I certainly benefit from my knowledge of German, since we test German participants and present them with German linguistic material. However, given the fact that research is international, one needs to be fluent in English and most scientific publications have to be written in this language, so in my everyday life I use both languages. I do experimental research on language comprehension and processing; for example, I study how quickly emotional words are understood compared to neutral words and which areas of the brain are activated in response to these stimuli. The idea behind it is that, no matter which specific language we investigate, the cognitive processes involved should be the same, i.e., emotional words should be more salient, capture attention and be understood more quickly than neutral words, in every language.

How are you applying this concept in your latest research?

In our latest study we concentrated on taste representations. We wanted to investigate whether abstract concepts are grounded in sensory-motor representations in a similar way in which concrete concepts are. For example, if one reads about a 'sweet cake', this will activate brain regions associated with language processing (phonological representations, lexical representations and so on) as well as brain regions associated with the perception of a sweet cake, i.e., gustatory areas as well as visual areas. But what happens in the case of abstract concepts? When we say “she is a sweet girl” we imply that that girl is kind (abstract concept), not that she tastes sweet (concrete concept). However, since metaphors are thought to be mappings between abstract and concrete concepts and thus help us in understanding abstract concepts, it may be the case that, even when we mean 'sweet' in a metaphorical, abstract way, we still get partial activation of the taste-related/ gustatory areas of the brain.

So how did you carry out these tests?

We tested this by having people silently reading metaphorical sentences as well as their literal counterparts, for example, “That was a bitter vs. sad break up for him”, while lying in the scanner. Our sentences were carefully matched for a range of psycholinguistic and affective features known to affect reading and were rated as highly similar in meaning. Thus any difference in their evoked brain activity could only be due to them being metaphors (or taste-related).

Did you get the expected results or were there a few surprises?

A mixture of both: our results did indeed show involvement of the primary and secondary gustatory cortices during comprehension of taste-related metaphors when compared to their literal counterparts, therefore supporting the idea that even abstract concepts are grounded in somatosensory representations.

In addition, we found something interesting and unexpected: our metaphorical sentences also activated the anterior portion of the left hippocampus and the amygdala. These regions have been associated with the processing of intense emotional stimuli. This led us to conclude that the metaphorical formulations we used may be more emotionally engaging than their corresponding literal formulations.

This is a very novel finding given the fact that the neuroscientific literature on figurative language did not take emotion into account! In addition, this finding is in line with some old behavioural studies showing that figurative expressions are more often used when formulating complaints or when describing one's own emotional reaction to an event compared with the description of what happened (Drew & Holt , 1988; Fainsilber & Ortony, 1987). Finally, our result is in line with a recent meta-analysis by Bohrn et al. (2012), which reviewed many neuroimaging studies on figurative language and showed involvement of the left amygdala when contrasting figurative and literal expressions.

As smell is related to taste, would a metaphor of a smell described as a 'perfume' or, negatively, a 'stench' have the same effect on the brain as metaphors referring to taste?

Yes, such words would also activate primary and secondary cortices, but the ones related to smell (olfactory regions), not the ones related to taste. Taste words such as “cinnamon” may activate olfactory areas too, so it would be interesting to specifically investigate whether taste-words with a strong olfactory component do affect smell-related areas more strongly than taste-words not related to smell, e.g., “sugar”.

As I know this is your next research area, do you believe there are also relationships between metaphors regarding touch and hearing and activations of areas of the brain, and if so, what can this lead us to theorise?

Other work on the embodied grounding of metaphors has shown activation of motor regions in response to action idioms, e.g., “to kick the bucket”; “to grasp the idea” (Boulenger et al., 2009), and activation of texture-related regions in response to metaphors such as “she had a rough day”; “he is a slimy person” (Lacey et al., 2012). These few studies, together with our latest study, all support an embodied view of conceptual representation, according to which even abstract concepts are represented by means of more basic, sensory-motor and perceptual processes.

Now turning more specifically to the nature of these sensory-related metaphors, are you able to recognise a difference in the effect on the brain between a metaphor with a negative implication (such as “being shot through the heart with an arrow”, or being “broken”) and those with positive implications?

Good question. We know that the emotional content of verbal material, such as emotional words in isolation as well as emotionally-laden texts, do elicit the activation of brain regions associated with emotion processing (Citron, 2012; Ferstl et al., 2005). However, this work has been conducted with literal language, and there is very little work on emotional figurative language. Preliminary results from a recent study conducted in our lab with positive, negative and neutral idioms, for example, “she was in seventh heaven”; “he blew his stack”; “it came to my mind”, show that our brain is quite good at differentiating idiomatic expressions with different emotional valence (positive vs. negative vs. neutral), but not quite as good at differentiating these emotional categories in the case of literal sentences.

In this regard then, is there a similar effect between somebody using plain (literal) language with certain words spoken emphatically/ dramatically or written so that they stand out, and the use of emotional metaphors? I would suggest the example of the financial crisis “killing” jobs versus it “CAUSING PEOPLE TO LOSE” their jobs (capitalisation to suggest the speaker's emphasis or the words written as such).

Of course any means we use to make a message more salient or emotional will evoke stronger emotional reactions in the reader/ listener. So, the use of capital letters as well as the way we express things when we speak (angry vs. excited vs. calm voice) will affect comprehension. Figurative language is only one way in which we can render a message more salient.

The natural next question in my mind, then, would be regarding how figurative vocabulary and emphatic delivery of choice words can be used in advertising and political speeches, for example. What are the implications from the findings of your research?

Well, our results, together with previous older studies, may imply that, when we want to communicate more effectively or persuade our interlocutor, using figurative or formulaic expressions may help. On the other hand, we should be wary of being influenced by figurative language.

Are all the effects on the brain we have discussed still the same for the speaker/ listener if they are not of the same mother tongue as the language the metaphor has been spoken in? For example, will the brain of a non-mother tongue listener still engage emotionally with a sensory metaphor, or will the neutrality of simply having learnt and repeated or recognised the word mean it is no different for them to hearing a non-metaphorical alternative? In other words, is an emotional brain reaction something innate that requires an immersion since birth in the language the metaphor has been spoken in, or can it be 'acquired' by someone who has attained a high level of fluency and cultural understanding of the language?

These are all interesting questions that we aim to address in our future research on second language (L2) speakers. Even highly proficient L2 speakers struggle with figurative expressions and therefore we would expect them to show less of an emotional response while reading figurative expressions in their L2 compared to native speakers. However, we also aim to investigate different ways of teaching L2 speakers figurative expressions and then test whether successful learning will lead to the following: better comprehension, more effective communication, better integration in their group of peers and stronger motivation to further learn.

Do you believe it is necessary to understand and be affected by figurative expressions in order to be 'fluent' in a foreign language, or does fluency entail something else for you?

It means being confident about speaking and understanding that language as well as being able to communicate effectively.