The mean percentage of correct responses during Mah-Jong tile discrimination was 88.8 В± 6.9% (mean В± SD) in the expert group, which was significantly better than the control group (70.0 В± 19.3%, P = 0.01, two-sample t-test). Furthermore, as we expected the same effect even with unfamiliar materials, the Mah-Jong experts also underwent a tactile discrimination task using Braille characters, to which they were naГЇve. Our hypothesis was that the subjects who are well-trained on the tactile discrimination of Mah-Jong patterns would show more prominent activation in the visual cortex when performing this task, including the multi-modal ventral association visual cortex, than the naГЇve subjects. Some well-trained Mah-Jong players can identify the carved patterns by touch. Mah-Jong is a Chinese game similar to card games, involving two-dimensional plastic tiles with various marks carved on one side. ДЄnticipating a similar learning effect in visual-tactile cross-modal plasticity, we conducted fMRI studies in persons with normal sight who have had long-term tactile shape discrimination training on Mah-Jong tiles. The naГЇve and less well-trained groups did not show activation of the left PT during any of the tasks. Viewing the bimanual hand movements of a piano player making key presses actually activated the left planum temporale (PT) of well-trained subjects. Well-trained pianists are able to identify pieces of music by watching the hands touch the piano keys thus the visual information from observing the sequential finger movements was transformed into the auditory modality during 'key-touch reading'. Recently, audio-visual cross-modal plasticity due to learning has been shown while pianists observe piano playing.

These findings indicate that cross-modal plasticity may be induced not only by sensory deafferentation but also by learning. Therefore, visual deafferentation may cause less demand on bottom-up visual processing, which may in turn introduce an opportunity for the expansion of tactile representations in the visual association cortex. Recent studies indicate that both tactile and visual processing of objects are represented in the visual association cortex, indicating that visual and tactile processing may be competitively balanced in the association cortices where the inputs adjoin. Hence the activation of the association visual cortex of blind subjects while performing a tactile discrimination task may be due to sensory deafferentation, wherein a competitive imbalance favors the tactile over the visual modality. Additionally, the occipital cortex of blind subjects without Braille training was activated during tactile discrimination tasks, whereas that of control sighted subjects was not. A previous study has shown that non-Braille haptic processes activate the visual cortex of blind subjects who read Braille proficiently, which is consistent with the latter hypothesis. As a result, it is not clear whether the visual cortex activation is related to long-term Braille training or to visual deafferentation. However, such findings typically come from experiments in subjects who have been blind from a very early age and who also have undergone Braille training from a young age. Braille reading is known to activate the visual cortex of the blind, indicating remarkable cortical plasticity. In Braille, the visual perception of printed characters is replaced by the tactile interpretation of raised dots.

For visually-deprived persons, Braille is the most successful system for the transmission of written information. Visual-tactile cross-modal plasticity due to visual deprivation has been reported. The activation of V1 in subjects trained in tactile discrimination may represent altered cross-modal responses as a result of long-term training. Both the LO and V1 of the well-trained subjects were activated during Braille tactile discrimination tasks. In the naГЇve subjects, the LO was activated but V1 was not activated.

When participants performed tactile discrimination of Mah-Jong tiles, the left lateral occipital cortex (LO) and V1 were activated in the well-trained subjects. Furthermore, seven out of eight experts performed a tactile shape matching task with unfamiliar 2D Braille characters. Eight Mah-Jong experts and twelve healthy volunteers who were naГЇve to Mah-Jong performed a tactile shape matching task using Mah-Jong tiles with no visual input. Here we conducted an fMRI study with normally-sighted participants who had undergone long-term training on the tactile shape discrimination of the two dimensional (2D) shapes on Mah-Jong tiles (Mah-Jong experts). It has yet to be determined whether visual-tactile cross-modal plasticity due to visual deprivation, particularly in the primary visual cortex (V1), is solely due to visual deprivation or if it is a result of long-term tactile training.

Random factor mahjong