A systematic, large-scale study of synaesthesia: implications for the role of early experience in lexical-colour associations

For individuals with synaesthesia, stimuli in one sensory modality elicit anomalous experiences in another modality. For example, the sound of a particular piano note may be 'seen' as a unique colour, or the taste of a familiar food may be 'felt' as a distinct bodily sensation. We report a study of 192 adult synaesthetes, in which we administered a structured questionnaire to determine the relative frequency and characteristics of different types of synaesthetic experience. Our data suggest the prevalence of synaesthesia in the adult population is approximately 1 in 1150 females and 1 in 7150 males. The incidence of left-handedness in our sample was within the normal range, contrary to previous claims. We did, however, find that synaesthetes are more likely to be involved in artistic pursuits, consistent with anecdotal reports. We also examined responses from a subset of 150 synaesthetes for whom letters, digits and words induce colour experiences ('lexical-colour' synaesthesia). There was a striking consistency in the colours induced by certain letters and digits in these individuals. For example, 'R' elicited red for 36% of the sample, 'Y' elicited yellow for 45%, and 'D' elicited brown for 47%. Similar trends were apparent for a group of non-synaesthetic controls who were asked to associate colours with letters and digits. Based on these findings, we suggest that the development of lexical-colour synaesthesia in many cases incorporates early learning experiences common to all individuals. Moreover, many of our synaesthetes experienced colours only for days of the week, letters or digits, suggesting that inducers that are part of a conventional sequence (e.g. Monday, Tuesday, Wednesday...; A, B, C...; 1, 2, 3...) may be particularly important in the development of synaesthetic inducer-colour pairs. We speculate that the learning of such sequences during an early critical period determines the particular pattern of lexical-colour links, and that this pattern then generalises to other words.     

Mental imagery and synaesthesia: Is synaesthesia from internally-generated stimuli possible?

Previous studies provide empirical support for the reported colour experience in grapheme-colour synaesthesia by measuring the synaesthetic experience from an externally presented grapheme. The current study explored the synaesthetic experience resulting from a visual mental image of a grapheme. Grapheme-colour synaesthetes (N=6) and matched controls (N=10 per synaesthete) completed a visual mental imagery task that involved visualising a letter and making a size-based decision about it. The background colour that the grapheme was visualised against was manipulated so that it was congruent or incongruent with the synaesthetic colour for the grapheme being visualised. Compared to matched controls, an effect of colour condition was found for four of the six synaesthetes, although importantly the direction of the effect varied between synaesthetes. In addition, a significant effect of group was found, as the synaesthetes were all faster than the matched controls at the imagery task, regardless of background colour. We conclude that there is some support for subjective reports of imagery-induced synaesthesia, but there are important individual differences. These findings are discussed in relation to both the visual imagery and synaesthesia literature.     

Brain areas involved in synaesthesia: a review

Despite a recent upsurge of research, much remains unknown about the neurobiological mechanisms underlying synaesthesia. By integrating results obtained so far in Magnetic Resonance Imaging (MRI) studies, this contribution sheds light on the role of particular brain regions in synaesthetic experiences. First, in accordance with its sensory nature, it seems that the sensory brain areas corresponding to the type of synaesthetic experience are activated. Synaesthetic colour experiences can activate colour regions in occipito-temporal cortex, but this is not necessarily restricted to V4. Furthermore, sensory and motor brain regions have been obtained that extend beyond the particular type of synaesthesia studied. Second, differences in experimental setup, number and type of synaesthetes tested, and method to delineate regions of interest may help explain inconsistent results obtained in the BOLD-MRI (Blood Oxygen Level Dependent functional MRI) studies. Third, an overview of obtained results shows that a network of brain areas rather than a single brain region underlies synaesthesia. Six brain regions of overlapping results emerge, these regions are in sensory and motor regions as well as 'higher level' regions in parietal and frontal lobe. We propose that these regions are related to three different cognitive processes inherently part of synaesthesia; the sensory processes, the (attentional) 'binding' processes, and cognitive control processes. Finally, we discuss how these functional and structural brain properties might relate to the development of synaesthesia. In particular, we believe this relationship is better understood by separating the question what underlies the presence of synaesthesia ('trait') from what determines particular synaesthetic associations ('type').     

Searching for Shereshevskii: What is superior about the memory of synaesthetes?

Some individuals with superior memory, such as the mnemonist Shereshevskii (Luria, 1968), are known to have synaesthesia. However, the extent to which superior memory is a general characteristic of synaesthesia is unknown, as is the precise cognitive mechanism by which synaesthesia affects memory. This study demonstrates that synaesthetes tend to report subjectively better than average memory and that these reports are borne out with objective testing. Synaesthetes experiencing colours for words show better memory than matched controls for stimuli that induce synaesthesia (word lists) relative to stimuli that do not (an abstract figure). However, memory advantages are not limited to material that elicits synaesthesia because synaesthetes demonstrate enhanced memory for colour per se (which does not induce a synaesthetic response). Our results suggest that the memory enhancement found in synaesthetes is related to an enhanced retention of colour in both synaesthetic and nonsynaesthetic situations. Furthermore, this may account for the fact that synaesthetic associations, once formed, remain highly consistent.     

Searching for Shereshevskii: what is superior about the memory of synaesthetes?

Some individuals with superior memory, such as the mnemonist Shereshevskii (Luria, 1968), are known to have synaesthesia. However, the extent to which superior memory is a general characteristic of synaesthesia is unknown, as is the precise cognitive mechanism by which synaesthesia affects memory. This study demonstrates that synaesthetes tend to report subjectively better than average memory and that these reports are borne out with objective testing. Synaesthetes experiencing colours for words show better memory than matched controls for stimuli that induce synaesthesia (word lists) relative to stimuli that do not (an abstract figure). However, memory advantages are not limited to material that elicits synaesthesia because synaesthetes demonstrate enhanced memory for colour per se (which does not induce a synaesthetic response). Our results suggest that the memory enhancement found in synaesthetes is related to an enhanced retention of colour in both synaesthetic and nonsynaesthetic situations. Furthermore, this may account for the fact that synaesthetic associations, once formed, remain highly consistent.     

Frequency correlates in grapheme-color synaesthesia

Individuals with synaesthesia experience certain stimuli in more than one sensory modality. Most common is the linkage of letters and digits (graphemes) to colors. Whereas synaesthesia might be partly genetically determined, the linkages to specific colors are assumed to be learned. We present a systematic statistical analysis of synaesthetic color perception based on subjects' reproduction of individual colors for each grapheme, instead of simple verbal categorizations. The statistical analysis revealed that the color perceptions, measured with the HSL (hue, saturation, and luminance) scale, varied systematically among the different digits and letters. The frequencies of the digits and letters (in the German language) partly explained these systematic variations. However, digit frequency was more strongly related to color perception in the synaesthetes than was letter frequency. The results for digit and letter frequency indicate that experience with graphemes may shape synaesthetic color perception.     

Synaesthesia: learned or lost?

The question why synaesthesia, an atypical binding within or between modalities, occurs is both enduring and important. Two explanations have been provided: (1) a congenital explanation: we are all born as synaesthetes but most of us subsequently lose the experience due to brain development; (2) a learning explanation: synaesthesia is related to some learning process during childhood. Three recent studies provide conflicting support for these explanations. Two studies supported the idea that synaesthesia is learned by showing that the frequency of everyday language implicitly modulates the synaesthetic experience. Another study argued that synaesthesia reflects basic, innate magnitude representations. In this paper we reassess these points of view, and show that it is possible for both to be valid. These findings are integrated into an interactive specialization account of development in order to explain the neuronal mechanism underlying synaesthesia.     

Is synaesthesia an X-linked dominant trait with lethality in males?

In previous research the inheritance patterns of synaesthesia (eg experiencing colours from graphemes) has been studied and it was concluded that synaesthesia is most likely to be the outcome of a single gene passed on the X chromosome in a dominant fashion. In addition, it has been reported that the female-male ratio of synaesthetes is as high as 6:1 and the families of synaesthetes contain more female than male members. This raises the possibility that the gene may be associated with lethality in males. In this study we replicate and extend previous research by investigating the female-male ratio and inheritance patterns in a large sample of synaesthetic families (N = 85). We were able to verify the authenticity of grapheme-colour associates in at least one proband from each family using internal consistency. As before, our results show a female-male bias and are broadly consistent with an X-linked dominant mode of inheritance. However, there was no evidence of male lethality (eg synaesthetes are just as likely to give birth to sons as to daughters). Moreover, our female-male ratio of synaesthetes within families was 2:1--considerably lower than previous estimates. We speculate that men may be more reluctant to disclose synaesthesia than women (indeed, our female-male ratio based on self-referral was 3.7: 1). Finally, we discuss how the genotype may give rise to the phenotype in terms of changes in synaptogenesis or plasticity extending into childhood, to be subsequently shaped by the environment.     

The neural basis of illusory gustatory sensations: two rare cases of lexical-gustatory synaesthesia

Lexical-gustatory synaesthesia is a rare phenomenon in which the individual experiences flavour sensations when they read, hear, or imagine words. In this study, we provide insight into the neural basis of this form of synaesthesia using functional neuroimaging. Words known to evoke pleasant, neutral, and unpleasant synaesthetic tastes and synaesthetically tasteless words were presented to two lexical-gustatory synaesthetes, during fMRI scanning. Ten non-synaesthetic participants were also scanned on the same list of words. The synaesthetic brain displayed a different pattern of activity to words when compared to the non-synaesthetes, with insula activation related to viewing words that elicited tastes that have an associated emotional valence (i.e., pleasant or unpleasant tastes). The subjective intensity of the synaesthesia was correlated with activity in the medial parietal lobes (precuneus/retrosplenial cortex), which are implicated in polymodal imagery and self-directed thought. This region has also previously been activated in studies of lexical-colour synaesthesia, suggesting its role may not be limited to the type of synaesthesia explored here.     

Colored-Speech Synaesthesia Is Triggered by Multisensory, Not Unisensory, Perception

ABSTRACT— Although it is estimated that as many as 4% of people experience some form of enhanced cross talk between (or within) the senses, known as synaesthesia, very little is understood about the level of information processing required to induce a synaesthetic experience. In work presented here, we used a well-known multisensory illusion called the McGurk effect to show that synaesthesia is driven by late, perceptual processing, rather than early, unisensory processing. Specifically, we tested 9 linguistic-color synaesthetes and found that the colors induced by spoken words are related to what is perceived (i.e., the illusory combination of audio and visual inputs) and not to the auditory component alone. Our findings indicate that color-speech synaesthesia is triggered only when a significant amount of information processing has occurred and that early sensory activation is not directly linked to the synaesthetic experience.     

Varieties of grapheme-colour synaesthesia: a new theory of phenomenological and behavioural differences

Recent research has suggested that not all grapheme-colour synaesthetes are alike. One suggestion is that they can be divided, phenomenologically, in terms of whether the colours are experienced in external or internal space (projector–associator distinction). Another suggestion is that they can be divided according to whether it is the perceptual or conceptual attributes of a stimulus that is critical (higher–lower distinction). This study compares the behavioural performance of 7 projector and 7 associator synaesthetes. We demonstrate that this distinction does not map on to behavioural traits expected from the higher–lower distinction. We replicate previous research showing that projectors are faster at naming their synaesthetic colours than veridical colours, and that associators show the reverse profile. Synaesthetes who project colours into external space but not on to the surface of the grapheme behave like associators on this task. In a second task, graphemes presented briefly in the periphery are more likely to elicit reports of colour in projectors than associators, but the colours only tend to be accurate when the grapheme itself is also accurately identified. We propose an alternative model of individual differences in grapheme-colour synaesthesia that emphasises the role of different spatial reference frames in synaesthetic perception. In doing so, we attempt to bring the synaesthesia literature closer to current models of non-synaesthetic perception, attention and binding.     

Colours sometimes count: awareness and bidirectionality in grapheme-colour synaesthesia

Three experiments were conducted with 10 grapheme-colour synaesthetes and 10 matched controls to investigate (a) whether awareness of the inducer grapheme is necessary for synaesthetic colour induction and (b) whether grapheme-colour synaesthesia may be bidirectional in the sense that not only do graphemes induce colours, but that colours influence the processing of graphemes. Using attentional blink and Stroop paradigms with digit targets, we found that some synaesthetes did report “seeing” synaesthetic colours even when they were not able to report the inducing digit. Moreover, congruency effects (effects of matching the colour of digit presentation with the synaesthetic colour associated with that digit) suggested that grapheme-colour synaesthesia can be bidirectional, at least for some synaesthetes.     

Colours sometimes count: Awareness and bidirectionality in grapheme–colour synaesthesia

Three experiments were conducted with 10 grapheme–colour synaesthetes and 10 matched controls to investigate (a) whether awareness of the inducer grapheme is necessary for synaesthetic colour induction and (b) whether grapheme–colour synaesthesia may be bidirectional in the sense that not only do graphemes induce colours, but that colours influence the processing of graphemes. Using attentional blink and Stroop paradigms with digit targets, we found that some synaesthetes did report “seeing” synaesthetic colours even when they were not able to report the inducing digit. Moreover, congruency effects (effects of matching the colour of digit presentation with the synaesthetic colour associated with that digit) suggested that grapheme–colour synaesthesia can be bidirectional, at least for some synaesthetes.     

Handedness and calendar orientations in time-space synaesthesia

In one common variant of time-space synaesthesia, individuals report the consistent experience of months bound to a spatial arrangement, commonly described as a circle extending outside of the body. Whereas the layout of these calendars has previously been thought to be relatively random and to differ greatly between synaesthetes, Study 1 provides the first evidence suggesting one critical aspect of these calendars is mediated by handedness: clockwise versus counter-clockwise orientation. A study of 34 time-space synaesthetes revealed a strong association between handedness and the orientation of circular calendars. That is, left-handed time-space synaesthetes tended to report counter-clockwise arrangements and right-handed synaesthetes clockwise. Study 2 tested whether a similar bias was present in non-synaesthetes whose task was to memorize and recall the spatial configuration of a clockwise and counter-clockwise calendar. Non-synaesthetes' relative performance on these two sorts of calendars was significantly correlated with their handedness scores in a pattern similar to synaesthetes. Specifically, left-handed controls performed better on counter-clockwise calendars compared to clockwise, and right-handed controls on clockwise over counter-clockwise. We suggest that the implicit biases seen in controls are mediated by similar mechanisms as in synaesthesia, highlighting the graded nature of synaesthetic associations.     

Type-based associations in grapheme-color synaesthesia revealed by response time distribution analyses

Determining the nature of binding in grapheme-color synaesthesia has consequences for understanding the neural basis of synaesthesia and visual awareness in general. We evaluated type- and token-based letter-color binding using a synaesthetic version of the object-reviewing paradigm. Although mean response times failed to reveal any significant differences between synaesthetes and control participants, RT analyses with ex-Gaussian distributions revealed that the response facilitation in the synaesthesia group reflected type representations exclusively, while response facilitation in the control group, who learned letter-color associations, was dominated by token representations. Thus, letter-color associations in associator synaesthetes are type-based, and do not involve binding to object tokens, consistent with their subjective reports. Contrary to recent studies that failed to find differences between synaesthetes and non-synaesthetes with behavioral measures, response time distribution analyses indicate that color sensations in synaesthetes are not simply the extreme form of normal letter-color associations, and cannot be attributed to demand characteristics.     

"That's not a real body": identifying stimulus qualities that modulate synaesthetic experiences of touch

Mirror-touch synaesthesia is a condition where observing touch to another’s body induces a subjective tactile sensation on the synaesthetes body. The present study explores which characteristics of the inducing stimulus modulate the synaesthetic touch experience. Fourteen mirror-touch synaesthetes watched videos depicting a touch event while indicating (i) whether the video induced a tactile sensation, (ii) on which side of their body they felt this sensation and (iii) the intensity of the experienced sensation. Results indicate that the synaesthetes experience stronger tactile sensations when observing touch to real bodies, whereas observing touch to dummy bodies, pictures of bodies and disconnected dummy body parts elicited weaker sensations. These results suggest that mirror-touch synaesthesia is not entirely bottom-up driven, but top-down information, such as knowledge about real and dummy body parts, also modulate the intensity of the experience.     

Not all synaesthetes are created equal: Projector versus associator synaesthetes

In synaesthesia, ordinary stimuli elicit extraordinary experiences. When grapheme-color synaesthetes view black text, each grapheme elicits a photism-a highly specific experience of color. Importantly, some synaesthetes (projectors) report experiencing their photisms in external space, whereas other synaesthetes (associators) report experiencing their photisms "in the mind's eye." We showed that projectors and associators can be differentiated not only by their subjective reports, but also by their performance on Stroop tasks. Digits were presented in colors that were either congruent or incongruent with the synaesthetes' photisms. The synaesthetes named either the video colors of the digits or the colors of the photisms elicited by the digits. The results revealed systematic differences in the patterns of Stroop interference between projectors and associators. Converging evidence from first-person reports and third-person objective measures of Stroop interference establish the projector/ associator distinction as an important individual difference in grapheme-color synaesthesia.     

Colour fluctuations in grapheme‐colour synaesthesia: The effect of clinical and non‐clinical mood changes

Synaesthesia is a condition that gives rise to unusual secondary sensations (e.g., colours are perceived when listening to music). These unusual sensations tend to be reported as being stable throughout adulthood (e.g., Simner & Logie, 2007, Neurocase, 13, 358) and the consistency of these experiences over time is taken as the behavioural hallmark of genuineness. Our study looked at the influence of mood states on synaesthetic colours. In Experiment 1, we recruited grapheme‐colour synaesthetes (who experience colours from letters/digits) and elicited their synaesthetic colours, as well as their mood and depression states, in two different testing sessions. In each session, participants completed the PANAS‐X (Watson & Clark, 1999) and the BDI‐II (Beck, Steer, & Brown, 1996, Manual for Beck Depression Inventory‐II), and chose their synaesthetic colours for letters A‐Z from an interactive colour palette. We found that negative mood significantly decreased the luminance of synaesthetic colours. In Experiment 2, we showed that synaesthetic colours were also less luminant for synaesthetes with anxiety disorder, versus those without. Additional evidence suggests that colour saturation, too, may inversely correlate with depressive symptoms. These results show that fluctuations in mood within both a normal and clinical range influence synaesthetic colours over time. This has implications for our understanding about the longitudinal stability of synaesthetic experiences, and of how mood may interact with the visual (imagery) systems.     

Grapheme-color synaesthesia benefits rule-based Category learning

Researchers have long suspected that grapheme-color synaesthesia is useful, but research on its utility has so far focused primarily on episodic memory and perceptual discrimination. Here we ask whether it can be harnessed during rule-based Category learning. Participants learned through trial and error to classify grapheme pairs that were organized into categories on the basis of their associated synaesthetic colors. The performance of synaesthetes was similar to non-synaesthetes viewing graphemes that were physically colored in the same way. Specifically, synaesthetes learned to categorize stimuli effectively, they were able to transfer this learning to novel stimuli, and they falsely recognized grapheme-pair foils, all like non-synaesthetes viewing colored graphemes. These findings demonstrate that synaesthesia can be exploited when learning the kind of material taught in many classroom settings.     

Synaesthesia,creativity and art: What is the link?

It has been suggested that individuals with synaesthesia may show heightened creativity as a result of being able to form meaningful associations between disparate stimuli (e.g. colour, sound). In this study, a large sample (N=82) of people with various kinds of synaesthesia were given two psychometric tests of creativity (Remote Associates Test, Alternate Uses Test) and were also asked about the amount of time engaged in creative arts (visual art, music). There was a significant tendency for synaesthetes to spend more time engaged in creative arts and this was, at least in part, dependent upon the type of synaesthesia experienced. For example, synaesthetes experiencing vision from music were far more likely to play an instrument than their other synaesthetic counterparts. There was no relationship between this tendency and the psychometric measures of creativity, but synaesthetes did outperform controls on one of the two psychometric measures (Remote Associates). We conclude that the tendency for synaesthetes to be more engaged in art is likely to have a different mechanism to psychometric measures of creativity, and that there is no direct link between them. Although synaesthetes may well perform better on some measures of creativity, we suggest that synaesthetes have better bottom‐up access to certain associations, but are not necessarily better able to use them flexibly (in divergent thinking).