Dr. Devin B. Terhune is a Senior Lecturer in the Department of Psychology at Goldsmiths, University of London. He applies a range of methods to different facets of consciousness, with a focus on time perception and hypnosis.

Synaesthesia is a fascinating condition in which a stimulus in one sensory modality will reliably and automatically produce a secondary experience in a different modality. The most well-studied example of this phenomenon is grapheme-colour synaesthesia, in which numbers and letters (inducers) will involuntarily elicit experiences of colour (concurrents) in some individuals (synaesthetes). Although we typically assume that we all perceive the world in a relatively similar way, synaesthetes actually have a radically different conscious experience of their environment. These unusual experiences seem to confer benefits: synaesthetes display superior working memory, episodic memory, and colour perception. This condition may yield insights into a wide range of phenomena including automaticity, learning, and consciousness.

The most dominant model of synaesthesia proposes that synaesthetic photisms arise from excess structural connections between brain regions subserving inducer and concurrent processing. For example, an individual who experiences colours when looking at faces might exhibit heightened connectivity between the fusiform face area, a critical node in face processing, and neighboring V4, an important brain region for colour perception. Putative cases of induced synaesthesia, in which non-synaesthetes have transient crossmodal experiences resembling synaesthesia, pose a challenge to this account. If these individuals are having genuine experiences of synaesthesia, this would suggest that hyperconnectivity is not a necessary condition for synaesthesia as such connections are unlikely to form in a rapid manner. Rather, hyperconnectivity may just be one of multiple routes by which one could develop synaesthesia. Further still, it may merely be an artefact of living with this condition – a byproduct of the repeated experience of synaesthesia or the consolidation of inducer-concurrent pairs over time that does not have a mechanistic role in the occurrence of the condition.

Researchers tend to either credulously assume that induced synaesthesia-like experiences qualify as synaesthesia or outright dismiss them as unrelated to developmental synaesthesia. These oversimplifications seem to arise because these literatures have not been properly integrated and interrogated. Accordingly, in a recent book chapter, my colleagues and I sought to synthesize the available evidence regarding the induction of synaesthesia in non-synaesthetes and provide a clear assessment of the state of our knowledge. In particular, we review research studies that have examined whether synaesthesia can be induced in non-synaesthetes using drugs, cognitive training, and hypnosis.

Researchers have repeatedly documented that drugs, particularly psychedelic drugs that target the serotonin system, such as LSD and psilocybin, produce crossmodal experiences similar to synaesthesia. Although different studies have yielded fairly consistent results, they suffer from numerous methodological limitations; the data come almost entirely from surveys, anecdotal reports, and studies that lack experimental controls or use self-report measures of synaesthesia. To address this gap in the literature, my colleagues and I recently conducted the first placebo-controlled investigation of whether LSD produces synaesthesia. We found that LSD elicited synaesthesia-like experiences, such as reports of colours when presented with sounds. However, these associations did not meet multiple standard criteria for synaesthesia. For example, under LSD, sounds would elicit experiences of colour but the colour would vary from one presentation of the same sound to the next. By contrast, developmental synaesthetes exhibit very consistent associations. However, consistency might not be a fair adjudication criterion by which to assess the veracity of induced synaesthesias. It is possible that inducer-concurrent consistency is not a core feature of synaesthesia but rather a product of the consolidation of synaesthetic associations over time. Indirect evidence for this possibility comes from research on synaesthetic children, who are not as consistent in their associations as adult synaesthetes, and independent work showing that synaesthetic associations seem to be shaped in part by environmental constraints.

These last lines of evidence segue nicely into another induction method: cognitive training. Given the foregoing results, perhaps we can induce synaesthesia by repeatedly and dynamically pairing crossmodal stimuli? Some, but not all, studies have been able to reproduce behavioural characteristics of synaesthesia such as the tendency to automatically associate crossmodal stimuli. However, nearly all studies have failed to meet one or more criteria for synaesthesia, particularly the conscious experience of concurrents. One exception is a recent study that included dynamic training over multiple weeks. This training regimen produced reliable associations between crossmodal stimuli in participants that met various characteristics of synaesthesia and more than 50% of the participants reported having conscious experiences of concurrents. These results are very encouraging and suggest that cognitive training represents a valuable approach. Some interesting ways forward might be to examine whether cognitive training is more effective at inducing synaesthesia in the family members of synaesthetes, who may have a genetic predisposition to such experiences or related perceptual abilities, and to couple training with the administration of psychedelics, which seem to more rapidly induce concurrent experiences.

A final approach involves using posthypnotic suggestions, which are administered during hypnosis but only take effect afterwards, to produce synaesthetic associations. Although widely misunderstood due to perennial misconceptions in popular culture, hypnosis is increasingly being accepted as a reliable technique for modulating the contents of consciousness. Hypnosis has been successfully used to model a number of conditions including hallucinations, with many of these effects corroborated using neuroimaging. It has also been used to disrupt the behavioural and electrophysiological markers of synaesthesia in a highly suggestible synaesthete. One study showed that posthypnotic suggestion can be used to produce robust synaesthetic associations including conscious experiences of concurrents in highly suggestible individuals. Although preliminary, this method appears to be capable of reproducing all of the behavioural and phenomenological features of synaesthesia. A significant limitation of this approach is that it requires the use of highly suggestible individuals, who comprise a small minority of the population with a prevalence that is not substantially higher than that of synaesthesia. More broadly, a clear need here, and with the other techniques, of course is to use neuroimaging to compare the neurophysiological correlates of developmental and induced synaesthesia. Similar behaviour and phenomenology do not necessarily mean similar brain mechanisms.

So, what is the answer? All three of these approaches seem to be producing experiences that closely resemble synaesthesia. However, the available data do not as yet allow us to make firm statements regarding whether these induced phenomena qualify as genuine synaesthesia. Nevertheless, the preliminary data are encouraging and worthy of further attention. Moreover, even if these induced synaesthesias turn out to be mechanistically distinct from developmental synaesthesia, they may prove useful in studying certain features of this condition. For example, training may serve as a valuable model for experimentally studying the learning processes that shape the formation of synaesthetic associations.