Serendip is an independent site partnering with faculty at multiple colleges and universities around the world. Happy exploring!

Exploring the “Links” in the Brain that Give Rise to Synaesthesia

Bo-Rin Kim's picture

I first learned about synaesthesia in my cognition class last fall. I was fascinated yet perplexed by the idea that some people are able to see colors when they look at letters or taste foods when they hear certain sounds. I did not understand how a perceptual stimulus could evoke a response from a sensory mechanism different from its own. I soon learned that this effect arose due to “links” in the brain that connect the regions responsible for different senses. However, what exactly are these “links” in the brain and how do they give rise to synaesthesia?

Synaesthesia is defined as a condition in which the perception of a sensory stimulus involuntarily and consistently elicits a secondary response in another sensory modality (6). There are many different types of synaesthesia, involving different sensory stimuli and responses. Although the most common type is colors that are induced by letters, numbers or words, other types such as word-induced taste, vision-induced touch, and taste-induced shapes also exist (2). This cross-activation of senses has been observed to be unidirectional in that sensory stimuli of one modality will trigger a response in another sense but not vise versa (1).

There are two main theories that explain how this sensory cross-activation occurs in the brain. One theory posits that synaesthetic brains have extra neural wiring that connects the sensory areas affected by synaesthesia. This structural link allows a stimulus to be relayed to both sensory areas when it would only activate one area in a non-synaesthetic brain.  The other theory emphasizes functional, rather than structural, differences in the brains of synaesthetes. It posits that two regions of the synaesthetic brain that are constantly activated together are functionally linked although they are not physically connected (1).

Several researchers support the theory that synaesthesia is a result of neural connections between sensory regions of the brain. In their analyses of color-grapheme synaesthetes, Ramachandran and Hubbard found that while numerical graphemes induced color, Roman numerals did not. This suggests that perceptual, rather than the conceptual, stimuli induce colors in synaesthetes. From this, they proposed that the color-grapheme synaesthetic effect is due to the cross-wiring of the grapheme identification region (area V4) and the color identification region, which are both located in the fusiform gyrus. They identified the fusiform gyrus as the location of cross-wiring because it is involved in the simple perceptual processing of input before they are relayed to higher areas that are involved in more sophisticated processing (5).

Rouw and Scholte (2007) confirmed this hypothesis of Ramachandran and Hubbard by using diffusion tensor imaging (DFI) on synaesthetic brains. The method uses magnetic resonance imaging (MRI) to track the movement of water molecules, which take on a certain conformation (anisotopic) when exposed to white matter. They found several concentrated areas of anisotopic water molecules, including the area surrounding the V4 grapheme area. This shows that there is more white matter in this area, suggesting that there is greater structural connectivity in this region. Since the V4 area and the “color” area are located close together in the fusiform gyrus, this finding also suggests that these areas are structurally connected (1).

Rauschecker (1995) also experimentally proved that linkages between different sensory areas could occur in the mammalian brain. He found that when cats are deprived of visual input, the visual areas of their cortex begin to respond to auditory and somasensory stimuli. These findings show what it means, in terms of neural activation, when synaesthetes “see” sounds (4).

However, the popular explanation for how cross-wiring happens in humans is not cortical plasticity or rewiring, but abnormal neural pruning during development. The Neonatal Synaesthesia hypothesis states that all newborns, up to 4 months, process sensory input using all of their senses. Thus, an auditory stimulus would trigger tactile, visual, olfactory and gustatory responses. After 4 months, selective pruning of neurons that connect all these sensory regions create distinct specialized sensory processing areas. Synaesthetic cross-wiring comes about when this pruning does not properly take place, leaving some sensory regions physically connected (3).

While these physical connections that arise from cross-wiring enable one sensory modality to stimulate another one, it does not determine the exact sensory associations that synaesthetes make. That is, these structural links do not determine that “P” will induce the color pink or that “Virginia” will induce the taste of vinegar (6,8). These specific associations must be learned. Rich, Bradshaw and Mattingley (2004) found that out of the 13 color-letter matches found amongst synaesthetes, non-synaesthetic controls also displayed 11 of these matches when tested. However, synaesthetes had a significantly higher consistency rate (87%) when tested several months later compared to non-synaethetes (26%) (6). The higher consistency of synaesthetes suggests that their neural connectivity makes it easier for them to consistently make the same color-letter associations. However, synaesthetes and non-synaesthetes must share common learning experiences that motivate them to make similar color-letter associations.

Supporters of the functional, rather than structural, linkage theory emphasize that learning causes brain regions activate together and become functionally linked. Rich, et al. (2005) found that synaesthetes who experienced colors for only a small set of lexical stimuli tended to have this response to sequential words, such as days of the week, letters and numbers, compared to other word categories. For these non-sequential words, synaesthetes reported that the color of the initial letter in the word generalized to the entire word. However, the colors they saw for sequential words did not match the colors of their initial letters. This suggests that for these sequences, which are learned early in life, colors come to be associated with whole words. Interestingly, the colors of the months of the year, which are usually learned after days of the week, tend to be determined by the initial letters of the words (6). This indicates that after a certain age, children start breaking down sequential words into their individual letters rather than processing them as whole units. Thus, the way in which words are learned, in sequence or not and at what age, determines how they become associated with colors. While it is unknown why certain colors become associated with certain letters or sequential words, I speculate that cultural or emotional meanings attached to colors affect how they are linked to lexical stimuli. This functional linkage of lexical and color processing brought about by learning offers a more comprehensive explanation for sensory cross-activation than simple structural connectivity.

Rich, et al. (2005) also found that the meanings of lexical stimuli triggered colors such that for most synaesthetes, color words induce the denoted color rather than the color induced by the initial letter. Moreover, number words induce colors consistent with the matching digit grapheme (6). These associations are conceptual, and therefore must result from learning rather than neural cross-wiring that one is born with.

In their studies of lexical-gustatory synaesthesia, Ward and Simner (2003) also found semantic associations between the many of the word stimuli and the experienced taste. For example, “blue” would trigger an inky taste. They also observed that phonemes that are a part of food words tend to trigger the taste of that food even when they are part of non-food words. For example, “made” would induce the taste of marmalade and “Virginia” would induce the taste of vinegar. These findings suggest that knowledge of language elements, such as phonemes, and meanings of words are needed to induce this type of synaesthesia. These are concepts that must be learned and are specific to each language, which indicate that innate neural connections cannot be responsible for these lexical-gustatory associations. Moreover, a physical connection between the phonological and taste centers of the brain cannot account for this phenomenon as the phonological center would not be able to distinguish which phonemes come from food words, which are the specific phonemes that trigger tastes (8).

Despite these different explanations for how sensory areas of the brain become linked, supporters of both structural and functional connectivity agree that these connections are inheritable. In one sample of synaesthetes, 42% reported having a first-degree relative who also had the condition (2). Areas on chromosomes 2, 5, 6 and 12 have actually been identified to relate to neuronal activity and brain structuring, suggesting to researchers that these may be the genetic areas potentially responsible for synaesthesia (7).

Supporters of structural connectivity argue that genetics affect how neural pruning and therefore hard-wire the brain to be synaesthetic or not (1). Supporters of functional connectivity argue that people inherit the tendency to link two senses together and to over-activate the corresponding brain regions. For example, imagining tastes normally activates the cortical taste area. However, in synaesthetes, non-food words that share phonemes or semantics with food words bring about this activation of taste. Moreover, the taste area is hyper-activated, resulting in not just an awareness of the taste, but an actual gustatory experience (8).

So, it seems that the common cause of synaesthesia is a genetically-determined over-connected brain structure or an inherited tendency to easily link two senses and to over-activate the corresponding brain areas to the point of actually experiencing a sense. There is not enough evidence to conclude that one theory is correct. The Neonatal Synaesthsia hypothesis and the abnormal pruning explanation for structural over-connectivity cannot be proven unless brain structure analyses are done on newborns of synaesthetic families, who are likely to have the condition, before they can learn any associations between letters and colors or words and tastes. If their neural wiring is different from other newborn babies, then the structural connectivity hypothesis can be proven, indicating that synaesthetes inherit an over-connected brain. Similarly, the functional connectivity theory can only be proven if evidence is found that synaesthesia involving “higher” processing such as semantics and meaning does not result in neural rewiring. While structural connectivity has been found for perceptual color-grapheme synaesthesia, this type of analysis has not been done with other forms of the condition that deal with higher levels of processing. As it has not been proven that structural connectivity exists in these higher forms of synaesthesia, the functional connectivity theory stands. Until more evidence is found, the question of what exactly synaesthetes inherit will remain. Do they inherit a type of brain structuring or do they inherit the tendency to cross-activate their sensory areas?

My hypothesis to these questions is that genetics determine a general predisposition to synaesthesia, such that people inherit greater neuroplasticity and the tendency to cross-activate senses. Thus, synaesthetes are not born with pre-wired brains nor do they share the same brain structure as everyone else, but their neural wiring changes to match the functional links they repeatedly make between different senses. This also accounts for how synaesthesia is genetic, yet different forms of synaesthesia are found within families (2). This is occurs because while the general predisposition to be synaesthetic is inherited, the sensory associations that the individual learns and rewires into his or her brain determines the actual manifestation of the condition.


1. Bargary, G., & Mitchell, K. J. (2008). Synaesthesia and cortical connectivity. Trends in
Neuroscience, 31(7), 335-342.

2. Barnett, K. J., Finucane, C., Asher, J. E., Bargary, G., Corvin, A. P., Newell, F. N., et al.
(2008). Familial patterns and the origins of individual differences in synaesthesia.
Cognition, 106, 871-893.

3. Baron-Cohen, Simon. "Is There a Normal Phase of Synaesthesia in Development?" Psyche
2.27 (1996). 14 Apr. 2009 <>.

4. Costa, Luciano Da F. "Synaesthesia: A Real Phenomenon? Or Real Phenomena?" Psyche 1.10
(1996). 14 Apr. 2009  <>.

5. Ramachandran, Vilayanur S., and Edward M. Hubbard. "Hearing Colors, Tasting Shapes."
Free Republic. 14 Apr. 2003. 14 Apr. 2009

6. Rich, A., Bradshaw, J., & Mattingley, J. (2005). A systematic, large-scale study of
synaesthesia: Implications for the role of early experience in lexical-color association.
Cognition, 98, 53-84.

7. Robson, David. "Genetic Roots of Synaesthesia Unearthed." New Scientist Feb. 2009. 14 Apr.
2009 <>.

8. Ward, J., & Simner, J. (2003). Lexical-gustatory synaesthesia: Linguistic and conceptual
factors. Cognition, 89, 237-261.


Paul Grobstein's picture

approaches to synesthesia

We all tend to start with the idea that "senses" are actually separate, and hence that people who experience mixes of them are different in ways that bring them together in particular neural structures.  But there is another possible starting place, that senses aren't in fact as distinct as we are taught they are (and many people come to experience them), cf  From Five Senses to Synesthesia and Beyond.  Perhaps that approach could make better sense of existing observations and suggest new ones?