Neurosci Lett. 2015 May 6;594:155-160. doi: 10.1016/j.neulet.2015.03.043. Epub 2015 Mar 26. 

Brain activities of visual thinkers and verbal thinkers: A MEG study


Kazuo Nishimuraa, Takaaki Aokib, Michiyo Inagawab, Yoshikazu Tobinagac, Sunao Iwakid

a Kobe University, RIETI, Japan and Santa Fe Institute, USA

b Institute of Economic Research, Kyoto University, Japan

c Elegaphy Inc., Japan

d Human Technology Research Institute, National Institute of Advanced Industrial Science and Technology, Japan

 *Corresponding author.



In this study we measured activation patterns in the primary visual cortex and the frontal language areas and compared them in individuals with strong and weak capacities to mentally visualize information during spontaneous thinking. Subjects were first administered a 5-item questionnaire to assess their ability to create mental pictures, and were divided into two groups (strong and weak visualizers) on this basis. They then performed tasks requiring visual imagery and verbal recollection, and their local neural activities were measured, using magnetoencephalography (MEG). Notably in the high beta-band (25Hz), the visual area was more strongly activated in strong visualizers, whereas, the frontal language areas were more strongly activated in weak visualizers. Strong visualizers are considered to be visual thinkers, and weak visualizers are verbal thinkers.

KEYWORDS: Beta-band activities; Differences between groups; Magnetoencephalography; Spatial filter; Visual thinkers and verbal thinkers

PMID: 25818330



The ability to imagine objects or phenomena varies among individuals. For example, some people can memorize the faces of persons they have met for the first time, while others cannot. It is very important to examine such differences in visualization abilities among individuals in relation to brain activity. To the present, it has been considered more difficult to measure spontaneous nerve activity compared with that of cranial nerves evoked by stimulation from outside of the visual or auditory organs. However, in this study, we succeeded in measuring spontaneous brain activity and detecting differences in it among individuals using a SQUID system that allows high-sensitivity and noninvasive measurement.


We performed magnetoencephalography (MEG), using such a system and involving individuals divided into 2 groups based on their visualization abilities: high (Group-I) and low (Group-L). To clarify whether or not differences in visualization abilities lead to variations in brain activity for spontaneous thoughts, we presented them with multiple tasks.


The experiment was conducted at the National Institute of Advanced Industrial Science and Technology (Ikeda City, Osaka, Japan) within the period between August 2 and September 13, 2011, during which MEG was performed using a whole-cortex-type, 122-channel, direct current SQUID system (Neuromag 122, Elekta-Neuromag, Helsinki, Finland). The types of thought of the 13 (11 males and 2 females) subjects targeted for analysis had previously been identified through a questionnaire survey.


During the experiment, they implemented 6 tasks under 3 conditions: 1) visual condition: visualizing Kiyomizu-dera Temple/the Diet Building; 2) verbal condition: remembering the 12 signs of the Oriental zodiac/the content of today’s conversation with someone; and 3) rest condition: not thinking about anything/not thinking about anything. The 6 tasks were continuously repeated twice during 1 session. The time allocated for each task was 10 seconds, and there was no break between the tasks. As a benchmark, the ratio of the value obtained during each task to that under the rest condition was calculated.


On comparison of brain activity for spontaneous thoughts, clear differences were observed between Groups-I and -L, suggesting that variations in brain activity are associated more closely with the visualization ability, rather than the visual or verbal condition. Based on the results of measurement, individuals with strong and weak visualization abilities may show tendencies toward visual and verbal thoughts, respectively.


On examining the values of Group-I compared with -L, the activity level in the proximity of the visual area was higher in the former under all conditions. In contrast, in the speech area of the frontal lobe, the level was higher in the latter. This tendency was more marked in the high beta brainwave range (center frequency: 25 Hz).


It is strongly suggested that the type of thoughts, such as visual or verbal thoughts, markedly influences one’s thought pattern in daily life. It is expected that these findings will be effectively used for economic activities, education, and other fields.




Figure Plots of the sensor-specific group difference (the ratios of the Group I relative to Group L) in spectral densities under the visual condition relative to the resting condition, for 22.5–27.5 Hz band (central frequency 25 Hz).

Each SQUID sensor (square panel) is tested for random effect and is colored with its statistical significance (t-statistics). This figure shows that Group I was significantly more activated than Group L for the primary visual area under the visual condition (p value less than 5%). On the other hand, for the frontal language area, Group L was more activated than Group I.



Multiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier SchönmannMultiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier Schönmann