CBU researchers Ian Charest and Nikolaus Kriegeskorte have just published a study which introduces the novel approach of individual representational similarity analysis based on neuroimaging data. In order to reveal the representations of personally meaningful objects in individual brains, each participant viewed photos from their personal lives alongside unfamiliar photos while in a magnetic resonance imaging (MRI) scanner.
For each object, they measured a brain-activity pattern representing that object. The similarity of the brain-activity patterns reflected judgments of similarities among the objects. They then compared the brain-activity similarities between subjects and found significant differences. Individual idiosyncrasies in brain-representational similarities reflected individual idiosyncrasies in similarity judgments, supporting the conclusion that the brain representational differences matter for brain function and subjective experience.
This paper links subjective experience to brain representations at a novel level of specificity: where every particular object and every individual is treated as a unique entity. Relating mind and brain is of broad interest and this study opens up a new dimension for basic and clinical research. Linking individual experience to individual brain representations might prove useful in understanding the altered subjective worlds of mental health patients, including conditions whose neuronal underpinnings have so far been elusive. Future studies using individual representational similarity analysis with clinical populations will be needed to assess the degree to which imaging can help characterise the biological underpinnings of the atypical subjective experience characteristic of a particular individual with a clinical diagnosis. This might help us understand an individual’s disorder as a unique point in a multidimensional space, and to track this point over time in the context of therapeutic interventions.
Paper entitled “Unique semantic space in the brain of each beholder predicts perceived similarity”, published in Proceedings of the National Academy of Sciences (online 22th of September) – see here .