Diagnosing synaesthesia
Individual Neurowiki by Amirah Momen as a part of the Synaesthesia group Neurowiki

Synaesthesia is a neurological condition characterized by the perception of a given sensory modality upon stimulation of another, typically independent, sensory pathway. To date, as many as 60 distinct variations of synaesthesia have been documented with individuals reporting mixed perception of every imaginable sensory combination[1] . The ability to identify and diagnose individuals with synaesthesia is critical to attaining a better understanding of both the prevalence and nature of this condition. Over time, research has equipped scientists and medical practitioners with increasingly sensitive tools for diagnosing and categorizing synaesthetes; however, there is still no true ‘litmus test’ for identifying effected individuals. Efforts to further develop current diagnostic tools are carried out in the hopes of discovering potential biomarkers for synaesthesia. Such biomarkers are critical not only in regards to characterizing atypical neurological functioning, but also to our understanding of the mechanisms underlying sense and perception in the normal brain.

1.1 Historical development of diagnostic tools

The earliest approach to diagnosing individuals with synaesthesia relied on self-report, a method used by Sir Francis Galton throughout his research of the condition in the 19th century. Galton stressed that to the individuals reporting synaesthetic experiences, they were very real; he also noted the highly subjective nature of synaesthesia along with the significant limitations associated with self-report[2] . Many scientists were quick to dismiss claims of synaesthetic experiences as 'imagined events' due to the absence of external symptoms or any means for third-person observation. [3] . Galton, a ‘pioneer’ of the mental testing movement, later abandoned his work with synaesthesia in favor of research that corresponded with the ‘behaviorist’ movement of psychology; the ideas of behaviorism did not correspond with the self-report methods used to diagnose and study synaesthetes as they relied too heavily on the invisible inner workings of the mind, the so called ‘black box’. Research later carried out by Karwoski & Riggs [4] relied on tests that made use of standard psychological principles of the times; these tests involved optical illusions and time latency measures to assess the nature of a given individual’s self-described 'sensory coupling'. Today, the diagnosis of synaesthesia is carried out with a more holistic approach that integrates tools developed by experts with backgrounds in the experimental, psychiatric and neurological foundations of the condition. Richard E. Cytowic [3] , a preeminent figure in the field of synaesthesia research, explains his commonly shared clinical approach to diagnosis of the condition as it conforms to the 'Cartesian' method. This method combines a person's history with neuropsychological tests that serve to 1. verify whether the criteria for the condition are met and 2. to localize the level of the 'lesion'. Patient history is of critical importance when considering whether an individual is truly a synaesthete; before any further testing is carried out, clinicians ask suspected synaesthetes about the following:
    • family prevalence of synaesthesia ;
    • age of the earliest remembered synaesthetic experience ;
    • prior history of drug use;
    • prior history of temporal lobe epilepsy

Fig1. Adapted from Baron-Cohen et al. [1996] Family pedigrees for six synaesthetes. Synaesthetes are shaded black; probands are indicated by arrows.

Once a family history has been taken, the clinician will ask the patient to describe the nature of their percepts and will assess whether their descriptions match the standard criteria for synaesthesia. Finally, the clinician can move on to more objective means of diagnosis i.e. neuropsychological testing. Current efforts are underway to develop diagnostic tools that rely on biomarkers for the various types of synaesthesia, however current methods of diagnosis remain dependent on the Cytowic's clinical approach.

1.2 Determining which processes to test: Perception versus Cognition

Before the creation of any diagnostic tests, early pioneers in the field of synaesthesia research had to come to an agreement about whether synaesthesia was a sensory-perceptual based condition or a product of higher cognitive processes.

Most of the evidence surfacing from this debate has supported the sensory-perceptual basis for synaesthesia. Such evidence includes observations of:
  • reduced percept vividness in accordance with lowered stimulus intensity. [5]
  • Ramachandran's 'pop-out' test shows grapheme-color synaesthetes better segregate amongst numbers due to their elicited color percepts. This supports the theory that cross-activation occurs early on in sensory processing.[2]
  • Synaesthetes can adapt to synaesthetic experiences. This is similar to how sensory input (in the form of a noisy room) can result in 'white noise', becoming part of an individual's unconscious sensory processing. [6] [7]
  • percepts are not elicited by abstract ideas but rather, by discrete stimuli.[8]
  • Brain imaging evidence in the form of ERPs. [9]
Many of the early experimental tasks used to settle the cognition versus sensory-perception debate have been developed into diagnostic tests for identifying synaesthetes. They are described below in section 2.2.

2.1 Criteria for synaesthesia

Synaesthesia is not listed in the DSM-IV (Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition) as it is not considered detrimental to the daily living of affected individuals. While there is no official 'universal criteria' for identifying true synaesthetes, the below criteria describing the characteristics of synaesthetic percepts ,as outlined by Richard Cytowic [3] , is widely used by clinicians and researchers alike.

1. Synaesthetic percepts are involuntary

2. Synaesthetic percepts are spatially extended, existing close to the body i.e. grapheme-color percepts are experienced as if projected onto a screen close to the body; not in the visual field exactly nor merely imagined (Note: this criterion has since been amended to account for the experiences of associator synaesthetes whose percepts are perceived within the 'mind's eye')[10]
3. Synaesthetic percepts are consistent over time i.e. sensory associations do not change with time. This is akin to the observation that synaesthesia shows 're-test reliability' over time[11]

4. Synaesthetic percepts are discrete in nature in that they are composed of simple elements and not complex scenes. This characteristic is what distinguishes synaesthetic percepts from hallucinations

5. Synaesthesia is memorable. This characteristic is detected in sensory recall tests where synaesthetic associations often serve as memory aids
6. Synaesthesia is emotional. Cytowic and Wood [12] suggested that in all forms of synaesthesia, sensory integration occurs via the limbic system; this hypothesis is based on the affective nature of the condition.

Of the criteria outlined by Cytowic [3] , the key characteristics of synaesthesia are that 1. percepts are consistent over time such that the association of one given stimulus always elicits the same cross-modal sensation and 2. that percepts are both involuntary and automatic in nature i.e. they are neither willfully suppressible nor inducible. In essence, a synaesthete will report experiencing specific and automatic cross-wired sense perceptions which are consistent and reproducible over a lifetime. In absence of these key characteristics, clinicians can rule out synaesthesia. Some conditions which may be confused with classical synaesthesia include drug-induced synaesthesia (LSD), sensory deprivation hallucinations, hallucinations induced by temporal lobe epilepsy, peduncular hallucinosis of Lhermitte and hypermnesia. [3]

A depiction of the involuntary, automatic and consistent nature of synaesthetic percepts. Note how the same tone is continually associated with the same color. Even minute differences in tone can be distinguished based on consistent tone-color associations.

2.2 Distinguishing among types: testing approaches

2.2a Associator versus projector synaesthesia

One recent development in the field of synaesthesia includes the amendment to Cytowic's original criteria which stated that precepts are always spatially extended into the outside world. Over the past few years, researches have come to distinguish two broad forms of synaesthesia which render the spatial expansion principle correct for only one of the two forms. Today, synaesthetes are recognized as either lower synaesthetes termed 'projectors' or higher synaesthetes termed 'associators' . The terms 'higher' and 'lower' refer to the level of processing at which the sense-perceptions are crossed while the terms 'projector' and 'associator' refer to where the percept is sensed i.e. out in the world versus in the mind's eye. Dixon et al.[10] came up with a method for differentiating between these two types of synaesthetes which does not depend on purely descriptive measures. They used third-person objective testing in addition to self-report in order to develop standardized performance measures for distinguishing between these groups. Using color-naming and photism-naming Stroop tasks they were able to identify two different patterns of Stroop interference for projectors and associators. Using these tests, Dixon et al. identified whether synaesthetes' percepts were made up of early-stage 'perceptual' or late-stage 'conceptual' aspects of color. In doing so, they identified whether it was form or meaning that elicited synaesthetic experiences for any given synaesthete. Ramachandran and Hubbard [8] have proposed that associator grapheme-color synaesthetes have 'cross-wiring' in the angular gyrus where higher order functions are carried out while projector grapheme-color synaesthetes are expected to have cross-wiring in centers responsible for lower levels of sensory processing. It is suspected that the proportion of associators greatly outnumbers that of projectors, however researchers have yet to propose values to support this claim. Further development of objective diagnostic tools are necessary to come to any conclusions on the associator-to-projector ratio.

2.2b Tests used: Stroop-effect based tests

Test of genuineness: Developed in 1987 by Baron-Cohen et al. [13] and later revised [14] , the Test of Genuineness (TOG) provides an objective basis for diagnosing synaesthesia. The TOG measures the degree to which a suspected synaesthete's percepts adhere to the key criterion of consistency over time. Subjects are instructed to provide detailed descriptions of the percepts they experience in association with a series of test stimuli i.e. tone-color synaesthetes will be played different tones and are asked to describe the colors they elicit. They are later re-tested without warning using the same stimuli. Synaesthetes score with a consistency percentile between 70-90% across trials while non-synaesthetes are only 20-38% consistent. The delay between tests is usually within two weeks of the initial test however, true synaesthetes can usually score within the expected 'genuineness percentile' when the delay is on the order of months.[13] Merikle and Dixon[10] have since developed an online version of the TOG that has improved upon the confounds of memory-effects and inconsistent test conditions.

Stroop Interference Tests: these diagnostic measures test for the involuntary and automatic nature of synaesthetic percepts. During a standard Stroop test, non-synaesthetes will exhibit interference due to competition between two characteristics of a given stimuli on account of one characteristic being more automatically processed than the other; this is seen with the classic colored-word Stroop Test for which reading is the automatic process that interferes with naming the colors of words[15] . Modified Stroop tests take advantage of the automaticity of the percept which is bound to the test stimuli. An example of this is when Dixon et al. [13] asked suspected synaesthetes to name the color of words which were themselves associated with synaesthetic color percepts. The time latency due to this additional interference is used to diagnose individuals as synaesthetes. Stroop interference tests have been developed to test for every kind of synaesthesia by tailoring task stimuli to correspond with the sense modality the suspected synaesthete associates with their percepts.


Fig. 2 Adapted from Palmeri et al. [2001] (a) This local versus global form image is used to test for differences in the synaesthetic experiences of projectors and associators (b) A 'cyclopean' numeral made using randomized dot stereoscopy is seen differently by synaesthetes and non-synaesthetes (c) A depiction of a number that is defined by motion. The number is defined by the direction of motion of the dots. Note that colors and shading are depicted only to illustrate how the task looks to synaesthetes. Fig 3. Adapted from Palmeri et al. [2001] (a) Visual search experiment display (b) The same search display depicted using the synaesthetic colors of WO, a lexical-color synaesthete . (c) and (d) indicate correct target-present response times as a function of display size for non-synaesthetes and for synaesthete WO, respectively. WO is able to differentiate between 2's and 5's faster than non-synaesthetes because of the different color percepts they evoke.

Besides the TOG and Stroop Interference Tests there are a few other diagnostics which rely on similar principles. Visual search tests and Ramachandran and Hubbard's Pop-out test
[8] are notable examples. Individuals with synaesthesia perform better on tests which require them to discriminate between stimuli that appear indistinguishable to non-synaesthetes on account of their automatic percepts. Like with Ramachandran's pop-out tests, visual search tests rely on the consistency of percept associations as well as the automaticity evoked synaesthetic percepts. The association between two sensory modalities serves to reinforce the subtle differences between image, sound, movement, taste and tactile test stimuli.

Diagnosing motion-sound synaesthesia.

2.3 The future of diagnosis: Biomarkers


Current efforts to develop physiologically-based diagnostic tools for synaesthesia include the search for neuroanitomical substrates of the condition. Scientists are actively pursuing both structural and functional biomarkers by imaging clinically diagnosed synaesthetes with such technologies as MRI, fMRI, DTI [16] and PET imaging. Recently, research focused on color-grapheme synaesthesia has identified anatomical differences in the cortical thickness and volume of the V4 cortex [8] . Besides biomarkers for specific forms of synaesthesia, it is conceivable that there may one day be markers for both associator and projector synaesthetes seeing as the difference in bottom-up and top-down processing associated with the two has been correlated with differences in functional connectivity. [17] Scientists have also carried out imaging studies to explore the neural correlates of lesion-induced synaesthesia. One example of this is the study carried out by Naumer & Bosch who observed altered thalamocortical connectivity in touch-tone synaesthesia. [18] It is important to note that anatomical or functional differences do not indicate causality; researches are not sure whether these observations are responsible for the synaesthesia phenotype or whether they are a result of a lifetime of synaesthetic experiences. If the former is the case, using these findings as diagnostic biomarkers will be complicated by idiosyncratic developmental changes and neuroplasticity.

Genetic Screens
The apparent hereditary nature of synaesthesia has been recorded from as far back as the 19th century by Galton[2] however, without the proper molecular tools, little headway was made in the search for diagnostic genetic markers until recently. Family pedigrees composed by Baron-Cohen et al.[11] have served as incentive in the search to identify genetic markers of the various forms of synaesthesia. There is reason to believe that different forms of synaesthesia share the same genetic basis seeing as synaesthetes often report having first, second and third-degree relatives with some form of synaesthesia. [19] The notion that all forms of synaesthesia share a similar genetic basis suggests the potential for a universal diagnostic marker which can identify each form of the condition. Additionally, there has been evidence that genes on chromosomal positions 2q24, 5q33, 6p12, and 12p12 are linked to auditory-visual synaesthesia [20] . Others have suggested that color-grapheme synaesthesia is an X-linked dominant trait [21] . While the search for genetic markers of synaesthesia continues, certain conclusions can be drawn from these works: 1. the genetic basis of synaesthesia is complicated 2. it is unlikely that there will be any single 'synaesthesia gene' and 3. the nature of contributions made by the many implicated genes to the phenotype will be challenging to unravel.

3.1 Difficulties in diagnosing synaesthetes

3.1a Subjective nature of synaesthesia

The subjective nature of synaesthesia limits researchers and clinicians in regards to forming a cohesive clinical diagnostic criteria. It is not simply the lack of any external manifestations that poses the greatest challenge to diagnosis but, rather, it is both the inability of synaesthetes to agree on the nature of their percepts along with their heavy reliance on language capabilities to relay their experiences which remain the greatest barriers to accurate diagnosis.[3]

3.3b Individual differences among same-type synaesthetes

Historically, synaesthesia research has been carried out under the false pretext that synaesthetes with the same 'type' of the condition are a homogenous group. This is incorrect. Hubbard et al. showed that grapheme-color synaesthetes tend to perform differently on cognitive tests and further, that these behavioral differences show similarly contrasting neural correlates. Such individual differences, even among same-type synaesthetes, further complicates the search for diagnostic tools. [22] . Additionally, recent findings suggests that the different subjective experiences of same-type synaesthetes are the result of distinct neural mechanisms.[23]

3.3c Confounds: globally altered networks mask cross-wiring

The search for neurological biomarkers of synaesthesia is further complicated by findings which suggest that synaesthetic experiences are caused by globally altered brain networks as opposed to discrete, localized neurological changes.[24] These findings suggest that identifying a synaesthete will not be as simple as looking for discrete changes in activity, but rather that the neuronal manifestations of synaesthetic experiences are composed of complicated changes in activation patterns across broad neural networks. It may also be the case that higher cognitive functions, such as language, mask the underlying neuronal basis of synaesthesia.
  1. ^

    Sagiv, N. & Ward J. Crossmodal interactions: lessons from synesthesia. Prog Brain Res. (2006) 155:259-71.
  2. ^

    Harrison, J. & Baron-Cohen, S. Synaesthesia: reconciling the subjective with the objective. Endeavour.(1995) 19 (4):157-160.
  3. ^ Cytowic, R. E. (2002). A Union of the Senses. London: The MIT Press.
  4. ^ Riggs, L. A., & Karwoski, T. (1934). Synaesthesia. British Journal of Psychology.(1934) 25: 29-41.
  5. ^ Hubbard E.& Ramachandran V.S. The Phenomenology of Synaesthesia. Journal of Consciousness Studies. (2003) 10 (8): 49-57.
  6. ^ Palmeri, T. J, Blake, R., Marois, R., Flanery M.A. & Whetsell W. The perceptual reality of synesthetic colors. PNAS (2001) 99(6): 4127-4131.
  7. ^ Hubbard E., Brang, D. & Ramachandran V.S. The cross-activation theory at 10. //J Neuropsychol//. (2011) 5 (2):152-77.
  8. ^ Hubbard E.& Ramachandran V.S. Psychophysical investigations into the neural basis of synaesthesia. Proc. Biol Sci. (2001) 268 (1470): 979-83.
  9. ^ Jäncke, L., Beeli, G., Eulig, C. & Hänggi. The neuroanatomy of grapheme–color synesthesia. J. European Journal of Neuroscience (2009) 29 (6): 1287-1293.
  10. ^

    Dixon, M.J., Smilek, D., & Merikle P. M. Not all synaesthetes are created equal: Projector versus associator synaesthetes. Cognitive, Affective, & Behavioral Neuroscience.(2004) 4 (3):335-343.
  11. ^


    Baron-Cohen, S. Synaesthesia: prevalence and familiality. Perception.(1996) 25:1073-9.
  12. ^

    Cytowic, R.E. & Wood, F.B. Synesthesia. I. A review of major theories and their brain basis. Brain Cogn. (1982) 1 (1):23-35.
  13. ^

    Baron-Cohen, S., Wyke, M.A. & Binnie, C. Hearing words and seeing colours: an experimental investigation of a case of synaesthesia. Perception. (1987) 16 (6):761-7.
  14. ^ Asher, J.E., Aitken, M.R.F, Farooqi, N, Kurmani, S. & Baron-Cohen S. Diagnosing and Phenotyping Visual Synaesthesia: a Preliminary Evaluation of the Revised Test of Genuineness (TOG-R). Cortex. (2006) 42(2): 137-146.
  15. ^

    Stroop, J. R. Studies of interference in serial verbal reactions. Journal of Experimental Psychology . (1935) 18 (6): 643–662.
  16. ^

    Ro T, Farnè A, Johnson RM, Wedeen V, Chu Z, Wang ZJ, Hunter JV, Beauchamp MS.Feeling sounds after a thalamic lesion. Ann Neurol (2007) 62: 433–441.
  17. ^

    van Leeuwen T. M, den Ouden H. E & Hagoort P. Effective connectivity determines the nature of subjective experience in grapheme-color synesthesia. J Neurosci. (2011) 31(27):9879-84.
  18. ^ Naumer, M.J. & van den Bosch J.J. Touching sounds: thalamocortical plasticity and the neural basis of multisensory integration. J Neurophysiol. (2009) 102(1):7-8.
  19. ^

    Barnett, K.J, Finucane, C. , Asher, J.A., Bargary, G., Corvin, A.P., Fiona N. Newell, F.N. & Kevin J. Mitchell, K. J.Familial patterns and the origins of individual differences in synaesthesia. Cognition. (2008) 106:871–893.
  20. ^

    Asher, J. A., Lamb, J.A., Brocklebank, D., Cazier, J. Maestrini, E., Addis, L., Sen, M. Baron-Cohen, S. & Anthony P. Monaco A.P. A Whole-Genome Scan and Fine-Mapping Linkage Study of Auditory-Visual Synesthesia Reveals Evidence of Linkage to Chromosomes 2q24, 5q33, 6p12, and 12p12. Am J Hum Genet. (2009) 84(2): 279–285.
  21. ^ Ward, J. & Simner J. Is synaesthesia an X-linked dominant trait with lethality in males? Perception. 2005;34(5):611-23.
  22. ^

    Hubbard, E.M. & Ramachandran, V. S. Neurocognitive Mechanisms of Synesthesia. Neuron. (2005). 48: 509-520.
  23. ^

    Rouw, R. & Scholte HS. Neural basis of individual differences in synesthetic experiences. J Neurosci. (2010) 5; 30(18):6205-13.
  24. ^ Hänggi J, Wotruba D & Jäncke L. Globally altered structural brain network topology in grapheme-color synesthesia. //J Neurosci//. (2011) 31(15):5816-28.