11 Explanatory Models

Four different models for explaining synesthesia have been developed based on previous results:

  1. The local crossactivation model assumes that a synesthete retains a larger share of the connections between certain brain regions that would normally be eliminated: While the human brain is developing, extensive sprouting of initially less-differentiated nerve synapses occurs. This is followed by the pruning of nerve synapses, so that only the really useful nerve synapses are preserved. It may be that in synesthetes, this process is more restricted and additional synapses are preserved. If one region is active, it could also simultaneously activate other regions. This hypothesis is, for example, supported by Hubbard and colleagues, who found that when they showed letters to grapheme-color synesthetes, not only was the word-form region activated, but the neighboring color region as well.
  2. In contrast, the disinhibited feedback model says that the number of synapses in the brain remains unchanged; however, due to a lack of inhibition, a heightened synaptic transmission between the brain region occurs. Grossenbach and colleagues postulate that this lack of inhibition comes about due to marginal activity of the tempero-parietal-occipital cortex—an intermodal field of association assumed to integrate visual, spatial, and acoustic signals.
  3. Re-entrant theory postulates that during the signal-processing cascade of primary into higher associative regions there is a retroprojection into presynaptic regions that is more marked in synesthetes and involves activation in these regions of previous sensory processing.
  4. The hyperbinding model says that in synesthetes, the brain regions that perform the binding of various sensory information to produce uniform perception are hyperactive, and this leads to an excess of links. According to Estermann, the critical region for this is to be found in the parietal cortex (Estermann et al. 2006); other authors hold the limbic system as principally responsible for this process (e.g., Schiltz et al. 1999, Cytowic 1989, see above).

These models are not necessarily mutually exclusive, so that a combination of the different theories is conceivable. Thus it is possible that the re-entrance of activity in presynaptic regions could occur via the disinhibition of feedback projections, or the pruning model could be the basis for hyperbinding or re-entrant processes.

In summary one can say: it is accepted that synesthesia is not an illusion in the minds of those concerned but a genuine neurophysiological phenomenon that can provide science with unique insights into human perception.