A recent study published in Consciousness and Cognition suggests that people with a unique perceptual trait called synesthesia tend to experience different thematic patterns in their dreams compared to those without the trait. The findings provide evidence that our individual cognitive styles shape our imagination and mental life even while we are asleep.
Synesthesia is a neurological condition where information meant to stimulate one of your senses triggers another sense at the same time. For example, a person with synesthesia might see specific colors when they hear music or taste certain flavors when they read words. People with this trait often score higher on measures of imagination, emotional reactivity, and openness to new experiences.
Scientists conducted this study to understand if these waking mental habits influence the subconscious mind during sleep. The continuity hypothesis of dreaming proposes that our dreams act as a mirror for our everyday thoughts, emotional tendencies, and personality traits. From this perspective, dreams are not just random static generated by the sleeping brain but are meaningful expressions of an individual’s underlying mental structure.
Prior research provides evidence that synesthesia involves an increase in structural connectivity between different regions of the brain. This heightened connectivity allows for greater blending of concepts and perceptions during waking hours. The researchers predicted that this unique mental organization might naturally spill over into the dreaming state.
“My work looks at large collections of dreams to understand how different groups of people experience and make sense of the world. This study is one step in a broader effort to see what dreams can tell us about stable differences in how people think, feel, and interpret their environments,” explained study author Emily Cook, chief science officer at The Center for Organizational Dreaming.
“Synesthesia was a compelling case for this, because it is a clearly defined trait where certain kinds of information automatically evoke extra perceptions or ideas—for example, seeing colours when reading letters, tasting flavours when hearing words, or picturing numbers laid out in space. People with synesthesia also tend to score higher on traits like fantasy proneness, absorption, vivid imagery, and openness to experience.”
“The study builds on the idea that dreaming is continuous with waking cognition, so if synesthesia is linked to a distinctive style of perception and association in waking life, some trace of that should appear in dream content as well,” Cook said. “I wanted to test whether, when you gather many reports together, people with synesthesia tend to dream in systematically different ways from people without synesthesia, and whether those differences align with what is already known about synesthetic imagination and association.”
To explore this question, the scientists analyzed a massive dataset of dream journals shared on the social media platform Reddit. They gathered 2,337 dream reports in total. This included 1,169 reports from individuals who self-identified as synesthetes by posting in a dedicated synesthesia community.
To ensure a fair comparison, the researchers matched these reports with 1,168 control dreams posted by non-synesthetes. They matched these control dreams to the exact same days the synesthete dreams were posted. This step helped eliminate any biases related to specific world events or trends that might have influenced what people were dreaming about at the time.
Instead of reading and categorizing the dreams by hand, the scientists used advanced computational linguistics. These are computer programs designed to process and analyze large amounts of natural human language. The software converted the text of each dream into complex mathematical representations.
This process mapped out the semantic relationships between words in the dream reports. It allowed the computer to group the text into thousands of highly specific topics. Eventually, the program condensed these smaller topics into twenty broad, overarching themes for the researchers to compare.
The researchers also trained machine learning models to see if a computer could guess whether a dream belonged to a synesthete or a non-synesthete. These classification models achieved a modest level of accuracy. This performance indicates that the differences in language use between the two groups were subtle and spread out across many different words.
The thematic analysis revealed that synesthete dreams systematically differed from control dreams in four distinct categories. People with synesthesia were more likely to describe dreams involving digital life. This theme included references to scrolling, screens, computer accounts, and routine technology use.
Synesthetes also reported more dreams centered on interpersonal regret. This theme featured scenarios involving guilt, moral conflict, missed opportunities, and urgent apologies. The scientists note that this aligns with the heightened emotional reactivity and memory retention frequently observed in people with synesthesia.
The third prevalent theme in synesthete dreams was diverse worlds. This category included shifting environments, cultural settings, and complex or dystopian landscapes. Because synesthetes tend to score high in openness to experience, they may possess a more flexible cognitive style that supports the construction of richly detailed and varied dream settings.
Finally, the violent conflict theme appeared more often in the dreams of synesthetes. This theme involved fictional threats, horror imagery, and words associated with intense physical clashes. The researchers suggest that individuals with enhanced memory abilities, a common trait in synesthesia, might be more likely to incorporate intense waking experiences into their dreams.
“One surprise was how rarely synesthesia was named directly in dreams,” Cook told PsyPost. “People with synesthesia rarely described classic synesthesia pairings, such as colours for letters or tastes for words, in their dreams. Given how central those links can be in waking life, it would be easy to expect more obvious synesthesia content in dreams.
“In the paper, we note that this likely reflects that our analysis focused on broad themes rather than specific sensory links, and that sensory details in general are often under-described. We suspect that if we collected dream information differently (for example, by asking synesthetes directly whether they have tasted colors in their dreams), we might find more evidence of ‘dream synesthesia.'”
Together, these results provide evidence that stable brain traits manifest themselves in unstructured, self-generated imagination.
“The main message is that people with synesthesia tend to dream in somewhat different ways than people without synesthesia,” Cook explained. “They were more likely to describe dreams about digital life, guilt and repair, violent conflict, and diverse or shifting worlds than matched controls.”
“Instead of treating synesthesia as merely an unusual way of sensing, the findings support a view of synesthesia as a broader style of imagination and association that shapes many aspects of mental life, including dreaming. The study shows that dreams can contain information about how different groups move through and make sense of their worlds, which is the broader aim of my work.”
While the study offers new insights into the dreaming mind, there are a few limitations to keep in mind. Because the data came from anonymous Reddit users, the scientists could not clinically verify that the participants actually had synesthesia. The researchers relied on self-identification, which means some people might have been misclassified.
“One of the most interesting aspects of this study was that we were able to collect dreams from 1169 synesthetes,” Cook said. “This is a huge number- small sample sizes have historically limited both dream content research and synesthesia research. As only 1-4% of the population has synesthesia, it can be hard to find enough participants.”
“We achieved a large sample size by gathering dream reports shared anonymously online from people who self-identify as synesthetes. The downside of this approach is that we could not use lab-style checks to ensure every single synesthete met the full criteria for synesthesia.”
Another factor to consider is the potential demographic differences between the two groups. For instance, the higher prevalence of digital themes might just mean the synesthete group was slightly younger on average, as younger people tend to use technology more often. The researchers also did not control for factors like gender or cultural background.
Additionally, relying on self-reported dream journals introduces a degree of selection bias. People are usually more likely to share bizarre, narratively structured, or highly emotional dreams online than boring ones. This tendency could skew the types of themes the computer programs detected in the text.
“My long-term goal is to build a richer map of how different groups dream and what that can tell us about their ways of seeing and understanding the world,” Cook told PsyPost. “I am planning studies that combine very large, anonymous online samples with smaller cohorts who complete in-depth assessments, so we can see whether the same patterns appear when dreams are collected and verified in different ways.”
“For me, this study shows how quickly dream research can advance now that we have large public dream collections and tools to handle them. When we bring thousands of dreams into one analysis, patterns in how different groups dream become much easier to see than in small, hand-coded samples. More broadly, it highlights that dreams carry rich information about how people experience the world, and that this information is only now becoming available in a way that lets us study group differences at scale.”
The study, “Synesthesia is associated with distinctive patterns in dream content,” was authored by Kyle Napierkowski and Emily Cook.
