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

Trusting Vision: Language’s Influence on Perception

Jessica Krueger's picture

    We, as humans, have long been enamored of our perceptive abilities, particularly our spectacular color vision. Heavily worn adages such as “I’ll know it when I see it,” and “Seeing is believing,” allude to our trust that our perception to delivers something of a reliable truth about the world around us; the idea being that something which is “out-there” bounces light into our eyes, which in turn “see” that external thing in an objective manner and that “seeing” can be validated by appealing to another set of eyes, confirming our own perception verbally. Unfortunately, perception is not so linear and the “reality” within which we operate on a daily basis may in fact not only be more subjective than previously assumed but also subject to influence from those in whom we seek confirmation. To investigate this idea further we will investigate a once forgotten but now very popular “disorder” of perception, synesthesia, and examine its neurophysiological substrates. We will then turn our attention to the influence of language on perception and how what we are told and what concepts are available to us can actually drive perception.

While it is commonly accepted that there are many ways for one to “see” things, this acceptance is usually limited to an acknowledgment of different viewpoints rather than wholly other ways of seeing. Most of us are comfortable with the fact that what I see has a different vantage point from what you see, but we seem to assume that this process is similar to aiming multiple cameras at one object and comparing the outcomes of identical processes. Synesthesia, a form of perception characterized by additional sensations loaded onto an independent sensory modality, gives serious reason to reevaluate this “camera” view of vision and seeing. While presumably any combination of sensory modalities is feasible, the most commonly reported (or at least the most frequently investigated) is color-grapheme synesthesia: letters and words are reliably associated with colors (1). Color-grapheme synesthetes view the same letters most humans do and can discuss the “true” color of those figures, but they see more, indicating camera-like homogeneous perception within the species is not to be taken for granted. Richard Cytowic, a neurologist who reintroduced synesthesia to inquiry after a hundred year dormancy, argues that although adult prevalence is low synesthesia is the mammalian state of perception and that adult synesthetes experience the world in a vestigial manner (2). Research specifically into the perceptual nature of other mammals remains lacking, though developmental studies in humans seem to indicate that neonates are in some manner synesthetic. Babies will attend to objects they’d explored orally more, even when they had never seen them before. Temporal lobe activity generalizes to the occipital cortex when babies hear speech, and neuroanatomical investigations in other neonate mammals has shown temporary connections between sensory cortices (3). Neonatal brains express a greater number of neurons, which are pruned around the third or fourth month in human babies (2). If these connections were left as they were, conceivably the neonatal synesthesia would progress on into adulthood, and fairly recent techniques allowing researchers to investigate the hitherto indiscernible white matter confirms that synesthetes show a higher degree of cortical interconnectivity. In particular color-grapheme synesthetes showed greater connectivity between regions of the brain responsible for the perception and categorization of graphemes and color perception (1). Furthermore, there appears to be a great deal of variety within synesthetes about how and what they perceive; projectors actually see their synesthetic colors projected out in front of them, whereas associators just “know” or see the color in their mind’s eye (4). The same study which found greater connectivity in synesthetes also found that the degree of connectivity had some correlation to the synesthetic experience, with higher connectivity resulting in color perception more analogous to that of ‘real’ perception (1). Synesthesia is a form of perception which results from an anomalous architecture within the brain, indicating that our experience of reality is mediated by, if not dependent upon, how the brain is constructed. The idea that what we all see is the same consequently must be rejected, as two individuals could draw very different conclusions about the “objective reality” which they are observing based on their own internal experience.

Differential neural architecture is not the only confound placing the experience of an objective reality in question; social influences can have profound effects on how and what we see, and arguably the most powerful of these is language. True, other animals and even mammals communicate in highly specific and interesting ways (indeed their vocalizations provide the much needed foundation from which language could evolve), they for the most part do not do so in a linguistic manner. The human brain has dedicated regions to the acquisition and production of language not found in other creatures, such as Wernicke’s and Broca’s area, and damage to these regions impair specific aspects of language, comprehension and production respectively. Meylination, a process whereby neuronal communication is facilitated, begins in the motor cortices, progresses to the limbic system and ends in the association cortices, predicts the slow progression of language acquisition with basic comprehension preceding production (5). Thus, any effects language might have on perception will be specific to human perception. Language itself is a complex and loaded concept, and sadly brevity precludes a full exploration of the individual aspects of language, from phoneme structure to syntactical organization, and their influences on perception. Instead, “language” in this discussion will serve as an umbrella term for the vocal, social communication humans engage in.

In one of the first ever documented etiologies of synesthetic colors associations, it was shown that synesthete AED’s letter-color associations were largely determined by a colored alphabet refrigerator magnets. This is not to say that the colorful magnets caused her synesthesia, but rather that her special form of perception was sensitive to the implied linguistic demand present by the colored letters. Other synesthetes have reported semantic influence on their perceptive experiences, such as tasting cake when hearing words with a similar hard “k” sound (6). Descriptions of synesthetic experiences have alternately been proposed as the basis of metaphorical construction and as subject to the same “ semantic processes and fashioned by time and cultural elements, much like other metaphors” (7). So it would seem that synesthetic perception is in some way beholden to language, but does this effect generalize to the “normal” population? Debate about the influence of language on perception has been waged for centuries, beginning with the Sapir-Whorf hypothesis, also known as the sociohistorical hypothesis, which asserts that language determines how we perceive our world, rather than language describing hard and fast external realities (8). Research into group influences, by way of the social nature of language, supports a weaker form of this hypothesis, which recognizes a profound but perhaps not ultimate effect of language on perception. Muzafer Sherif showed in 1936 that individual observations of an “auto-kinetic” event, a dot of light being perceived as darting about a dark room, would conform to a group average and maintain over time (9). Further research has shown that if a minority is consistent in their asserted perception of blue as green, test subjects will at the very least conform their report of the color they see (10). It could, however, be argued that these participants were merely changing their answer in order to avoid appearing incorrect; indeed synesthetes often do not advertise their altered perception for fear of or having been labeled “crazy” (1). A recent study, however, has shown that not only can people learn new color categories, but that training can effectively wipe out old color distinctions, contradicting the assumption that color categories reflect some absolute, quantifiable distinction between color (11). It has also been shown that color perception, particularly in ambiguous cases (when differences not well articulated by an individual’s socioculural history), activates brain regions responsible for language processes such as word search (12). In her account of autistic consciousness, Temple Grandin explained that individuals with autism do not think in terms of language, but rather in a serious of mental images. What makes this argument particularly interesting is another characteristic of autism referred to as sensory overload.

People with very severe sensory processing problems loose their body boundary when they become             overloaded with too much sensory stimulation.... They cannot attend to, or integrate both auditory             stimulus and visual stimulus at the same time. It appears that the brain processes information in a             compartmentalized manner. Seeing words, hearing words, thinking about a word and speaking a word         activate different brain regions....[and a person with autism] looses some of the ability to extract                 meaning from sensory input. (Grandin, 1998)

It would seem, then, that without the ability to break up reality along socially circumscribed lines, people with autism experience something akin to unmitigated synesthesia; Grandin further relates an anecdote from a friend who could not differentiate between herself and a cat on her lap, with the black color invading every aspect of her conscious experience. True, Grandin and her friend both have access to language, but they do not think in terms of language (13). Social demand and language are not as critical to autistic individuals as they are to the rest of the population, so while sensory overload does not illuminate the specifics of how language can shape experience, it does seem to imply that without it something akin to synesthetic perception would occur. The close temporal proximity of the advent of language acquisition in babies (four to six months, 5) and the cellular apoptosis which presumably precludes synesthetic perception (three to four months, 1) would also seem to indicate some sort of relationship between language and perception, with language somehow breaking up perception into separate sensory events. It has been suggested during the 2008 Spring Semester meeting of Neurobiology and Behavior at Bryn Mawr College that the I-function, a story-telling box within the nervous responsible for conscious experience, cannot handle the volume of perceptual information processed by the nervous system (14). Perhaps language is one method of parsing out “reality” as perceived by the nervous system, allowing the story-teller to construct on account in terms in has either dictated or can understand. Clearly, further investigation, such as observing neuronal activity in language regions of the brain in synesthetes and individuals with autism during perceptive tasks, would go a long way to clarifying this conjecture.

     What we perceive when we look out onto our world does not reflect some inherent truth about said world; it reflects the combined effects of our brains and our culture. Language in particular has been shown to mediate the perception of color, both in the normal population and in synesthetes whereas sensory overload in autistic people may stem from their lack of reliance upon linguistic thought. This is not to say that the faith with which we invest our vision is misplaced, indeed compartmentalized perception would not have evolved were there not some adaptive benefit to experiencing the world in specific sensory modalities (3). Instead it bespeaks the need for caution when we assert truth based solely on what our eyes tell us. In some instances, even if seeing is believing, it isn’t true.


Works Cited

1.Rouw & Scholte. (2007). Increased structural connectivity in grapheme-color synesthesia, Nature Neuroscience, 10(16), 792-797.

2. Cytowic. (2004). Synesthesia Encyclopedia,

Baron-Cohen (1996). Is There a Normal Phase of Synaesthesia in Development?,

4. Hubbard & Ramachandran. (2005). Neurocognitive mechanisms of synesthesia, Neuron, 48, 509-520.

5. Konner, (2002) The Tangled Wing: Biological Constraints on the Human Spirit Second Edition, Henry Holt, New York

6. Witthoft & Winawer. (2006). Synesthetic colors determined by having colored refrigerator magnets in childhood, Cortex, 42, 175-183.

Baron-Cohen. (1996). Synesthesia & Synesthetic Metaphor,

8. Ratner. (1989). A Sociohistorical Critique of Naturalistic Theories of Color Perception, Journal of Mind and Behavior, 10, 361-372.

9. Muzafer Sherif: A Study of Some Social Factors in Perception: Chapter 3

10. Minority Influence and Moscovici

Benson. (2002). Different shades of perception, Monitor on Psychology, 33(11), 28.

12. Language And Color Perception Linked In Human Brain (2008). ScienceDaily,

13. Grandin. (1998). Consciousness in Animals and People with Autism

14. Bio 202, Spring 2008 Notes, April - End,


Paul Grobstein's picture

Color, language, autism, and synesthesia

Very interesting linking together of different areas of inquiry. An intriguing possibility is that what is at issue is not language per se but rather social consensus, however achieved, and that "compartmentalized perception" isn't as much a function of changing "connections" but rather of continuing influences of social cueing?